The Camera Intelligence Caira is something between a dedicated camera and a phone accessory. It has its own Micro Four Thirds mount, an 11.4MP sensor and a Snapdragon processor to handle computational tasks like multi-exposure image stacking, and autofocus and white balance calculations. However, it only has a single control – a shutter button – and no display. To actually use it, you need an iPhone (specifically, as it won’t work with an Android phone), which physically connects to the back of the Caira using MagSafe and digitally connects to it via Wi-Fi for liveview and control.

It’s also a camera that’s all-in on AI for both shooting and editing. It uses the tech to control autofocus, white balance and exposure while you’re taking a picture, and, after you’ve captured the scene, you can change it using an editor powered by Google’s “Nano Banana” generative AI model. The app gives you preset changes, like removing clutter from a scene, changing eye-lines and applying a VHS-style filter, as well as a chatbot to make more arbitrary changes, or to get feedback on the image you’ve taken. However, that bits that deal with generative AI are optional (and, indeed, require a monthly subscription), so we’ll mainly be focusing on how the Caira behaves as a camera.

The Caira is available for pre-order for $1000, though there’s more to the availability story.

Key Specifications

• Micro Four Thirds mount
• 11.4MP ~Four Thirds (244mm²) BSI CMOS sensor w/ Quad Bayer
• Brightness-map HDR JPEG output
• Qualcomm Snapdragon processor
• MagSafe mount for iPhone
• AI exposure, white balance and autofocus
• Built-in generative AI-powered editor
• 64GB internal storage

Note: this review was written after using the camera with several beta firmware versions provided by Camera Intelligence, and extensive use with the latest version.

Body, handling and controls

Mitchell Clark

The Caira’s body is made out of aluminum, and it’s a nice-feeling physical object. There are only a few features on it: a (disappointingly vague) shutter button, the lens mount, a power button, a USB C port for charging and 3.5mm microphone jack. It also has a cold shoe for mounting accessories.

While I found it relatively comfortable to hold in one hand, using my thumb to control it and index finger to press the shutter, I couldn’t help but worry about the connection between the camera body and my phone. Apple’s MagSafe is relatively robust, but it still just relies on magnets, which are only strong in certain directions. When I’m using the Caira, it’s hard not to imagine bumping my phone wrong and sending it crashing to the ground.

Mitchell Clark

Speaking of my phone, most of the camera’s interface is actually in its companion app. It connects to the camera via Wi-Fi, and provides controls for various parameters: aperture, shutter speed, ISO, exposure compensation and focus. You can also set the controls to “Advanced” to change lens distortion correction and white balance temperature / tint.

By default, the controls are collapsed, and the camera handles exposure and white balance automatically. If you tap a button to gain access to the controls, though, you can then tap the icon for each parameter to turn it into manual mode, and control it using a slider. I’ve found this to be okay for controlling focus – the camera has some decent assist tools – but not so great for exposure; the sliders make it easy to make big adjustments, but fine-grained ones like picking F4 instead of F3.5 or 4.5 require a fair amount of concentration. This is a camera best used in auto mode, with manual intervention only when necessary.

Mitchell Clark

I found that the Caira works best with smaller, lighter lenses (the kind that are thankfully abundant in the Micro Four Thirds ecosystem). I used it a bit with the OM System 12-40mm F2.8 II Pro, and found it a difficult to keep steady with how front-heavy it was; it’s difficult to compose when the preview is jittery and bouncy. It was still workable, but I don’t think I’d want to shoot with anything much bigger, and preferred using small primes.

Image Quality

Before we dive into image quality, a note on the sensor. Very much like the Panasonic’s GH5S, the Caira uses a slightly oversized Four Thirds sensor, that allows you to use different region of the sensor to produce different aspect ratios. But while the Panasonic went up to 10.2MP in its 4:3 mode, the Caira allows you to go up to 11.4MP in its 3:2 mode. This essentially uses the standard Micro Four Thirds sensor height but extends around 10% further than usual, horizontally.

This is because, rather than limiting itself to using the sensor area covered by the Micro Four Thirds imaging circle, it instead uses that extended region, warning you that your lenses may or may not be able to project light onto the whole thing. So far, all the lenses I’ve tested with it have performed well, with no notable vignetting in the corners, but it’s worth keeping in mind that there’s a potential trade-off between maximum compatibility and resolution.

And you may want to think twice before giving up any resolution, as the Caira captures less detail than most other cameras you can buy today. That’s not to say that it’s insufficient – the photos it takes will look great on social media and most computer monitors and small prints – but you don’t have to zoom or crop in far before things start to look a little soft. It also has a base ISO of 400, so the cleanest images you can get out of it will still be 1EV noisier than most other Four Thirds sensor cameras, at least in its single-shot mode. However, there’s a good reason why.

One of the Caira’s tricks is that, as we’ve seen with the Sigma BF and Hasselblad X2D II, its JPEGs contain embedded gain maps to produce an HDR effect. When shooting a high-contrast scene, the camera is reducing exposure to retain as much highlight information as possible (a task it didn’t always do successfully; I have a lot of pictures with clipped highlights). Then, it’s raising the shadows in the JPEG to produce an image that looks correct, but that includes dazzling highlights.

This is actually a method of shooting we often advocate for, given the strong dynamic range performance of modern sensors, so it’s interesting to see a camera that actually does it with relative success, even if it doesn’t seem like this sensor is the absolute best fit for that style of shooting. It would also explain why the camera’s base ISO is 400, as it’s trying to capture one additional stop of highlights, compared with the ISO 200 modes on most Micro Four Thirds cameras designed to shoot in SDR.

While the camera’s low-light mode can stack up to 16 images together, the results are usually still relatively noisy, without a ton of detail. The process also won’t work with a moving subject. Still, in some circumstances it can result in decent images, but it’s definitely worth taking the name of the feature seriously: it’s a low light mode, not a no-light mode. It also reduces the amount of manual control you have, understandably locking shutter speed and ISO into auto, as those parameters start to mean less as you stack exposures.

Both cameras were set to auto and told to set focus and aperture on the same flower. Note that the iPhone is set to shoot using its 24MP mode.

It’s hard to directly compare a more traditional camera with a phone, as the latter is always doing a lot of heavy lifting with computation, where the Caira currently only uses a single exposure outside of its low-light mode. However, I found that my iPhone 15 Pro was typically better at focus and making the lightness and white balance more true-to-life. What the Caira could do that it can’t, though, is use Micro Four Thirds lenses that provide more flexibility in focal lengths and deliver pleasing bokeh that comes from optics, rather than computation. Even when I didn’t prefer how the Caira interpreted the scene (or was annoyed that it missed focus), the photos had an undeniable look to them that phones still can’t replicate.

Autofocus and performance

Mitchell Clark

The Caira has three autofocus modes: single, continuous and manual. It has a single size autofocus point, and you tap on the screen to choose its location. It can recognize faces, which it can use to track a person as they move through the frame, though you can also tell it to focus on a non-human subject even when there’s a face in the frame. Outside of tracking faces, though, the autofocus point doesn’t attempt to stick to the object you put it over if it or the camera moves; it stays in the same position in the frame.

I’d be tempted to complain about the lack of generic subject tracking, but I won’t. The camera seemingly used to have it prior to the continuous autofocus update (that, or it was just extremely overeager when it came to identifying things as faces), and it was incredibly frustrating, wandering off the subject and resetting to its original point just before I was going to take a picture, with no way to turn it off. It’s a pity to not have it as a function, but better to have a system that predictably works for the things you ask it to do than one that gives you false hope.

Camera Intelligence gives you several options for how you want the focus to behave when you half-press the shutter, as well as manual focusing assistance tools, like a tiny, low-contrast “rangefinder” patch that you can line up with your subject to ensure focus, and a much more reasonably-sized magnified window that appears when you half-press the shutter in manual focus mode (though it’s rendered in black-and-white, rather than color).

Autofocus performance

The Caira doesn’t use a traditional contrast-based or phase-detection system for AF. Instead, it uses “custom-trained neural networks running on the camera to estimate from the preview feed exactly how far and in which direction to move the lens to bring the image into focus,” according to Liam Donovan, the company’s CTO. The camera also uses different algorithms for single-shot and continuous AF.

I found the single-shot mode to be, broadly, quite frustrating. It’s sometimes capable of nailing focus, but never feels confident. I have sequences of images where it would get the first exposure sharp, then drift out of focus in subsequent shots, and it typically ends up hunting quite a lot.

This behavior will still happen occasionally in the continuous mode, but it’s overall less indecisive and my hit rate with it was much higher. That’s not to say that its autofocus is as good as most of the other cameras I test – there were plenty of relatively easy situations where it still completely missed focus – but it was at least usable, which isn’t how I’d describe the single-shot algorithm.

There’s another issue with the focus though, which makes it even harder to trust. The live view preview, transmitted over Wi-Fi, is compressed to the point where it’s very difficult to tell if the image is soft because it’s completely out of focus, or if it’s just the low-quality video stream. This would be one thing if I could trust the AF system to get it right all or most of the time, but since I can’t, it makes getting sharp images out of the camera that much more difficult.

Speed

The Caira app is responsive on my iPhone 15 Pro, though that isn’t surprising given it’s a relatively modern flagship phone. What isn’t as snappy is the camera’s readout speed. The camera doesn’t have a mechanical shutter, so you’re stuck using the E-shutter, which takes 50.7ms to read out in stills mode. This means that any fast-moving subjects will have substantial rolling shutter distortion, and there’s a risk of banding when shooting under artificial lighting.

Using Wi-Fi, the pictures can also take a beat to transfer from the camera to the phone, and the app can occasionally get confused during this process if you try to review a photo too early, showing you your previous photo rather than the low-res preview of the incoming one.

A camera tied to a phone

Mitchell Clark

That aforementioned blocky, soft autofocus preview is just one symptom of the Caira’s total reliance on the iPhone it’s attached to. Shooting with it made me realize that I take the integration of a traditional camera for granted. If I see a shot I want to take, I can pull the camera out and, typically with a flick of my finger, turn it on and take the picture.

This was not the case with the Caira. Even in the best-case scenario, capturing a moment with it requires unlocking my phone (a process that, occasionally, means putting in a passcode), and waiting a beat for the Caira app to get back up to speed.

But just as often, it meant doing that, then having to switch back to the app from whatever I had previously been doing on my phone – looking at a map, texting my wife, sharing photos, or whatever else I needed my phone for – and having to reconnect to the Caira’s Wi-Fi, since it only stays connected to your phone for a (thankfully customizable) set period of time. And that’s assuming I even had my phone attached to it, which I wouldn’t always, since having a camera and lens stuck on the back makes it hard to use as, you know, a phone.

As a reviewer, I typically try to set aside my personal preferences when evaluating a product. Even if I, say, prefer 35mm to 28mm, I won’t hold the choice of the latter against a compact camera. However, the multi-step process required to take a picture with this camera was genuinely a massive barrier to me wanting to use it. I frequently saw something that I wanted to take a photo of, but ended up not bothering because of the hassle of having to wait for the camera to turn on (which takes around 15 seconds), then wait for my phone to connect to its Wi-Fi; a process that I mustn’t start even a second before the camera is ready, otherwise I’ll have to wait for the phone to fail to connect before telling it to retry.

It’s also worth saying explicitly: the only reason I’m even able to have these complaints is because I’m an iPhone user. If I had an Android phone, I couldn’t use the camera at all. It’s completely unusable without its iOS-only app.

This approach isn’t without its benefits, though. The wireless connection between camera and phone means you have near-unlimited flexibility when it comes to how you frame your shots, since you can pull your phone off the body while maintaining the preview, as long as you don’t stray too far away from the Caira. You can also remotely adjust its settings and trigger the shutter.

Despite how good many camera apps have become, the Caira is by far the easiest camera to share photos from, as they’re downloaded as soon as I take them. While some camera apps have a similar function, it requires manually starting the app and connecting it to the camera; of course, the Caira requires this too, but you’ll never forget to do it, as it’s a prerequisite for even being able to use the camera.

It has its own storage, but the photos – Raws and JPEGs – are also dumped straight into my Apple Photos library, where they’ll automatically sync with all my devices via iCloud. It’s really nice coming back from a shoot, opening my laptop, and immediately having access to all the pictures I just took.

It’s also rare that I’ve used a dedicated camera with a screen as good as my iPhone 15 Pro’s. Not only is it larger and higher-resolution than the typical camera screen and bright enough to be visible in most conditions, it can also easily display the HDR images the Caira takes in their full glory.

The future of cameras?

Mitchell Clark

Unsurprisingly, given the Caira’s ties to phone photography, it joins the relatively small ranks of cameras that shoot for HDR output by default. We’ve seen some other camera manufacturers start to adopt similar techniques, especially now that it’s possible to create backwards-compatible files that will display as standard JPEGs if your hardware or software doesn’t understand or can’t display the gain map. However, it goes further than any other camera I’ve tested, though, providing several “levels” of HDR effect, ranging from “Subtle” to “Extreme.”

When enabling the latter, the app even warns you that images taken with the setting will be displayed using your screen’s full capabilities, and that it could even compromise quality on older devices. Despite this, and somewhat ironically given the name, I found the differences in the images taken at the various settings to be quite subtle, and the camera doesn’t appear to adjust its metering (and hence highlight capture) based on how much HDR effect you’re asking it to apply. The differences are small enough to make me wonder if the feature is even working as intended, as is the fact that setting HDR to “Disabled” still results in images with an embedded brightness map that light up my MacBook and iPhone’s screen.

Whichever setting it’s at, the Caira’s HDR implementation has become appreciably more subtle and pleasing after a few software updates. Donovan says the company redid it to make sure it got things right, and represented colors accurately.

Mitchell Clark

Because so much about the Caira is defined by software, it’s comparatively easy for the Camera Intelligence team to update it, often in substantial and fundamental ways. During my time testing the camera, the company has improved the image processing in a way that genuinely leads to much nicer-looking pictures, and has added features I consider to be vital, like an exposure compensation control.

There are still improvements to be found – I found the camera still occasionally gets white balance and colors quite wrong, and some of my night mode shots ended up with corrupted rows of pixels – but, unlike other camera companies, I strongly suspect Camera Intelligence will keep working on it.

While the Caira was arguably released in a much less finished state than most cameras, with its barely-working AF and lack of critical features, I wouldn’t mind seeing other manufacturers focus more on addressing user feedback and making improvements after a camera’s release. While there are some that do – Panasonic and Nikon deserve a shout out for their most recent models – it’s rarely with the same speed as Camera Intelligence (which, in fairness, only has to support a single model).

I also appreciate the phone-like use of stacking to improve image quality in low-light, with no requirement to assemble the image in post, as we’ve seen with other cameras. While there are definitely benefits to the single-shot exposures, having the option to do multi-shot stacking would be a nice addition to smaller-sensor cameras (and, indeed, is a welcome inclusion on cameras from companies like OM System and Panasonic). Camera Intelligence also says that it’s working on a multi-shot high-resolution mode which, again, I would expect to be very easy to use, not requiring a desktop app to put it all together.

Finally, there’s the camera’s voice control mode, which theoretically lets you issue commands to your phone to control the camera in everyday language. I say theoretically because half the time it either missed what I said, or just decided to ignore it. When it worked, though, I could see the bones of a useful feature. I could do basic commands like “take a picture” or “shoot in black and white,” but I could also ask it to do things like “make the background less blurry,” and watch it adjust the aperture.

Mitchell Clark

That could be really powerful for people who don’t know photography terms, but was unfortunately a very mixed bag. Basic asks like “make this picture brighter” yielded no results, and requests that it did understand were sometimes clunky. If the aperture was in auto mode, asking it to change anything about it would only switch the aperture setting to manual. It didn’t fully implement my request until I repeated it, making it a two-step process. And sometimes when I asked it to make my background less blurry, it’d pick something sensible, like F7.1. Other times, it’d stop the lens all the way down to F22. It did this in the same shooting conditions, so it’s not the result of it being clever about how much light was in the scene.

Oh, and if I did try to do things manually and tell it to “set my aperture to F4” or F8, it consistently set it to something other than what I asked. It would show on the screen that it heard me say F4, then I’d watch as the aperture slider moved to… F5. F8 was F7.1, and a request for F16 got me F14.

Beyond a reliance on an external device, there are other aspects of the Caira that I wouldn’t like to see carried over to traditional cameras. While the company says it’s added additional guardrails to the Google Nano Banana generative AI editing, beyond what the model itself enforces, I’m still not a fan of changing the reality of what was captured. I believe the imperfections in images, and the challenge of working around less-than-ideal shooting conditions, are vital parts of the craft. I don’t like the idea of a future where photographers get used to just fixing it in post the way filmmakers creating a fictional narrative do, or where its the norm to assume that every photo you see has been altered with AI (though it may be too late for the latter).

The Kickstarter of it all

I suspect most people won’t come away from this review planning on buying the Caira, but if you have, a word of caution. Before working on the Caira, the same team made a similar product called the Alice camera, which was initially sold via crowdfunding site IndieGoGo. Despite the camera being available in retail stores, there are backers who have complained as recently as March of this year that they haven’t received the cameras they paid for in 2021. When we asked the team about the unfulfilled orders late last year, they said clearing their backlog was their “highest priority,” and detailed the logistical and legal hurdles that had led to the shipments being delayed.

While the Caira’s Kickstarter is far less overdue, its comments section is still full of people asking where their camera is, and bemoaning the lack of communication from the company. Camera Intelligence is still putting out updates to backers, but they’re not terribly frequent. It’s also selling preorders on the dedicated Caira website, which a bit of an uncomfortable look when the team is still working on fulfilling its Kickstarter backlog for two separate projects.

Conclusion

Reviewing a camera like the Caira is hard, because it’s such a moving target. If I had written this a few months ago when I first received the camera, it would’ve been a very different conversation. It’s been impressive seeing the camera go from something that, frankly, barely worked, to something that’s genuinely okay to use.

However, over that time I’ve also discovered firsthand why previous attempts at this kind of camera/phone accessory hybrid from companies like Sony and DxO haven’t really taken off, despite being released in an era where smartphone’s cameras were far less capable on their own. The vast collection of AI features and tech built into the Caira hasn’t fixed the inherent clunkiness to the experience of an app-reliant camera. True, it improves on the smartphone by giving you access to a larger sensor and interchangeable lens mount, but so would buying a traditional ILC, which will give you a far more enjoyable shooting experience.

Mitchell Clark

And that’s a shame, since there are some really good ideas in the Caira, even if many of them are still at the “has promise” stage of implementation. And the team behind it clearly has talent; the fact that they built a working AI-powered autofocus and exposure system and multishot-stacking algorithm from the ground up is impressive, despite still feeling like they need a lot of work.

Camera Intelligence bills the Caira as a camera for creators, for those who want good photos and video without having to understand how to make those. Unfortunately, it’s not there yet. Even as an experienced photographer, I often struggled to get it to do what I wanted, largely thanks to the autofocus system and controls. The results it produced weren’t enough of a step up from my (much easier to use) phone to justify the extra bulk of carrying it around, and the multi-step process to take a picture. Plus, the lack of stabilization combined with its somewhat awkward ergonomics can make previewing your image a bit of a pain unless you have rock-steady hands.

Still, my time with the camera was enough to make me hope that Camera Intelligence is able to stick around to keep working on the Caira so it can reach its full potential, and to continue talks with other camera makers who may be interested in using their technology. However, it’s hard to avoid the feeling that the former may be more of a proof of concept or beta test for the latter, rather than an actual consumer product, despite the creator-focused marketing. Even if those talks go nowhere, though, I’d still like to see the company stick around, at the very least so everyone who ordered one gets it.

Sample Gallery

Please do not reproduce any of these images on a website or any newsletter/magazine without prior permission (see our copyright page). We make the originals available for private users to download to their own machines for personal examination or printing (in conjunction with this review); we do so in good faith, so please don’t abuse it.

A note on captions: I’ve shot with three major firmware versions while testing this camera, all of which have come with relatively fundamental changes to the image processing and/or shooting experience. Images taken with the original firmware have it noted in the caption, while images marked “Shot with Continuous Autofocus Update” were taken using the latest firmware available at time of writing, and include the updates to the image pipeline that came with the second version of firmware.

The Leica SL3-P is a high-speed, high-resolution L-mount full-frame camera built around a 44MP BSI CMOS sensor. It’s the latest in the company’s SL range and perhaps the most polished and all-round capable camera in the series so far.

Key specifications

• 44MP BSI CMOS full-frame sensor with phase detection
• up to 40fps shooting with AF in 12-bit mode. Up to 15fps in 14-bit.
• Subject recognition AF
• 8K open-gate video up to 24p (7.2K open-gate up to 30p)
• Full-width 8K (UHD or ∼DCI) up to 30p
• Full-width 4K up to 30p, 4K/60 and 4K/120 with increasing crops
• Content Credentials image certification
• Battery rated to 383 shots per charge
• IP54-rated all-metal construction

The Leica SL3-P will be available immediately at a recommended price of $6695 excluding tax in the USA, or €5990, including sales tax.

Index:

• What is it?
• How it compares
• Body and handling
• In use
• Video
• Autofocus
• Initial impressions
• Image quality
• Conclusion
• Specifications
• Leica press release

What is it?

Thus far, Leica has produced high-resolution SL models and high-speed SL-S versions. The SL-P appears to be something of an attempt to merge the two lines, offering both high speed and high resolution.

The result is something akin to Sony’s recent a7R VI or, perhaps more pertinently, the Panasonic S1RII: not an out-and-out sports camera but more a studio and landscape camera that’s happy to venture further afield. That Panasonic comparison is unavoidable as the 44MP, 40fps headline spec is shared between both cameras, as are their processors, even if the branding of them differs.

However, our experience of the SL3-P and the S1RII reminded us of how different you can make two cameras, even if you start with many of the same components and have similar destinations in mind.

Leica talks a good game about building cameras that last, and that’s immediately apparent in the feel and heft of the SL3-P. But what’s really striking is how few control points the Leica seems to have. That minimalism, honing the camera down to its core functions and little more extends to the menus and interface, meaning it gives a very different experience to the Panasonic. Or, indeed, any other mainstream camera.

How it compares

Although it offers both high speed and high resolution, the SL3-P doesn’t match the do-everything levels of capability of the Nikon Z8 and Canon EOS R5 II, with their stacked CMOS sensors. Instead it’s closer in ethos to the Sony a7R VI or, perhaps unsurprisingly, the Panasonic S1RII, essentially delivering a studio and landscape camera that isn’t scared of a little action, rather than a camera you might pick for sports work.

We’ve also compared it here to the SL3, which offers higher resolution but less speed, if absolute image quality is more important to you than the added speed.

It probably shouldn’t come as a surprise to find that the SL3-P’s specs are a very close match for those of the S1RII. The differences between the two cameras aren’t something readily captured in a specs table.

The original SL3 offers a slight boost in resolution capture but at the cost of a lot of the new camera’s flexibility. Sony’s a7R VI offers both though: higher resolution and competitively fast shooting with excellent autofocus and slightly wider dynamic range. It’s hard to imagine how two user interfaces and user experiences could be more different, for two cameras ultimately trying to do the same thing, though.

Body and handing

The SL3-P uses the same body as the existing SL3 models: a solid, angular block of a thing with a few large control points and a substantial grip to wrap your hand around. The Leica branding across the viewfinder hump is so beautifully proportioned that it’s easy to assume that this series (and, perhaps, a series of film cameras before it), have always looked this way. So it was a shock to see the original SL in the company’s archives and see how far the design has evolved.

There are two buttons on the front of the camera, two on the top and one on the back, none of which have marked functions. Instead they can be re-programmed with a long press. In addition there are three marked buttons on the back of the camera: Play, Fn and Menu.

Then there are the three large command dials: one towards the front of the top plate, one protruding from the rear of the right shoulder and a third on the top left of the camera (which I kept forgetting about, as it’s such an unorthodox place to find one).

Beyond this there’s a small rubber joystick on the back of the camera, and that’s about it. Relatively few control points and still fewer dedicated ones. No switches for focus mode or AF area, not even the seemingly obligatory stills/movie switch: that’s controlled by swiping left or right on the rear screen.

We’ll touch more on the user interface in the ‘In use’ section of the review, but Leica is continuing to refine the rather elegant and focused interface it introduced in the SL3. For instance, in this latest camera, the company has re-worked the way you select a video mode: letting you select the parameters that are most important to you: frame rate, resolution, codec, etc, then filtering down the available list from a seemingly unmanageable 193 down to the handful that fit your requirements. It sounds obvious once you’ve seen it done, but it’s cleaner and simpler than most cameras we use.

Viewfinder and screen

The SL3-P has the same viewfinder as the other SL3 models: a 5.76M dot OLED panel (1600 x 1200px) with 0.76x magnification and a 21mm eyepoint. A large circular rubber cushion makes the camera comfortable to use if you wear glasses and there’s a -4 to +2 diopter adjuster set behind that cushion.

The rear screen is a 2.33M dot (1080 x 720px) panel mounted on a tilt up/down mechanism.

Ports

The SL3-P has the same port arrangement as its sister models, with 3.5mm headphone and mic sockets and a full-sized HDMI socket under a large, rubber flap on the left flank of the camera. Below these is a USB-C socket that can be used for data transfer (10Gbps), tethering (including to an iPhone) or charging.

Battery

The camera uses the same BP-SCL6 battery as its siblings: a 15.8Wh unit that powers the camera to a rating of 383 shots per charge, when tested to the CIPA protocol. As always, this tends to significantly under-represent the actual number of shots you can expect to get, but a rating of nearly 400 shots per charge is the sort of number that should means an enthusiast photographer rarely has to worry about topping-up on a day’s shooting. It’s only intensive shooting such a weddings, or anything involving a mixture of video and stills that are likely to call for you to have a spare to hand.

In use

One of the Leica’s most compelling aspects isn’t something you can see from product photos or interpret from the spec sheets: the user interface.

The SL3-P uses the latest iteration of the back-to-basics interface Leica has been refining over the past few years.

At its heart, it’s designed around putting the key photographic parameters at your fingertips and minimizing the extent of, and need for, menu options. For instance, holding down a customizable button or long-pressing any of the on-screen icons brings up a list of other functions you can assign to it. In fact one of the few options present in the main menu is one to filter-down which options are presented to you, when you long-press a control point to customize its function.

The result is an interface very focused on the fundamental settings of photography. It affords you the luxury of just ignoring what’s going on in the interface, much of the time, and just letting you concentrate on the settings assigned to the two big dials, and what you’re seeing in your viewfinder.

the rest are designed to be configured (and reconfigured) to suit your shooting style and preferences.

This minimalist approach extends to the body and its control points, too, with a smattering of custom buttons, a joystick and three large, solid-feeling command dials being your points of contact. It’s remarkable how few dedicated control points there are and while, for my shooting, I found myself sometimes wishing for an AF mode selector, with a little work it’s very likely that you can set the camera up in a way you find very quick and easy to work with.

Because it’s such a different way of working, the interface takes a little bit of getting used to, though. It rewards you for taking the time to adjust to its design philosophy. One aspect of which is learning to save your setup as a user profile, or risk discovering that selecting the default profile erases any customization you’ve started applying.

It’s also worth recognizing what is and isn’t included in these profiles. Exposure mode is, but specific exposure settings aren’t, so my attempt to build a profile for freezing the action and another one, with a slower shutter speed, for panning ended up being unsuccessful. Likewise, it’s worth carefully planning your profiles before saving them: I thought it would be handy to define a custom button to switch between profiles. Which was great until I pressed the button and switched to a profile with different custom buttons settings, so couldn’t get back.

Overall it works very well, though, giving a very photo-focused shooting experience where, for all the cleverness that modern cameras have brought, you end up remembering that the only things that really matter are shutter speed, aperture value, composition and where you’re going to focus.

Video

Despite us saying that the SL3-P has a very photo-focused interface, it’s probably better to say that it’s a very exposure-focused. Swipe left on the camera’s settings panel and you get a similar but now video-centric display, with yellow, rather than red accents.

The SL3 offers an incredible range of video modes: 157 in total. These range from 8K open-gate capture, that uses the full extent of the sensor, via 8K in either 1.89:1 or 16:9 aspect ratios and on to 4K capture, again with a choice of aspect ratios. But the camera also offers a selection of other resolutions for each of these capture regions, letting you strike a balance between resolution capture and speed.

Like Panasonic’s S1RII, the SL3-P doesn’t quite have enough pixels to capture DCI 8K (8192 x 4320px) footage so instead outputs 8132 x 4288px footage. UHD 8K is no problem, though, so you can also select the more conventional 7680 x 4320 resolution, if you prefer.

As you’d expect of a modern camera, you get the choice of 10-bit Log or HLG, or 8-bit if you don’t plan to color grade, and you’re targeting SDR displays.

Video options:

As with the Panasonic S1RII, the SL3-P has a Dynamic Range Expansion mode in video, which reads each exposure in both low conversion gain mode for maximum highlight capture and in high conversion gain mode, for cleaner shadows, and combines the two, to boost DR.

This is available in any resolution mode but only at frame rates up to 30p. In each case it increases the readout rate by around 33%. So, for instance, the UHD 8K/24 mode increases from 23.9ms to 31.9ms.

The SL3-P’s video options are, as you might expect, very similar to those of the Panasonic S1RII. It lacks the ProRes RAW option of that camera (though the 1.3x crop on the Panasonic’s Raw mode means it’s not the camera’s best feature). It also misses out on the waveform display option that the S1RII offers. Realistically, given its lower cost and the inclusion of a fan and more flexible rear screen, the S1RII is a better stills/video camera than the Leica. But, looked at the other way: as a primarily stills-focused camera, the SL3-P has a very broad set of very competent video features.

Autofocus

The Leica’s autofocus behavior is as innovative and unconventional as the rest of the UI.

The camera has three AF area modes, but offers more than three AF areas. The first is called “Spot / Field,” the second is called “Multi-Field / Zone” and the last is called Tracking.

The clue is in the forward slashes. Spot / Field lets you select a single AF point of various sizes, while Multi-Field / Zone lets you choose between an adjustable AF zone and all-area AF, where the camera selects a subject from anywhere within the frame. Once you’ve chosen an AF zone mode, you can adjust it by pressing the AF area on your rear screen or long-pressing anywhere on the live preview. This then lets you use the rear dial to adjust the focus area size.

Like Panasonic’s recent cameras, the camera will track recognized subjects when you have subject recognition turned on, but won’t track non-recognized subjects at all. So you’ll need to turn subject recognition on and off depending on what you’re trying to photograph. We’d definitely recommend configuring a button to do this.

By default, the camera’s Tracking AF mode gives you a small rectangle that resets to the center of the frame at the end of each burst. However, while you can’t change the size of the initial target, you can configure the tracking target to revert to wherever you last triggered it. So, for instance, if you want to track a car down through a series of curves, across your frame, then have the camera jump back to where the next car is likely to arrive, you can.

Autofocus performance

We were pretty impressed with the SL3-P’s autofocus performance. It doesn’t quite match the best of its peers in terms of apparent infallibility, but its subject detection is good and highly tenacious. Close examination showed that, while the camera is very quick to focus on fast-moving subjects and usually very effective at following them within the frame, it’s not always perfect at driving the focus to exactly the correct distance. We used a selection of primes and zooms with the SL3-P and got an excellent hit rate but motorsports felt like they were at the very limits of what the SL3-P can do.

Like the Panasonics, subject recognition is a separate mode from general tracking AF, so you’ll need to toggle in and out of a subject recognition mode if you want the camera to give its best performance for recognized subject types but still be willing to focus track the rest of the time. Interestingly, we didn’t encounter the failure to find focus that we saw very occasionally with the S1RII (though not the S1II, post firmware update).

It’s all a significant step forward for the SL series and helps contribute towards the sense that the SL3-P can handle pretty much anything. It’s only when compared with the very best (often pro-sports tested) systems that it falls a fraction short.

Initial impressions

The Leica SL3-P is the fastest-shooting camera Leica has ever made and should, in principle, be the only SL model they need to persist with. It offers an awful lot of what the SL3 offers, in terms of resolution, and it outperforms the more speed-focused SL3-S.

The subject recognition AF performs impressively well: finding subjects very quickly and following them well (especially when it’s a subject the camera has been trained to recognize). Just as the user interface as a whole takes some getting used to, it’s worth digging around in the menus and familiarizing yourself with the SL3-P’s AF logic before you go out shooting. Without an instruction manual it wasn’t obvious that Leica had grouped its AF area modes, or how to switch between them, nor that you can override the camera’s default behaviour when it comes to AF tracking.

Some issues you can’t overcome, though. The sensor is fast, but it’s not as quick as the best stacked CMOS sensors. So the camera will start showing a slideshow of the images you’ve just shot, when you’re shooting at its fastest rates, making it very difficult to keep pace with the action. Similarly, we were advised to shoot Raw-only and to disengage Content Credentials, to get the best performance.

Richard Butler

And, even with that done, the camera’s approximately 20ms (1/50 sec) readout rates in 12-bit mode mean that rolling shutter will be apparent if you’re trying to follow really fast action.

In this respect it’s a lot like the Sony a7R VI we recently reviewed: very fast for a high-res studio/landscape camera, meaning it’ll handle itself very confidently in a much broader range of circumstances than that. It’s not quite the high-res, high-speed all-rounder that Canon’s EOS R5 II and Nikon Z8 are, but it’ll still be more than enough camera for a very broad range of photographers, taking a very broad range of photos.

The challenge, as ever with Leica, is the price. It’s a beautifully built object, constructed in Portugal and Germany, that wears the Leica name and is likely to sell in relatively small numbers, when compared to the volumes of the large Japanese manufacturers. As such, the price is high.

For that money you get the fruits of some very clever design work, in a body that feels like it’s designed to last decades. However, parts availability and the continued march of technology (we can already see that sensors and AF will continue to get faster), will make it difficult to match the sense of longevity that a Leica film camera gave you. So, as always, part of the calculus comes down to: how important are those five letters to you?

Image quality

Studio test scene

Our test scene is designed to simulate a variety of textures, colors, and detail types you’ll encounter in the real world. It also has two illumination modes, full even light and low directional light, to see the effect of different lighting conditions.

As you might expect, there’s no meaningful difference between the Leica and the Panasonic in terms of detail capture (the very slight difference shows how good the comparison tool is at highlighting the smallest of differences in focus), or for that matter, the Nikon Z8. However, the 66MP Sony a7R VI clearly captures finer detail than its rivals.

A quick check in JPEG mode shows that the default color rendering of the SL3-P is distinct from that of the Panasonic, though, with differences appearing in the blues and caucasian skintone patch, in particular.

Beyond this, there aren’t any real surprises: noise performance is solid in Raw, very closely matching the performance of its peers. JPEG noise reduction looks similar to that of the Panasonic, with what appears to be more sharpening added, by default, to pull out detail. The Auto white balance also looks to be working harder to neutralize the tungsten light used for our low-light mode.

Conclusion

The SL3-P is an inherently difficult camera to review. Leica’s luxury pricing always makes its cameras hard to assess, because the concept of value has a personal component. This is made more pertinent when there are direct, like-for-like rivals at lower prices, and still more stark when some of these less-expensive rivals share significant amounts of their hardware and capabilities.

The SL3-P’s works rather differently from any other camera in its class, despite that commonality with other cameras. Its large, prominent finger and thumb command dials and comparatively small number of other control points, along with the large, relatively uncluttered viewfinder make clear that this is a camera centered on the most fundamental aspects of photography: exposure and composition. Everything beyond that is secondary.

It takes a while to get used to the SL3-P’s way of working, but once you have, and have set up user profiles for your different types of shooting, it’s really quite elegant. It means the SL3-P doesn’t have every possible bell or whistle that its rivals do, but it is refreshingly good at letting you just get on with your photography. And it’s a very capable camera: it delivers a good amount of resolution and dynamic range while also delivering Leica’s best-yet autofocus.

Unexpectedly, for a camera so focused on photography, the SL3-P has a strong video feature set. Swipe on the screen and you find a camera with 10-bit and ProRes 422 capabilities, a false color display for setting exposure and the ability to upload LUTs, either to give a corrected preview or to bake into your footage. It’s hard to tell how many SL3-P buyers will make use of the video capabilities, but those who do are unlikely to be disappointed, except, perhaps, by the battery life.

Richard Butler

Ultimately, though, there’s no getting away from the SL3-P’s price. Those five large, bold letters along the front of its viewfinder hump mean it’s priced unlike anything else on the market. Leica is an exclusive, luxury-priced brand: you either buy into its history, its mythos and its pricing model or you don’t. As the line of cameras with the most direct mainstream competitors, comparison is unavoidable, and it’s implausible, almost to the point of impossibility, that the SL3-P could be objectively so much better than its peers as to justify the price difference on rational grounds. No SL camera could ever be good enough to win over those people that don’t buy into the Leica brand.

Which is a shame as, aside from price, it’s one of my favorite cameras in its class. It’s supremely solid feeling, does a wide range of things very well and feels like a camera designed and developed by people who’ve prioritized photography over technology. But most of us aren’t in the position to ignore the price. And, because our rewards include a consideration of value, we can’t sensibly give it an award. Some people are going to love this camera: probably the best SL camera yet, but we can’t ignore the fact that you can buy a more technically capable camera for half the price.

Scoring

Scoring is relative only to the other cameras in the same category. Click here to learn about what these numbers mean.

Compared to its peers

The Sony a7R VI is the rationalist’s choice, compared with the Leica. It has a similar stress on resolution while still offering very decent speed and video capabilities, but it does so while offering more resolution and slightly better dynamic range. It also has a higher resolution viewfinder and significantly longer battery life, it also uses the best-supported modern lens mount. However, we find it hard to imagine anyone who’s used both concluding that the Sony is better designed or nicer to use.

On paper, particularly in light of the L² technology sharing agreement, the Panasonic Lumix S1RII and SL3-P seem to be extremely similar cameras. In practice, it’s remarkable how different they are to use. The S1R II is covered in buttons and switches and its menus are full of options, whereas the SL3-P is much more pared-back. The Leica is also a fair bit larger and more substantial in the hand. Other than lens mount, using the Panasonic is much more like using a Sony or Canon than like using the Leica. And that user interface philosophy is one of the main distinctive things the SL3-P offers, compared to its rivals. The S1R II isn’t a half-price Leica and the SL3-P isn’t a modified S1R II.

If you’re really into sports photography or video production, the Nikon Z8 or Canon EOS R5 II are probably better choices. Both have stacked CMOS sensors that read out much faster than the Leica’s, meaning less rolling shutter impact in both stills and video shooting. Both also offer Raw video capability and, at the professional end of the market, an extensive pro support network. Both cameras offer a good degree of backward compatibility with their extensive SLR-era lens mounts but also exist in manufacturer-restricted but increasingly extensive lens ranges.

In many respects, the Leica SL3 is the SL3-P’s only direct peer, because it’s the only one that also buys you a Leica. The SL3 has the higher image quality, with a fraction more resolution and dynamic range, but without the video or autofocus performance of the faster SL3-P. The SL3 is perhaps the better choice if you’re primarily a landscape, studio or portrait photographer, but the SL3-P gives you greater flexibility if your shooting is more ad hoc and less prescribed.

Sample gallery

Please do not reproduce any of these images on a website or any newsletter/magazine without prior permission (see our copyright page ). We make the originals available for private users to download to their own machines for personal examination or printing (in conjunction with this review); we do so in good faith, so please don’t abuse it.

Updated August 2, 2026

Travel can expose you to once-in-a-lifetime sights and experiences, and a good camera can be the perfect way to preserve those memories. We’ve looked at a range of models that offer great image quality, good autofocus and excellent video, so you’ll be prepared to capture whatever you run into.

We’ve also tried to select relatively small cameras so they don’t interfere with your trip, though it’s worth touching briefly on lens size. If you choose a system with a Four Thirds or APS-C sensor, you’ll be able to use lenses that are lighter and more compact than their full-frame counterparts. Full-frame sensors, however, offer the potential for even better image quality, so it’s worth considering whether portability or image quality matters more to you and doing some research on what types of lenses you’d like to use and how big the full-frame versions are.

Our recommendations

• Portable full-frame: Sony a7CR
• All-round compact capability: Sony a6700
• The great travel kit: Fujifilm X-T50 with 16-50mm F2.8-4.8
• The ready-for-anything option: OM System OM-3
• The stylish travel companion: Nikon Zf
• Best pocketable travel camera: Ricoh GR IV
• The classic choice: Fujifilm X100VI
• The fun-to-use compact with a zoom: Panasonic L10
• The portable option with a ton of reach: Sony RX10 V
• The best video camera for travel: DJI Osmo Pocket

Portable full-frame: Sony a7CR

Richard Butler

The Sony a7CR is a relatively compact full-frame camera with an image-stabilized 61MP CMOS sensor. You sacrifice the convenience of an AF joystick and get a pretty small and basic viewfinder to keep the camera’s size down, but there’s little else that delivers this much image quality in such a compact package.

You’ll need to pick your lenses carefully to keep the camera portable but the a7CR’s autofocus means it can respond rapidly to just about anything you encounter on your travels.

Read our full Sony a7CR Review

See the Sony a7CR studio scene

All-round compact capability: Sony a6700

Richard Butler

The Sony a6700 is built around a 26MP APS-C-sized sensor. Its excellent autofocus performance means it excels at both stills and video capture. There’s a good choice of relatively compact lenses available, too. We’d suggest avoiding the really small 16-50mm F3.5-5.6 though: it’s not the lens to get the most out of this camera.

Fast, dependable autofocus with a selection of subject recognition modes means the a6700 will help maximize your chances of capturing the unexpected moments on a trip.

Read our full Sony a6700 review

See the Sony a6700 studio scene

The great travel kit: Fujifilm X-T50 w/ 16-50mm F2.8-4.8

Richard Butler

The X-T50 can be purchased with Fujifilm’s excellent 16-50mm F2.8-4.8 kit lens, which provides a versatile focal length for travel and should be able to keep up in darker environments. Fujifilm also sells a wide variety of relatively fast, compact primes.

Read our full Fujifilm X-T50 review

See the Fujifilm X-T50 studio scene

The other APS-C options

If you’re looking for an APS-C camera that’s even smaller than the X-T50, the Fujifilm X-M5 may be worth considering. It’s substantially cheaper, though it’s easy to see where the company has saved money: it uses a lower-resolution 26MP sensor, which isn’t stabilized, and there’s no EVF. It does have the same film simulation dial and autofocus subject recognition modes, though, so those tradeoffs may be worth it if size or price are your top priority.

The Nikon Z50II is also a strong all-around contender, but its lens selection isn’t as strong as Sony or Fujifilm’s, nor does it have a stabilized sensor. However, it does have solid autofocus performance, a viewfinder, and a sub-$1000 price point.

The ready-for-anything mirrorless: OM System OM-3

Mitchell Clark

The OM-3 is well-suited to traveling, as the Micro Four Thirds system lets you use especially compact lenses, and its Stacked sensor gives it fast autofocus, incredible burst rates for shooting even the fastest subjects, and the ability to use computational modes to adapt to a wide variety of situations. It also sports an IP53 rating for dust and water resistance, so you won’t have to worry if it starts raining.

Read our OM System OM-3 review

See the OM System OM-3 studio scene

The other Micro Four Thirds options

OM System’s OM-5 is substantially cheaper than the OM-3, though you give up the speed, USB-C port and advanced subject recognition autofocus, as well as a few of the computational modes. It is, however, slightly more compact if space is at a premium in your pack.

The stylish travel companion: Nikon Zf

Richard Butler

The Nikon Zf is a classically styled full-frame camera built around a 24MP full-frame sensor that’s used by many of its peers. The angular 80s-style body isn’t the most comfortable to hold in your hand for extended periods, but the lack of stick-out grip makes it smaller than many of its rivals, making it a tempting travel companion. However, like with the a7CR, many of the full-frame lenses available for it will greatly increase its size and weight.

At least in its all-black form, there’s a chance of it being mistaken for a film camera, which may help avoid the wrong kind of attention when you’re out and about in an unfamiliar setting.

Read our review of the Nikon Zf

See the Nikon Zf studio scene

Best pocketable travel camera: Ricoh GR IV

Richard Butler

The Ricoh GR IV is the latest entry in the company’s series of classic compact cameras with large APS-C sensors and a 28mm equivalent lens. It’s not the most flexible camera, but it has engaging controls and is one of the most pocketable models with a large sensor, which is probably why the GR series has built up a dedicated following of photographers who find it a joy to shoot with.

Our one concern with using the GR IV for travel is that some users of the previous generation model reported dust getting into the lens. Ricoh has said it’s made a few changes that should help prevent that, but still doesn’t make any claims of weather sealing, so it may be worth keeping it in protective bag and putting it away in particularly unpleasant weather.

There’s also the GR IIIx , which has the previous generation sensor and processor, but with a longer, 40mm equivalent lens. This may be a little tight for documenting your travels, but it’s historically a popular focal length. Ricoh also sells a GR IV Monochrome, though a camera that only shoots black and white probably wouldn’t be our first pick for travel photograph.

Read our Ricoh GR IV review

See the Ricoh GR IV studio scene

Other compacts

Canon has recently announced that it plans to increase its production of compact cameras, citing an increasing number of orders for the PowerShot G7 X III , which has a 24 – 100mm (equiv.) zoom lens, providing a solid range for travel photography. Its autofocus isn’t

The Leica D-Lux 8 features a larger sensor than the G7 X III – it uses up to up to 83% of the area of a Four Thirds (17.3 x 13 mm) sensor, while the G7 X III has a Type-1 (13.2 x 8.8 mm) sensor – and has a photo-focued interface that we particularly like. However, its slow zoom, only decent image quality and substantial price tag make it hard to recommend to all but the most Leica-faithful.

The classic choice: Fujifilm X100VI

Richard Butler

Like the Fujifilm X100V before it, the Fujifilm X100VI is a classically styled fixed lens camera with a 35mm equivalent F2 lens. Updated with in-body image stabilization rated at up to 6EV of correction and a 40MP BSI CMOS APS-C X-Trans sensor, the X100VI is a significant upgrade over its predecessor despite looking virtually identical. And those improvements in part explain the X100VI’s $200 increase in list price to $1,600.

It’s not smallest or most inconspicuous camera, but its fixed focal length means you develop an ‘eye’ for the photos it’ll take, essentially making it a better quality, more engaging alternative to snapping away with your phone.

The addition of subject-detection autofocus has given the X100VI the ability to recognize animals, birds, automobiles, motorcycles and bikes, airplanes, and trains though human (face/eye) detection is a separate mode. However, the lens the X100VI shares with the X100V is not the fastest to autofocus, prioritizing sharpness over speed.

The X100VI brings a level of polish you would expect from a sixth iteration of a camera. Fujifilm has done a great job keeping the X100-series up to date without messing with the formula that’s made it so popular. No surprise, this is the best X100 yet.

Unfortunately, despite ramped-up production, demand for the X100VI seems to have outstripped supply with the camera already on backorder at virtually all dealers.

Read our Fujifilm X100VI review

The fun-to-use compact with a zoom: Panasonic L10

Mitchell Clark

The L10 takes a lot of what the Ricoh GR IV and Fujifilm X100 VI get right – it’s relatively small, engaging to use, and has great image quality – and pairs it with a zoom lens instead of a prime one. While primes definitely have their place, it can be hard to tell what you’ll run into while traveling. If you want the experience of a compact but the flexibility of a 24-75mm equiv., the L10 is the camera to pack.

Read our Panasonic L10 review

See the Panasonic L10 studio scene

The portable option with a ton of reach: Sony RX10 V

Richard Butler

The Sony RX10 V is built around a Type 1 (116mm2) 20MP Stacked CMOS sensor, and features a far-reaching 24-600mm equiv. F2.4-4 lens.

For many of us, traveling somewhere and looking at faraway things is a big part of vacationing. And with its 24-600mm equiv. zoom, blistering fast burst rates and best-in-the-business autofocus, the Sony RX10 V is the perfect tool for photographing things in the distance (while still being engaging for capturing nearer subjects). It’s not the smallest compact ever, but almost anything else with this kind of reach would be much bigger, heavier and more expensive.

While we haven’t completed our full review of the camera yet, we’ve used it enough to know that it belongs on this list. It provides good image quality, is engaging to use, and features Sony’s latest autofocus tech, which is still among the best in the business.

Read our Sony RX10 V review

See the Sony RX10 V studio scene

The best video camera for travel: DJI Osmo Pocket

Richard Butler

DJI’s Osmo Pocket series of cameras feature a stabilized handheld vlogging camera, designed for on-the-move vlogging. The main cameras on them are built around a type 1 (13.2 x 8.8mm) sensor mounted on a tiny gimbal, meaning it can deliver stabilized footage with smoother motion than any conventional camera or phone.

When you’re traveling in an unfamiliar place, there are a million things to focus on; the Pockets let you capture it all, without requiring much effort on your part. It can also natively film in horizontal or vertical, which can be useful if you’re looking to share footage from your travels on social media.

The reason we’re referring to multiple cameras is that the Pocket’s availability will vary depending on what region you’re in. If you’re in the US, the latest version you have access to is the Pocket 3. In the rest of the world, though, the Pocket 4 is now available, adding integrated storage, high framerates and better controls, along with a few other improvements. There’s also the Pocket 4P, which adds a second 60mm equiv. camera with a Type 1/1.28 (75mm²) sensor and features better dynamic range on its main camera.

The Pockets’ small sensors mean they works better in well-lit situations, though a reasonably bright F2.0 lens on the main camera mean they’re usable indoors, too.

It can shoot 4K 16:9 video or, by rotating the screen into the upright position, 3K vertical 9:16 footage. A small joystick lets you control the camera’s movement, and tapping the joystick button flips between the camera pointing toward you or away from you.

The older Pocket 3’s reliance on contrast-detection autofocus means its focus can be prone to hunting, though the cameras’ face-detection mode works well enough that this is rarely a problem when talking to the camera.

The video quality isn’t always the highest, and the autofocus can occasionally lead to frustration, but overall the quality and convenience of the Pocket as a package is really hard to beat.

Read our hands-on impressions of the Osmo Pocket 3

Why you should trust us

This buying guide is based on cameras used and tested by DPReview’s editorial team. We don’t select a camera until we’ve used it enough to be confident in recommending it, usually after our extensive review process. The selections are purely a reflection of which cameras we believe to be best: there are no financial incentives for us to select one model or brand over another.

Want to stay up to date on the latest product news and releases? Add DPReview as a preferred source to ensure our independent journalism makes it to the top of your Google search results.

Updated August 1, 2026

Not long ago, it seemed like the market for new compact cameras with zoom lenses had mostly dried up, despite the demand from enthusiast photographers who like the form factor and people looking for a nostalgic photographic experience. This year, though, the market has picked up, and not just with subtle refreshes that remove as many features as they add.

You may still have to hunt to find certain models in stock, and if you haven’t bought a compact in a decade or two you’re almost certainly in for some sticker shock. But each category has at least one or two very capable options, and the cameras of today can still offer an experience, image quality or both that can’t be matched even by top-end phones.

Our recommendations:

• Enthusiast camera
  • Best enthusiast compact: Panasonic L10
  • Best long-zoom compact: Sony Cyber-shot DSC-RX10 V

• Waterproof cameras
  • Best waterproof compact: OM System Tough TG-7

• Vlogging cameras
  • Best vlogging compact: Canon PowerShot V1

Enthusiast compacts

The final niche in which it’s difficult for a smartphone to compete is for enthusiast photography. No matter how good smartphone image quality gets, it’s difficult for a phone to provide the feeling of connection to the process of taking photos that an enthusiast compact with lots of direct controls can offer.

We’ve picked cameras with large sensors that deliver good image quality, zoom lenses and direct controls. Some have built-in electronic viewfinders for shooting when it’s bright outside, again giving a more pleasant experience than a smartphone.

Best enthusiast compact: Panasonic L10

Mitchell Clark

The L10 combines an engaging control setup with modern connectivity and features and an ambitious lens with a focal range that’s standard for a reason. It’s not particularly pocketable, but is still small enough that you won’t have any reason not to take it with you on a trip, to an event or even just out on the town.

Read our Panasonic L10 review

See the Panasonic L10 studio scene

Honorable Mentions

If you need something that’s more pocketable, the Canon G7 X III offers a similar focal range to the L10 in a much smaller package. It also has engaging controls, but its smaller sensor means there’s a bit of an image quality trade-off, and enthusiast photographers may not prefer its EVF-less design. Its biggest downside, though, is availability. Despite multiple assurances from Canon that it’s working on increasing production, it’s still quite hard to find. If you can get your hands on one, though, it offers a good experience at a great price.

The Sony RX100 VII is similarly pocketable, but features a pop-up EVF and a lens that reaches further, going out to 200mm equiv. Even at its widest, it’s not as bright as the one on the L10 and G7 X III, putting it at an image quality disadvantage, especially as light levels fall, and the camera is seriously expensive thanks in part to the stacked sensor that lets it achieve blistering burst rates. But if you need a little extra reach in something that’s pocketable, it’s still a very capable pick.

Best Long Zoom Compact: Sony RX10 V

Richard Butler

The Sony RX10 V is built around a Type 1 (116mm2) 20MP Stacked CMOS sensor, and features a far-reaching 24-600mm equiv. F2.4-4 lens.

It may be pretty much the only camera left in this category that doesn’t use a minuscule sensor, but it’s not just on this list by default. Sony has equipped it with its latest autofocusing system, menus and battery, and has reworked the ergonomics and controls.

While we haven’t completed our full review of the camera yet, we’ve used it enough to know that it belongs on this list above other cameras that offer similar reach. It provides good image quality and, while not small, is much easier to carry around that most other setups with this kind of reach.

Read our Sony RX10 V review

See the Sony RX10 V studio scene

Waterproof cameras

Waterproof cameras are another specialty that smartphones can’t necessarily displace. Many flagship phones are waterproof to a decent degree but you shouldn’t try to submerge them to any appreciable depth or use them in salt water. Waterproof cameras also tend to be pretty rugged, designed to withstand a lot more mistreatment than a phone will withstand, meaning you can carry them with you at all times, without ever having to worry about it. Great for bikers, climbers and custodians of small, inquisitive but not necessarily careful children.

Best waterproof camera: OM System Tough TG-7

The OM System Tough TG-7 is a rugged, waterproof compact with a 25-100mm equivalent zoom lens. It is fully waterproof down to a depth of 15m (50 ft).

The OM System isn’t the only rugged waterproof still on the market, nor the cheapest, but it has several major factors in its favor. The first is that it lets you shoot Raw images, which gives you the ability to correct the white balance, which even the best cameras tend to get wrong when shooting underwater. Another factor is that there is a range of accessories for the TG-7, including a light guide that directs light from the built-in flash into a circle around the lens, letting you illuminate close-up objects.

Beyond this there are a few details that different users may find handy: it has an unusually wide-angle lens with a bright maximum aperture, helping the performance underwater and in low light. It also has a built-in GPS, compass and pressure sensor that lets you log your adventures. These all add up to a camera that’s worth the added expense, we feel.

Read our hands-on with the OM System TG-7 article

Vlogging cameras

Smartphones are great for vlogging, but a dedicated vlogging camera can shoot better quality video, have more sophisticated microphones (and provide options for connecting better mics), and can autofocus very reliably, dependably delivering YouTube-ready footage.

Best vlogging compact: Canon PowerShot V1

It offers a competitive suite of vlogging features, a built-in ND filter and good stabilization options. It’s also comfortable to hold pointing towards you, and lets you access key settings from the touchscreen. It includes microphone and headphone jacks and a fan for extended recording sessions.

The V1 is a capable option with a useful zoom range, and while its slower sensor may not be the greatest for capturing fast action, it has all the features you’d need to start capturing the world around you.

Read our full PowerShot V1 review

Why you should trust us

This buying guide is based on cameras used and tested by DPReview’s editorial team. We don’t select a camera until we’ve used it enough to be confident in recommending it, usually after our extensive review process. The selections are purely a reflection of which cameras we believe to be best: there are no financial incentives for us to select one model or brand over another.

As we push on with our Leica SL3-P review, we’ve run it through our standard studio tests. To a large extent, we already have a pretty good idea what to expect, given we’ve seen this sensor / processor combination before. But it’s always worth testing these things, just to be certain.

Studio test scene

Our test scene is designed to simulate a variety of textures, colors, and detail types you’ll encounter in the real world. It also has two illumination modes, full even light and low directional light, to see the effect of different lighting conditions.

As you might expect, there’s no meaningful difference between the Leica and the Panasonic in terms of detail capture (the very slight difference shows how good the comparison tool is at highlighting the smallest of differences in focus), or for that matter, the Nikon Z8. However, the 66MP Sony a7R VI clearly captures finer detail than its rivals.

A quick check in JPEG mode shows that the default color rendering of the SL3-P is distinct from that of the Panasonic, though, with differences appearing in the blues and caucasian skintone patch, in particular.

Beyond this, there aren’t any real surprises: noise performance is solid in Raw, very closely matching the performance of its peers. JPEG noise reduction looks similar to that of the Panasonic, with what appears to be more sharpening added, by default, to pull out detail. The Auto white balance also looks to be working harder to neutralize the tungsten light used for our low-light mode.

The Sony FX5 is a compact cinema camera capable of up to 4.5K120 and Raw video capture, aimed at independent videographers and small productions. It builds on the legacy of the FX3, adding a host of features we’ve not previously seen from Sony.

Key features

• 16.6MP Full-Frame stacked CMOS sensor
• 4K/60 from 5K capture
• 4K/120 from 4.5K crop
• 4K/240 capability via firmware due in late 2027
• Open gate shooting
• Combined dual gain mode for increased DR
• Three “base” ISO steps
• Internal Raw capture
• Subject recognition autofocus
• 32-bit float audio compatibility
• Large 16:9 rear LCD
• Waveform display
• Optional EVF

The FX5 will be available, body-only, for $4899, or $5499 with the XLR top-handle, which allows 32-bit float audio capture. Other optional accessories include a removable, tiltable viewfinder and a USB-C-to-BNC timecode cable. It will be available from August.

What’s new

The FX5 is a surprisingly significant upgrade compared to the FX3, suggesting Sony is still working out what this type of camera needs to offer, and iterating. It nominally sits above the FX3 but looks like it’s trying to do what the FX3 did, plus a lot more, making it most directly comparable to the likes of Canon’s C50 cinema camera.

The most visible addition is its ability to capture Raw footage, but that’s just the most dramatic departure from the existing camera. There are also a host of features and UI changes that have come down directly from the company’s pro cinema products, to the degree that it represents not just a major upgrade but will also present an appreciable learning curve for existing FX3 users.

Don’t call it Raw

Interestingly, Sony doesn’t refer to the camera’s Raw mode in those terms, instead settling on the snappy X-OCN, which I’m sure we can agree is clearer. They tell us it’s a 16-bit linear Raw format, which is a lot more detail than we’re typically given about the other Raw video codecs. Linear encoding is a rather inefficient way of doing things but, in theory at least, gives the maximum flexibility in terms of highlight editing and reconstruction.

This “EXtended tonal range Orginal Camera Negative” gives you more flexibility in terms of correcting white balance, applying noise reduction, highlight and tonal adjustment and color grading. It’s a version of the format already used by the Burano and Venice cameras, so should be compatible with DaVinci Resolve, Adobe Premiere Pro, Avid Media Composer and Baselight software.

The FX5 uses the X-OCN LT (Light) version of the codec, used on the Burano cinema camera (and that’s even an option on the company’s flagship Venice 2 model).

New sensor

James Tirado

The FX5 is based around a new stacked CMOS sensor. Sony quotes the pixel count as 16.6MP which, given the FX3 turned out to have a 48MP behind a 12MP color filter array, should make us wonder how similar it is to the 66.8MP sensor in the recent a7R VI.

Sony wouldn’t be specific about any differences, but the significantly higher output speeds make us assume that there’s DRAM or more complex readout circuitry built into the back of the sensor. We know that the a7R VI’s 4K video comes from combining quartets of pixels, so placing that same pixel architecture behind oversized color filters that extend across sets of four pixels would make a lot of sense. Even if you’re not that interested in video, this ability to produce variations of a sensor with different capabilities might still offer an interesting insight into the products we’ll see in the coming years.

In practice, the FX5 is faster at reading out a comparable region of its sensor (around 5.5ms in full-width, 16:9 5K modes, around 24% quicker than the 7.2ms that the a7R VI takes in its ‘5K sampled’ 16:9 modes). Open gate mode read out the full height of the sensor in 8.3ms (1/120 sec).

Sony says the camera has a “Dual Gain” mode that boosts dynamic range by combining the low gain and high-gain signals being read off the sensor. It is available in the camera’s lowest native gain mode (ISO 800) and the modes derived from it (ie up to EI 3200). It can be used at up to 60p in most resolutions or up to 30p in open gate mode.

We’re told it may operate differently from the combined dual gain mode the a7R IV offers in video, though. We were briefed that the image is analysed and the high gain (low noise) data is only utilized in the shadow regions of the image, making the process faster than combining both gain settings for all pixels. Engaging the Dual Gain mode (via the ISO menu) increases the readout time by a factor of 2.1x, which is very similar to the proportion of increase we measured on the a7R VI.

Three “base” ISO settings

Interestingly, Sony also says the FX5 has three “base ISO” settings: 800, 4000, 12,800. To make sense of this, it’s worth noting that there isn’t a fixed or universally shared definition of what ‘base ISO’ means. What it doesn’t directly refer to is conversion gain steps at the sensor level. So, for instance, the FX3 and a7S III had dual conversion gain sensors, but Sony says it does not consider them to be cameras with dual base ISO.*

James Tirado

Instead, Sony tells us its definition of base ISO relates to modes that offer the same dynamic range distribution between highlights and shadows, and have a near-identical noise floor, allowing footage to be intercut between the modes. This makes most sense when working in the camera’s Cinema EI mode, where the camera is locked to one of these “base” states and then delivers ISO settings up to two stops higher or lower by essentially under- or over-exposing the ‘base’ state, then including a metadata tag saying how light or dark the footage is meant to be rendered. This really comes into its own in a Raw workflow, as this is how it’s possible to adjust the “ISO” of footage, after the fact.

The decision of Sony to offer three “base ISO” states shouldn’t lead you to conclude that there are three distinct gain steps at the hardware level, nor should it be confused with the camera’s optional “Dual Gain” mode that combines the output of different hardware states.

*Just to make clear how slippery the concept is, the FX3’s firmware V2 added Cinema EI, including a setting called Base ISO, which has two settings. We are told the higher of these is officially considered a High Sensitivity mode, not a second base ISO.

New UI, wider range of features

Beyond these, the most radical change is to the user interface, with the FX5 gaining an all-in-one settings display borrowed almost directly from the Burano (and closely mimicking the displays that Panasonic borrowed from Arri, etc, etc). Its menus have also been re-worked to make them look more like Sony’s pro cinema models, which will add a bit of a learning curve for existing FX users but brings it closer into line with other production cameras.

The company says it will be adding a vertical version of the UI in firmware 2.0.

James Tirado

Also added are a host of tools that other brands have started offering but that have been absent from previous Sonys at this point in the market. So, in addition to options like the focus map, that highlights blocks of the image that are in and out of focus, the FX5 offers a false color display and waveforms for helping to judge exposure, and a vectorscope display for assessing color distribution. These come on top of tools such as zebras, focus peaking and a histogram, that were already available.

James Tirado

The camera is based around Sony’s latest Bionz XR2 processor, so gains all the subject recognition AF modes that we’ve become used to on the stills side of things, boosting autofocus dependability for run-and-gun shooting.

Body and handling

James Tirado

At first glance, the FX5 looks a lot like the FX3, but the changes reveal themselves, the closer you look. It’s a fraction larger, for a start, having grown a couple of mm in each direction but significantly the body of the camera itself is deeper, within those boundaries, so it feels bulkier. It’s also a fraction heavier, though Sony says it’s a less than 5% increase.

The layout of the camera’s six programmable buttons has been reworked and relabeled to better match what you’d find on the company’s higher-end cinema cameras. It also gains the multi-function dial from the Sony Burano, which can be used to navigate through menus, and make quick changes from the live view screen.

James Tirado

The FX5, like the recent Alpha cameras, gains a second USB-C socket. This can be used for a range of things, from tethering, powering the camera, data transfer or for accepting a timecode signal. As with other recent Sony cameras, the second USB-C socket is actually a USB 2.0 (480Mbps) socket, so can only be used for power and less data-intensive tasks.

The camera’s connectivity is further updated with an upgrade to Wi-Fi 6E, and the addition of an Ethernet port, which will likely find use in many houses of worship.

Display

Sony has upgraded the rear tilting/articulating display to a 3.5″ model with a video-friendly 16:9 aspect ratio. Sony says the latter change makes it a much bigger upgrade to the 3″ display on the FX3 than it appears at first glance, as that screen’s 4:3 aspect ratio meant you were looking at a video that was really only 2.5″. Now, it can fill the entire screen, provided you’re not using an open gate or DCI (“17:9”) recording mode.

James Tirado

The FX5 has two slots, each of which take either CFexpress Type A cards or the slower UHS-II SD cards. The two card types don’t connect to the same connectors: one socket it essentially sited within the other.

Battery

James Tirado

The FX5 becomes the second camera to adopt Sony’s new 20.9 Wh NP-SA100 battery, following the a7R VI. This not only provides higher voltage and capacity than the NP-FZ100 batteries the FX3 used, but should allow for slightly longer record times. Sony says it can shoot for 130 minutes at 4K/60, while the FX3 caped out at 113. Of course, this change means that those with a large collection of batteries will need to spend quite a bit to rebuild it (provided they were running off an internal unit, rather than powering their cameras via an external V or Gold mount battery).

Despite the increased battery capacity, and smaller tweaks to the body like the addition of a second 1/4-20″ socket on the bottom plate to reduce the risk of rotating, Sony says the new camera is close to the FX3’s weight, coming in at 734g (25.9oz), versus 715g (25.2oz)

Accessories

While the camera doesn’t have an integrated EVF – likely a concession to keeping its body as protrusion-free, and therefore gimbal-friendly as possible – Sony is making a removable one as an optional accessory. It features a 3.68M-dot panel that can reach up to 2500 nits peak brightness, and supports a wide color gamut to provide a more accurate preview. It’ll plug into a new port right above the multi-interface shoe, meaning you can’t have it attached if you’re also using the shoe. However, Sony says it’s working with a third-party to create an extension cable for it that will let you place it in more ergonomic position. The EVF, named the DVF-EL1, costs $698.

Another accessory is the XLR handle which, like the ones released for Sony Alpha-series cameras, will enable both the use of professional microphones as well as 32-bit float audio recording. The latter feature will an exciting addition for documentary and event shooters, as its different method for storing data essentially eliminates the risk of clipping, no matter what your gain levels are set to (though your microphone can still be overwhelmed). The handle can be purchased in a bundle with the FX5, or separately for $798

Video modes

Like most of Sony’s previous cameras, the FX5 delivers all its compressed footage in output-ready 4K resolutions: with in the 16:9 UHD or ~17:9 DCI formats. However, in a detail borrowed from cameras like the Venice and Burano they are named in terms of the pixel region used to capture them. So, although the camera has “Imager Mode” options called 5K 16:9, the output is still the standard 3840 x 2160 UHD footage.

James Tirado

The Raw footage, by contrast, arrives in your editor in the resolution in which it was captured. So 5K means 5K.

One of these Raw modes is a 3:2 open gate mode, capturing the full area of the sensor to make it easier to output both landscape and vertical cuts from the same clip (if you can find a composition that’s sensible for both, that is). The camera can also apply a real-time desqueeze correction for anyone using anamorphic lenses, with 1.3x, 1.5x, 1.6x, 1.8x and 2.0x squeeze factors being supported.

Raw recording

When recording to its internal Raw X-OCN codec, the FX5 can shoot at roughly 5K in 3:2, ∼17:9 and 16:9, at up to 60p. There are three levels of compression for the codec, with the standard being “LT,” and C1 and C2 recording at increasingly lower bitrates to reduce file sizes. Sony says C1 will be around 25% smaller than LT.

The company says it plans to release a firmware update sometime in 2027 that will allow for recording 4K/240 footage in Raw, though it hasn’t provided any details on whether the mode will have a crop.

The camera needs to reboot in order to switch into or out of Raw mode. As we’ve seen on other cameras, it also needs to reboot to switch between 24p, PAL and NTSC modes, as the system clock needs to reset to work in multiples of 24, 25 and 29.97 frames per second. Because there’s no separation of Raw/Compressed or 24p/PAL/NTSC, it’s very easy to accidentally prompt the camera to reboot if you select the wrong shooting mode.

Likewise, the camera will only playback footage if it’s in the right system clock mode, and will only show you the Raw clips if you’re in Raw mode and only your compressed clips when in compressed mode.

Compressed footage modes

The FX5 names its video modes by the areas that are sampled, rather than their output resolution. The camera’s “17:9” modes (5K and 4.5K) are output as DCI 4K (4096 x 2160), while the 16:9 modes (5K, 4.5K and 3.2K) are output as 3840 x 2160 UHD.

The DCI modes can only be shot in XAVC S-I (All-I H.264), whereas moving to 16:9 adds the option to shoot XAVC H (H.265) or XAVC S (H.264) with a choice of 4:2:2 or 4:2:0 chroma sub-sampling.

Sony says the FX5 will gain support for shooting 3:2 open gate in XAVC S-I will come as part of firmware 2.0, due in around October 2027.

Super35

The FX5 also provides an option of cropping in to an APS-C-sized region of the sensor, providing compatibility with lenses designed for Super35 cameras. While the FX3’s APS-C mode was sampled from an 1080p region of the sensor, the FX5’s higher-resolution sensor means it can capture a 3.2K (3264 x 1836) region in 16:9. However, this is still output as 1080p when shooting in compressed modes.

Support tools

The FX5 gains a whole host of video-specific tools meant to make it easier to get proper exposure and focus. It has a waveform monitor for clearer and more accurate exposure monitoring, a false color overlay that makes it easy to see at a glance whether your subject is properly exposed and whether any of your highlights and shadows are clipping, a vector scope for more information about your colors and a focus map overlay to assist with focus.

The company has also added more options for desqueezing the preview when shooting with anamorphic lenses, providing several options between a 1.3 and 2.0x ratio.

Initial impressions

James Tirado

The FX5 is an evolution of the FX3 concept that tries to accommodate a wider range of users, and we doubt any one user will want or need the full range of its capabilities. Where the FX3 was essentially an a7S III in a more video-friendly format (or, perhaps, the camera the a7S III should have been, in the first place), the FX5 is a much more capable machine that tries to address multiple different workflows.

Small scale cinematographers can adopt its Raw workflow, while others might choose to shoot open gate and some users will output H.265 UHD files in S-Cinetone. Some people will build it into a rig, others will mount it on a gimble, while others will leave it on ‘sticks.’ The fact that the XLR handle and EVF module are optional suggests that these are option that some people will find essential while others would resent the bulk and cost of something they don’t want. Then there’s the Ethernet port, which allows the camera to be mounted and operated remotely, for applications such as churches, we’re told.

In doing so, though, Sony has made the camera a fraction larger, but probably not to the point that it would alienate many of the run-and-gun shooters who came to love the FX3.

However, the major changes to the user interface, including its frequent decision to reboot, if you accidentally select the wrong frame rate, while changing settings, are likely to make the camera seem surprisingly daunting to existing FX3 users.

James Tirado

Sony talked about adding features and capabilities that have been requested by users, but it also feels like the FX5 is much more a product of the cinema department, like the FX6, rather than having evolved from an Alpha model, as the FX3 appeared to. A lot of the features (not least Raw) might have trickled down from the Burano, but an awful lot of the interface and operational logic has done, too, which is likely to be something of a shock to anyone primarily familiar with the FX3 or 2.

James Tirado

It its favor, it’s stayed in the relatively compact body that has made the FX3 a success, but this means there isn’t room for the variable ND filter system that slides in filters from below the sensor on the bigger camera, that a lot of users were hoping for. Additional options like a USB timecode sync cable aren’t necessarily going to excite a lot of independent FX3 users, but they broaden the camera’s appeal.

It’ll be interesting to see whether Sony decides that the FX5 (or, rather, the existence of both the FX5 and FX2) means that it doesn’t need to create a direct replacement for the FX3.

The Godox iT32 is a pocketable, $80 flash. And it essentially sweeps away all the hurdles that stood in the way of me using flash.

I suspect I’m not alone in having found the cost of flash systems off-putting, when I first started photography. I’d scrabbled together enough money for a camera body with a nice kit zoom and immediately turned my attention to additional lenses. Flash was always the next thing on my list. And, two decades later, it still is.

You learn early on that flashes are most effective, creatively when they’re positioned away from the camera, but this invokes all the costs of remote triggers and off-board flashes, or being limited by the length of cables, because I’d bought an entry-level DSLR that couldn’t act as a flash commander, and I didn’t feel confident enough to start with manual, rather than TTL-controlled flash.

Mitchell Clark

The iT32 solves that off-camera problem because it connects to your camera via a small, hotshoe-mounted radio-frequency controller. The cost of this controller? between $20 and $34, depending on which brand of camera you use.

The X5 controller slots into your camera’s hotshoe and the iT32 attaches magnetically to the controller. Or you can disconnect the flash and fire it remotely. The flash itself communicates using Godox’s own communication protocol and can pair with any X5 controller, providing TTL compatibility whatever brand camera you own. It also means that you can buy a single iT32 and make it work even if you have cameras from more than one brand.

Richard Butler

*Godox says Canon cameras with multi-interface hotshoes should use an adapter to mount the X5 C. It makes a “X5 Cm” version for these newer cameras but has limited its sales to China.

The X5 can only control flashes operating in a single group, so you can’t use it as the basis of a complex multi-group setup, but the iT32 itself can operate in different groups, so will still be usable, if you decide to expand your setup, later.

So far, I’ve just been using the X5 and iT32 combination and have found it to be a really low-risk place to start experimenting with flash. It’s a genuinely small unit, that makes it easy to keep with me. It has a guide number of 18m, so it’s not especially powerful, but it’s enough of a step up from any camera’s built-in flash that, especially combined with the ability to use off-camera, is enough to make it worth carrying around.

In-use

Mitchel Clark

For such a small unit, the iT32 has a surprising number of control points. There’s a power/flash test button, a Set button and a Menu button, along with a dial on the side, used for quick control. But the iT32’s small rear screen is also touch-sensitive, so it’s really quick to operate, rather than requiring odd combinations of button pressing and dial turns.

To turn the unit on you hold the power/test button for two seconds then scroll the control dial upwards to confirm that it’s not just pressed against something in your bag.

You can access the menu by pressing the Menu button or swiping down on the screen. In the menu you’ll find options for controlling the modelling lamp on the front of the flash, a control for switching between TTL, Manual and multi-flash (strobe) modes, as well as deciding whether the flash is in on-camera mode, sender or receiver.

Mitchel Clark

The iT32 automatically switches to receiver mode when you detach it from the on-camera X5, so you don’t need to give most of this much thought.

In use you can set exposure compensation by turning the dial on the side of the flash, by swiping left or right on the touchscreen or, when the iT32 is working remotely away from the base, by pressing the small + and – buttons on the back of the X5 controller.

If you have a pair of iT32s you can your their controls to set them to different manual flash values (or flash exposure comp values if in TTL mode). You can still use the X5 to increase and decrease these values remotely, but any increase or decrease will be applied to both flashes, so you can’t remotely adjust them independently, as you would in a proper system of grouped flashes.

Mitchel Clark

When working handheld, I found the touchscreen to be very sensitive and the control dial very easy to jog, so tended to make use of the ‘Lock’ function in the menu, to stop myself accidentally changing the flash settings. Once lock is engaged, you need to hold ‘Set’ for two seconds to re-activate the controls.

The iT32 has its own, built-in battery, so you don’t need to hunt around for a fresh set of AAs all the time. It recharges over its USB-C socket, which is also used for updating firmware for both the flash and the X5. The X5 unit has its own battery, which charges up when connected to the iT32. You can check the battery level on the menu of the flash, when it’s connected, or long-press the – button, where it will communicate its charge level with between zero and four flashes of its tiny red LED.

Richard Butler

Generally, getting started with the iT32 is simplicity itself. The two units pair with one another when physically connected, switching the flash to the correct camera brand’s protocol, and then auto switches between on- and off-camera operation. It’s only two swipes on the touchscreen to manually engage receiver mode, if you power the flash on again later and don’t want to have to physically re-connect them each time.

As you use the setup more, it’s worth scanning through the manual, as there are some differences between camera brands about how to access modes such as high-speed sync and rear curtain flash sync, but there’s nothing that a complete novice should find off-putting at all.

Overall, though, I found the X5 / iT32 pairing to be a remarkably cost-effective way to access the world of flash photography. And, significantly, the realm of off-board flash. This simple, sub-$100 set overcame pretty much all of my previous concerns about cost, about compatibility and around batteries, leaving me with no excuse.

Want to stay up to date on the latest product news and releases? Add DPReview as a preferred source to ensure our independent journalism makes it to the top of your Google search results.

The Sony RX10 V is an update to the company’s long-zoom large-sensor compact. It’s built around a 20MP stacked Type 1 (116mm²) sensor and a 24-600mm equiv, F2.4-4.0 stabilized zoom.

It’s a long-zoom ‘compact’ whose newfound processing power means it can shoot a range of subjects from landscapes to wildlife and maybe even a little sport. We’d be tempted to make the comparison to a Swiss Army knife, only the RX10 is far from pocketable.

Key specifications

• 20MP Type-1 (116mm²) stacked CMOS sensor
• 24-600mm equiv. F2.4-4.0 lens
• Up to 30fps shooting
• 4K up to 120p with crop (60p full width) in 10-bit
• 3.68M dot EVF
• 1.62M dot tilting rear LCD
• UHS-II SD card slot
• 630 shot/charge battery life rating

The RX10 V will be available in early August with a recommended price of $2,300. This is exactly in line with the US consumer price index inflation rate applied to the $1700 launch price of the RX10 IV, nine years ago. To use a different benchmark, it’s cheaper, relative to the a7 V and that camera’s peers, than the Mark IV was, relative to its contemporary a7 model.

Index

• What’s new
• Body and handling
• In use
• Initial impressions
• Autofocus
• Image quality
• Video
• Conclusion
• Sample gallery
• Specifications
• Press release

What’s new

It’s been eight years since the RX10 IV and Sony has made big steps forward in both technology and ergonomics in the meantime: the RX10 V benefits from both.

On the technology side, the RX10 V gains the Bionz XR processor and “AI processing unit” from the previous generation of Alpha cameras. This brings the full range of subject recognition modes, including birds, animals, cars, planes and the “human pose estimation” system that is designed to recognise people-shaped things in the scene, even when the face or eyes aren’t visible, to help the camera continue to track the same subject.

The added processing power allows the RX10 V to shoot at up to 30 frames per second with a blackout free viewfinder. You also gain the Speed boost function that lets you temporarily engage a faster (or slower) shooting mode, when you hold a customized button down, but you don’t gain the pre-capture function.

Autofocus has come a long way in the nine years since the previous RX10 was released, and the new model feels like a thoroughly modern camera. The other area of significant improvement over the same period is video, and the RX10 V is a bit step forward in that regard, too, with a host of Sony’s latest options including, significantly, the ability to capture 10-bit and H.265 footage, both of which make it more pleasant to work with.

Sony says it’s the same lens as on the RX10 IV but CIPA standards for reporting focal length have apparently changed in the meantime, meaning it’s quoted as a 9.1-210mm zoom, rather than the 8.8-220mm figures printed on the side of the previous two models. We suspect the figures now quoted are the literal focal lengths, rather than the effective focal lengths after corrections has been applied. The end result is the same, though: a 24-600mm equiv range with an F2.4-4.0 maximum aperture and in-lens stabilization.

It’s hard to get your head round just how much flexibility a 24-600mm lens gives you until you use it. There are cameras that offer even more range, but all of them are based on much smaller sensors, so you pay a price in terms of image quality, if any survives the heat haze.

The lens has a built-in shutter which can go as fast as 1/1000th of a second at wide apertures, and 1/2000 sec at F8 and slower. This is great for flash, as you can sync up to those maximum speeds but quickly becomes limiting in bright light. You can switch the camera to e-shutter mode which, with a rolling shutter rate of 15.7ms (1/64sec) is still pretty fast, but this requires you to switch to the lossy compressed Raw format. There’s no auto mode to let you use mechanical shutter up to its limit, then e-shutter beyond that.

Body and handling

Richard Butler

The RX10 V is a substantial camera but that’s to be expected for something trying to deliver a 24-600mm equivalent zoom range. It’s been a long time since we handled an RX10 IV but the grip shape seems improved, offering a sturdy grip on the dense, polycarbonate body.

There’s no front dial, the logic presumably being that you’ll use your index finger for operating the zoom rocker. This means the two command dials on the rear shoulder are both operated with your thumb. There’s also a rather fiddly dial on the rear plate of the camera but thankfully, with the provision of two customizable command dials and an aperture ring, there’s rarely need to use it (it’s not used in photo mode, by default).

Richard Butler

Unlike recent Sony cameras, the RX10 V has a 10Gbps USB-C socket for fast data transfer but then retains the USB Micro B ‘Multi’ socket for attaching external accessories. This can no longer be used for charging the camera, reducing its usefulness to being an accessory port.

The RX10 V still uses a single SD card but the interface now uses the slightly newer UHS-II standard, rather than the UHS-I slot on the previous model. This is quick enough to accommodate the 600Mbps (75MB/s) data rate required for 4K/60 video capture with All-I compression.

Viewfinder

Richard Butler

The viewfinder gets an upgrade to 3.68M dots (1280 x 960px), which is a decent step up from its predecessor. The resolution drops a little if you engage the high refresh rate mode, and battery life drops, too, but whichever mode you choose, the resolution is maintained, even when using C-AF, so you don’t get the potentially distracting drop in detail when you half-press the shutter.

Screen

The screen is also refreshed, with the Mark V using a 1.62M dot (900 x 600px) screen that matches the sensor’s aspect ratio, rather than the 4:3, 800 x 600px panel on the previous model. The screen is still mounted on a tilt up/down cradle, which gives a decent level of flexibility but isn’t well suited to portrait orientation shooting.

Battery

Richard Butler

Other than the improved autofocus, perhaps the biggest change on the Mark V is the move to the much larger NP-FZ100 battery. This is the 16.4Wh battery that has powered the various a7 models for the past nine years. It has a capacity more than twice that of the FW50 battery used in previous RX10 cameras, giving much better battery life in a way that seems appropriate for a camera so likely to be used for travel and long shoots outdoors.

In use

When you first start up the camera, you’re confronted with a very simple, compact-camera display, with five large touchscreen buttons arrayed down both sides of the screen. This ends up feeling somewhat nostalgic, having not encountered a mainstream point-and-shoot for some time, and rather at odds with the more Alpha-styled controls and main menu.

Anyone not wanting to use touchscreen buttons, or wanting a less cluttered display can remove them (the option to do so is under Touch Operation / Touch Panel Settings, rather than with the display options in Operation Customize).

With this done, the RX10 V feels a lot like an Alpha-series camera, with the new menus representing a big improvement over the ones in the previous models. It’s not without its foibles but for the most part it’s well grouped and organized, and with a little customization of the Fn menu and custom buttons, you can minimize your need to visit it.

The RX10 V has the same issue as other Sony cameras in terms of a couple of incompatible settings: you’ll need to disengage Raw and switch to HEIF format before you can engage the camera’s HLG Stills mode to shoot true HDR images, and you’ll have to switch your Raw format if you want to engage the electronic shutter mode. All of which leaves the risk that you forget to switch back.

Richard Butler

Also, somewhat disappointingly for such a travel-friendly camera, Sony remains the only major brand not to offer some kind of in-camera Raw conversion. So if you take a nice shot but want to tweak the white balance, color mode or DRO level before sharing it, tough luck, you’ll just have to messily hack the JPEG around on your phone.

There’s a lot to like, though. The dials and joystick give a good sense of direct control, letting you feel you have direct command over the camera’s really rather impressive capabilities. It has its limits, though: it will try to keep focusing if you zoom with the ring around the lens (you can’t use the zoom rocker), but it definitely struggles, which can limit some sports or action shooting. But the ability to quicky zoom with the rocker under your index finger and fine-tune the framing with the lens ring helps keep the power zoom feeling pretty engaging.

While the RX10 V’s interface is borrowed directly from the Alpha series, its wide range of features and capabilities means it somehow manages to be even more cluttered than on those cameras. So, in addition to the settings and icons that surround all four edges of the screen, there’s a Stabilization indicator that shows up in the main preview, along with warnings about the AF illuminator and a huge “FULL” icon, towards the middle of the display. This has nothing to do with you running out of card space, though; it’s just reporting the essential information that the focus range limiter isn’t engaged.

Initial impressions

Richard Butler

It’s been nine years since the last RX10 model was released and, although things have been pretty quiet on the compact front, the rest of camera technology and performance have continued to make advances. Much like watching a young relative you rarely get to see grow up through a series of infrequent visits, the prolonged gap really makes you notice how much has changed in the meantime.

The RX10 V is a dramatically more advanced camera than its predecessor: faster to focus and much more dependable at subject tracking. The menus and ergonomics have come a long way, too, and its video capabilities are leagues ahead of where the RX10 IV had brought us to.

I’m not generally a fan of long-zoom cameras, as I’d usually prioritize compactness over that additional reach that I rarely find myself wanting to use. But the enhanced AF performance and speed capture add a lot to what was already a camera with a strong following. They vastly expand what the RX10 V can do, and the changes to the controls and battery are significant benefits, too.

Unlike the recent L10, that doubling of battery capacity hasn’t led to much of a change in body size, and it wouldn’t have made much difference anyway, on a body so dominated by that vast lens.

There can be a few delays, here and there, waiting for the lens to initialize, and it was disappointing to have to swap shutter modes (and Raw filetype) or stop down when hitting the 1/1000 sec limit of the mechanical shutter. But overall I’ve found the RX10 V to be a pleasant travel companion whose lens constantly challenged me to think of photos that I couldn’t otherwise take.

Whether that makes it worth $2300 is for each individual to decide, but it’s striking to get to see what nearly a decade’s worth of improvements, delivered all in one go, look like.

Autofocus

The RX10’s autofocus interface resembles that of Sony’s Alpha series ILCs, meaning you get a choice of 11 different AF area modes, which are supplemented with tracking version of each area mode, when you switch to AF-C mode. There’s no longer a switch to select AF-C on the front of the camera (just an AF/MF switch on the lens), but for the vast majority of subjects, you can comfortably leave the camera in AF-C all the time.

In addition to the area modes the camera has its subject recognition system, trained to recognize six different subject types. There’s also an Auto subject mode that will hunt through all its subject recognition modes to see if it can identify your subject. On top of this, there’s the option to select which subject types are included in this Auto mode, to boost both speed and accuracy.

The RX10 V’s autofocus is a big step forward from its predecessor. Faster processing and subject recognition algorithms mean the RX10 V can recognize and track subjects very reliably, and these capabilities extend across to video, in a way they didn’t, before.

However the system isn’t completely flawless: when asked to react quickly with a distant subject, aircraft detection mode didn’t find the planes in the following picture, instead focusing on the Space Needle. Overall, it’s beyond what we’ve previously seen from a compact camera, though.

The fact that it’s awkward to zoom while you’re shooting bursts (the lens ring continues to work but the zoom rocker doesn’t) and the camera isn’t able to maintain focus while zooming puts an upper limit on what you can expect the RX10 V to achieve, in fast-moving situations, but it’s still one of the most sports-capable fixed-lens cameras we’ve ever used.

Image quality

Studio test scene

Our test scene is designed to simulate a variety of textures, colors, and detail types you’ll encounter in the real world. It also has two illumination modes, full even light and low directional light, to see the effect of different lighting conditions.

Unsurprisingly, the RX10 V performs very similarly to the RX10 IV when looking at its Raw images at base ISO, which is a good thing. It captures a good amount of detail, and at 88mm equiv., its lens actually holds up quite well compared to other cameras with Type 1 sensors and lenses with much less ambitious focal lengths. It also provides more detail with less noise than the Nikon P1000, which offers much more reach but uses a lower-res and much smaller 16MP 1/2.3″ (28mm²) sensor.

The RX10 V’s JPEGs are where the decade of progress becomes apparent. The colors are much more pleasing than those of the IV, with more natural skin tones, and the vibrancy of colors turned up where they were lacking, and turned down where they were overbearing. We also found this to be the case in the sample photos we shot with the V. Sony’s progress on this front has been slow and steady, but with so much time between the two models, it’s much more noticeable.

This class of cameras has never been the best pick for low light photography, the RX10 V’s Raws again look as we’d expect. Switch over to JPEG and it’s clear that Sony has upgraded the camera’s noise reduction to some extent, but in higher-contrast, more detailed areas, we’d be hard-pressed to tell its performance apart from its predecessor’s. However, it does outperform Panasonic’s similarly-specced (and now discontinued) FZ-1000 II from 2020 in this regard, and is significantly better at low light than the P1000.

The camera’s reasonably fast F2.4-4.0 maximum aperture means it can maintain good image quality when there’s plenty of light, but it doesn’t take much for the light levels to start chipping into the image quality, as you can see here, where the

Video

The previous RX10 V’s video topped-out at 4K/30, despite the use of a fast, stacked CMOS sensor. The Mark V pushes way beyond this, with the option to shoot 4K/120 with a significant 1.38x crop or oversampled 4K/60 from the full width of the sensor. As you’d expect from a modern camera, this can all be done in 10-bit, making Log, and in particular, the more ambitious S-Log3 curve, much more usable.

Richard Butler

But video improvements over the past nearly decade go further than speed. The RX10 V lets you shoot in the attractive, S-Cinetone color mode and it lets you upload LUTs for previewing your preferred color grade, or to burn directly into the footage. It has Sony’s Auto Framing mode that identifies a subject, punches in and follows them around, upscaling the footage and giving a camera operator dynamism to what are actually locked-off shots. There’s also the Active Mode IS that corrects a wider range of movement, available at up to 4K/60.

Then, on the audio side of things, the RX10 V can capture up to four-channel audio, if you attach an accessory to its multi-interface connectors in the hotshoe. A 72mm filter thread allows you to attach ND filters as needed.

Above all of this, though, the RX10 V’s video also benefits from its improved autofocus system, which is much more tenacious and dependable in terms of subject tracking, especially if your subject is of a type the camera has been trained to recognize.

Conclusion

The RX10 V is an extremely capable all-in-one camera whose extreme flexibility, paradoxically, risks limiting its appeal to a relatively small audience.

It’s a huge step forward from its predecessor, with its updated autofocus in both stills and video, and the addition of an Alpha-style joystick makes it more engaging and usable.

Richard Butler

The AF improvements widen the range of circumstances in which the camera will confidently deliver, expanding its wildlife capabilities and also making sports shooting much more plausible than its predecessor. And this is a case of broadening the horizons of a camera that, with its huge lens range, could already do a lot.

There are some odd quirks, though: it feels strange that Sony didn’t bring its pre-capture mode that would boost its action-capturing abilities. And the lack of a mode that auto-switches shutter mode when the camera hits the upper limit of its mechanical shutter is frustrating (even if it’d mean being stuck in or switching to the lossy Raw mode used for E-shutter).

Some existing RX10 users are likely to miss the top-plate settings display, whereas our bigger concern was that all its command dials are positioned to be controlled with your thumb, rather than thumb and forefinger. But the addition of the joystick and larger battery make for a net improvement, we felt.

The RX10 V’s self-contained nature is very much part of its appeal. You never have to think about which lens to buy or bring with you: you already have a combination that extends from wide-angle to super telephoto. However, to achieve this range means the camera has a moderately sized sensor which, combined with its modest aperture range, puts an upper limit on image quality.

If you need the full breadth of the RX10 V’s features in a single package, it’s genuinely excellent, and it’s hard to think of a more responsive, engaging ‘compact.’ However, if you only need a subset of what the RX10 can do, you may well be able to find a different combination that offers better image quality or an even greater aptitude for sports at a comparable price.

What you won’t find is a camera that offers the RX10 V’s combination of range, capability and size, for the same price. If your shooting needs everything that the RX10 V offers, it’s the only game in town, so for those users, it’s a Gold.

Scoring

Scoring is relative only to the other cameras in the same category. Click here to learn about what these numbers mean.

8.0 out of 10

8.5 out of 10

8.8 out of 10

8.8 out of 10

8.5 out of 10

8.1 out of 10

8.5 out of 10

8.5 out of 10

9.2 out of 10

8.8 out of 10

9.5 out of 10

8.5 out of 10

5.0 out of 10

Show All Scores

Overall Score

85%

Scores reflect competition at the time of release; cross-year comparisons may be misleading

Expert Summary

The RX10 V is a hugely capable self-contained package that can lend itself to a remarkable range of photography. You probably have to need everything it offers to make its price make sense, but it's perhaps the most flexible 'compact' camera we've ever seen.

Good For:

A bit of (almost) everything

Not So Good For:

Sports, Maximal image quality

Reviewed on August 24, 2026

Read In-Depth Review

Sample gallery

Please do not reproduce any of these images on a website or any newsletter/magazine without prior permission (see our copyright page ). We make the originals available for private users to download to their own machines for personal examination or printing (in conjunction with this review); we do so in good faith, so please don’t abuse it.

Sony DSC-RX10 V specifications