AVerMedia DualCam (PW313D) Review - Seeing Double 3

AVerMedia DualCam (PW313D) Review - Seeing Double

Value & Conclusion »
The AVerMedia DualCam uses two separate camera modules. The left one is built around a 2 MP CMOS sensor with a maximum resolution of 1920×1080 at 30 FPS and a 71° diagonal field of view. The right one uses a 5 MP CMOS sensor with a maximum resolution of 2592×1944 at 30 FPS and a slightly wider 76° diagonal field of view. Both are autofocus units, which is essential for the document camera role, as the distance between the lens and the subject changes constantly. Higher framerates aren't supported on either camera, even at lower resolutions.

Both cameras share a single USB 2.0 connection, and that turns out to be the most consequential technical decision AVerMedia made with this product, because the two feeds can't both run at their maximum resolution at the same time. Set the left camera to Full HD and the right one immediately drops to 1280×720. Push the right camera to its native 2592×1944 and the left one drops to 1280×720 instead. There isn't even enough headroom for two simultaneous Full HD feeds: setting the right camera to 1920×1080 is already enough to knock the left one down to 720p. In other words, one camera always ends up at 1280×720, and you get to choose which one.

What makes this more frustrating than a simple bandwidth limitation is that framerate has nothing to do with it. I dropped both cameras all the way down to 5 FPS, which should leave a USB 2.0 connection with bandwidth to spare, and the exact same resolution downgrade occurred. The allocation is evidently fixed in firmware rather than negotiated according to what the connection can actually carry, so there's no combination of settings that will let you trade framerate for resolution. You're stuck with the combinations AVerMedia decided on, regardless if they make sense for your use case.

In practice, this matters less than it sounds, provided you think about which camera you actually need the resolution on. My recommendation after living with the DualCam for a while is to leave both cameras at 25 or 30 FPS and allocate resolution according to what you're showing: left camera at 1280×720, right camera at 1920×1080 or its full 2592×1944. The logic is straightforward, as the whole reason you bought this device is the second perspective, and sharpness matters far more on a document, a circuit board, or whatever else you're holding up than it does on your own face. A talking head at 720p is perfectly acceptable in any videoconferencing context, and most platforms would recompress your Full HD feed down to something similar anyway. Reverse the priority only if you're doing something where your face is genuinely the main subject, at which point you should probably be asking yourself whether you needed a dual camera in the first place.

On the audio front, the DualCam has dual mono omnidirectional microphones with a specified pickup distance of 3 meters (9.8 ft) and AI-powered noise reduction.

Picture Quality

The lighting in this first sample is deliberately hostile: a glass wall behind me and a single softbox to one side, with both cameras left on Auto. Backlighting of that severity is the classic webcam killer, and the DualCam handles it about as well as anything in this price bracket, which is to say it survives rather than triumphs. What's worth extracting from the sample is that the face camera's 63° horizontal field of view is narrower than most webcams offer. Manufacturers usually treat a wide angle as a selling point, but for a device meant to frame one person at a desk, narrow is the better choice, and it saves you from broadcasting your entire room.

The keystone demonstration in the second half deserves a word of explanation. What you're watching isn't optical correction but a software transform, which means the pixels being stretched back into a rectangle come from a shot where the far edge was sampled at lower effective resolution than the near edge. That's why the 5 MP sensor matters more here than the raw resolution figure suggests: the correction spends some of that detail to buy you the geometry. It also explains why the feature is restricted to the right camera, and why it assumes a flat subject. Anything with real height gets stretched along with the plane it's sitting on, which looks wrong in a way that's hard to unsee.

This is the DualCam in the conditions it was built for, and the contrast against the previous sample is the more useful comparison of the two. Give it an evenly lit room and the right camera turns in genuinely solid work: text on the page is readable rather than merely identifiable as text, which is the line separating this product from a conventional webcam pointed at a desk.

The page turns expose that the autofocus reacts to my hand entering the frame, not to any change in subject distance, since the book never moves. That's the algorithm doing exactly what it was designed to do and being wrong about it, and it's the reason I'd steer anyone in this scenario towards manual focus. Set it once on the page and the problem disappears entirely, because there was never a focusing problem to solve; the distance from lens to desk is fixed. Autofocus is worth having when you're holding objects up at varying heights, and worth switching off the moment your subject lies flat.

Everything demonstrated in this video comes down to one firmware decision, and it's the most consequential finding in this review. Autofocus on the DualCam only functions with Auto Exposure enabled. Take exposure into manual and the lens freezes where it stands. I verified this in both Streaming Center and OBS, on both cameras, and on two different computers, which rules out an application bug and places the behavior in the camera itself. AVerMedia states that the two systems are independent by design, and I have no reason to doubt that; the refocus trigger may simply lean on the metering data automatic exposure produces, leaving it without a signal once exposure is locked. Whatever the mechanism, the consequence is easy to state: the two controls you'd most want to combine can't usefully be used together.

That would be a footnote if the automatic exposure algorithm were exceptionally competent. It isn't. It pursues brightness above all else, and with shutter speed unavailable as a user-adjustable parameter, the only lever it has is a slower shutter. Motion blur follows automatically. The irony is hard to miss, since the right camera exists specifically so you can move things around underneath it, and the exposure algorithm is optimized in the wrong direction for that job. Manual exposure produces a better image, but the control is cruder than its appearance suggests. A slider marked 0 to 100 implies you can fine-tune; in practice it resolves to 10 steps, so you're picking from a preset list and living with whichever value lands closest. Combined with the loss of autofocus, that makes manual exposure unintuitive to use.

The result is a loop with no clean exit. Auto exposure buys you focus and costs you sharpness in motion. Manual exposure buys you sharpness and costs you focus. Neither branch is wrong for every scenario, and as the book sample demonstrates, a static setup happily takes the manual route. But the dynamic scenarios are exactly the ones this loop penalizes, and a firmware update decoupling the two controls would fix it.

This sample is engineered to be difficult, and it's included less as a fair test than as a demonstration of where the DualCam's limits sit. The red cast on my face is the auto white balance taking a monitor at its word. A screen is a narrow, unnatural light source, and a webcam with no reference point for what skin should look like has little chance of correcting for it. Most cameras in this class fail the same test, and the practical takeaway is that a monitor-lit face is not a scenario any of them are built for.

The keyboard shows the exposure algorithm's priorities more clearly than any of the earlier samples. Starved of light, it slows the shutter until the frame is bright enough, and every hand movement smears as a result. There's no setting that would let you tell it you'd rather have a darker, sharper image, because shutter speed isn't yours to control. This is the same behavior as in good light, just amplified to the point of being unmissable. It also confirms that the blur is a deliberate trade rather than a sensor limitation, which is worth knowing, because it means adding light solves it completely. Even a single table lamp helps a lot.

Microphone Quality

The DualCam carries dual mono omnidirectional microphones with a specified pickup distance of three meters and AI noise reduction. The dual mono designation is worth pausing on, since two capsules spaced this far apart could have given AVerMedia a stereo image, and they chose not to take it. That's arguably the right call for a conferencing device, where mono is the safer format and stereo can confuse platforms that downmix anyway, but it does mean the second microphone buys you coverage rather than width.

Isolating the channels turned up something I didn't expect. The two microphones don't sound the same. The left one is noticeably cleaner, and the difference between them isn't a matter of level but of tonality, as though the two capsules were voiced differently or sit in acoustically different pockets of the housing. On a device that sums both channels into mono this is less consequential than it would be on a stereo microphone, and in normal use you'd never isolate them the way I have here, but it's an odd thing to find on a product where matching the two should have been trivial.

In the dual mono configuration you'll actually be using, the result is fine. Across all the samples on this page, recorded from a normal seated position, voice pickup is clear and adequate for calls and online teaching, which is the realistic bar for microphones built into a webcam, though not much beyond it. The DualCam won't have you retiring a headset, and anyone recording anything more demanding than a meeting should plan on an external microphone.
Next Page »Value & Conclusion
View as single page
Nov 17th, 2026 20:15 CST change timezone

New Forum Posts

Popular Reviews

TPU on YouTube

Controversial News Posts