X-T5 vs X-H2s & X-H2 Initial Impressions (Store demo)

I had a chance to go back to the store today with an SD card and some more experience with the X-T3 under my belt.

First, I want to begin by saying that my initial impressions overstate the degree of improvement in autofocus at least as far as the X-T5/XH-2 are concerned vs the X-T3 and X-T4.

Coming from almost six years with an X-Pro 2, vs only about a week with the X-T3, I was still fairly unfamiliar with the autofocus options available on that camera. In the intervening time, I spent some more time getting to know the camera, did a large photoshoot (20+ people, multiple poses, ~4 hours) with it, and also took it on a brief, two day vacation. With the correct settings (that better approximate how the X-T5 and X-H2S were set up in store), the X-T3 holds up much better than I thought for AF, face detection, tracking, pre-AF, etc. The X-T5 is still a clear improvement, but it may well be a smaller one than I initially led myself to believe.

Next, I want to discuss burst speeds and buffer depth as I feel like this is a real issue with the X-T5, and an area where it is noticeably worse than not only the X-H2s and X-H2, but also the X-T3. In a lot of ways the X-T5 reminds me of the X-T10: A relatively fast, lower priced camera with flagship features, hobbled by an inadequate buffer.

Seemingly no matter what the settings, when shooting full resolution raw, the X-T5 can only burst for about 1.5 seconds (~22 frames) before it slows down significantly. It then takes a significant amount of time to write out to the SD Card. While it will likely benefit from faster SD cards here (I ended up having to test the burst speed with a slower card), as someone who shoots action on occasion, I feel this cut hurts the camera a lot. However, I have some ideas for workarounds here which I will discuss in the next section.

Next up, I want to discuss file size and lossy compressed raw, something I have not really seen discussed here or elsewhere (such as in reviews) at all despite it being super relevant to a camera like the X-T5, and honestly to anyone who shoots action and needs more flexibility than JPEG but doesn’t necessarily need the full range offered by lossless compressed RAW. Lossless compressed RAW files shot on the X-T5 can clock in anywhere from around 39MB to a whopping 60(!)MB per file at high ISO. Compressed RAW files by comparison seem to land between 28.5MB to 30.5MB per file, cutting file size in half in the most extreme cases. Now, I haven’t had time to process any of these files yet, so I can’t comment yet as to what you’re losing, but what I can say is that this does seem to extend the X-T5’s paltry buffer by around 0.5 ~ 1 second (~30 frames with JPEG, ~37 frames without JPEG) at low ISO. This can be further extended by another ~1.2 seconds (~57 frames) using the 1.29x crop mode (or presumably/hopefully by lowering the resolution by a similar amount, although I did not test this.)

The X-H2S obviously has much better burst depth even when shooting with the same relatively slow SD card, maintaining 40FPS for ~3.5 seconds (146 frames) before slowing down (albeit by not nearly as much). From what I have read online, the burst depth is even deeper when using a CF card for RAW. That said, the XH-2S can also benefit from smaller file sizes (lossless RAWs are ~30-40MB, slightly larger than X-T3, compressed RAWs by contrast are ~20MB.) I’d imagine you can get an even deeper buffer using compressed RAW but didn’t have enough SD card space left to test it out by that point. That said, on the X-H2S the file size difference isn’t quite as dramatic, so I’d imagine this is really only useful if for some reason you’re not using the CF slot (for example if its full and you’re left shooting on SD) or if you’re really tight for storage.

Another quick note, HEIF can save a fair bit of space compared to JPEG (20~30%) although I didn’t have time to test in detail.

I can’t say anything too definitive about image quality yet although I tried to take a fair shot between ISO125/160 to ISO 6400~12800. Hopefully, I’ll have time to spend playing with the files over the next few days to get a better feel for resolution/flexibility on the X-T5/X-H2. That said, while the X-T5 is obviously higher resolution but I’m not sure I can really judge how important that is with pictures taken inside the store.

A few final things of note for this update. At least so far I'm still really not a fan of the ergonomics on the X-H2 series. The joystick is larger and more tactile yet somehow feels less precise (easier to move the AF point two stops when I only meant to move it by one for example,) while the lack of physical controls is just really annoying. Also by moving the joystick, Fuji has made it hard to go back and fourth between an X-H2 body and an X-T body. Not. Great ( performance sure is though!) Also, I stand by what I said last time. The X-H2S feels faster in general operation (menus, etc) although this may be because the X-T5 is on beta firmware (1.0).