sebjmatthews
Offline
•
Upload & Sell: Off
|
I have both the R5 and R50, so I can do some direct image quality comparisons. Do bear in mind that just because the R50 and R10 have the same sensor and processor does not necessarily mean they will have identical image quality, as we have no idea if the ADC is the same, and other small adjustments and optimisations can result in differing images between cameras with the same sensors.
I don't check these boards often and my time online is limited, but if there's something really specific you'd like to see compared in a very specific way, detail it as much as possible and I'll see what I can do. For now, I've uploaded some very simple shots just of a ColorChecker under household 'warm white' LED lighting, using ISO 1600-6400. This should give you an idea of how the two compare under typical light you might find anywhere, and in the ISO range where noise can be visible even after resizing.
Flickr album here.
Each pair of photos is arranged with the R5 version first and the R50 equivalent second.
The first pair are simply ISO 3200 files converted to jpg, not resized or edited in any way. You can click through to see and download the full size files, and compare them however you like.
The next pair (images 3 and 4) are ISO 3200 underexposed by one stop then pushed back up, given noise reduction, and resized a little smaller than the others (1200x800) as a kind of 'worst case scenario' example. I think both cameras get away with this.
After that are four pairs resized to 1800x1200, which is approximately the size displayed by Instagram, Twitter, Facebook, etc. In order the pairs are ISO 1600, 3200, 6400, then 3200 underexposed and pushed a stop.
Lastly there's a pair of crops of the pushed files, not resized, if you want to compare their noise fully.
Files were processed/resized/simply converted in Lightroom, using Adobe's 'camera standard' profile, which is 99% the same as Canon's own Standard colour profile. I used the EF 100mm f/2.8L macro lens with all lens corrections turned off. Highlight priority was left off in both cameras. Electronic shutter was used for both because the R50 doesn't have a full mechanical shutter, and the improved IQ the R5 has in full mechanical mode only happens below ISO 800 anyway.
I feel a daylight test is a bit pointless as anything up to and including ISO 800 in both cameras is clean enough for anything other than the most technically-demanding of archive work. Similarly, I feel testing over 6400 is pointless because quite frankly I don't think any image from any camera is at all usable beyond that. If you really want to see an ultra-clean or an ultra-dirty test then I can do that, but I can save you the time and just tell you that 800 and below is fine on both and >6400 is rubbish on both.
—
Having used both cameras extensively next to each other, here's how I rate their image quality:
- For ISO 100-200, the R50's IQ is good enough that with the bump in resolution, it flat-out beats the R5 in crop mode, in every way. Even if you're resizing, the extra resolution still results in better clarity.
- At 400-800, I feel it's dead even between the two. Obviously you get slightly more detail with the R50, but the R5's lower density does result in what's probably about a quarter or third of a stop less noise. When there's not much detail in the shot anyway, the R5 comes out looking better; when there's a lot of detail all over the place, the R50 looks better. After resizing (and I always resize a little bit, even when printing large) the R50 generally comes out on top, but not by much, and not always.
- For 1600-3200, I think the cropped R5's IQ is better, but not by enough to put me off using the R50. It's a minor difference in IQ, which is far smaller than the difference in size and weight.
- For 6400 I think both cameras can be made acceptable for web use, but the R50 takes more work to get there and the 1200x800 example I've put in that album is the limit of how big I'd use that file; the R5's file doesn't need quite as much noise reduction, can handle a fraction more sharpening, and could be used a tiny bit larger. I would not print from either of them.
- If you have to crop into the file further than I prefer the R50 file, at any ISO.
- At all ISOs the cropped R5 has about half a stop less chromatic noise than the R50, though as that is the easiest noise to non-destructively remove in raw processing software, it won't be an issue for many people. For in-camera jpg users, though, it's noticeable, since Canon's noise reduction only targets luminance noise.
- Overall, the R5 crop and the R50 seem to have the same dynamic range (how much contrast can be smoothly captured and seen before processing), but the R5 has a fraction more exposure latitude (how much the file can be pushed/pulled/recovered in processing) in desktop raw processing software. For in-camera users it's the other way around, as the R50 seems to handle post-capture exposure adjustments better. I don't know if that's a quirk of the hardware or the software, and I don't know if the R10 will be the same.
Overall, I don't think raw image quality is a reason to pick between the R5 in crop mode vs the R50 (or R10). If you ever want to use in-camera jpg processing then the R50 (and presumably the R10) does more clearly beat the cropped R5. Of course, the full R5 beats the smaller sensors easily, no matter how you shoot.
I think it's also worth noting that while the R5's tracking focus does get better in crop mode, the R50 still beats it in both speed and accuracy. I have to assume the R10's tracking autofocus is at least as good as the R50's, i.e. also better than the cropped R5. This isn't something which ever gets mentioned in reviews or marketing material, so I thought it's worth highlighting. (Single-shot focus is identical in both cameras.)
—
FWIW, later this year I'm going to be going on a long trip overseas; I intend to bring a camera with me, and I'm not even considering the R5. I'll be taking either the R50, a Fujifilm X-T30 II, or an Olympus E-P7. For 'travel', and general photography, all three of them are preferable to the R5. I imagine the R10 would be, too. I like my R5, but I only use it when I actually need the extreme performance and will get significant use out of the whole sensor to justify the extra bulk. Nineteen times out of twenty, the smaller cameras are more enjoyable to carry, more enjoyable to use, and will generate the same result.
YouTube and forums are full of people who will always tell you to only use the biggest, most powerful, and most expensive gadget, no matter the situation. At least when it comes to cameras, and especially image quality for day-to-day use, bigger does not necessarily equal better.
|
.)