RoamingScott wrote:
I suppose the linked above A/B tests aren't QUITE A/B given the difference of the lenses...
Looking at these in LR, what immediately jumps out is the color cast on the Sony images. The Fuji whites are just...white. The Sony has some yellow and green in the white that stubbornly hangs around.
On high key areas, you can see the smoothness that the highlights roll off, and on the Sony, it's a more stark demarcation.
The Fuji colors pop more off the jump too. Whether that is realistic or not...who can say?
Then you return to reality and stop looking at 100% and they are, for all intents, the same photo. No one on IG, my website, FM, or otherwise would ever notice the differences between the two....Show more →
Okay, so what system are you going with?
gear-nut wrote: RoamingScott wroteI suppose the linked above A/B tests aren't QUITE A/B given the difference of the lenses...
Looking at these in LR, what immediately jumps out is the color cast on the Sony images. The Fuji whites are just...white. The Sony has some yellow and green in the white that stubbornly hangs around.
On high key areas, you can see the smoothness that the highlights roll off, and on the Sony, it's a more stark demarcation.
The Fuji colors pop more off the jump too. Whether that is realistic or not...who can say?
Then you return to reality and stop looking at 100% and they are, for all intents, the same photo. No one on IG, my website, FM, or otherwise would ever notice the differences between the two.Okay, so what system are you going with?
I haven't even gotten the Sony in my hands. Either way, I'll make great photos with whatever
This was an interesting read through. TL;DR, the GFX 100mp sensor is considerably better at recovering shadow detail, and the 61mp Sony is better at recovering highlight detail. He mentions that the new AWB in the A7R5 is much more accurate than the A7R4 and that it actually oversaturates blues of all things now.
The GFX100 II in comparison delivers top-notch image quality, colors and tones at a more affordable price. The weight of the camera is manageable, and the electronic viewfinder is outstanding. It also costs twice as much as the Sony which overall is still a great performer.
After testing the Fuji camera, I have decided to stick to the A7R V for now for my architectural and interior work. Switching system always ends up being a costly exercise and I am satisfied with the quality produced from the Sony. I found the GFX100 II's overall image quality to be better, but only by a small margin...For most people seeking the best colors and image quality, The GFX100 II is currently the best choice.
A tripod based arch shooter opting to stay with the Sony system despite the visible gains of GFX, and he doesn't even get to the top of mountains with it
He's got A/B sample RAWs, checking those out....Show more →
Thanks for the lead. I downloaded the RAW files. Amazing that he was willing to share these!
It's not quite an A/B comparison because the lenses are different. I know the Fuji GF 30mm t/s is superb. I don't know if that Sony lens is any good, and I also have not used the S-K 35mm for Phase One. If we can assume that the Sony and S-K lenses are good, I'd buy the Fuji for my work and live with the compromises -- no question. (Full disclosure: I did buy the Fuji GFX 100S and I am living with the quirks and compromises. Now I'm just happier about that choice.)
I came to this conclusion using the f/8 "Sharpness" RAW files. I cropped the 4:3 files to 3:2 and then exporting the three RAW files as TIFF to 11,648 pixels. I realize that's not a fair fight because that leaves the Fuji file in its native format, while the Sony and Phase One IQ260 files are uprezzed. Nonetheless, if I was printing all three files to the same size, that's what would happen -- so it's not an unreasonable approach.
Comparing the three TIFF files side-by-side in Photoshop at 100%, the Fuji file is far and away the sharpest, cleanest image. It's not close.
Would anyone but a pixel-peeping photographer notice the difference? I highly doubt it. The three processed JPEG files that are all 5,000 pix on the long edge are the proof of that. The Fuji version still looks a bit better to me, but all three versions are excellent and should satisfy any client of this kind of work.
FYI Rob, he was adapting the old Canon TS-E to the Sony. Who knows what all that changes here.
I suppose you can look at the experience of editing/pixel peeping as the modern equivalent of the darkroom experience. What one sees and feels while in the darkroom influences the final product and output.
Does having this extra information available to us during post processing drastically affect the output? Probably not all that much.
Do we spend a whole bunch of our lives (if we're dedicated to the modern craft) in this digital darkroom? Yes, and perhaps there is something to be said for improving that aspect even if the compressed, downrezzed output is nearly identical. A fine timepiece tells the same time, but in a different way. A fancy car still drives the same road, but the experience from A to B is different and improved.
This was an interesting read through. TL;DR, the GFX 100mp sensor is considerably better at recovering shadow detail, and the 61mp Sony is better at recovering highlight detail. He mentions that the new AWB in the A7R5 is much more accurate than the A7R4 and that it actually oversaturates blues of all things now.
The GFX100 II in comparison delivers top-notch image quality, colors and tones at a more affordable price. The weight of the camera is manageable, and the electronic viewfinder is outstanding. It also costs twice as much as the Sony which overall is still a great performer.
After testing the Fuji camera, I have decided to stick to the A7R V for now for my architectural and interior work. Switching system always ends up being a costly exercise and I am satisfied with the quality produced from the Sony. I found the GFX100 II's overall image quality to be better, but only by a small margin...For most people seeking the best colors and image quality, The GFX100 II is currently the best choice.
A tripod based arch shooter opting to stay with the Sony system despite the visible gains of GFX, and he doesn't even get to the top of mountains with it
He's got A/B sample RAWs, checking those out.Thanks for the lead. I downloaded the RAW files. Amazing that he was willing to share these!
It's not quite an A/B comparison because the lenses are different. I know the Fuji GF 30mm t/s is superb. I don't know if that Sony lens is any good, and I also have not used the S-K 35mm for Phase One. If we can assume that the Sony and S-K lenses are good, I'd buy the Fuji for my work and live with the compromises -- no question. (Full disclosure: I did buy the Fuji GFX 100S and I am living with the quirks and compromises. Now I'm just happier about that choice.)
I came to this conclusion using the f/8 "Sharpness" RAW files. I cropped the 4:3 files to 3:2 and then exporting the three RAW files as TIFF to 11,648 pixels. I realize that's not a fair fight because that leaves the Fuji file in its native format, while the Sony and Phase One IQ260 files are uprezzed. Nonetheless, if I was printing all three files to the same size, that's what would happen -- so it's not an unreasonable approach.
Comparing the three TIFF files side-by-side in Photoshop at 100%, the Fuji file is far and away the sharpest, cleanest image. It's not close.
Would anyone but a pixel-peeping photographer notice the difference? I highly doubt it. The three processed JPEG files that are all 5,000 pix on the long edge are the proof of that. The Fuji version still looks a bit better to me, but all three versions are excellent and should satisfy any client of this kind of work. ...Show more →
I think comparing sharpness between systems is a largely pointless exercise. Far and away the biggest determining factor of sharpness when you're comparing sensors with the same pixel size (3.76 microns in the case of the GFX100 and A7RV) is going to be the lens used. He has used different lenses in this test obviously, and what's more the lens used on Sony isn't even a native mount lens, but instead a Canon lens from 2009 versus a Fuji design 15-years newer! A sharpness comparison between the two systems might make more sense if one used the sharpest available lens within a given focal range (adjusting for FOV differences between sensors) for each mount, but he has not done that. The Canon 24mm TS-E is by no means the sharpest 24mm for Sony. At best this perhaps shows if you insist on having a TS-E lens with a 24mm FOV, the GFX is the better choice since no such native lens exists for Sony so the 2009 Canon is the best alternative.
By all means compare on factors like dynamic range, noise, etc, but testing sharpness is largely redundant.
RoamingScott wrote:
FYI Rob, he was adapting the old Canon TS-E to the Sony. Who knows what all that changes here.
Good to know that it was the Canon TS-E 24mm. I went straight for the RAW files and didn't read his text so I missed that. I haven't used that lens but the consensus among people who have used it on GFX is "good" but not top notch like the newer Canon TS-E designs.
With that lens, it wasn't a fair fight for the Sony.
RoamingScott wrote:
I suppose you can look at the experience of editing/pixel peeping as the modern equivalent of the darkroom experience. What one sees and feels while in the darkroom influences the final product and output.
Does having this extra information available to us during post processing drastically affect the output? Probably not all that much.
Do we spend a whole bunch of our lives (if we're dedicated to the modern craft) in this digital darkroom? Yes, and perhaps there is something to be said for improving that aspect even if the compressed, downrezzed output is nearly identical. A fine timepiece tells the same time, but in a different way. A fancy car still drives the same road, but the experience from A to B is different and improved. ...Show more →
You are preaching to the choir.
It's a kind of madness to be carrying 25 lbs of digital view camera around for seven hours of field work like I did last Friday, but here we are. To be happy in my work, I just have to have the movements, and I'm willing to suffer the weight, cost and hassle to get them.
RoamingScott wrote:
Jack has a good point. It's all about finding that balance between IQ, portability, etc. It's made harder by the A7R5, 20-70, and 70-200 costing me less than a 100S II after tax.
It would take a bit to recoup the extra cost of the GFX system via sales, but I'm not all that concerned about that. I'm more concerned about going to a place once and missing a shot because I was worried about changing lenses in tough conditions, or worse, simply lacking the needed focal length and cropping into and winding up with a 50mp shot on my 100mp camera. ...Show more →
You also made it clear that you were going to carry this stuff around, on hikes, sometimes a far distance, and wanted a great kit, but not an overly heavy kit. If you now think something like a 70-200 is in play, that changes the ballgame a bit on the GFX side.
But, the GFX is probably a better all around landscape camera, on the image quality side. It might be a lot better in some peoples hands than a 45-60MP FF sensor, but maybe only a little better in more experienced landscapers hands.
61mp might be the ultimate that FF sensors can be, and some think think 50mp is where the FF sensor should be before it suffers from some issues (like signal to noise, blah blah). Is 100mp the best than MF can be, or is 120mp? Not that you need it, or would ever need it, but, is the MF sensor as far advanced as the FF ones are?
And yeah, given the lens on the GFX, a cropped 60MP image from it might be better in the corners than a full version from the Sony. Might be your last big camera purchase, and if you want to be in the giant landscape art space, the GFX is made for that more than the Sony is.
Aztatlan wrote:
By all means compare on factors like dynamic range, noise, etc, but testing sharpness is largely redundant.
It's difficult to compare sharpness, but it's only redundant if you don't care about it. I don't think it's unreasonable as a comparison point if we're comparing high end systems like these.
Of course there are other points of comparison that can be much more important. In most cases, we are well past the point where the camera and lens is the limiting factor in the quality of the work produced.
rdeloe wrote: Aztatlan wroteBy all means compare on factors like dynamic range, noise, etc, but testing sharpness is largely redundant.It's difficult to compare sharpness, but it's only redundant if you don't care about it. I don't think it's unreasonable as a comparison point if we're comparing high end systems like these.
Of course there are other points of comparison that can be much more important. In most cases, we are well past the point where the camera and lens is the limiting factor in the quality of the work produced.
You can care about sharpness all you want, the point is that the test is telling you nothing useful about it when the GFX100 is being tested with a brand new $4000 lens and the Sony is being tested with a Canon lens from 2009. About the only useful bit of data in terms of sharpness is that if you must use a 24mm tilt shift on Sony, this is where you would land. If you just need a 24mm lens on the other hand, there are much sharper choices available for the mount than was used here.
p.5 #11Sony A7R5 vs GFX 100S (et al) for landscape
Aztatlan wrote: rdeloe wroteAztatlan wroteBy all means compare on factors like dynamic range, noise, etc, but testing sharpness is largely redundant.It's difficult to compare sharpness, but it's only redundant if you don't care about it. I don't think it's unreasonable as a comparison point if we're comparing high end systems like these.
Of course there are other points of comparison that can be much more important. In most cases, we are well past the point where the camera and lens is the limiting factor in the quality of the work produced.You can care about sharpness all you want, the point is that the test is telling you nothing useful about it when the GFX100 is being tested with a brand new $4000 lens and the Sony is being tested with a Canon lens from 2009. About the only useful bit of data in terms of sharpness is that if you must use a 24mm tilt shift on Sony, this is where you would land. If you just need a 24mm lens on the other hand, there are much sharper choices available for the mount than was used here.
Sure. But with that argument, comparison is pointless unless you're using the same lens, and a lens that works on both cameras will not be native. It won't even be the same field of view if the sensors are different sizes, as they are in Scott's scenario.
Scott still has to make a decision. If he had both cameras in hand with the lenses he wanted to use on each of them, he'd still want to compare the files they produced if he's interested in sharpness, even if they used different lenses.
p.5 #12Sony A7R5 vs GFX 100S (et al) for landscape
RoamingScott wrote: gear-nut wroteRoamingScott wroteI suppose the linked above A/B tests aren't QUITE A/B given the difference of the lenses...
Looking at these in LR, what immediately jumps out is the color cast on the Sony images. The Fuji whites are just...white. The Sony has some yellow and green in the white that stubbornly hangs around.
On high key areas, you can see the smoothness that the highlights roll off, and on the Sony, it's a more stark demarcation.
The Fuji colors pop more off the jump too. Whether that is realistic or not...who can say?
Then you return to reality and stop looking at 100% and they are, for all intents, the same photo. No one on IG, my website, FM, or otherwise would ever notice the differences between the two.Okay, so what system are you going with?I haven't even gotten the Sony in my hands. Either way, I'll make great photos with whatever of course, but you’ll have more fun doing it with the GFX — especially when you’re processing them out 🤣🤣🤣
p.5 #13Sony A7R5 vs GFX 100S (et al) for landscape
rdeloe wrote: Aztatlan wroterdeloe wroteAztatlan wroteBy all means compare on factors like dynamic range, noise, etc, but testing sharpness is largely redundant.It's difficult to compare sharpness, but it's only redundant if you don't care about it. I don't think it's unreasonable as a comparison point if we're comparing high end systems like these.
Of course there are other points of comparison that can be much more important. In most cases, we are well past the point where the camera and lens is the limiting factor in the quality of the work produced.You can care about sharpness all you want, the point is that the test is telling you nothing useful about it when the GFX100 is being tested with a brand new $4000 lens and the Sony is being tested with a Canon lens from 2009. About the only useful bit of data in terms of sharpness is that if you must use a 24mm tilt shift on Sony, this is where you would land. If you just need a 24mm lens on the other hand, there are much sharper choices available for the mount than was used here.Sure. But with that argument, comparison is pointless unless you're using the same lens, and a lens that works on both cameras will not be native. It won't even be the same field of view if the sensors are different sizes, as they are in Scott's scenario.
Scott still has to make a decision. If he had both cameras in hand with the lenses he wanted to use on each of them, he'd still want to compare the files they produced if he's interested in sharpness, even if they used different lenses.
You're kind of right, but in practical terms as long as one is making a comparison using the sharpest available comparable focal length lenses for a system, you've got useful data. Eg if you take the sharpest 24mm for Sony, and the sharpest 30mm for Fuji, and compare them, you have the same FOV, and using the sharpest lens on both systems means you are using the best possible data to compare. But yes, it isn't perfect as there is still differences in lenses but this at least tells you it's the sharpest possible output for both systems at a given FOV.
When comparing wildly different lenses, as has been done here, it's far more of a lens comparison than a sensor/system comparison.
p.5 #14Sony A7R5 vs GFX 100S (et al) for landscape
Aztatlan wrote:
You can care about sharpness all you want, the point is that the test is telling you nothing useful about it when the GFX100 is being tested with a brand new $4000 lens and the Sony is being tested with a Canon lens from 2009. About the only useful bit of data in terms of sharpness is that if you must use a 24mm tilt shift on Sony, this is where you would land. If you just need a 24mm lens on the other hand, there are much sharper choices available for the mount than was used here.
I'm with you. I quit paying attention to that article as soon as I saw the lens lineup. I would think one would need to find at least the current lens equivalents per system otherwise it seems pointless. I have Sony and have a friend who uses Fuji, and while I like the platform, the lens choices and overall costs are prohibitive for me. What I like about Sony is size, weight and that I was able to find my jam in lenses. I find the Fuji experience to be limiting because the lenses are very expensive and you're kind of locked in. I'm not sure if I had been stuck with Sony brand lenses only if I would have continued on with that platform. These days these sensors are sharp and powerful and it's the logistics of the platform (size, weight, buttons, menus) that make the real differences and lenses are a huge part of that. Not to say that if I had the cash I wouldn't get along with Fuji, my experience with the files was positive. Yet, as a manual focus shooter who detests focus by wire with a passion....
I think if I was going to do cost no object architectural or landscape photography, I would opt for Alpa or Arca-Swiss with a Hassy 100mp back...
But, I get along with my A7RIV and am trying to tell myself that I don't need the upgrade to the A7RV....
p.5 #15Sony A7R5 vs GFX 100S (et al) for landscape
I did landscape comparisons a few years ago, but they are all slightly odd matchups: GFX 23mm and 30mm versus Nikon Z 14-24/2.8, GFX 45-100 versus Nikon Z 35/1.8 and adapted CV 65/2. The midrange comparison is further complicated by the 45-100 not being a great landscaper at 45mm.
I wish I could dump them all online somewhere, but it's about 10 gigs of raws...I might be able to put a few up on dropbox. It's difficult to compare between systems though.. after upscaling everything to a common res (don't just upscale one), turning off all sharpening, and matching the crops...matching the exposure and contrast is still tough, and the eye is enormously persuaded by exposure and contrast. Then add in things like the higher aliasing the Z7+CV65 combo creates, and remembering that aliasing affects the ability to add sharpness in practice...
p.5 #16Sony A7R5 vs GFX 100S (et al) for landscape
letchhausen wrote:
I think if I was going to do cost no object architectural or landscape photography, I would opt for Alpa or Arca-Swiss with a Hassy 100mp back...
But, I get along with my A7RIV and am trying to tell myself that I don't need the upgrade to the A7RV....
Unfortunately the Hasselblad CFV 100C has a major limitation for that use case. The wide technical lenses produce strong banding when shifted. Owners who bought it to use on technical cameras and digital view cameras are not happy, especially now that it's been documented that Hasselblad knew about the issue.
This is really unfortunate because it is otherwise a great option. There's no timeline from Hasselblad on a fix (firmware or Phocus).
p.5 #17Sony A7R5 vs GFX 100S (et al) for landscape
RoamingScott wrote:
Well, until someone coughs up a better A/B test, this is the best I could find
You may have answered this question already in the thread Scott, but have you considered renting what you need to try each camera with at least one lens that you would definitely get for each one (with some focal lengths between them)? That way you can make some meaningful comparisons.
I'm always a bit envious that folks in the US have LensRentals. We don't have anything like that in Canada, so the stakes are a bit higher when you want to choose between two options.
p.5 #18Sony A7R5 vs GFX 100S (et al) for landscape
rdeloe wrote: RoamingScott wroteWell, until someone coughs up a better A/B test, this is the best I could find You may have answered this question already in the thread Scott, but have you considered renting what you need to try each camera with at least one lens that you would definitely get for each one (with some focal lengths between them)? That way you can make some meaningful comparisons.
I'm always a bit envious that folks in the US have LensRentals. We don't have anything like that in Canada, so the stakes are a bit higher when you want to choose between two options.
I could do that for sure. Usually I just buy the two things used and sell off the loser, since it's about the same price as a rental and I get to keep the gear far longer.
My wife, I fear, would not abide by such an experiment in this case
p.5 #19Sony A7R5 vs GFX 100S (et al) for landscape
RoamingScott wrote: rdeloe wroteRoamingScott wroteWell, until someone coughs up a better A/B test, this is the best I could find You may have answered this question already in the thread Scott, but have you considered renting what you need to try each camera with at least one lens that you would definitely get for each one (with some focal lengths between them)? That way you can make some meaningful comparisons.
I'm always a bit envious that folks in the US have LensRentals. We don't have anything like that in Canada, so the stakes are a bit higher when you want to choose between two options.I could do that for sure. Usually I just buy the two things used and sell off the loser, since it's about the same price as a rental and I get to keep the gear far longer.
My wife, I fear, would not abide by such an experiment in this case
p.5 #20Sony A7R5 vs GFX 100S (et al) for landscape
rdeloe wrote: Aztatlan wroteBy all means compare on factors like dynamic range, noise, etc, but testing sharpness is largely redundant.It's difficult to compare sharpness, but it's only redundant if you don't care about it. I don't think it's unreasonable as a comparison point if we're comparing high end systems like these.
If you shoot high resolution sensors through a low resolving optic, the predictable result is that the higher resolution sensor gains you rather little. So using that to evaluate the actual relative merits of those sensors is pointless.
Of course another great leveler is aperture. For landscapes you'd usually wana shoot closed up apertures for it's larger DOF, but that will also limit the achievable resolution. And this effect is independent of sensor size: as you close the larger and the smaller sensor systems to equivalent apertures, the limiting effect of diffraction will also be equivalent.