T
Tom Ames
Guest
Very good point!
-
Welcome to the new forums! Please read this first. For known issues we are working to resolve, click here.
Studio Scene at 4:3 downsampled to 4K, Z8 and GFX 100
- Thread starter JimKasson
- Start date
It was Greg’s assertion that there were great differences between those images.
Ap0ll0n
Veteran Member
Ok, the second looks much better than the first on some image pieces, not all.OK, how's this?
Exported from Lightroom at 2160 pixels high, no sharpening. JPEG Quality 100.
--
Apollon
Greg7579
Forum Pro
LOL, but I already know the answer. Fly to San Antonio and you can blind test me on my 6K monitor. I will nail it every time. 100% hit rate if at full res if we can somehow get properly prepped raw files with the 60 MP Q3 with the fantastic 28 and the 100 II with a close focal equivalent on a tripod at equivalent settings and enough DOF to geta lots of the pixels in focus on a deep scene.
It will take me 5 seconds on each image on my 6K monitor to at 1:1 to pick the GFX file every time. 100% success rate.
But we need scenes with depth and lots of detail. I want a guy sipping coffee 400 meters away next to a car where I can see the license plate.
I'm disappointed though New Orleans. I'm going to have to add you to my list of 7 MF Board monitor tech deniers. But that list is now down to 4. And I have those 4 rank-ordered on how stubbornly radical they are on their dedication to belief in this false myth, and some of them don't shoot 100 MP. LOL.
But all kidding aside, I understand the math and of course realize that there is downsizing when I view my GFX 100 MP raw at full screen fit view on a 6K monitor.
But those GFX files look so good and sensor size and res matters, and yes, I can of course see it on my monitor.
But so can everyone else. I think even the monitor tech deniers are starting to come around.
Hey, we are all brothers and sisters in high-res photography, and that includes FF at 50 and 60 MP.
So few of us left.
I don't care if a few of you don't believe me when I say what I can see on my monitor with my GFX files.
We all know high-res FF and GFX / Hassy MF is pretty darn good.
It will take me 5 seconds on each image on my 6K monitor to at 1:1 to pick the GFX file every time. 100% success rate.
But we need scenes with depth and lots of detail. I want a guy sipping coffee 400 meters away next to a car where I can see the license plate.
I'm disappointed though New Orleans. I'm going to have to add you to my list of 7 MF Board monitor tech deniers. But that list is now down to 4. And I have those 4 rank-ordered on how stubbornly radical they are on their dedication to belief in this false myth, and some of them don't shoot 100 MP. LOL.
But all kidding aside, I understand the math and of course realize that there is downsizing when I view my GFX 100 MP raw at full screen fit view on a 6K monitor.
But those GFX files look so good and sensor size and res matters, and yes, I can of course see it on my monitor.
But so can everyone else. I think even the monitor tech deniers are starting to come around.
Hey, we are all brothers and sisters in high-res photography, and that includes FF at 50 and 60 MP.
So few of us left.
I don't care if a few of you don't believe me when I say what I can see on my monitor with my GFX files.
We all know high-res FF and GFX / Hassy MF is pretty darn good.
just Tony
Veteran Member
I jumped right to the etching where Grandma looks thrilled to see the painting on the easel; those walls are unforgiving. It was too easy to identify the Z. My gear list shows what I'm familiar with firsthand.
Yes. In auto racing, you can't beat cubic inches. In aliasing, you can't beat a high sampling frequency.
Greg7579
Forum Pro
Yes, true that. Always trying to get better. Usually results in mediocrity or failure, but sometimes a hit!
Greg7579
Forum Pro
Can't deny that math, but a lot is going on with these new monitors besides just the pixel math. But yes, looking at a GFX 100 MP file at full screen fit, it is of course downsized. But it is less downsized at 4K and 6K and I can't wait to see 8K.If you look at the whole image on a 32" 4K monitor, you are looking at an 8MP image. That's a big downsize from 100MP.I don't see any arguments here. Greg, like me, is saying that there is an obvious difference when the images are viewed prior to any downsizing, at least on a large-ish monitor. Jim is saying that I identified those the other way around by looking at the downsized images, a screenshot at such.You are twisting my words yet again, Greg.All images look better on them by a significant margin. It is much easier to see the fine details and amazing image fidelity of our GFX raw files and everything else on these monitors. And 6K? Wow! Forget about it. There is not an argument here. If you saw it you would drool. Just like I did.
Downsizing, cropping and sensor dissection of large pieces of smaller sensors against smaller pieces of big sensors is interesting mathematically but unrelated to anything I'm saying, which is that I can see the increased image fidelity of GFX on my monitor, and anyone who says I can't need to go get one of these babies or make a trip to the ophthalmologist. LOL. Kidding, but I am baffled that me saying this simple truth upsets 5 people (as if it were an assault on FF), including my hero and good friend - my mentor and teacher - The Great Jim Kasson (who by the way has had 5,000-dollar monitors for the past decade!). LOL.
Now don't get mad at me Jim. You know I luv ya!
First off, I never claimed no difference. I said the differences weren't night and day. In support of that, note that the Z8 image was identified as the better image in this post:
https://www.dpreview.com/forums/post/67763266
Now, I'm not claiming that the Z8 image is better; I like the GFX 100II image better. But the fact that someone could consider it better means to me that the differences aren't that great, once downsampled by Ps.
I think the secondary point (sure, not how the thread started but certainly how it evolved) is that I don't think anyone would downsize any photo prior to viewing it on their monitor. The main being that when viewing normally, on a large-ish monitor, I think based on this thread at least, most folks would find that there is a significant difference.
I denied that math in this thread, and my calculations yield 6.2MP, but I take your meaning.Can't deny that math,If you look at the whole image on a 32" 4K monitor, you are looking at an 8MP image. That's a big downsize from 100MP.I don't see any arguments here. Greg, like me, is saying that there is an obvious difference when the images are viewed prior to any downsizing, at least on a large-ish monitor. Jim is saying that I identified those the other way around by looking at the downsized images, a screenshot at such.You are twisting my words yet again, Greg.All images look better on them by a significant margin. It is much easier to see the fine details and amazing image fidelity of our GFX raw files and everything else on these monitors. And 6K? Wow! Forget about it. There is not an argument here. If you saw it you would drool. Just like I did.
Downsizing, cropping and sensor dissection of large pieces of smaller sensors against smaller pieces of big sensors is interesting mathematically but unrelated to anything I'm saying, which is that I can see the increased image fidelity of GFX on my monitor, and anyone who says I can't need to go get one of these babies or make a trip to the ophthalmologist. LOL. Kidding, but I am baffled that me saying this simple truth upsets 5 people (as if it were an assault on FF), including my hero and good friend - my mentor and teacher - The Great Jim Kasson (who by the way has had 5,000-dollar monitors for the past decade!). LOL.
Now don't get mad at me Jim. You know I luv ya!
First off, I never claimed no difference. I said the differences weren't night and day. In support of that, note that the Z8 image was identified as the better image in this post:
https://www.dpreview.com/forums/post/67763266
Now, I'm not claiming that the Z8 image is better; I like the GFX 100II image better. But the fact that someone could consider it better means to me that the differences aren't that great, once downsampled by Ps.
I think the secondary point (sure, not how the thread started but certainly how it evolved) is that I don't think anyone would downsize any photo prior to viewing it on their monitor. The main being that when viewing normally, on a large-ish monitor, I think based on this thread at least, most folks would find that there is a significant difference.
NAwlins Contrarian
Forum Pro
If you're even half-serious, then I'd love to try to have the forum members put together a set of matched pairs of images for people to compare / try to tell apart. Maybe Jim has the technical expertise, access to equipment (especially through LensRentals), integrity, and universal respect to head that up. The logistics would be daunting, but not impossible. And although it's been like a quarter-century since I've been to San Antonio,* there's actually no need: everything could be shared online. 'Here, take a look at these files, report to The Tabulator which is GFX and which is Z8.' Then when all the participants had had a chance to compare files and report what they think, the Encrypted Answer Key could be opened and applied to the results. Then we'd know whether you and the other participants really can pick out a ... can we call it 'medium format look'? distinct from the large resolution advantage.LOL, but I already know the answer. Fly to San Antonio and you can blind test me on my 6K monitor. I will nail it every time. 100% hit rate if at full res if we can somehow get properly prepped raw files with the 60 MP Q3 with the fantastic 28 and the 100 II with a close focal equivalent on a tripod at equivalent settings and enough DOF to geta lots of the pixels in focus on a deep scene.
It will take me 5 seconds on each image on my 6K monitor to at 1:1 to pick the GFX file every time. 100% success rate.
But we need scenes with depth and lots of detail. I want a guy sipping coffee 400 meters away next to a car where I can see the license plate.
I'm disappointed though New Orleans. I'm going to have to add you to my list of 7 MF Board monitor tech deniers. But that list is now down to 4. And I have those 4 rank-ordered on how stubbornly radical they are on their dedication to belief in this false myth, and some of them don't shoot 100 MP. LOL.
But all kidding aside, I understand the math and of course realize that there is downsizing when I view my GFX 100 MP raw at full screen fit view on a 6K monitor.
But those GFX files look so good and sensor size and res matters, and yes, I can of course see it on my monitor.
But so can everyone else. I think even the monitor tech deniers are starting to come around.
Hey, we are all brothers and sisters in high-res photography, and that includes FF at 50 and 60 MP.
So few of us left.
I don't care if a few of you don't believe me when I say what I can see on my monitor with my GFX files.
We all know high-res FF and GFX / Hassy MF is pretty darn good.
Basically, I'm looking for a more tightly-controlled, comparably-take version of this excellent thread from four years ago:
https://www.dpreview.com/forums/post/63578554
With even more photos. To compare with GFX, I'm fine with Nikon Z, but would be equally fine with Sony E or Canon R. We could discuss what pairs of lenses to use, to try to take out of the equation that the GFX system is populated with excellent lenses and zooms that are low-ratio compared with what's typical in other sensor sizes.
And best would be two sets of comparison photos: one set at UHD monitor size (2880x2160 pixels) and one at some higher resolution that is still enough to mask the considerable resolution differences and mitigate the effects of downsampling settings and techniques, the idea being to compare at 100%--maybe something like 4400x3300 pixels. Anything higher-res than that probably gives too much advantage to the sheer pixel counts on the GFX100 sensors, which obviously are a real thing and a real issue.
In such a tests, if there are differences that are substantially visible to you, then you should be able to correctly say which is GFX and which is Z8 (or whatever) 95% of the time. It would be fascinating to see who (if anyone!) could do that.
If you really, really want to move forward, I don't mind drafting a protocol and seeing whether a few key players want in.
*Which I enjoyed, but that landing was one of the two times I thought I was actually at non-trivial risk of death by airplane crash.
Last edited:
You will have problems with selection bias. Anyone who is involved enough in this debate, to be willing to participate in such a test by selecting the necessary photos, will be biased in their selection.If you're even half-serious, then I'd love to try to have the forum members put together a set of matched pairs of images for people to compare / try to tell apart. Maybe Jim has the technical expertise, access to equipment (especially through LensRentals), integrity, and universal respect to head that up. The logistics would be daunting, but not impossible. And although it's been like a quarter-century since I've been to San Antonio,* there's actually no need: everything could be shared online. 'Here, take a look at these files, report to The Tabulator which is GFX and which is Z8.' Then when all the participants had had a chance to compare files and report what they think, the Encrypted Answer Key could be opened and applied to the results. Then we'd know whether you and the other participants really can pick out a ... can we call it 'medium format look'? distinct from the large resolution advantage.LOL, but I already know the answer. Fly to San Antonio and you can blind test me on my 6K monitor. I will nail it every time. 100% hit rate if at full res if we can somehow get properly prepped raw files with the 60 MP Q3 with the fantastic 28 and the 100 II with a close focal equivalent on a tripod at equivalent settings and enough DOF to geta lots of the pixels in focus on a deep scene.
It will take me 5 seconds on each image on my 6K monitor to at 1:1 to pick the GFX file every time. 100% success rate.
But we need scenes with depth and lots of detail. I want a guy sipping coffee 400 meters away next to a car where I can see the license plate.
I'm disappointed though New Orleans. I'm going to have to add you to my list of 7 MF Board monitor tech deniers. But that list is now down to 4. And I have those 4 rank-ordered on how stubbornly radical they are on their dedication to belief in this false myth, and some of them don't shoot 100 MP. LOL.
But all kidding aside, I understand the math and of course realize that there is downsizing when I view my GFX 100 MP raw at full screen fit view on a 6K monitor.
But those GFX files look so good and sensor size and res matters, and yes, I can of course see it on my monitor.
But so can everyone else. I think even the monitor tech deniers are starting to come around.
Hey, we are all brothers and sisters in high-res photography, and that includes FF at 50 and 60 MP.
So few of us left.
I don't care if a few of you don't believe me when I say what I can see on my monitor with my GFX files.
We all know high-res FF and GFX / Hassy MF is pretty darn good.
Basically, I'm looking for a more tightly-controlled, comparably-take version of this excellent thread from four years ago:
https://www.dpreview.com/forums/post/63578554
With even more photos. To compare with GFX, I'm fine with Nikon Z, but would be equally fine with Sony E or Canon R. We could discuss what pairs of lenses to use, to try to take out of the equation that the GFX system is populated with excellent lenses and zooms that are low-ratio compared with what's typical in other sensor sizes.
And best would be two sets of comparison photos: one set at UHD monitor size (2880x2160 pixels) and one at some higher resolution that is still enough to mask the considerable resolution differences and mitigate the effects of downsampling settings and techniques, the idea being to compare at 100%--maybe something like 4400x3300 pixels. Anything higher-res than that probably gives too much advantage to the sheer pixel counts on the GFX100 sensors, which obviously are a real thing and a real issue.
In such a tests, if there are differences that are substantially visible to you, then you should be able to correctly say which is GFX and which is Z8 (or whatever) 95% of the time. It would be fascinating to see who (if anyone!) could do that.
If you really, really want to move forward, I don't mind drafting a protocol and seeing whether a few key players want in.
*Which I enjoyed, but that landing was one of the two times I thought I was actually at non-trivial risk of death by airplane crash.
Someone who falls into the „there is a difference“ camp would select photos where he believes the difference can be seen. And the opposite for the „you can’t see a difference“ camp.
Even if these photos are taken for this specific purpose, there will be bias in the motive selection.
I think the shots provided by dp review do the job just fine. I see a number of areas in those shots, that show a clear difference between Z8 and GFX100. For me that’s enough, I don’t need further testing really.
NAwlins Contrarian
Forum Pro
If a skilled photographer creates pairs of carefully-matched photos (same subject, same tripod, etc.) using comparably-good lenses on e.g. a GFX 100 and a Nikon Z8, that should be very close. Of course you have to crop them to the same composition and aspect ratio, and scale them down somewhat to reduce the effect of the difference that everyone agrees is there, resolution at 100% viewing or in large prints. Then you strip our the EXIF data.
Then you administer the review double-blind: neither the person serving up the pairs of photos, nor the person evaluating them, knows which is which. That squeezes out the bias and tests whether you're reliably seeing a difference.
This is a standard way to test claims of whether two things are really / functionally the same, or whether one is really better than / preferable to another.
Ap0ll0n
Veteran Member
Agreed, no need for a complicated or expansive test. Those samples already contain multiple and diversified subjects in them. One just needs to measure how many of those subjects appear better for one camera vs. another, if the difference is tight, else just observe which camera resolves most subjects better and report. Or just observe each image overall as it looks on screen and let the subconscious decide which one stands out more.
Sure, there may be some bias from some folks but if the difference in IQ reported by the users is not large enough, and larger than any reasonably expected bias, then that is a finding right there.
Jack Hogan
Veteran Member
Hi Ap0ll0n,I was doing the same just nowBelow is what I am seeing; on the left is a screen capture of the studio's GFX 100 photo (when viewed uncropped) and right is the best one I could pick from your images. I think we can safely say that what I am seeing based on the studio image is a lot better than based on your downsampled image.
Late to the party but the left image above jumped out at me because it is a good demonstration of how a poor downsampling algorithm can mess up a perfectly good raw capture: for instance the texture on the wall is mostly false detail introduced by poor (or no) filtering during downsizing.
Aliasing allows higher frequencies to sneak in as non-existent lower ones. What a great example, I'll save it for future reference.
Jack
Last edited:
Greg7579
Forum Pro
Don't worry Boaster, that kind of test is never going to happen on this forum. It would be impossible to conduct on the internet and no one would get the scene or equivalencies right and it would be a pincushion for argument, accusations, abuse and controversy. I would never fall into that trap. It will never happen, and if it does, I won't play that game.You will have problems with selection bias. Anyone who is involved enough in this debate, to be willing to participate in such a test by selecting the necessary photos, will be biased in their selection.If you're even half-serious, then I'd love to try to have the forum members put together a set of matched pairs of images for people to compare / try to tell apart. Maybe Jim has the technical expertise, access to equipment (especially through LensRentals), integrity, and universal respect to head that up. The logistics would be daunting, but not impossible. And although it's been like a quarter-century since I've been to San Antonio,* there's actually no need: everything could be shared online. 'Here, take a look at these files, report to The Tabulator which is GFX and which is Z8.' Then when all the participants had had a chance to compare files and report what they think, the Encrypted Answer Key could be opened and applied to the results. Then we'd know whether you and the other participants really can pick out a ... can we call it 'medium format look'? distinct from the large resolution advantage.LOL, but I already know the answer. Fly to San Antonio and you can blind test me on my 6K monitor. I will nail it every time. 100% hit rate if at full res if we can somehow get properly prepped raw files with the 60 MP Q3 with the fantastic 28 and the 100 II with a close focal equivalent on a tripod at equivalent settings and enough DOF to geta lots of the pixels in focus on a deep scene.
It will take me 5 seconds on each image on my 6K monitor to at 1:1 to pick the GFX file every time. 100% success rate.
But we need scenes with depth and lots of detail. I want a guy sipping coffee 400 meters away next to a car where I can see the license plate.
I'm disappointed though New Orleans. I'm going to have to add you to my list of 7 MF Board monitor tech deniers. But that list is now down to 4. And I have those 4 rank-ordered on how stubbornly radical they are on their dedication to belief in this false myth, and some of them don't shoot 100 MP. LOL.
But all kidding aside, I understand the math and of course realize that there is downsizing when I view my GFX 100 MP raw at full screen fit view on a 6K monitor.
But those GFX files look so good and sensor size and res matters, and yes, I can of course see it on my monitor.
But so can everyone else. I think even the monitor tech deniers are starting to come around.
Hey, we are all brothers and sisters in high-res photography, and that includes FF at 50 and 60 MP.
So few of us left.
I don't care if a few of you don't believe me when I say what I can see on my monitor with my GFX files.
We all know high-res FF and GFX / Hassy MF is pretty darn good.
Basically, I'm looking for a more tightly-controlled, comparably-take version of this excellent thread from four years ago:
https://www.dpreview.com/forums/post/63578554
With even more photos. To compare with GFX, I'm fine with Nikon Z, but would be equally fine with Sony E or Canon R. We could discuss what pairs of lenses to use, to try to take out of the equation that the GFX system is populated with excellent lenses and zooms that are low-ratio compared with what's typical in other sensor sizes.
And best would be two sets of comparison photos: one set at UHD monitor size (2880x2160 pixels) and one at some higher resolution that is still enough to mask the considerable resolution differences and mitigate the effects of downsampling settings and techniques, the idea being to compare at 100%--maybe something like 4400x3300 pixels. Anything higher-res than that probably gives too much advantage to the sheer pixel counts on the GFX100 sensors, which obviously are a real thing and a real issue.
In such a tests, if there are differences that are substantially visible to you, then you should be able to correctly say which is GFX and which is Z8 (or whatever) 95% of the time. It would be fascinating to see who (if anyone!) could do that.
If you really, really want to move forward, I don't mind drafting a protocol and seeing whether a few key players want in.
*Which I enjoyed, but that landing was one of the two times I thought I was actually at non-trivial risk of death by airplane crash.
Someone who falls into the „there is a difference“ camp would select photos where he believes the difference can be seen. And the opposite for the „you can’t see a difference“ camp.
Even if these photos are taken for this specific purpose, there will be bias in the motive selection.
I think the shots provided by dp review do the job just fine. I see a number of areas in those shots, that show a clear difference between Z8 and GFX100. For me that’s enough, I don’t need further testing really.
The only guys that want to do something like that are the very few deniers (some of whom do not shoot GFX or Hassy or 100 MP) who actually believe you have to print big to see the difference between FF and MF, and that I can't see the difference on my monitors (not you Jim) and now some of these guys are taking it even further - MF and APSC are the same if you chop and crop enough.... LOL. I mean it just could not get any funnier.
Besides, I don't need it because I see it every day. I don't care if the 4 or 5 guys that actually believe what they say they believe continue not to believe what I say (which is the simple truth). They are good guys and entitled to their opinion. Four of them are guys I really like and have been conversing with for years here. We just disagree on this point, and it isa very important point on the Medium Format Forum. Hello..... Boys.... This is the Medium Format Forum.
Furthermore, I'm willing to stop harping on this and actually decided not to keep saying this, but then new threads pop up with cherry-picked pixel math, guys start challenging me to prove it, or get me to agree to silly amateurish impossible to conduct comparison "tests" and then try to convince new potential Hassy and GFX customers and forum members of their narrative (and that narrative is dead wrong)
No one believes that stuff.
BTW, I can't believe the number PMs, emails, texts and Flickr contacts I'm getting on this issue. I know you all agree with me, but I don't care to be honest. Say it here or just let it go. No biggie.
And James, this is not aimed at you. I'm not talking about you. So don't get mad at me or all defensive. You've explained yourself quite well the last 30 hours. I need you so don't want to get on your bad side. Who else around here is going to tell me what gear to buy? You are the only one I trust.
Greg Johnson, San Antonio, Texas
Greg Johnson’s albums | Flickr
www.flickr.com
Peace, Greg. That is a heavy responsibility. I remember one guy who'd made way too much money at Rolm admiring my 1980 928 (before Porsche uglied it up) and telling me that he thought it was great, but he really wanted a convertible. With a devilish grin, I told him that there was a company in Germany that would take the top off a 928 and sell it to him for prices starting at double the 928 price and going way up if you got carried away with the options list. He bought one, and drove it for many years.
A few years later, he asked me about stone massage tables, and, with the same evil grin, I suggested a malachite one. He didn't bite.
Jim
Ap0ll0n
Veteran Member
Hi Jack, which detail and are you talking about the photo on the left? I see a similar wall on both photos as shown above (still different from each other) even when zooming DPreview's RAF files at 1:1.Hi Ap0ll0n,I was doing the same just now
Late to the party but the left image above jumped out at me because it is a good demonstration of how a poor downsampling algorithm can mess up a perfectly good raw capture: for instance the texture on the wall is mostly false detail introduced by poor (or no) filtering during downsizing.
Aliasing allows higher frequencies to sneak in as non-existent lower ones. What a great example, I'll save it for future reference.
Jack
--
Apollon
Jack Hogan
Veteran Member
Who are you going to believe Ap0ll0n, me or your lying eyes? Sometimes false detail looks real. You have provided one of the best examples I've seen yet of how undersampled higher detail frequencies can sneak in through aliasing, producing non existent lower frequencies in the image. Here is the actual, well defined detail top right, captured in the raw file:Hi Jack, which detail and are you talking about the photo on the left? I see a similar wall on both photos as shown above (still different from each other) even when zooming DPreview's RAF files at 1:1.Hi Ap0ll0n,I was doing the same just now
Late to the party but the left image above jumped out at me because it is a good demonstration of how a poor downsampling algorithm can mess up a perfectly good raw capture: for instance the texture on the wall is mostly false detail introduced by poor (or no) filtering during downsizing.
Aliasing allows higher frequencies to sneak in as non-existent lower ones. What a great example, I'll save it for future reference.
Jack
PNG from raw, view at 100% by clicking on 'original size' below and compare to screen capture above left
Aliasing and false detail is instead everywhere in your screen capture, indicating poor or no filtering during downsizing. They do not call that card the aliasing nightmare portion of the Studio Scene for nothing
Jack
Last edited:
WGHalvorsen
Leading Member
To put legs on the snake, here are two images of the GFX 100 "aliasing nightmare" card downsized with antialiasing.Who are you going to believe Ap0ll0n, me or your lying eyes? Sometimes false detail looks real. You have provided one of the best examples I've seen yet of how undersampled higher detail frequencies can sneak in through aliasing, producing non existent lower frequencies in the image. Here is the actual, well defined detail top right, captured in the raw file:Hi Jack, which detail and are you talking about the photo on the left? I see a similar wall on both photos as shown above (still different from each other) even when zooming DPreview's RAF files at 1:1.Hi Ap0ll0n,I was doing the same just now
Late to the party but the left image above jumped out at me because it is a good demonstration of how a poor downsampling algorithm can mess up a perfectly good raw capture: for instance the texture on the wall is mostly false detail introduced by poor (or no) filtering during downsizing.
Aliasing allows higher frequencies to sneak in as non-existent lower ones. What a great example, I'll save it for future reference.
Jack
PNG from raw, view at 100% by clicking on 'original size' below and compare to screen capture above left
Aliasing and false detail is instead everywhere in your screen capture, indicating poor or no filtering during downsizing. They do not call that card the aliasing nightmare portion of the Studio Scene for nothing
Jack
/Bill
Image resized with a scale of 0.25 by bilinear method with antialiasing.
Image resized with a scale of 0.25 by bicubic method with antialiasing.
Last edited:
Similar threads
Keyboard shortcuts
- f
- Forum