Pages: [1]   Go Down

Author Topic: The Proof is in the Printing Part III of A Slew of Similar Sony Sensors  (Read 2926 times)

jaypalmer

  • Newbie
  • *
  • Offline Offline
  • Posts: 2

I truly appreciate LL for considering the entire photographic process from imaging, through post, and finally print. This is applicable to all working photographers, regardless of the equipment they have.

I am not a professional photographer, just an amateur who enjoys the process and has done some paid work on occasion.

In my professional work, I'm aware of process. I think this article is informative, intriguing, and it's great to see some medium format camera systems appearing at an approachable price. And, particularly from Fuji, which brings an innovative twist on digital imaging.

Being aware of process, I have to consider that outlined in the article -- from sensor, to post, to conversion to printing, and the final print output, which is presented as the final document for comparing the 2 systems -- has some unanswered questions, if not weaknesses. Every step in that introduces changes in the original image taken, and the step from printer driver to printer, in particular, seems quite dependant upon the maker's support for the system at issue.

Then, let's backup to the beginning of the process. In each case, a zoom lens is chosen for comparison. So, it would seem that we are comparing the quality of each lens as much as we are the sensor.

I shoot on a humble Sony A7Rii. I shoot almost entirely vintage optics. I pixel peep -- and typically keep only lenses that resolve at or above that of the sensor. At 100%, the best lenses resolve down to the pixel. So, the 100% magnifications look to me more limitations of the lenses than of the sensor.

I recognize that we're talking a significant investment in lenses. But, for a fair comparison, I'd suggest, even though it would involve some effort or expense, using primes. In particular, on the Sony, how about a Voightlander 50mm f2 APO-Lanthar?

Although suggested that the 16 bit output doesn't appear to factor in an improvement, I would posit that this is equally as important as the increase in resolution in micro-contrast and micro-tonality.

In any case, keep up the good work!
« Last Edit: July 10, 2021, 05:09:34 pm by jaypalmer »
Logged

leuallen

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 427

Lenses are important no doubt. But they are not the only criterion. Things such as the capturing the peak moment, the emotion brought by the photo, sentimental associations with the image, and so on are equally important if not more so. I want a camera that allows me to capture those moments, if they are uber sharp great, if not I will still cherish the photos.
Logged

jaypalmer

  • Newbie
  • *
  • Offline Offline
  • Posts: 2

I couldn't agree more with your comments.

I understand that the article proposes that the superior resolution of the Fuji sensor is visible even in enlargements of 16x20 compared to FF and crop cameras.

My response is that the many levels of processing to create a print changes/degrades the resolution of the sensor -- particularly the lens used -- so that the method of comparing resolution based on the final print is a flawed process.
Logged

leuallen

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 427

I don't really care for comparisons.  A lens is what it is, you either like it or not. I use m4/3 and my two favorite lenses for music concerts are the 75mm 1.8  and the 10-25mm 1.7. The 75 is a great usable focal length for tight shots and head shots. At 1.8 it is plenty sharp.

I got the 10-25 used for a good price considering its list. Was ambivalent about it at first but I sure have come around. It has opened up new vistas in image composition and expanded the way I see. The zoom is a must as the quarters are often tight. It is plenty sharp at 1.7 which is where it lives. It is the only lens like it, a 1.7 zoom which is actually sharp enough to use.

I have other favorite lenses depending upon what I am shooting. Results count and if you are happy with them so much the better.

Bigger engine, faster, but do you need to go faster?
Logged

vjbelle

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 601

From all of my years of printing I was more than a little shocked to see how badly the Sony 7RM4 image compared to the Fuji 100s that you posted.  I own both of those cameras along with a Phase 4150.  Most of my recent printed images came from my first Fuji (50s) and held up extremely well even at 40 inches which is the small end of my normal print sizes.  I'm talking about extreme inspection by anyone. 

I decided to run a small test of my own.  I took two images - one with my 100s and 110mm lens and the other with my 7RM4 and a Leica 90mm Aspherical.  Both of those lenses are extremely sharp.  The subject matter was a pot of recently planted flowers which had a fair amount of depth.  I needed that so that there would be some overlapping areas of identical sharpness.  I shot the 110 at f8 and the 90mm at f5.6. 

I upsampled the Sony image in GAI to closely match the pixel dimensions of the Fuji.  The upsampling amount was just about 1.22% which produced a file size of just around 102MP.  Close enough for my experiment.  Both images were at 300ppi and were sharpend identically with Focus Magic ( amount of 1 with slower/finer precision ).  I cropped each image to 16 inches on the long side and printed both images on my Canon 8400 at 600ppi with the Canon print plugin. Both files were printing at near 12000 pixels on the long side - the Sony file having two upsamplings and the Fuji just one.  The paper used was Fine Art Pearl which I happened to have available. 

Visually upon the closest native inspection I could not see a difference between the two prints other then slight point of focus.  I then inspected the prints with a Schneider 6X6 3X loupe and although I could detect subtle, very subtle differences it was difficult and at times I would favor each one over the other.  I then inspected the prints with a 10X Peak loupe and even though I could again find some subtle differences the fact that they were so subtle and needed a 10X loupe to even see convinced me that the prints for all intents could be considered indistinguishable under native close inspection. 

Maybe my approach is different from yours but I just don't see the dramatic differences that you posted. Upsampling is upsampling - it can be done ahead of time or at the printer which in my case is the Canon plugin which works extremely well.  I have tested it numerous times against GAI doing all of the upsampling and if there is a difference I could not detect it.  GAI doing all of the upsampling (taking the image to 600ppi at the desired print size) was very slow and produced no better results than my normal workflow. 

I'm also fairly convinced that all upsampling could have been done in PS which has lately shown itself to be a worthy competitor to other methods and extremely fast. 

Just my 2c.

Regards,

Victor B.

Logged

John Hollenberg

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1121

For a professional level comparison of image quality of Sony a7R IV vs Fuji GFX100 see:

https://blog.kasson.com/gfx-100/a7riv-vs-gfx-100-for-landscapes/

https://blog.kasson.com/gfx-50s/fuji-50-3-5-on-gfx100-sigma-35-1-2-on-a7riv/

Jim Kasson uses excellent prime lenses for each system to compare as closely as possible the contribution of the camera/sensor to image quality.

Logged

kers

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3897
    • Pieter Kers

For a professional level comparison of image quality of Sony a7R IV vs Fuji GFX100 see:

https://blog.kasson.com/gfx-100/a7riv-vs-gfx-100-for-landscapes/

https://blog.kasson.com/gfx-50s/fuji-50-3-5-on-gfx100-sigma-35-1-2-on-a7riv/

Jim Kasson uses excellent prime lenses for each system to compare as closely as possible the contribution of the camera/sensor to image quality.

That Jim Kasson does well, but the title is :
Quote
the proof is in the printing....

To see everyting at 200% on screen is one way to see the tiniest differences; down on a print the results will be less different, or you have to print at many meters wide and put your nose at the wall.

Since the sensors are essentially the same (7IV and Fuji GFX100); the difference will be simply be the sensor size difference... when using good lenses on both camera's



Logged
Pieter Kers
www.beeld.nu/la

John Hollenberg

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1121

That Jim Kasson does well, but the title is :
To see everyting at 200% on screen is one way to see the tiniest differences; down on a print the results will be less different, or you have to print at many meters wide and put your nose at the wall.

Since the sensors are essentially the same (7IV and Fuji GFX100); the difference will be simply be the sensor size difference... when using good lenses on both camera's

The key difference is that Jim Kasson used outstanding prime lenses.  The test in the article is close to useless because the quality of the lens may be (and probably is) the determining factor in the differences seen rather than differences in the cameras.  Others who have done testing have generally concluded that seeing a difference at the size of print mentioned in the article is unlikely.
Logged

vjbelle

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 601

In my example the size of my prints was done only to save paper.  The resolution did not change.  I could have printed them at full size which would have been 38.82 inches in the long side at 300ppi.  The two prints still would have been indistinguishable at any viewing distance.  A 40 inch print would be no different since it's such a small upsample.  At 48 inches on the long side there may be some very small difference upon very close inspection. 

Victor B.
Logged

Benny Profane

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 281

I think the Adobe Super Resolution algorithm has changed the game big time. I am super impressed, and pretty happy, because, suddenly, My XT3 captures are greatly improved. I go no bigger than a 17 inch Epson P800, and really do not want to, because, well, enough is enough in space and material needed for such sizes. I am in this for the pleasure of it all, not to make money. I have worked with 24 and 44 inch Epsons in studios, and, no thanks for the home. If I ever want something bigger, well, I'll just make a proof and say, match it, to the printmaker.

Never worked with the large Fuji 100s or similar, but, I did do a weekly rental of a Sony Ar7iV early in the pandemic, and, well, very nice camera, beautiful files, but, the larger lens just made me appreciate my smaller Fuji lenses, a lot. I tend to walk five miles a day, sometimes more, on European jaunts and hike for miles in nature, and, the less weight, the better. The XT3 with the little 16-80 just fits in my hand so nicely, too. I almost dropped the Sony setup a few times. I couldn't imagine using the Fuji except for images that I didn't have to walk or hike very far to, so, at that expense, eh, no thanks, not yet. Still not versatile. If Adobe continues to improve it's AI development, and come up with a Super Duper Resolution, which, I bet they will, I can stay at the smaller ASPC body and lens size, and make prints that are, well, compared to the best film technology pre digital, superb. Then I can just concentrate on making beautiful prints, at an affordable price. Maybe the digital race is over, and this is as good as it has to be. But, ha, maybe just...a...little...better? Keep the body and lens size, though.
Logged
Pages: [1]   Go Up