1DsII vs Scanned 4x5 redux

[David R. Gurtcheff]

Hi Jack:
Glad to hear of your results with the MK II. I don't have a Mk II (yet), but I have some matted and framed 20"x30" prints from the 1DS, and they are sharper and have more detail than my scanned Pentax 645 prints. I use Fred Mirand's 1Ds sharpening action, and stair step interpolation action. I can believe what you are saying because the detail in my 1Ds prints in fantastic; I can only imagine your Mk II results! Thanks for sharing.....
Dave Gurtcheff
www.modernpictorials.com

[didger]

I realize a few regulars on this site may be offended by the implication that the 1DsII is generating clearly superior prints to the 1Ds from a detail standpoint, but the fact is I'm now getting results from my 1DsII that simply blow me away

I suppose I'm one of these "few regulars", though "offended" hardly applies, probably not to anyone.  Skeptical, maybe. Perceived print quality has so much to do with pre-print processing, that it muddies the issue a bit.  Maybe someone that's very keen to show the superiority of 4x5 (or even 1ds) could come up with a pre-print processing recipe that would prove this in a "blow ya away" fashion, at least for some subject matter, at least to themselves.  Since I probably won't have a chance to personally see a direct comparison of 1ds/1ds2 prints of the same size using exactly honestly comparable processing recipes and subjects, I'd still some day like to see some ordinary "pixel" peeping crops on my monitor for comparison; something like the Akiss images that showed no real difference between 1ds/1ds2 with the lens and aperture that he used for that type of image.  If no clearer difference than that can be demonstrated with perhaps other lenses and/or subject matter with simple small crops that can be evaluated on a monitor, I'll remain among the skeptical "few regulars"; not skeptical that 1ds2 is superior by some margin with some lenses at some apertures, but skeptical about how big the margin is and how often you'll see that margin in real life shooting situations where you may not always be able to use only world class lenses at optimal aperture.

As for shadow detail and noise;  I tried a test and posted pictures.  You can check my thread on the necessity of ETTR.  I see no problem with 1ds shadow detail or noise whatsoever for conditions that I would ever likely have to deal with.  True, very deep shadows and high ISO would be another matter, but not relevant to me.

Hey, Jack, do you realize what a humorous turn-arount this is?  I've got those "everybody knows they're the best" Zeiss lenses and you were an annoying buzz in my ear about that until it became apparent that the Zeisses you tested had to have been defective and not merely bad samples and also that your 16-35mm lens is better than anything my research indicated was possible from any Japanese ultrawide zoom manufacturer.  Well, now I'm the annoying buzz in your ear about that "everybody knows it's best" 1dsMKII.  I don't yet know at all that it would be best for my needs or for the money or for real life shooting situations with real life lenses and apertures, where resolution is concerned.  It's nice to see your enthusiasm, but, hey, people are generally enthused about their new toys.  I'd have to personally see some really amazing direct comparison results before I'd give up my plan to skip any more 35mm format DSLR upgrades rather than hold out for MF digital in a year or two; same weight, cheap high quality lenses, definitely WAY more resolution and no caveats about extreme requirements for absolute best lenses at optimal apertures.

[didger]

1Ds is probably way more camera than you need right now anyway...

Well, considering that my first marketing "target" idea is a coffee table book, 1ds is just about exactly right.  Considering that my ultimate "target" is to try to come as close to LF print quality as I can, even the best that MF DSLR's are ever likely to get will probably fall a little short.  But then, it's hard to avoid shortfalls in life in just about every area.  Even my beloved Sierra Nevada has way too many mosquitos, and more than ever and more at higher altitudes as global warming progresses.

[RobertJ]

Even my beloved Sierra Nevada has way too many mosquitos, and more than ever and more at higher altitudes as global warming progresses.

Totally OT... but the reason for the increase in mosquitos in your beloved Sierra-Nevada has nothing to do with global warming...  (In fact, mosquitos tend to reproduce just as well in mild climates as they do in the troics -- as long as the female gets her drop of blood.  Just ask anybody that lives in Alaska  )

The increase has more to do with the California endangered species act now restricting the use of pesticides that used to be used (and are required) for mosquito abatement.  A prime example is "Mosquito Lake" in Yosemite...  It got that name long before the original abatement programs went into effect -- back in the 20's when the "globe" was a "cooler" place.  

Cheers,
Jack

Very interesting...

Anyway, back on topic, these print tests are very much appreciated, Jack.  It looks like I may start shooting 4x5 anyway, because there's no way in #### I can afford the 1ds2...yet.

T-1000

[didger]

How can 22 Mpixels, no matter how pristine and pure and accurate each and every one of those pixels is give you the detail of a 4x5 negative?  Come on!!  Even if you scan at a resolution way way below where any grain will show, you'll have a whole lot more than 22 Mpixels' worth of high quality detail.  Maybe soon someone will come out with a $100,000 22 Mpixel sensor that will outdo 8x10.

Uh, actually, come to think of it, I think I'll start saving up for the ultimate extension of this technology, $1,000,000 8 Mpixel P&S that outdoes 8x10 film.

[Jack Flesher]

How can 22 Mpixels, no matter how pristine and pure and accurate each and every one of those pixels is give you the detail of a 4x5 negative?  Come on!!  Even if you scan at a resolution way way below where any grain will show, you'll have a whole lot more than 22 Mpixels' worth of high quality detail.  

Good points Didger...  

While I definitely agree, I think the issue comes down to how the images look when comparing the final prints, and here is where the waters get muddied...

From a purely technical POV, more detail does not necessarily mean more percieved sharpness in the final image.  Contrast plays a very large roll in our perception of image "sharpness," so at some point the advantages from one begin to outdo the other in that perception.  

(As an interesting aside, this has historically been a large difference in lens design philosophies of Japanese versus German lens designers.  The problem was, in lens design you can have ultimate contrast or ultmate resolution, but not both together -- they cancel each other out at some point in the design process.  In the end, the Germans tended to go with raw resolution while the Japanese tended to go for the higher contrast.  WHich is better is left to the eye of the beholder, but this may in part explain why Zeiss lenses on Canons high resolution cameras are so popular.)

Cheers,
Jack

[Jack Flesher]

Pardon me if I've missed something somewhere along the line, but doesn't this back have to be part of a system that requires a laptop and lots of other stuff in the field?  That at least has been to ongoing complaint about MF digital.  I was under the (perhaps false) impression that the Mamiya ZD was the first MF digital capture option that was at all "convenient".

Consider yourself pardoned -- The P25 is totally self-contained, uses readily available rechargeable camcorder batteries and writes to a CF card. FTR,  Kodak backs have been self-contained for several genreations.

I don't feel I can be sure if you're comparing sharpenining and other pre-print recipes or comparing sensors.  That's why I like to peep at raw 100% comparison crops on my monitor.

Peep away -- But for me, the proof is always in the final print...

Nothing in the world can do anything about missing data.

Agreed -- and I used to profess the same mantra daily.  HOWEVER we are now at the point where the differences between scanned 4x5 detail and direct digitally captured detail (assuming the best lenses are used in both systems) cannot be easily seen in a 20x print unless you look at both under a loupe...  Obviously, if you go to 48-inch print then the scan will produce the clearly superior image.  

So IMO the real issue comes down to your "normal" maximum print sizes...  

Cheers,
Jack

[Jack Flesher]

 He does really big really expensive really high quality prints and that's what his customers expect.

Then my advice is that if you have that inroad use it now... Bypass digital altogether, get yourself a 4x5 camera and a few lenses and start building your image library.

You can buy a great 4x5 for less than $1000, and plan on another $1000 for three good lenses.  Another $500 for meter, darkcloth, readyload holder, film and loupe, and you will be set for anything the landscape has to throw at you.

I can recommend my 4x5 camera; Made by Dick Phillips, it weighs 3.8 pounds, has enough bellows extension to infinity focus a 450mm non-tele lens, yet has enough flexibility to focus a 47mm hyper-wide without changing bellows.  It, 6 lenses and all of the other required paphenalia fits in a medium-sized backpack and weighs less than 20 pounds.

[RobertJ]

he's expecting a collection of images to review about next Spring and he knows they're 1ds.

Didger:
Are you going to be the one scanning film and printing?  If not, then shooting 4x5 isn't a bad idea.  If all you have to do is send a collection of transparencies for review, and let others do the scanning/printing/whatever they want, I don't see the need to spend the money on a 22megapixel DSLR or back.  For your personal stuff, the 1Ds is probably adequate for now.  Am I right?  If I'm wrong, then ignore this.

Anyway, if you do decide to go for the Mamiya, I'm not sure I would buy the Medium Format DSLR.  I'd probably go for the 645AFD and the ZD back instead.  Just seems like trading in old backs for new backs is better than trading in whole camera bodies for new camera bodies, if indeed, Mamiya comes out with a whole new body in the future, like 35mm digital does every 6-24 months.

My brain hurts.

T-1000

[didger]

I for SURE would not be doing either the scanning or printing.  The guy has a zillion bux worth of scanners and printers and I have zilch.  He's close enough I can go in any time to collaborate on scans and print pre-view, etc. Problem with 4x5 for me is not cost or weight, but losing the whole charm (for me) of the digital work flow.  Moreover, I need to have a lot of finished 1ds images for a portfolio before this thing ever becomes more than loose vaporware.  ZD decisions will depend on so many variables that at present are impossible to assess; mainly cost of various options and weight of the DSLR vs 645AFD body and back.  At the moment I can just about afford a few more months of shooting, sleeping in my truck all the time or backpacking and never eating out.  No urgency about making any sort of decision very soon, except the decision to go shoot some more and get some sort of portfolio together in the next few months.

"My brain hurts"  Maybe you need a frontal lobotomy.  Just make sure you have your brain surgeon and not your dentist do it.  You may also consider the wisdom of a friend who said "I'd rather have a bottle in front 'o me than a frontal lobotomy".

[Graham Welland]

All this theoretical discussion about resolution and image quality is fine but ultimately if you want to know which medium produces the "BEST" images then shoot a portfolio of identical images with your 1DsII, 1Ds, 4x5, MF digital etc, and produce a set of prints of the highest quality that you can manage and compare them side by side.

Pixel peeping is fine, but it is well known that what looks good at 100% in PS isn't necessarily what looks great on paper. Fact.

If you go through this test then you can make an informed decision about print quality regardless of media. Couple this with any convenience, portability and workflow issues and you'll be able to settle on the best set up for you. I don't think that much else matters.

[charles354]

All this theoretical discussion about resolution and image quality is fine but ultimately if you want to know which medium produces the "BEST" images then shoot a portfolio of identical images with your 1DsII, 1Ds, 4x5, MF digital etc, and produce a set of prints of the highest quality that you can manage and compare them side by side.

Pixel peeping is fine, but it is well known that what looks good at 100% in PS isn't necessarily what looks great on paper. Fact.

If you go through this test then you can make an informed decision about print quality regardless of media. Couple this with any convenience, portability and workflow issues and you'll be able to settle on the best set up for you. I don't think that much else matters.

I agree in the fact of what ever your final need is that is what should be tested. For me it is the print I sell my prints in Art Shows and Galleries. I still have prints that I am selling that I print from my  MF color negs in my color wet darkroom. Almost all my new  work from the last 18 months is digital from my 1DS. I print that in my Digital Darkroom ( epsons 7600 with atkinson profiles,4000 with imageprint,and a2200 for note cards, with 4 mac computers. My wet darkroom 4x5 omega and a 20" morpro st processor. I will also say for me I much prefer the digital darkroom over the wet darkroom. You might say I see the light now.
thanks charles 354. wow thats the most I have said since I have been on here.

[Jack Flesher]

A brief follow up to the post where I declared scanned 4x5 a clear winner in large, 48"x72" prints...  

I have had a bit more time to  "learn" the 1DsII files, and now with C1 and ACR available to convert, I re-did a head-to-head print comparison. This time I chose a more realistic 32 inch wide print size, since that is as large as I usually ever print.

For this size print, I printed the scanned 4x5 without interpolation at 240PPI.  For the 1DsII file I first converted the file with C1, applied a proprietary sharpening routine, and then had to interpolate this file up by 400% to get a comparable print size (at 300PPI).

I won't bore you with actual pixel views of the results, because they are not significantly different than before -- the scan has more detail, period.

HOWEVER, now that I have the raw conversion process for the 1DsII files better tuned, the final  digital 32x48 inch print looks sharper than the 32x40 scanned 4x5 print -- moreover, this is with my corrective reading glasses on and nose in the print!  Frankly, I was stunned by this fact and went back to make sure I had the files labeled properly...  I did.  Only when you put a loupe to the print, can you see more detail in the 4x5 verison, yet the 1DsII image definitely looks better -- being both sharper and cleaner -- to the naked eye, even up close.  

FWIW, I NEVER got anything this good out of the 1Ds...  So I am going to slide carefully out on a limb and declare the 1DsII an INFREAKINGCREDIBLE camera!  

(I realize a few regulars on this site may be offended by the implication that the 1DsII is generating clearly superior prints to the 1Ds from a detail standpoint, but the fact is I'm now getting results from my 1DsII that simply blow me away. I have a local friend who has a 1Ds, so I am planning a follow up report on a head-to-head, same image, same time of day, same lens, comparison of the 1Ds and 1DsII.  With a little luck, we may even have a P25 available at the same time to include in that comparison  )

Happy holidays to all!
Jack

[charles354]

Jack thanks for doing a follow up with the new raw converters. I just picked up my 1DSmkII today. At first I was not going to upgrade. But with all the  good things being said about it, especially in the shadow areas I had to do it. I have had the 1DS scince it came out and have been always happy with it. I also have the 1Dmkll that I have for a back up. So I will sell it and use the 1DS as backup.
Thanks charles354

[Jack Flesher]

Didger:  

Glad you are happy with you 1Ds and current lenses.  And to be sure, the 1Ds is still a great camera.  

I am experienced in gettnig optimum results form 1Ds files, and I am learning how to get more out of the 1DsII as we speak.  As for finding an exact processing recipe that optimally processes each file, and then standardizing that so others can replicate it, forgetaboutit!  LOLOLOL!  Aint gonna hapen-- at least it aint gonna come from me.  I've got barely enough time to do these comparisons for myself.  Readers can either take or shake my results, totally their perogative...

Frankly though, at least for you, since you mentioned in another thread that you don't even own a printer, the 1Ds is probably way more camera than you need right now anyway...

Cheers,
Jack

[Jack Flesher]

Even my beloved Sierra Nevada has way too many mosquitos, and more than ever and more at higher altitudes as global warming progresses.

Totally OT... but the reason for the increase in mosquitos in your beloved Sierra-Nevada has nothing to do with global warming...  (In fact, mosquitos tend to reproduce just as well in mild climates as they do in the troics -- as long as the female gets her drop of blood.  Just ask anybody that lives in Alaska  )

The increase has more to do with the California endangered species act now restricting the use of pesticides that used to be used (and are required) for mosquito abatement.  A prime example is "Mosquito Lake" in Yosemite...  It got that name long before the original abatement programs went into effect -- back in the 20's when the "globe" was a "cooler" place.  

Cheers,
Jack

[Jack Flesher]

T-1000:

Thanks for the kind comments!

And FWIW, I think you'll find shooting 4x5 a great experience -- it is one of the major steps I took in really learning photography

[Ray]

Interesting comparison, Jack! I guess we shouldn't be too hasty in drawing conclusions when doing such comparisons. It's also interesting that Michael in his field report of the P25 says the following:-

But, my honest opinion is that this back, coupled with the Contax's Zeiss lenses, produces images as detailed as those that I've ever seen from any photographic system that I've used in some 40 years as a photographer – and I include 4X5" film, either darkroom printed or drum scanned.

No mention of print size, though. Do you reckon Michael's talking about 48"x64" or 32"x48"?

[Jack Flesher]

Interesting comparison, Jack! I guess we shouldn't be too hasty in drawing conclusions when doing such comparisons. It's also interesting that Michael in his field report of the P25 says the following:-

But, my honest opinion is that this back, coupled with the Contax's Zeiss lenses, produces images as detailed as those that I've ever seen from any photographic system that I've used in some 40 years as a photographer – and I include 4X5" film, either darkroom printed or drum scanned.

No mention of print size, though. Do you reckon Michael's talking about 48"x64" or 32"x48"?

Hi Ray:

I don't want to put words in Michael's mouth, but I reckon Michael is referring to the maximum print sizes he regularly prints to -- which I believe he's stated somewhere as being on the order of 24x36.  Moreover when he states "produces images" I think he is referring to the final print at the above size, and not the pre-print image at actual pixel view on his computer screen.  But perhaps he will elaborate himself...

Lastly, a comment on the Contax 645 itself.  Unless you have used one, it is difficult to convey how crisp an image its Zeiss glass produces -- on film or digitally. It is equally difficult to describe the framing and composition advantages (compared to standard DSLR's) of the relatively HUGE and exceptionally bright viewfinder.  Even if the images it plus the P25 back produce are technically no more detailed than the 1DsII, the fact they are physically larger makes it a very alluring package to consider.  Finally, factor in the convenience of direct digital capture with that system, and it would be difficult to consider ever scanning film again...  At least assuming you can afford the system in the first place.

,
Jack

[didger]

factor in the convenience of direct digital capture with that system

Pardon me if I've missed something somewhere along the line, but doesn't this back have to be part of a system that requires a laptop and lots of other stuff in the field?  That at least has been to ongoing complaint about MF digital.  I was under the (perhaps false) impression that the Mamiya ZD was the first MF digital capture option that was at all "convenient".

I think the issue comes down to how the images look when comparing the final prints, and here is where the waters get muddied...

Very muddied indeed, whether you're comparing p25 with whatever, or whether you're comparing 1ds/1ds2.  I don't feel I can be sure if you're comparing sharpenining and other pre-print recipes or comparing sensors.  That's why I like to peep at raw 100% comparison crops on my monitor.

In the end, the Germans tended to go with raw resolution while the Japanese tended to go for the higher contrast.  WHich is better is left to the eye of the beholder

Ah, but Photoshop changed all that.  Nothing in the world can do anything about missing data.  You can do interpolation so that the missing data is not faked unpleasantly and you can do a bit of sharpening, but you can't manufacture detail out of nothing.  On the other hand increasing contrast in Photoshop is a piece of cake, so I think the Germans unwittingly went the better route for today's processing technology.