Luminous Landscape Forum
Equipment & Techniques => Digital Cameras & Shooting Techniques => Topic started by: Jonathan Wienke on December 23, 2004, 01:36:21 pm
-
It's interesting reading people's opinions, theories, and experiences. It would be a lot more interesting (and convincing) to hold sample prints in one's hands and see the differences.
You can click here (http://visual-vacations.com/images/2004-07-31-0068L.jpg) to download and print a 1Ds images shot with the 70-200/2.8L IS zoom. Download it, print it, and draw your own conclusions.
-
I know of no where that I can download samples of the same scene shot under the same conditions using different cameras.
I'll try to post my 10D vs MF samples whwen I can. I'm afraid it won't be for a few weeks as I'll be away (photographing some of the time, I hope).
Eric
-
Hello Jose, I bought LivePicture rescently on ebay ($31), it vanished from the face of the earth in the late 90's just as I was about to pay several thousand dollars for it, and would dearly love to make contact with other users. I will be ordering Josh's book in the new year. Do you have details of the user list that you mentioned, and I would be very interested in any tips you could pass on. Seasons greetings and best regards.
Jason Berge.
-
This was about cost- the upshot tbeing that IF you already have a MF film camera, and IF you are on a limited budget, AND you want to retain excellent image quality for large prints, THEN it would be wiser to buy one of the newer scanners, scan the film and wait until 22MP camera prices fall into a reasonable range.
First off, let me compliment you on posting a well stated POV and clear question
As for cost, I think it really depends on how much you use the camera:
120 chrome emulsions run about what, $4.50 per roll? Last time I processed a roll of E6 120 it cost me about $7.00. Assuming you get 10 6x7 frames per roll of 120, then the per-shot cost comes out to be around $1.15 per frame. Assume you bought a new 1Ds for $7500 when they first came out two years ago, and assume you were going to sell it today, you'd get about $4000 for it used. At 1.15 per frame, youd only have needed to shoot about 3050 images total in those two years to "break even" on your $3500 loss on the 1Ds. Anything over that and you're ahead with digital -- and 3000 frames in two years is not a very difficult feat. Factor in the time spent scanning the images you want to print, and that number drops even lower. I don't think we need to factor in the cost of an MF scanner, since if you ever shot MF film, you will likely want to get a scanner to digitize those (plus all your other historical film) images anyway.
EDIT: I just did a quick comparison for myself on P25 MF capture to scanned 4x5 costs. Figuring a $35,000 investment for the P25 and camera system with a few lenses would depreciate to half that, or $17,500 in three to five years, I'd have to shoot only 3500 4x5 images to break even... Add in my time spent scanning the 4x5 keepers, and that drops further. Factor in the conveniences, and I can now even justify the obscene initial outlay
Anybody want to buy a great 4x5 system? :laugh: :laugh: :laugh:
-
I like Roger Hicks. I like a lot of what he writes about traditional 'wet' photography. However, his opinions on digital equipment do not seem to be based on any objective testing that I have ever seen, although he has recently become a Nikon D70 owner. Some might point out that he sports a large white beard and wears a monocle and many will draw their own conclusions from that.
-
MF lenses can be used, although you'll get a narrower FOV since the 1Ds sensor is smaller than MF film. But you're also using the sweet spot of the lens.
There's an interesting perspective here. A 90mm lens is a 90mm lens is a 90mm lens, whatever its image circle. From the perspective of the 35mm user, an MF 90mm lens will behave like any other 90mm lens designed for 35mm (apart from loss of automatic functions of course); same FoV, except the edges should be sharper.
-
I read with interest the review of the P25 back vs Ids MkII. Now, allow me to be a bit controversial and provocative here and then you can let me have it ...
I'm assuming here that in comparing apples and oranges, Michael is doing this from the perspective of a pro landscape photographer looking for the best tool. I can't imagine who else might be trying to make a decision between these two very different machines. Yes, I suppose that spending $30,000 on a back gets you better image quality than an $8,000 camera. Not very relevant to the 99.9% of folks who will never have to make this weighty decision. But interesting nonetheless.
Let's leave aside the obvious advantages of digital for a moment. Digital has many advantages but two of them are not resolution or dynamic range. Michael says that dynamic range is linked to noise. Sure it is. But I like to think more about how dynamic range is linked to resolution. The lower the resolving power of the chip, the less able to reproduce detail and the greater the tendency to lose detail in shadows and blow out highlights. This is where all digital cameras lose out (still) to MF film.
IMHO (and in that of many other photographers), a 22 MP back still does NOT have the resolution or dynamic range that one can still get from medium format film. From a cost/performance perspective, a $35,000 investment in a digital back, MF camera body and a good selection of lenses make no sense whatsoever unless you are a pro studio that requires a smooth, fast and efficient workflow and time is money. For the landscape photographer, it's a big stretch to see how this investment could be justifed no matter how many galleries your work appears in. A MF film camera will give you richer tonality, more resolution and require less time in photoshop. Not to mention at a fraction of the cost (assuming you do most of your work IN the camera, and not in photoshop).
Also, scanner cost/performance is improving so quickly that it makes more and more sense, FOR THOSE WHO NEED TO MAKE LARGE PRINTS WITH HIGH RESOLUTION, to keep their medium format cameras and scan the film in a good scanner. The newly introduced EPSON GTX-800 flatbed scanner is a good example. ( Most of you haven't seen this scanner yet- it just appeared in Japan) This scanner has a dMax of 4.0 at a cost of about $600. Even a $2,000 Coolscan 9000 (dMax 4.8) is still more cost effective. You will end up with more resolution, broader tonal range at a lower total cost per picture. And perhaps even save time too.
BTW, I own a measly Canon 20D and love it. I am seriously going to be looking at the Mamiya ZD MF camera too. But for now, for landscapes, I believe that digital is still inferior to film in these respects.
-
Thanks, Michael. In view of the extreme disappointments I've had looking at 100% crops of my 35mm slide scans, I'm very inclined to aim for MF digital rather than 4x5 film. I'm also not eager to do complex multi-layer Photoshop processing with 4x5 scan files. Does Photoshop CS even run on a Cray and isn't that even more expensive than a p25 sensor?
I want to be able to go up to 30x40, not just 20x24, but in view of the fact that Galen Rowell suffered no loss of credibility with some of his huge prints from 35mm Velvia, I reckon my customers and I can probably live with prints that size from 22 MP digital (4x 35mm film res and no grain), though it won't be p25. ZD will be hard enough to manage.
-
n every meaningful way, the 1Ds can match or beat 6x7 film, so a P25 with twice the pixels and better dynamic range isn't going to have any problem completely outclassing 6x7.
Well, quite some time ago I asked Michael if he felt that 1ds had reached MF level in every way and he said clearly no as far as absolute resolution for very large prints, but yes and then some every other way. Now, as far as p25 vs MF film, that's about like 1ds vs 35mm film; no brainer.
For me, however, it's Mamiya ZD vs 4x5 and I assume that for very large prints (30x40) with very good shooting, scanning, and production (and a lot of patience if you're not using a Cray) the 4x5 would ace out any 22 MP sensor. However, real life is something else. 4x5 would have to be consistently phenomenally better and that doesn't seem likely. In any case 50% of what I'm turning into being as a photographer happens after the exposure is made. I don't see routine complex multi-layer Photoshopping of those huge 4x5 scans as a likely scenario until the terahertz dual G10's gets a little cheaper.
Unless I see some truly persuasive 4x5 arguments (and more importantly, real life proof) I'll bite the bullet and take some time out to make the money for ZD, once it's clear that it fulfills expectations. It's hard to imagine that it won't and the issue of quality, price, and light weight of lenses is already quite clear.
-
I found my LP disks and manuals. I have 2.5. Was that the last release?
I'm wondering what I'd be missing with LP that Photoshop CS has. 10 years of Photoshop development surely must have ended up being good for something. I find the Photoshop selection tools particularly powerful and also very critical for my approach to image manipulation; magnetic lassoo, magic wand, add or subtract mode for all selection tools are particularly important. Quick mask and all the powerful easy painting and gradient fill selection manipulations in quick mask are also essential to me. The adjustment tools I use most often are curves, levels, hue/saturation, color balance. Then there's the matter of plug ins like Color Mechanic. Can you truly dispense with Photoshop altogether and retain the huge functionality or do you have to accept substantial compromises for the sake of the speed of LP?
-
Didger, in the drum sacanning days I used PS for color management ( assign, convert profiles ) , spotting just because spotting dust is faster in PS, now spotting is almost non existent so I use very seldom PS, but I know it quite well, if that can be said of PS!!
As I told you one can not compare LP and PS as far as image handling is concerned because the LP way is totally different, just an example the selection as you know it in PS just dont exist in LP, however LP is so much powerfull getting the same goals but in much more elegant and effective way. For the description of your needs one can easily tell that you´ll feel totally at home once mastered LP. If you have still the original user guide and tutorial start with it, but by all means get the Karson book it is the eye opener on the power of LP. Develop a dayly routine of working with LP just for one hour, and be constant, you can easily get caugth in the trap of thinking that PS is easier, it is not, it is just that you already know it.
If I had to describe LP I´d say it is the ultimate tool to bring photography closer to painting, in the sense of being able to fine tune tone and color to the most sutile levels ( in the LP CD there is a pdf by Joe Holmes about color correction ).
LP was from scratch conceived as a photographer´s tool and it shows.
The best way for you to get convinced is just see it working in the hands of a experienced operator
In my opinion LP is the best imaging tool that there is, period. and it is a shame for all the owners it have had since Live Picture Inc. disapeared that they have not even attempted to carbonize the app to run native on OSX. though it does pretty well in Classic.
Jason. this is address of the Lp list owner
http://www.calverley.co.uk/LP-GROUP/default.html (http://www.calverley.co.uk/LP-GROUP/default.html)
once you get in the list, you can access the list archives
The last version is 2.6.2 but in my opinion the best is 2.5.1 it is the one written by the original programmers, for 2.6 another team took charge and did not did it well, the fiasco was repaired trugh 2.6.1 and 2.6.2, but for me it s 2.5.1
Regards
Jose
-
I really don't know what that guy's talking about. It's pretty clear now that even the 1Ds Mark II can flat-out beat 645, and probably match 6x7: http://photo.nemergut.com/equipment/canon1ds/markii.html (http://photo.nemergut.com/equipment/canon1ds/markii.html) .
So he's wondering if a 22 megapixel medium format device can match 645 film? Gee, I wonder...
T-1000
-
You're overlooking the obvious point that many people who would use MF lenses on a 1Ds formerly used them attached to a MF camera, and they aren't going the get the same FOV on the 1Ds as they did with that lens on a Contax 6x7.
-
All I can add in response to this is that theory and practice are not the same thing.
Countless photographers have made the comparison between scanned film and digital (even conservative Popular Photo magazine), and come to the conclusion that scanning grain at higher and high resolution does not produce higher resolution images.
The reality is that since digital sensors have inherently higher resolution than film grain particles, along with essentially zero noise at normal apertures, that battle was over several years ago.
All it takes is the desire to make the comparisons for onself and a little work. Making judgements based on "theory" isn't appropriate, since so many people have already done the empirical tests for themselves. It's a bit like saying that bumble bees are too heavy to fly. Tell that to the bee.
As a rule of thumb, 6MP cameras (I'm talking DSLRs, not digicams) outperform 35mm film, 11MP+ cameras outperform medium format film, and 22MP MF backs outperform 4X5" film when it comes to overall image quality, including resolution. One can quibble over the details, (drum scans vs flatbed), film speed, DR, etc, but enough working pros and studios, as well as fine art photographers (including many on this forum) have done the comparisons and drawn their own conclusions from them.
There's little point in taking a position based on opinion rather than first-hand emperical evidence.
Cost is another story, though as we've seen with the drop in DSLR prices – now to under $1,000 for a high quality 6MP body, this is a constantly changing target.
Michael
-
It's interesting reading people's opinions, theories, and experiences. It would be a lot more interesting (and convincing) to hold sample prints in one's hands and see the differences.
It would be great to be able to download some sample crops from full-frame digital, MF digital and MF film and print them out.
I'm never going to spring for a $5,000+ camera but I would spend a few dollars to experience the quality first hand.
-
Michael did exactly such a comparison between the 1Ds and 6x7 film and posted the results. Search the site for "ultimate shootout" and see the side-by-side comparisons for yourself. (That article is one of the reasons I decided to get the 1Ds, BTW) I don't have MF film gear and have no interest in buying some just to redo a comparison that's already been done to death.
In every meaningful way, the 1Ds can match or beat 6x7 film, so a P25 with twice the pixels and better dynamic range isn't going to have any problem completely outclassing 6x7.
-
Digital capture is so much more convenient.
For me it's becoming far more than just convenience. I'm discovering that the things you can do with Photoshop can change an image in so many "artistically" versatile and interesting ways, that convenient "Photoshoppability" is as important as any other variable (cost, resolution, portability) in juggling things to decide on a format. Shooting and scanning 4x5 is not only very expensive, but those large file sizes (if you want to realize that huge resolution) are also very cumbersome to process if you routinely do very complex multi-layer manipulations in Photoshop. If I switch from 1ds to MF digital, processing times will double and I'll need to upgrade from my G4 to a G5. What do I upgrade to if I go to 4x5 scans, a Cray?
-
"Faith-based decision making" Too funny! Scary, but still funny. "Faith: unquestioning belief"
Disclaimer: I just unloaded all my MF (6X7) gear.
As Michael says, "The proof is in the print". I sit and look at 13X19 prints from 25 years of MF and recent stuff from my D70 and the difference is plain. The D70 stuff is better. The images are cleaner, smoother and just as detailed.
There's also a lot more of 'em. At a over a buck a frame *just to purchase the film*, for me shooting MF film is OVER.
Man! are we lucky!
Peter
-
At a over a buck a frame *just to purchase the film*, for me shooting MF film is OVER.
Yeah, and for me MF and LF film ain't starting. If D70 13x19 looks as good as folks are saying, then how good will Mamiya ZD look printed a whole lot bigger than that?
Maybe Michael ain't such a dummy or lying about MF digital.
"Faith: unquestioning belief"
That's all there is. We trust our senses, though they're great deceivers. We trust the engineering of airplanes and bridges without any understanding it, etc.
-
I´d like to add my two cents to this thread. I´ve been a pro for more 35 years, and through all these years I´collected quite a collection of the best gear that one can dream, from Leica Ms outfits, Hasselblads, linhof, Sinar p2 etc. all of them with the best glass. when the transition to digital postprocessing started I bougth a drum scanner to extract the last drop of quality from my chromes ( 10.000 ppi and 4.6 Dmax ) . From the moment that I got a D30 a few years back I knew that the status of photography as we knew it was to change very fast. I found out quickly that a D30 could match a 6x7 drum scanned( I must add that I´m a very competent scan operator) and printed to to 8x12 inches. As soon as the D60 came out and I got one plus the three TSE lenses, since then my drum scanner is iddle, in fact I have it connected to a old laptop and installed in the armoured room where I keep all my cameras, rigth now it is more than a year that I have not used it. With the D60 and the TSE lenses I get flat stitched files of more than 16 Mp and even more when I attach it to my Sinar P2 and using my Hassel lenses. I get quickly shot tiles files of 9 images that give me more than 30 Mp files with zeiss quality. I´m getting a 1DSMk2 next month to simplify things a bit or if using the same stitching methods get much more bigger files than now.
In my opinion the digital images are vastly superior to film , specially if you are like me a grain avoider. for me there is no way back. specially since last summer that under a retro attack I took my Master Technika and three lenses to a outing. to make short it, never more.
Didger. I understand that you are a Mac user, there is a simple and beautiful solution for the problem of handling big files with personal computers. it is called Live Picture a software that was discontinued a few years back, but still runs pretty well on OSX Classic, it was very advanced in its time, all the wonderfull tricks that you have learnt lately with layers and layers mask were routine for us LP users when Photoshop 3 even did not have layers, and in real time with any size image , and I mean any size, with just 64 Mb of ram,. When it came out it was over 7000$ and now you can have it sometimes in ebay for as low as 25 $. It is a bit different working philosophy, but if you get to master it, you will not want anything else. I´ve been using it the last 11 years and plan to do it as long as I last :-) it is a very nice way of skipping the need for upgrading computers
Well I wish you all the best of ligth for the next year
Jose
-
I do not understand why we need to debate this point. I have looked at the post that started this thread and all I can see are assumptions based on scanner resolution and Dmax. No other facts or valid comparisons. Resolution for example is not just about numbers of megabytes. My only interest in this discussion is, can you take a better picture more easily with a digital camera most of the time than you can with a film one? In my experience, the answer would have to be yes and this is what most people seem to have discovered in the last two or three years.
The reason for starting this discussion was not to start another silly film vs digital debate. That is much too broad and simplistic but unfortunately, this is how some people choose to perceive it. I am committed to digital (Canon 20D) and certainly am familiar with its advantages and disadvantages.
Again, my point was that, at this point in time and FOR LARGE PRINTS (ie: bigger than A3) does it make sense to spend $8,000 or even $30,000 to assure yourself that you are getting superior image quality?
As a photographer, I can evaluate this subjectively without the need to quote technical details. The "proof" is in the perception, not the "facts", which as you know can support any contention you care to imagine.
For thoughtful people who are open to different opinions, we have debates. In dictatorships, only one view counts. Which do you prefer? It is not beyond question that an 16 MP camera is better in every respect than MF film. No matter how often Mr. Didger offers his entusiastic support for Michaels' views, I still think there is a debate here, as evidenced by the opinions of other very experienced photographers in this same forum.
This year, I won an international photo competition in Japan with a 6x7 image that bested many 12MP and up digital photos. Does this prove 6x7 is better? No, of course not. What it does show however, is that making a good picture is determined much more in composition and technique, and finally by perception. It's not in the number of megapixels. (and yes, I agree digital workflow is superior) Which brings me to the final point I was trying to make.
This was about cost- the upshot tbeing that IF you already have a MF film camera, and IF you are on a limited budget, AND you want to retain excellent image quality for large prints, THEN it would be wiser to buy one of the newer scanners, scan the film and wait until 22MP camera prices fall into a reasonable range.
I guess that means I respectfully disagree with Michael that an 11 MP image is "better" than 6x7. All I can say is I there is a legitimate time and ocassion to still use this format. Viewing the images from both formats side by side, I made my own subjective evaluation.
-
I decided I might as well offer some images as "proof" to back up my claim above.
The 1Ds2 image was taken with a 35mm f1.4 L lens at f8, a very sharp lens. The 4x5 image was taken with a 90mm Rodenstock lens, which is NOT the sharpest 4x5 lens in my bag. I used the 90 since I wanted to equal the 35mm FOV on 6x9cm format film. The film was scanned at a fairly low 1800PPI on my Imacon scanner, to further make my point.
Here is the full frame shot showing the 50x50 pixel (high contrast) crop area from both images. Note that with my naked eyes from where I am standing, I can just barely read the street sign -- last time I had my eyes checked, I had slightly better than 20-20 distance vision:
(http://jack.cameraphile.org/albums/album06/full_image_001.sized.jpg)
Here is the 50x50 crop from the 1Ds2 file at actual pixel view, after careful conversion in C1. With proper sharpening you can almost, but not quite make out the street name on the sign:
(http://jack.cameraphile.org/albums/album06/50x50_pixels_1Ds2.jpg)
Here, in the effective 6x9 film scan, you can CLEARLY make out the street name:
(http://jack.cameraphile.org/albums/album06/50x50_scanned_4x5.jpg)
My only point is that scanned MF film can generate more detail than 16MP digital capture -- I am NOT claiming one is better than the other. In fact, 32" prints of this image from the 1Ds2 file look better to me from an overal image quality standpoint than the film print does. Even when closely inspected at critical viewing distances -- meaning my nose in the print! I cannot actually see better detail in the scanned film print until put a loupe on it. And FWIW, my Epson printer is good enough to replicate those details from the files -- under the loupe I cannot read the street sign in the digital print, but I can read it in the scanned film print.
(Please note that I do not consider this an artistic image. While the building has the potential for that, I use this shooting position as it provides a variety of qualities to look at for comparison purposes.)
Cheers,
Jack
-
Some of the world-weary "digital fundamentalists" on this forum wonder why they have to keep evangelizing us "film pagans" about the coming of the digital messiah. I offer you a comment by Roger Hicks which is pretty typical of the views of most professionals and industry watchers on film vs digital
In the January Shutterbug magazine review of digital and medium format developments at Photokina notes:
" In digital “medium format,” too, there were two surprises. One was a new Mamiya ZD (www.mamiya.com), a big brute of a camera but not a lot larger than the current generation of vast 35mm and multi-megapixel SLRs. It offers 22 megapixels. This genuinely does compete with or even surpass 35mm, but it is (to be generous) disputable whether 22 megapixels begins to approach 645 film in potential resolution, let alone 6x7cm or bigger." See: http://shutterbug.com/show_reports/0105sb_cameras/ (http://shutterbug.com/show_reports/0105sb_cameras/)
-
The Canon mount is such that just about any lens can be attached to an EF mount camera with appropriate adapter. MF lenses can be used, although you'll get a narrower FOV since the 1Ds sensor is smaller than MF film. But you're also using the sweet spot of the lens.
-
No, I'm not overlooking that. I'm assuming that professional photographers are familiar with the range of focal lengths for 35mm cameras, which have been around for a long, long time and that the same confusion that has applied to the smaller format DSLRs that use 35mm lenses is less likely apply to 35mm cameras attached to MF lenses, although the cropping principle is exacly the same.
-
Michael, thanks for that combination of common sense and credible personal experience report. I'm surprised that it's still ever necessary to convince anyone that digital is a suitable medium for landscape photography.
I have, however, been wondering about comparing 4x5 film with 22 MP MF, even very very good 22 MP. With all your experience now, would you say that the MF back can absolutely rival 4x5 film in terms of the sharpness you'd see in a 30x40" print? Does a really accomplished 4x5 shooter like David Muench really have no advantage at all over a 22 MP digital photographer of comparable skill for very large prints?
I'm seriously considering Mamiya ZD, but I was figuring that for extremely large prints even the best 22 MP outfit with the best lenses and best technique all around would still visibly fall short of 4x5 film for very large prints. Not so, you think?
-
I'd say it's a push as to whether a properly produced 22MP file from an MF camera would be slightly better or worse than a 4X5" drum scan.
The 4X5" scan at 4000ppi could be printed larger, of course, but then ressed up so could the digital file. It's a matter of ressing up vs enlarging grain.
It's a quibble really, but at print sizes up to 20X24" there isn't much to choose between them, except that the 4X5" is harder to produce both pre and post exposure, and the 22MP costs more in terms of gear.
Michael
-
You can click here to download and print a 1Ds images shot with the 70-200/2.8L IS zoom. Download it, print it, and draw your own conclusions.
You missed my point. Eric (see earlier post) has at his disposal MF film and a 10D. He made comparable shots and printed them out.
I can go here and there and collect various files and print them. But I know of no where that I can download samples of the same scene shot under the same conditions using different cameras.
As the old saying goes... "Seeing is believing.".
If such samples were on line the Doubting Thomases could judge for themselves.
-
I think that once you start looking at 4x5 your motives are not just in terms of image resolution/quality but should also be based on the flexibility of the platform (i.e. movements) to capture the absolute 'best' image of the scene. You could argue that perhaps something like the Hartblei T/S lens for MF can achieve this, although only at a limited focal length. I hope to get one in the new year some time to compare - maybe that'll be another reason to consign the LF to the cupboard again (although I actually enjoy the 'process' of LF capture on occasion).
In terms of workflow, no contest. Digital capture is so much more convenient.
-
The 35 mm film/6-8 meg dSLR, MF film/1Ds type questions frequently occur all over the web (at least on photography forums). People respond with numbers or with what they've observed from their tests.
That leads to faith-based decision making. Is Michael a god, or is Michael a liar? I've seen him portrayed as both depending on the individual's need to support their personal beliefs. I'm thinking that if it were easy to hold the evidence in one's hand....
The Doubting Thomases who most interest me are the long term MF film shooters who are just encountering the idea of digital perhaps being an equally valid way of producing a print. They've often been put off by the limited abilities of lesser digital cameras.
These people most certainly know how to evaluate a print. Put something in their hands and let them decide for themselves.
There are a few isolated places where one can download and print some examples.
I've been thinking for a while that it would be nice if there were a repository of comparison shots. Someone like Eric could post his test 10D/MF film shots, Michael some 22 meg/MF film shots, .... (Especially I'd like to see a 'reasonable' scanner/drum scanner 35 mm film/20D/1Ds study.)
When the 'Which is better?" type questions arise people could be pointed to the site. They could print them out at their convenience.
Perhaps the samples should be coded (Tree1, Tree2) and the codes tied to camera on a separate page. (I'd really like to print out a MF film and 22 meg MF back and hang them on the wall for a few days without knowing which is which.)
-
I´d like to add my two cents to this thread. I´ve been a pro for more 35 years, and through all these years I´collected quite a collection of the best gear that one can dream, from Leica Ms outfits, Hasselblads, linhof, Sinar p2 etc. all of them with the best glass. when the transition to digital postprocessing started I bougth a drum scanner to extract the last drop of quality from my chromes ( 10.000 ppi and 4.6 Dmax ) . From the moment that I got a D30 a few years back I knew that the status of photography as we knew it was to change very fast. I found out quickly that a D30 could match a 6x7 drum scanned( I must add that I´m a very competent scan operator) and printed to to 8x12 inches. As soon as the D60 came out and I got one plus the three TSE lenses, since then my drum scanner is iddle, in fact I have it connected to a old laptop and installed in the armoured room where I keep all my cameras, rigth now it is more than a year that I have not used it. With the D60 and the TSE lenses I get flat stitched files of more than 16 Mp and even more when I attach it to my Sinar P2 and using my Hassel lenses. I get quickly shot tiles files of 9 images that give me more than 30 Mp files with zeiss quality. I´m getting a 1DSMk2 next month to simplify things a bit or if using the same stitching methods get much more bigger files than now.
In my opinion the digital images are vastly superior to film , specially if you are like me a grain avoider. for me there is no way back. specially since last summer that under a retro attack I took my Master Technika and three lenses to a outing. to make short it, never more.
Didger. I understand that you are a Mac user, there is a simple and beautiful solution for the problem of handling big files with personal computers. it is called Live Picture a software that was discontinued a few years back, but still runs pretty well on OSX Classic, it was very advanced in its time, all the wonderfull tricks that you have learnt lately with layers and layers mask were routine for us LP users when Photoshop 3 even did not have layers, and in real time with any size image , and I mean any size, with just 64 Mb of ram,. When it came out it was over 7000$ and now you can have it sometimes in ebay for as low as 25 $. It is a bit different working philosophy, but if you get to master it, you will not want anything else. I´ve been using it the last 11 years and plan to do it as long as I last :-) it is a very nice way of skipping the need for upgrading computers
Well I wish you all the best of ligth for the next year
Jose
-
Didger, if you decide to go Live Picture route, contact me to tell you how configurate the OS9 Classic for a stable running of LP
Thanks. At the moment I'm running a day late getting my stuff together for a big cold solo Sierra shooting/camping trip. Then a bit of business stuff in January and some more cold shooting with a friend., then February in India, and so it goes. Nepal some time. Patagonia some time. Utah again some time. Too many fun and interesting software options (ray tracing and really far out image processing not even started yet), too many interesting places to go, too little time, too little money. Familiar scenario to anybody else?
-
This year, I won an international photo competition in Japan with a 6x7 image that bested many 12MP and up digital photos.
I guess that means I respectfully disagree with Michael that an 11 MP image is "better" than 6x7.
I think most of us know that a good image requires good technique, but all of these "1ds2 vs. blahblah" threads are questioning the technical aspects of one main thing: A photographic PRINT. Whether the photograph is artistically acceptable or not, most of us are talking about actual detail that we see in prints. So far, from Michael's tests, along with some other people, it appears that digital looks better than film in prints - up to a certain printable size. (13x19, and in some cases, larger)
As for 6x7 being better than 11 megapixels? Well, I guess a drum-scanned 6x7 will result in a file that allows you to make a larger print. However, file size and dimensions do not necessarily have anything to do with "Resolution."
T-1000
-
I guess that means I respectfully disagree with Michael that an 11 MP image is "better" than 6x7. All I can say is I there is a legitimate time and ocassion to still use this format. Viewing the images from both formats side by side, I made my own subjective evaluation.
Me too... I've said from day one that the 1Ds was a tad better than scanned 645 and a tad behind scanned 6x7, though all are very close.
Also, Michael stated earlier in this thread:
"As a rule of thumb, 6MP cameras (I'm talking DSLRs, not digicams) outperform 35mm film, 11MP+ cameras outperform medium format film, and 22MP MF backs outperform 4X5" film when it comes to overall image quality, including resolution."
I would have agreed with him except for the last two words, "including resolution." The problem is no comparison of digital to film resolutions is complete without also stating the contrast ratio of the detail areas at the same time. Film's resolution limits increase with contrast and I have too many examples where my Imacon-scanned 4x5 Astia outresolves my 1DsII in these areas to be convinced otherwise. Film simply has more resolution capability than a 16MP FF or 22MP MF digital sensor in certain situations, period.
HOWEVER, I do believe that the "normal" 16MP FF or 22MP MF file is capable of producing images that regularly appear more detailed than the "normal" scanned 4x5 file will, up to say 32x40/48 in print size. Mind you, it is not always in fact more detailed -- putting a loupe to the print will confirm this -- but the larger spatial frequency areas do appear crisper and thus give the entire print the illusion of better detail. And customers vote with their checkbooks... (I'm sure Michael and I would agree on this point :: )
Lastly, as a practical matter, 32x48 is as large as I'm ever likely to print (at least for photo-quality) since I can only frame them out to 40x60 maximum. (I might go larger on canvas, but I don't really think canvas cuts it for a true fine-art print... But that is a debate for another thread :: ) Hence, my use for film -- more accurately 4x5 film -- will be restricted to special projects where its assets outweigh its inconvenience and per-shot cost.
Cheers,
Jack
-
FWIW, Steve Hoffman has some comparisons on his web site with MF, 4x5 and a 1dsM2.
http://www.sphoto.com/techinfo/dslrvsfilm.htm (http://www.sphoto.com/techinfo/dslrvsfilm.htm)
In addition, there was a decent article in Nature Photographer with comparisons between 4x5, a P25, and a 1dsM2, with the 4x5 narrowly out-resolving the P25 which out resolved the Canon. But they did not post photo comparisons in that article and the reader is just left to take the writer's word for it.
Don Allen
www.donallen.net
-
And all of the auto focus/exposure stuff that you are paying for won't work (?).
I think if I were going to spend $8K for a 1ds plus adapter for lenses that wouldn't work quite right, I might be inclined to wait for a back that would let the MF camera/lenses work as designed. Just an unprofessional opinion though.
-
This does seem to be supported by PhotoDO's measurements of MTF for 35mm and MF lenses, where the 35mm lenses generally do better. Note that they test MF and 35mm lenses at the same 10, 20 and 40 lp/mm, and at the same range of distances from center, 0, 3, 6, 9, 12, 15, 18, and 21 mm. Thus their results correspond very directly to an asessment of performance when used with 35mm format and its 21mm image circle radius.
There are a couple of issues here that are not clear.
(1) Photodo applies a greater weighting to centre performance than to edge performance when arriving at their over all rating for a lens. Therefore, the superior performance expected of MF lenses at the 21mm mark may not be fully reflected in the rating figures.
(2) Because MF lenses are generally used at smaller apertures than 35mm for DoF purposes, some of the lenses tested by Photodo might have been designed for optimal performance at f11 or even f16 instead of f8 which is the smallest aperture at which Photodo does its tests.
I can think of at least one lens where this is the case; the Russian made Hartblei 45mm MF tilt & shift lens. The figures on it's website indicate this lens is equally sharp at f16 and f22 (at least very close) and noticeably sharper at f22 than at f8.
Another example of a lens being sharper at apertures that Photodo doesn't use in its test, is the Canon 100-400 IS. This lens appears to be best at f11, according to a comaprison done by Michael with the 400/5.6 prime.
Once again, unless manufacturers provide comprehensive, individual MTF charts with their lenses, we can only speculate on such matters.
-
I am familiar with the attachment of a lens designed for MF to a 35mm camera, and the "sweet spot" arguement.
Frequently, a good MF lens will not be as "sharp" as a good 35mm, even in the sweet spot. Yes, the MF lens might produce the same sharpness to the corner of a 35mm frame, but the sweetest spot on the 35mm might be better. Would a 35mm format lens on a less than full frame 35mm camera be better than a MF format lens on the same body (generally speaking of course)?
Can such a practice lead to internal flair and "fogging" since the lens with the larger image circle will create more "stray" light than a 35mm lens with a smaller image circle? Are 35mm bodies designed to handle the stray light?
-
After reading all of the discussions of digital vs. film on the LL forum some months ago I decided to do my own modest comparisons, just as Michael has suggested. Just a few years ago I had gotten rid of my 4x5 (and much earlier my 8x10) because I was getting too old to want to lug it around, and I settled on a Pentax 67II for landscape work. Last January I got my first digital, a Canon 10D. The rest of my equipment is at the budget end of LL readers: Epson 2200 printer and Epson 2450 scanner (neither the 4870 nor the 3200 had come out yet when I bought it). Lenses at the time: prime 55/4, 135/5 (macro), and 200/4 on the Pentax; 17-40/4L and 70-200/4L on the D10.
The question in my mind was basically: how much more detail will I get on a large (for me) print from the Pentax than the 10D. So I set up my Gitzo by a small pond with houses and trees all around it on a good day with lile wind and good light. I took several shots of several scenes using the same aperture and same angle of view with both cameras and various lenses. The film was Provia 100.
I scanned the transparencies using the 2450's transparency unit at the highest resolution, and "developed" the D10 images in Capture One LE. The scans and the digital images were minimally processed (necessary USM) in PSCS, and 8.5x11 sections of each (to save paper) were printed to full-frame size of 13x19", the biggest my Epson 2200 will do.
At that size, the film and digital images looked different, of course, but the differences were like using Kodak film vs Fuji, rather than differences in detail or resolution. To my utter astonishment, there was no difference in visible detail! None whatsoever! And the D10 isn't even full-frame! I was shocked.
Yes, I used MLU on the Pentax, but the Pentax was using primes while the Canon had zooms. The probable limiting factor in my workflow was the scanner. If I had the films drum-scanned, or even if I got a good MF film scanner I'm sure there would be a noticeable difference. And my test didn't investigate differences in Dmax, where film would surely win.
But the bottom line is: for my workflow and the equipment I can afford, MF film has no practical advantage over the dinky D10.
But yes, I still plan to use the Pentax (at least for a while). It's just a different medium and a different way of working.
Eric
-
Right on, Drew. It's hard enough to choose equipment for your own use wisely without realizing 6 months later that some of that hard earned money was spent poorly, in spite of maximum conscientious effort to get the best information possible. If we also take it upon ourselves to educate everybody else how they should make they wisest choices, when will there ever be any time to take pictures?
-
Thanks Jose. Such detailed info from the perspective of a 35 yr. pro is worth maybe even more than $.02, for those that perhaps think that Michael has an axe to grind or just doesn't know what he's talking about.
I have Live Picture, diddled with it a bit some years ago and then never had occasion again. At this point Photoshop CS is working very comfortably for me with my 1ds files and I'm not eager to tackle the Live Picture learning curve again. When (if) I get a Mamiya ZD, I'll have to face the choice of G5 or Live Picture. At any rate, if I can find my Live Picture CD ROM, I'll install it and see how long it takes me to make sense out of it and see if it really matches what I can do with PS CS.
-
Also, Michael stated earlier in this thread:
"As a rule of thumb, 6MP cameras (I'm talking DSLRs, not digicams) outperform 35mm film, 11MP+ cameras outperform medium format film, and 22MP MF backs outperform 4X5" film when it comes to overall image quality, including resolution."
I would have agreed with him except for the last two words, "including resolution." The problem is no comparison of digital to film resolutions is complete without also stating the contrast ratio of the detail areas at the same time. Film's resolution limits increase with contrast and I have too many examples where my Imacon-scanned 4x5 Astia outresolves my 1DsII in these areas to be convinced otherwise. Film simply has more resolution capability than a 16MP FF or 22MP MF digital sensor in certain situations, period.
HOWEVER, I do believe that the "normal" 16MP FF or 22MP MF file is capable of producing images that regularly appear more detailed than the "normal" scanned 4x5 file will, up to say 32x40/48 in print size. Mind you, it is not always in fact more detailed -- putting a loupe to the print will confirm this --
Jack:
Thank-you for your cost analysis and excellent illustration of the MF film vs digital issue. Your posting of the crop images between film and digital should clarify for everyone that the issue of comparing these two formats is not so simple as some would have us believe. The comparison you showed echo's similar ones I've seen before demonstrating that resolution is still slightly better with 6x7. This is important IF you want maximum detail in large prints- which is what is important in my work. For anything up to A3 however, I am more than happy with my digital camera.
I also agree that a digital 16 MP or higher digital image will APPEAR the better image. Michael usually points out that it is OVERALL image quality he is talking about and I agree that digital CAN be "better". It also depends on what aspects of visual perception you value. Some photographs look "better" with grain for example (ie: look at Bresson's work). Some people have commented to me that digital photos sometimes look "unreal" or " too real". These are lay persons comments that some might laugh off or blame on poor technique, but perceptions count- especially if you are showing in galleries and selling your work.
So I think we can agree that "better" is a subjective term and for now, MF still has a legitimate place in a photographer's toolkit, especially for those on a budget. I look forward to the point when MF digital cameras are sub $10,000 in price range-the arrival of the Mamiya ZD early next year will be closely watched.
Jeff
-
It's pretty clear now that even the 1Ds Mark II can flat-out beat 645, and probably match 6x7: http://photo.nemergut.com/equipment/canon1ds/markii.html (http://photo.nemergut.com/equipment/canon1ds/markii.html) .
Looking at the len's that are being used, the canon uses a 50 and the MF an 80, how does the images look at the plane of focus . The tower is clearly behind the house and can its fuzzyness be explained by being far away from the plane of sharp focus
Bob
-
... an MF 90mm lens will behave like any other 90mm lens designed for 35mm ... except the edges should be sharper.
And except that maybe the center of the field might be less sharp than with the 90mm lens designed for 35mm format. Lens designers have to make trade-offs that potentially include improving sharpness at the center of the field at the expense of less sharpness or more vignetting further from the center, or vice verca. An MF design, constrained by the need to maintain good performance over a larger image circle, might have to make compromises that slightly harm central performance.
This does seem to be supported by PhotoDO's measurements of MTF for 35mm and MF lenses, where the 35mm lenses generally do better. Note that they test MF and 35mm lenses at the same 10, 20 and 40 lp/mm, and at the same range of distances from center, 0, 3, 6, 9, 12, 15, 18, and 21 mm. Thus their results correspond very directly to an asessment of performance when used with 35mm format and its 21mm image circle radius.
More generally, I am inclined to believe that the designers of high end lenses know their business far better than any of us do, and that a lens designed for a specific purpose (e.g. 35mm image circle) will in general be as good or better for that purpose than one designed with somewhat different goals.
-
Jeff,
The dynamic range of digital is superior to slide film of any kind in my experience, and if you use a tripod you can considerably further enhance the dynamic range by taking two or three different exposures and then combining them in Photoshop, a trick that's hard to replicate using film because of registration errors.
There are economic reasons why purchasing a 22mp back might not make much sense to many, but you can get 95% of the perfomance at 25% of the cost, so all is not lost.
My workflow with digital is so much more economic and pleasant.
With film it would be: buy film (ouch!), take shot(s), take film in for development, wait a day or two (or pay a premium for fast turnaound), pay throught the nose for reliable development (ouch!), carefully mount trannie on drum with oil, overlay, tape etc, scan. spot scan in photoshop, realise I flunked the exposure slightly (it happens), finally get to work on the image several days after I took the shot.
Witb digital its take the shot, check exposure and re-shoot if required, transfer to PC (cold beer) and decode raw files, and begin work - ofter all on the same day. No cost, no hassle, more focus on the image not the process.
Quentin
-
Bob's point about doubting Thomas is exactly right. I have another which is you can argue about whether something is a dog or not, but if its furry, barks at you and wags its tail, it almost certainly is.
I do not understand why we need to debate this point. I have looked at the post that started this thread and all I can see are assumptions based on scanner resolution and Dmax. No other facts or valid comparisons. Resolution for example is not just about numbers of megabytes. My only interest in this discussion is, can you take a better picture more easily with a digital camera most of the time than you can with a film one? In my experience, the answer would have to be yes and this is what most people seem to have discovered in the last two or three years.
As for the doubting Thomas, I would suggest to just let him carry on doubting. An example,...I have a respected photographer friend who asked my advice as to whether I thought a monitor calibration tool was worth investing in. I gave what I thought would be standard advice from someone who ought to know, which is that I thought it was close to being essential if you wanted to be certain that you were carrying out meaningful and accurate edits on your softproof images. I then bumped into him again a month later and asked him if he had bought one? No, he had spoken to another friend who had bought one, had not been able to get it to work and had given up on it. Therefore he was not going to buy one. I just smiled and shrugged. Life is too short to be constantly dragging horses to drink when they will not.
-
Didger, if you decide to go Live Picture route, contact me to tell you how configurate the OS9 Classic for a stable running of LP. Try to find a book titled Live Picture Revealed by Josh Karson, it is esssential to learn easily LP. There is also a list group full of professionals that still use LP. such as Gerald Bybee, Joe Holmes, Lee Varis etc
Regards
Jose
-
I have just rec'd my P25 for my Hassy H1's. I have been a Hassy shooter for 30 years and the reason that I still use Hassy is that I can shoot MF Film and Digital on the same platform. I have 3 H1's. I have dedicated 1 for the P25 back. 1 has been dedicated to film and 1 is a backup that will probably never see the light of day. The P25 will never be taken off except to clean it.
I will be taking a couple months to build my custom computer platform and to learn how to use the P25 properly. Then, I'll go to Colorado to actually take a 2-day course on how to perform with the P25 from shooting and post-processing to printing. I want to max out my information and skills before I step into the arena with the P25 at my side. I shoot instinctively. I know what the exposures for film will be based on the heat from the sun. I always use the handheld meter for verification but I want to be able to "feel" my shots without being concerned...at all...about whether or not I got the shots. I do my prep before I do the shots.
I will have a boatload of side-by-side tests under every possible condition, color and activity...all shot with the Hassy H1's, same lenses exactly, different bodies but exact lenses. All will be tripod mounted and metered with documented histories for comparisons. The MF film will all go to PCI lab, in Irvine, for custom printing and Crystal Archive matt papers. If there is a better, more complete comparison available, I'd be surprised. I'll post everything from data to images and my final critical opinions.
I need to know that I made the right decisions to pick up the H1's and P25, sight unseen.
Stay tuned.
Paul :cool:
-
The cost analyses for changing from an existing MF to a new 1ds system neglect the price of the lenses. I am assuming that the MF lenses will not work with the 1ds, and I will need (or want) a fleet of new Canon lenses. So the $8K 1ds price a couple years ago would be higher, and there may be more incentive than 3000ish frames to make me want to switch.
-
"More generally, I am inclined to believe that the designers of high end lenses know their business far better than any of us do, and that a lens designed for a specific purpose (e.g. 35mm image circle) will in general be as good or better for that purpose than one designed with somewhat different goals." Agreed, but with no data. With a high speed computer, it is relatively easy to design an excellant lens for a given purpose. It then becomes a matter of compromising the design for manufacture at the right price.
Knowing that MF lenses are to be used on MF cameras (and likely their own), makers appear to compromise center performance to get better center to edge performance. A 35mm lens maker does not need to compromise as much since the edge isn't as far away. I would guess that for a MF lens maker, having very large edge fall off in sharpness and illumination would be very bad. A reasonably sharp edge will look much better compared to a reasonable sharp center, compared to a fuzzy edge compared to a razor sharp center. It is the entire image that counts, not just the center. I would also guess that the Hasselblad lens market for the Canon 1ds is next to nil, and not a design consideration at all.
"An MF design, constrained by the need to maintain good performance over a larger image circle, might have to make compromises that slightly harm central performance." I think this is generally supported looking at the MTF for, say, Hasselblad and Canon lenses.
Given all this, and the fact that the final image will never be as good as either the lens or sensor can produce, I wonder why one would want to adapt (even though possible) a usually more expensive but lower performing MF lens onto a 35mm digital camera body.