Topic: Want – Need – Afford (Read 44269 times)

Mark D Segal · « **Reply #80 on:** August 27, 2009, 06:03:42 pm »

If you can convince Nikon that they can profitably market a HIGH VOLUME 65 MP MFDB solution for $10K all included: camera, back and lens (recall Mamiya has that in 21 MP), and they actually turn around and do it, you'll have done the community a huge service.

BernardLanguillier · « **Reply #81 on:** August 27, 2009, 08:39:01 pm »

Quote from: MarkDS

If you can convince Nikon that they can profitably market a HIGH VOLUME 65 MP MFDB solution for $10K all included: camera, back and lens (recall Mamiya has that in 21 MP), and they actually turn around and do it, you'll have done the community a huge service.

I would really love to achieve this, but I am afraid that my meager abilities are not up to the task. Don't get me wrong though, I have nothing against our current MFDB providers in themselves, just about their current practises, I'll become a big supporter of them the day they turn around and move on to a strategy that enables them to bring to the market more affordable products. They do bring creativity and innovation.

As far as the package you mention, I don't think though that we could reasonnably expect the lens to be part of the deal at that price point, except for a basic standard lens, but in terms of production costs, the pixel count has little influence so Canon and Nikon being arguably a lot better than mamiya at cost control and process optimization there is zero doubt that they can do it.

The key here, as always, is the size of the market and the amount of units you can spread the R&D cost on.

Now are they going to do it is the only question. It wouldn't surprise me if the head of Canon and Nikon camera divisions - who must play golf together more often than we think - challenged each other on such a topic... for the sheer fun of it.

Cheers,
Bernard

Mark D Segal · « **Reply #82 on:** August 27, 2009, 08:48:40 pm »

Bernard, my recollection of the Mamiya a couple of years ago is that the normal prime lens was included in that $9900 package, at least in NYC.

Yes, I think you're right - they probably do play golf, scope the planet and the stars together and probe eachother on where they think the next big push is going to come from.

It is all a chicken-egg business and hard to say where they would break into the viscious circle of high costs-small market-high costs. Probably bringing the costs down in whatever ways their inventive minds can conjure-up, because in this economic environment they aren't going to take big risks on market. There are those who think very high cost MFDB is inevitable, and I can see why, but I too wonder about that from time to time.

Ray · « **Reply #83 on:** August 27, 2009, 08:51:54 pm »

Quote from: MarkDS

Ray, I'm there, and from what I see in the Overview, the Phase P65 ranks within a quibble the same as a Nikon D3x except for "Low Light ISO" where the Nikon comes out on top hands-down. Their overall DxO mark is also extremely close. Interestingly, however, once one looks at the individual tabs underlying the overview, the Phase seems to perform less well than the Nikon. So without further explanation of how the whole is derived from the parts for this particular comparison, it all seems a bit contradictory to me.

This I think is a worthwhile question in the context of this discussion because the price disconnect between the two formats is huge and what we "need" or "want" should be informed by some notion of value-added in respect of results.

Mark,

I think the explanation for this apparent contradiction has to do with image/print size. It makes little sense to compare image quality at different sizes. Whether one makes a comparison on screen or on print, one should view equal size images from the same distance, which means either the D3X image will be upsampled more, or the P65+ image will be downsampled more.

In this context, the P65+ is marginally better, except for low light sensitivity, as you mentioned. Interestingly however, even at equal print sizes of 8"x12", the D3X still retains a 2/3rd EV advantage in dynamic range, although the P65+ takes the lead in the other parameters of SNR, Tonal Range and Color Sensitivity.

In the different context of comparing stitched D3X images of approximately the same pixel count as the P65+, one is in effect comparing image quality at the pixel level, in which case when both images are viewed at equal size there is an equal degree of upsampling or downsampling.

The individual tabs in the DXOMark comparison provide the option of comparing results at the pixel level and at a downsampled 8"x12" size. Unless the stitching process were to degrade the images with regard to tonal range, color sensitivity, sharpness etc, it would be reasonable to deduce that the final stitched result from the D3X images, of equal pixel count to the single P65+ image, might actually be better than the P65+ image in all departments. However, it's quite likely that such differences might be insignificant for most practical purposes, except for dynamic range. The 1.33 EV DR advantage of the D3X (at the pixel level) is surely significant.

Christopher · « **Reply #84 on:** August 27, 2009, 09:18:23 pm »

DXOMark mark is for people who have to much time and want to waste it on strange numbers compared to real world images.

Ray · « **Reply #85 on:** August 27, 2009, 09:59:41 pm »

Quote from: Christopher

DXOMark mark is for people who have to much time and want to waste it on strange numbers compared to real world images.

Wrong! DXOMark is for people who wish to make equipment comparisons when real world images are not available, or when the real world images that are available are not suitable for comparison purposes for a variety of reasons, such as the subject and lighting being different, the nature of the subject not being suitable to differentiate the differences of camera/lens performance, the technique of the cameraman not being adequate etc etc.

For example, if you want to compare the DR of two differents systems using real-world images, then the scene being photographed should have an SBR (subject brightness range) which is significantly greater than the DR capability of the weakest system. In addition, both images should be equally exposed to the right, otherwise the results could be misleading.

You will notice that the subject of ETTR keeps cropping up again and again on this site. That's because it really is difficult to get an exact ETTR without going to some trouble. It is a safe bet that the people at DXO know what they are doing and use a consistent approach to get their test results. You can whinge all you like about DXOMark results, but I've yet to see a P65+ shot that displays a greater DR than a D3X shot of the same high SBR scene, at base ISO.

Mark D Segal · « **Reply #86 on:** August 27, 2009, 11:46:51 pm »

Quote from: Ray

Mark,

I think the explanation for this apparent contradiction has to do with image/print size. It makes little sense to compare image quality at different sizes. Whether one makes a comparison on screen or on print, one should view equal size images from the same distance, which means either the D3X image will be upsampled more, or the P65+ image will be downsampled more.

In this context, the P65+ is marginally better, except for low light sensitivity, as you mentioned. Interestingly however, even at equal print sizes of 8"x12", the D3X still retains a 2/3rd EV advantage in dynamic range, although the P65+ takes the lead in the other parameters of SNR, Tonal Range and Color Sensitivity.

In the different context of comparing stitched D3X images of approximately the same pixel count as the P65+, one is in effect comparing image quality at the pixel level, in which case when both images are viewed at equal size there is an equal degree of upsampling or downsampling.

The individual tabs in the DXOMark comparison provide the option of comparing results at the pixel level and at a downsampled 8"x12" size. Unless the stitching process were to degrade the images with regard to tonal range, color sensitivity, sharpness etc, it would be reasonable to deduce that the final stitched result from the D3X images, of equal pixel count to the single P65+ image, might actually be better than the P65+ image in all departments. However, it's quite likely that such differences might be insignificant for most practical purposes, except for dynamic range. The 1.33 EV DR advantage of the D3X (at the pixel level) is surely significant.

Ray, I understand this problem - image size issue - but when you start resampling you are introducing another variable that can bias outcomes - exactly how and how much being the questions.

But I'd have to go back and look at the DxO stuff in the context of what you are saying here. If indeed the results are on the whole proximate, even save for one variable, it does raise a question about what one is paying for. While Christopher is wrong about the usefulness of DxO for the reasons you state, I would not rule out some kind of flaky issue that may have influenced their results in the case of the Phase camera. Perhaps there needs to be some discussion with them about this comparison, if they care to talk about it.

Schewe · « **Reply #87 on:** August 28, 2009, 12:32:12 am »

Quote from: MarkDS

Ray, I understand this problem - image size issue - but when you start resampling you are introducing another variable that can bias outcomes - exactly how and how much being the questions.

So, unless you have shot something with a 1Ds MIII (sorry, I don't do Nikons) and a Phase One P65+ and compared the results side be side, you just can't really appreciate the differences between 21+ MP and 60+ MP. With the 1Ds MIII at 240 PPI (the "minimum I would print with) I can get a print that is 23.4" x 15.6" uncropped and not resampled. With the P65+ uncropped an unresampled the print at 240 PPI would be 37.4" x 28". Now, I gotta tell ya, the difference between 23" whatever by 15" whatever and 37" x 28" is friggin huge...

I just shot a cello for an Epson video where I shot with a Sinar 4x5 with the P65+ back with a 300mm APO lens and had to upsample (a bit) to get to 38" wide...the print which ended up being 38" x 58" at about 240PPI looked better than ANYTHING I HAVE EVER SHOT ON 8X10....

Screw contact prints bud...who cares about stuff sitting on your lap. What does the image look like when it's 3'x4' and you have to back off to see the whole thing?

Bernard's stuff is really nice looking...I love it. I've got some 5-10 image panos I've assembled from 1DsMIII's in Antarctica that are real nice...but I just shot some 3 & 5 shot panos with the P65+ and I'll be shooting with Martin Evening in Sept in southern Utah where I'll be shooting panos with the P65+ and the Phase One and the 28MM lens (you can buy a real nice used car for the price of that lens). Guess what...multiply the same sort of size relationship between a single 1DSMIII and a single Phase One P65+ when contemplating stitching medium format...3-5 or 7 stich from a P65+? Huge...

So, here's the funny payoff for all this...in the most recent Real World Image Sharpening book I compared a shot done with an iPhone to a shot done with a P65+. Ya know what? For a reproduction of about 3"x4", there's no real difference in halftone repro. An iPhone that costs about $199 can do about as good a job as a P65+ at $40K. If all you need is a 3"x4" halftone image...

Does that mean I would tell commercial (or fine art) photographers to go out and buy an iPhone fore their serious stuff?

Sure, if all they are gonna print is 3"x4" stuff...

Size is extremely relative and contrary to what your girlfriend (or wife) may say, size does matter...a lot.

Sorry, deal with it...

stewarthemley · « **Reply #88 on:** August 28, 2009, 01:56:35 am »

The last time I supported Jeff some "person" jumped in with an attack on both him and me. But then he withdrew his remarks. So, assuming the coast is clear (!) I agree Jeff. (Dives for cover.)

Of course size matters. The only reason I had to go from the Mk3 (Canon) to a 39mb back was because my biggest client, who knows lots of other prospective clients, wanted something big to look good close up. I hired a MFDB/camera and after staring at the glaringly obvious quality difference we both decided there was no going back. I bought the camera, he now has lots of stuff done big, is happy to pay a fair percentage more, and we're both happy. So are my new clients that he recommended me to.

Seriously, if your clients can't see the difference between a Canon/Nikon and a MFDB then keep using the DSLR. Just take care if a competitor approaches them with a decent MF print.

(punctuation edit)

Ray · « **Reply #89 on:** August 28, 2009, 02:02:22 am »

Quote from: Schewe

So, unless you have shot something with a 1Ds MIII (sorry, I don't do Nikons) and a Phase One P65+ and compared the results side be side, you just can't really appreciate the differences between 21+ MP and 60+ MP. With the 1Ds MIII at 240 PPI (the "minimum I would print with) I can get a print that is 23.4" x 15.6" uncropped and not resampled. With the P65+ uncropped an unresampled the print at 240 PPI would be 37.4" x 28". Now, I gotta tell ya, the difference between 23" whatever by 15" whatever and 37" x 28" is friggin huge...

I just shot a cello for an Epson video where I shot with a Sinar 4x5 with the P65+ back with a 300mm APO lens and had to upsample (a bit) to get to 38" wide...the print which ended up being 38" x 58" at about 240PPI looked better than ANYTHING I HAVE EVER SHOT ON 8X10....

Screw contact prints bud...who cares about stuff sitting on your lap. What does the image look like when it's 3'x4' and you have to back off to see the whole thing?

Bernard's stuff is really nice looking...I love it. I've got some 5-10 image panos I've assembled from 1DsMIII's in Antarctica that are real nice...but I just shot some 3 & 5 shot panos with the P65+ and I'll be shooting with Martin Evening in Sept in southern Utah where I'll be shooting panos with the P65+ and the Phase One and the 28MM lens (you can buy a real nice used car for the price of that lens). Guess what...multiply the same sort of size relationship between a single 1DSMIII and a single Phase One P65+ when contemplating stitching medium format...3-5 or 7 stich from a P65+? Huge...

So, here's the funny payoff for all this...in the most recent Real World Image Sharpening book I compared a shot done with an iPhone to a shot done with a P65+. Ya know what? For a reproduction of about 3"x4", there's no real difference in halftone repro. An iPhone that costs about $199 can do about as good a job as a P65+ at $40K. If all you need is a 3"x4" halftone image...

Does that mean I would tell commercial (or fine art) photographers to go out and buy an iPhone fore their serious stuff?

Sure, if all they are gonna print is 3"x4" stuff...

Size is extremely relative and contrary to what your girlfriend (or wife) may say, size does matter...a lot.

Sorry, deal with it...

Jeez, Jeff! How can you miss the point so much? Do you drink? (If you do, you're excused)

There can be no doubt whatsoever that a single P65+ shot will have superior resolution to a single D3X shot of the same scene, same FOV etc.

For those working in a studio producing poster size images of models flinging their hair, or any scene where significant movement takes place, the P65+ will reign supreme for very large prints.

However, the thread is about 'want-need-afford'. Landscape in particular tends to be a static subject. Therefore, if one needs a large panoramic poster, perhaps more often than not, the subject lends itself to a stitching process. If it does, then it seems superfluous to spend so much money on such an expensive, heavy and cumbersome piece of equipment as the P65+ with body and lenses.

I always like to illustrate my point with a practical example. If you and I, Jeff, were hiking up Poon Hill in Nepal, to get the following dawn shot of the Himalayas, who would likely get there first, you with a P65+ (and paraphernalia) or me with a 5D2?

[attachment=16266:For_LL.jpg]

John Camp · « **Reply #90 on:** August 28, 2009, 02:13:44 am »

There are obviously uses for which a 39 or 50 mp MFDB will provide a higher-resolution image than a D3x, such as fashion shots with a moving model, or any close-up, moving subject...if you want to print really big. If you're printing magazine pages, I doubt that anyone could see a difference. However, we're seeing more and more gargantuan images -- the model shots you see on billboards and bus-stop boards, etc., where people can get really close to a large image. For those, I don't think a MFDB could be beaten.

For basically static landscapes, I think Bernard has a strong point, and I essentially agree with him. If you can stitch, and you're good at it, a D3x can match a MFDB. In some circumstances, a stitched D3 (not x) image could blow the doors off a MFDB, because of its high ISO capabilities. It can do things that no MFDB can do.

As for the 8x10 argument, the way I see it, 8x10 was a limitation that Edward Weston had to deal with; it wasn't a feature, it was a bug. Though it's traditional, it's too small. I've seen 8x10 contact prints by famous photographers at the Minneapolis Museum of Art and to appreciate that good-ol' contact-print resolution, you have to have your nose a quarter inch from the print. Guess what -- in any museum, that print is going to be behind glass, and usually behind filtered glass, and you won't be able to appreciate the detail because you can't see it. By the way, one reason almost all oil paintings you see in museum are larger than 8x10, when they could, in fact, be almost any size, is that most of them are "human sized" --- made to be easily seen and appreciated in human living and working spaces. So they are typically larger than about 20x12, and a very larger number are about 30x40. Photographs were 8x10 because of limitations in the medium, not because it was a magic size for photographs. Digital is breaking through that barrier.

I have a gorgeous copy of "Moonrise" hanging in the next room, and it really is gorgeous; but I've seen digital prints that are technically as good. And I say that as a collector of B&W silver prints.

Christopher · « **Reply #91 on:** August 28, 2009, 03:12:25 am »

Well Ray if you want to compare systems, than you go out and rent them. I still don't see the sense of DXOMark. If I'm interested in a system i try it. End of discussion. DXOMark is for people who don't really care about real world images and more about numbers and scores.

Another thing, I bet that my Phase System is not heavier than any D3X system....

Ray · « **Reply #92 on:** August 28, 2009, 06:32:52 am »

Quote from: Christopher

Well Ray if you want to compare systems, than you go out and rent them.

Not if you are a busy person with a sense of economy and efficiency and are smart enough to understand the figures, tests and reviews which you seem to so despise.

I've never, ever selected a camera by first hiring two or more cameras and then spending a lot of time comparing them with thorough tests, which I am capable of doing, by the way, but which I know is very time consuming.

Mark D Segal · « **Reply #93 on:** August 28, 2009, 08:29:45 am »

Quote from: Ray

Jeez, Jeff! How can you miss the point so much? Do you drink? (If you do, you're excused)

There can be no doubt whatsoever that a single P65+ shot will have superior resolution to a single D3X shot of the same scene, same FOV etc.

For those working in a studio producing poster size images of models flinging their hair, or any scene where significant movement takes place, the P65+ will reign supreme for very large prints.

However, the thread is about 'want-need-afford'. Landscape in particular tends to be a static subject. Therefore, if one needs a large panoramic poster, perhaps more often than not, the subject lends itself to a stitching process. If it does, then it seems superfluous to spend so much money on such an expensive, heavy and cumbersome piece of equipment as the P65+ with body and lenses.

I always like to illustrate my point with a practical example. If you and I, Jeff, were hiking up Poon Hill in Nepal, to get the following dawn shot of the Himalayas, who would likely get there first, you with a P65+ (and paraphernalia) or me with a 5D2?

Well, a couple of prizes for the guy who gets there first , but quite seriously, from what I've seen, there isn't that much difference in heft and weight between a MFDB and a Canon 1DsMk3. When it comes to adding lenses to the bag, then it becomes a different story. So you add some sessions with a PT to the price of the MFDB and get into better shape at the same time. A win-win for the MFDB and PT industries!

ErikKaffehr · « **Reply #94 on:** August 28, 2009, 09:30:31 am »

Hi,

It depends a bit. Are relevant tests available? If you spend a lot of money it may be a good option to lend stuff and test it. I guess that DSLRs are essentially in the same league so it may not matter a lot. Lenses are a different thing.

Christian's view makes a lot of sense to me...

Best regards
Erik

Quote from: Ray

Not if you are a busy person with a sense of economy and efficiency and are smart enough to understand the figures, tests and reviews which you seem to so despise.

I've never, ever selected a camera by first hiring two or more cameras and then spending a lot of time comparing them with thorough tests, which I am capable of doing, by the way, but which I know is very time consuming.

Mark D Segal · « **Reply #95 on:** August 28, 2009, 10:35:42 am »

Quote from: ErikKaffehr

Hi,

It depends a bit. Are relevant tests available? If you spend a lot of money it may be a good option to lend stuff and test it. I guess that DSLRs are essentially in the same league so it may not matter a lot. Lenses are a different thing.

Christian's view makes a lot of sense to me...

Best regards
Erik

DSLRs are not in the same league; they handle differently, they have different feature sets and they produce different results -sometimes subtle, sometimes not so subtle. How different brands handle exposure is a key case in point. Please see my article Some Noise About Noise on this website. Hence I agree with your view that renting equipment or finding some way of getting a "trial-run" before buying can be a very good precaution. The more you know before you buy the better.

As far as Christian's view is concerned - rubbish - this is an insult to the serious scientific effort of a justifiably prestigious organization and what they are doing is complemetary to the hands-on testing and image viewing one would like to do before buying. Again, the more one knows before making expensive commitments the better, and the data provided by DxO is another kind of information which, provided they got it right, is no less valid than "your-own-hands-on" information. It makes a lot of sense to me to be able to test my own field observations against their laboratory observations. If the two cohere, it is reinforcing. If they don't cohere, then one needs to re-examine either their stuff or my stuff to understand why. It's part of improving knowledge.

Ray · « **Reply #96 on:** August 28, 2009, 10:45:24 am »

Quote from: MarkDS

......but quite seriously, from what I've seen, there isn't that much difference in heft and weight between a MFDB and a Canon 1DsMk3.

Mark,
There also isn't that much difference in performance specs between a 5D2 and a P65+ at the pixel level. The P65+ has about 1/3rd of a stop greater DR, but only at base ISO.

At ISO 100 and above the 5D2 is better in all departments except single shot resolution. With an actual base ISO of only 44, I think one would have to add the weight of a tripod to that trip up Poon Hill.

Schewe · « **Reply #97 on:** August 28, 2009, 10:56:49 am »

Quote from: Ray

I always like to illustrate my point with a practical example. If you and I, Jeff, were hiking up Poon Hill in Nepal, to get the following dawn shot of the Himalayas, who would likely get there first, you with a P65+ (and paraphernalia) or me with a 5D2?

I don't know...would the naked ladies be there too? What would they be carrying?

Actually, the Phase One 645 and a couple of lenses isn't really so much worse than a 1Ds MIII and a couple of lenses...yes, the 5D MII would be much lighter (a reason I often shoot candids with a Canon Rebel).

The point I was making (since it seems to have zoomed by your head) is that you use the tool that you need to use to get what you want out of an image. If you need a 3" x 4" image, you would be foolish to buy a P65+. If you need a 3' x 4' image, your 5D MII is gonna come up short compared to a P65+. And, if you don't get what you need because you can't afford it, I don't know what to tell somebody other than lower your expectations (and make smaller prints).

Mark D Segal · « **Reply #98 on:** August 28, 2009, 11:19:01 am »

Quote from: Schewe

I don't know...would the naked ladies be there too? What would they be carrying?

Actually, the Phase One 645 and a couple of lenses isn't really so much worse than a 1Ds MIII and a couple of lenses...yes, the 5D MII would be much lighter (a reason I often shoot candids with a Canon Rebel).

The point I was making (since it seems to have zoomed by your head) is that you use the tool that you need to use to get what you want out of an image. If you need a 3" x 4" image, you would be foolish to buy a P65+. If you need a 3' x 4' image, your 5D MII is gonna come up short compared to a P65+. And, if you don't get what you need because you can't afford it, I don't know what to tell somebody other than lower your expectations (and make smaller prints).

The way I interpret that scene is that the naked ladies should be there already awaiting your arrival and shouldn't be burdened beforehand carrying stuff. It wouldn't be "chivalrous", and they may not be as fresh and happy as they appear in Ray's image.

Back to serious, are you saying Jeff, that if one's print size were going to be limited say to Super A3 (13*19), apart from the luxury of a huge amount of cropping flex you'd get from a 65MP image, there wouldn't be much IQ difference between a 1DsMk3 and a P65, say in the range of ISO 100-400, using pro-grade lenses on each?

Schewe · « **Reply #99 on:** August 28, 2009, 12:16:13 pm »

Quote from: MarkDS

...are you saying Jeff, that if one's print size were going to be limited say to Super A3 (13*19), apart from the luxury of a huge amount of cropping flex you'd get from a 65MP image, there wouldn't be much IQ difference between a 1DsMk3 and a P65, say in the range of ISO 100-400, using pro-grade lenses on each?

You would prolly have to look REAL HARD to see any substantial difference if the print size for each camera was considerably smaller than "native" resolution at say 360PPI. So, if all you make are "smallish" prints, the reason for getting a P65+ VS a 1DsMIII (or 5DMII) would have to be something other than just IQ. There are reasons...cropability is one, plane of focus on a view camera another. But then on the flipside, for the P65+ you need more light, and higher ISO would go to the 5DMII (you end up with a 15MP capture if you use the P65+ to do pixel binning at ISO 1600).

It really all comes down to rero size...

Author Topic: Want – Need – Afford (Read 44269 times)

stewarthemley