Measuring Megabytes

[Mark D Segal]

After reading all the posts in this thread, I think it is important to dis-entangle ones' personal circumstances from the world of objectivity. This isn't about "I don't need it so it isn't important", "I'm satisfied with what I have so it doesn't matter", or "I can't afford any of that gear so it's irrelevant". The way I see it, this is about carefully observing the evolution and state of photographic capture technology in a practical and scientifically objective manner. For all those people who aren't interested in the technical progress made, say between the 1Ds and the 5D, or the value-added of a medium format back versus a 1DSMk2, no-one is forcing them to read this stuff or to buy the disc. For those of us who ARE interested in it - if only for its own sake, this is truly a unique and remarkable resource. But as well, I think it may also help me decide about my next camera, because I will see the new stuff in context of what came before it and what is for sure or likely coming next. That's partly how I opted for a used 1Ds when the MK2 hit the market. I'm really looking forward to receiving this disc - a ten dollar investment in knowledge that will long outlast a couple of Mocca-Latte's from my favorite Starbucks, no matter how fast the technology evolves.

[Tim Gray]

I think, in part, what surprises Michael is the time folks spend contributing (and I intend no sarcasm) to these forums - discussing relatively expensive hardware.  

If their time is worth more than $.50 per hour, I don't see why someone (like me) who never aspires to MF wouldn't fork out $10.00 (I did) to get some real life files to play with.  If their time is worth less than $.50 per hour then I guess $10.00 is a bit steep.

[drm]

Well I suppose I haven't really expressed my reaction terribly well.  My intention was to point out there is a context to this stuff, and as long time readers of this forum may have noticed, I do have a tendency to delight in playing the Devil's advocate.  I still believe that the asymptotic cost of high end photographic equipment is an issue, and that it has a social context too, but this forum is not the appropriate place to air that view. I also have a slightly better grip on business plans than I do on photography, and I'm not totally convinced that the numbers add up in the majority of cases.  But .... if a PhaseOne P45 and Hasselblad H1 turned up on my doorstep, would I turn my nose up at it ? Like hell I would :-)

So, Michael, I apologise for going way off topic, and in particular for any offense to you, your co-authors, or other contributors to this topic.

David.

[BJL]

You don't think even 3600 US$ is a fortune ?  I'm sure many would disagree....

Photography - in the artistic sense - is rapidly losing grip on economic reality.
[a href=\"index.php?act=findpost&pid=65683\"][{POST_SNAPBACK}][/a]

Amen; but I think that it is only some forum participants that have lost track of economic realities, talking of cameras that cost far more than many good medium format film options as if they were widely affordable options ofr photgaphic enthusiasts.

This does not seem to apply to the SLR buyers who are overwhelmingly choosing sub-$1000 DSLR, and buying the $1700 D200 at about four times the rate of the $3000 5D. (A market reality check: even Pentax apparently sells more of its DSLR's than Canon does of its 35mm format models, and Olympus sells twice as many again, while Canon's own EF-S mount bodies outsell those 35mm format DSLRs almost twenty to one).

My guess is that for many photographers who are disposed to work in the traditional large format view camera mode with stationary subjects and many minutes spent making an image, very good results good could be achieved for far less than US$3,600 by stitching images from a quite modestly priced DSLR and a good sharp lens. With stitching, the gap between even the new low-price champion Pentax K110D ($600 with kit lens, probably under $500 when the body-only option comes out) and something for more expensive like a D200, 5D, D2X or 1DsMkII is only having to take about twice as many frames for the stitch-up job.

[benInMA]

Well the thing is all the cameras in this comparison look great and you have to pixel peep or look at large prints to see any difference.

It's all in your given needs & budget.  The 5D might seem expensive when compared against the 30D or D200 but in this company it is a smoking bargain.

Since this comparison already covers such a monstrous price range it would have been interesting to see a 1.5x/1.6x sensor camera thrown in just for the heck of it.  My guess is we should have been able to pick out the small sensor DSLR even with the JPGs on the website.

Everything electronic seems to have a price/performance curve.  At some point around the D200 or 5D the price/performance curve skyrockets... huge amounts of money spent for diminishing increases in image quality.

[Ben Rubinstein]

Oy BJL, hands off my 5D    

Seriously though if canon are willing to fill that niche then they are still selling the 5D well to the wedding/portrait/landscape market with the lenses that go with it.

Let's face it, the entry level market is where the real money is, it's no secret. I'm sure the D2X and 1Ds mkII or even the 1D mkII also sell fractions in comparison to the D50/70 and 350D but that is the point, they need to be there for the pros and for the companies street cred.

Lets face it, even if the original groundbreaking 1Ds was a loss leader, it did an incredible amount for canon's prestige, for the canon name, no doubt sold more canon lenses than any other camera ever made (considering it pulled in so many nikon, minolta, med format shooters, etc) and even now is considered a ground breaking classic. The 1D and mkII has insured that the sidelines of any sports game, press scrum, catwalk or indeed any where you see a journalist there are a bevy of white lenses. Did not the 1Ds and it's successor ensure that medium format prices on ebay are just silly? Did it not contribute to the death of Bronica, Contax, possibly Mamiya? I would not doubt that it was these bodies far more than the digital medium format niche which killed those manufacturers and believe me Canon are milking it for all it's worth.

Another point, anyone who is going to spend the money on a stitching setup (software, tripod setup, brackets, learning curve, more computer memory etc) might well forgo the bother and spend the extra. Stitching is not for everyone and as you say, more suited for those used to a large format type shooting style with the added pain of all the extra time at the computer afterwards. I would of course be useless for a majority of subjects where stitching just wouldn't be an option and that is pretty limiting.

I also think that it's important to reiterate what I said earlier here. Many many people here are shooting these so called 'OTT expensive' digital solutions because it makes sense financially for them. Yes for enthusiasts the math may not add up, but hey how many of them are even shooting a 1Ds mkII never mind a medium format back. Of course many of them do have the 5D but once you've had the 20D it's only a step up in price isn't it? There I do agree with you wholeheartedly, if it isn't making economic sense then it's just paying lots for nice toys and we know how vulnerable us blokes are for that kind of thing not least with cameras!

[Quentin]

Hmm.  I am not surpised only 200 or so have shown an interest in the disk; the market for these backs is vanishingly small. I know several full time pros and well-heeled amateurs who would (and some do) laugh at the idea of spending mortgage-level sums of money on a technology that is changing so rapidly.

So while I am fascinated to read such well executed comparisons on line, and genuinely grateful to Michael and chums for doing the tests, I regret that I'm not interested enough (and plainly I am not in a minority in thinking this way) to pay to look at the raw files on disk.  

Quentin

[BernardLanguillier]

Amen; but I think that it is only some forum participants that have lost track of economic realities, talking of cameras that cost far more than many good medium format film options as if they were widely affordable options ofr photgaphic enthusiasts.

My guess is that for many photographers who are disposed to work in the traditional large format view camera mode with stationary subjects and many minutes spent making an image, very good results good could be achieved for far less than US$3,600 by stitching images from a quite modestly priced DSLR and a good sharp lens. With stitching, the gap between even the new low-price champion Pentax K110D ($600 with kit lens, probably under $500 when the body-only option comes out) and something for more expensive like a D200, 5D, D2X or 1DsMkII is only having to take about twice as many frames for the stitch-up job.
[a href=\"index.php?act=findpost&pid=66366\"][{POST_SNAPBACK}][/a]

Good points. Besides, using even a 6MP DSLR at its full potential isn't as easy as it seems, and many of us could probably still gain something by sticking with our existing gear and focussing on knowing it and using it better.

On the other hand, there are IMHO objective reasons to invest in DSLRs like the 5D/D2x/1ds2.

1. IMHO, very few high end amateurs own printers that go larger than A2. With a D2x/1ds2, you have enough pixels on the long end of the frame to do one row stitches that have a quality high enough to use the potential of A2+ paper rolls.

A 6MP DSLR would require you to do 2 row stitching to get there, which requires a more complex pano head and implies a slower stiching routine in the field, furhter reducing the attainable scope for stitching.

Besides, we all know that stitching isn't the solution for all situations, and it makes sense to have a DSLR that can yield images with a resolution high enough to be compatible with large prints without requiring stitching.

2. I feel that it takes that level of resolution to clearly out do 35 mm film and get into MF realm, the traditionnal sweet spot for landscape in terms of compromise between usability and image quality. For those photographers that are able to resist the pressue of buying the best and the latest at each new release, this means that a 5D/d2x/1ds2 could remain a credible solution for many years to come...

3. The 5D/d2x/1ds2 cameras are expensive, but as I wrote before, so are cars for instance. I haven't owned a car for 10 years, which allowed me to save between 20.000 and 40.000 Euros. Even then, I coudl hardly afford a P45, but I can easily afford a D2x.  Granted, not all people are in a position not to own a car.

Regards,
Bernard

[Blake]

Hi Michael,

     Re: your "personal note" article, I must say that I admire your professionalism in addressing those who denigrate the efforts of people like you (as well as Charlie and Bill) who make the net a better place by sharing their invaluable experience with us all; I'm not sure if I would have been able to maintain such candor in a similar situation.

     I know of more than one instance where such disrespect has worn down the creator of a site or list to the point where they simply didn't want to deal with it anymore so they gathered up their marbles and went home, and I would hate to see such a thing happen here.  Thank you once again for providing the excellent photographic resource that is The Luminous Landscape.

 -Blake

[BJL]

Bernard,

   your analysis makes sense to me. In the rarified realms calling for resolution beyond the now common-place 8MP, there are two cases
a) stitching, which it is probably better to do in one direction only, so that you want the long dimension of your sensor to match the resolution needs of the short dimension of your prints. The ultimate digital panoramic camera might use a linear CCD as in a scaning back, scanning the image with a Nobilex style rotating lens system.
 single shot, where of course pixel counts of 12MP or 16MP and beyond have their place, though only for those who are likely to make some uncommonly large prints, like A2 and up, and view the mfrom uncommonly close for that print size.

My observation is that viewers of even huge prints rarely get much closer than the short dimension of the print, and about 3000 to 4000 pixels on that short dimension seems to be about enough for that: 6-12MP for stitching, 12-14MP for single shot satisfying most but not all needs?