Having read a while ago how good the 1Ds was picturewise, when it was the only full frame around and seen some full size pictures then, I'm curious to hear how good the current 1DIII is compared to the old 1Ds. I mean, old 11mp versus new 10mp, shouldn't make that kind of a difference resolutionwise and picturewise. But from what I have seen, the cropped sensor 1DIII still looks like a cropped sensor image factory. Shooting Nikon currently, I'm debating with myself if I should go for a 5d or for a 1DIII. I don't need speed, but I'm seriously impressed with the live view of the 1DIII, I mean a well focussed 10mp is always better than a badly focussed 5D pic, isn't it? From my former Kodak 14n I remember that 3D look, but I haven't seen it yet in the samples of the 5D and 1DIII, is it the aa fliter maybe?
Thanks in advance,
Bart

I've owned every 1D model since the original and currently have a Mark III as well as a Mark IIn and a 1Ds2. The 1Ds was a great camera for the time. Though I have not yet shot extensively with the Mark III, based on past experience with the 1Ds and projected experienced (based on the specs of the Mark III), the newer camera should be superior in most respects. The 1Ds started being grainy (noise) at 400. The Mark III is probably the equal at 1600. (I rarely shoot much above 200, and then it's usually only at 400.) The best aspect is that the hi-bit images are based on 14-bit, rather than 12-bit imaging, so you should have better, smoother images less likely to blow out highlights.

Your real question to answer is: Can you live with less that "full-frame"? I suppose you've answered that in a way since you've been using Nikons. I wholly agree with you, regarding the 5D. I just finished a shoot where I had to rent a 5D for a week because the shutter in my 1Ds2 crapped out. I've just finished doing the digital imaging for the shoot, of which a number of images were going up to 8' tall. The 5D images were no better, or only marginally better, than my 1D2n. Though I shot with quality glass (mostly L series), I could not see a increase in quality of the 5D over the 1D, and it was WAY below the images I get from my 1Ds2. Plus I HATED the 5D handling, compared to the 1D. I know some people loved the camera, but after handling the 1D for years, it felt like a cheap toy. (As it is I prefer the old menuing system of the 1D over the new Mark III.)

So, my suggestion, if you don't want to spring for a 1Ds2, go for a Mark III over the 5D, especially if superwide angle is not a factor.

Nemo

Thanks Nemo,
You seem to be just the right person to talk to .
To rephrase my question: where the 1Ds was to be the one and only choice after it was introduced, is it still a better camera than the 1DIII at iso 100/200?
About the 5D, I don't really care how a camera looks like, but I like ergonomics, but most of all image quality is what I'm after. I remember talks where attending photographers were saying the 5D is as good if not better than the 1DsII picturewise. Hard to believe for me where I have only seen pics until now that had a bit of a plastic look to me.
I won't mind the cropped sensor of the 1DIII, since I have heard the viewfinder has been enlarged to the feel of full size finders and live view is something hard to ignore, if only the image quality would hold well against that of the 1Ds.
(Your portfolio looks very good by the way.)
Thanks,
Bart

I have 1Ds and 1DsII . The 1DsII is simply not good for people. I had a 1DII, it was worse.

If people here can post images of PERSONS comparing a 1Ds @ ISO 100 and a 1DIII at its native ISO we would know what the state of play is - otherwise I would say that the only thing which will image better inthe studio than a 1Ds is a digital back.

Edmund

What an extraordinary statement. Why do you say this?

Nill
~~
www.toulme.net

It was discussed a lot in days gone by; James Russell will confirm. The 1DsII is plasticky on skin, but extraordinary for objects. The 1Ds is a bit noisy over ISO 200 but has a beautiful skin rendering. Clients always chose the 1Ds shots over the 1DsII pix when they saw them side by side, until I stopped using the 1DsII completely for people. As for my 1DII, it was a train wreck and had to be sold. Of course all these cameras were first in their batch - the 1Ds too was not so good initially (noisy) but I had the sensor changed, and it became a beauty.

So, I use the 1DSII for objects and the 1Ds for people, and they are a good combo. I just have to use ultra-fast lenses on the 1Ds for the fashion shows, but the pictures get published.

The 5D units I tested seemd to combine the best of both the 1Ds and 1DsII - good skin tone and low noise. Frankly, I would recommend a 5D to anyone wanting a pro camera. This is with my own custom profiles - without them it doesn't look suite as good in C1.

Edmund

This sounds extraordinary to me too. Is there any rational explanation for this, Edmund? I mean, the idea of editing with programs like photoshop and plug-ins, is you can create virtually any effect you want. If you want a really platicky effect and a skin texture as smooth as a baby's bottom, then aggressive noise reduction in Neat Image will do the trick. If you want more noise or a film grain appearance, or a Velvia appearance with saturated colors, then there's an appropriate filter or plug-in for the job.

Are you by any chance referring only to jpeg shots with in-camera noise reduction and/or removal of compression artifacts?

I use a 5D, the 1D variation are too big for my needs, but something I've noticed from digital imagery is the plasticy skin referred to above and with high end cameras too. Looks a bit video to me - unsurprising as a digital image is just a video still in effect. Horrible.
Though I did see a nifty trick in a C1 the other day that stops skin looking so plasticy. I don't use C1 yet [waitng for v4] so cannnot remember how to do it. But it can be done, so ask around.

I would be interested in seeing a comparison of the 1DIII and 1DsII and getting some RAW files using the same lens for some comparisons. I do not expect the 1DIII to be better at the lower ISOs but would be interested to see.

Plasticky skin? Have you guys ever thought of getting back to reality?

This is grainy; a recent scan I did of a B&W shot taken 43 years ago with a Pentax Spotmatic.

Click to view attachment Click to view attachment

And this is plasticky.

Click to view attachment

I mean Raw processed to my abilities. As a color consultant I make judgment calls on renderings. I consulted with expert photographers (I'm more of a tech guy than a photographer) and none of us could find a solution to the 1DsII problem, so I just gave up and regressed to the 1Ds, James Russell sold his 1DsII cameras and moved to a back etc.

For the handful of fashion images I publish every year the 1Ds is more than adequate, although it is now on its third shutter and looks like the mailman's car

So far, what I've seen of the 5D I have liked.

Edmund

Every review I've yet read of the III states that it has the best image quality of any canon DSLR yet, even at low ISOs. The low ISO part is said to be due to minimal shadow noise and retained detailing.

Hopefully I'll be getting to see some shots with it soon, a friend's should be here within the week.

what reviews?

"A final word on image quality Even with the image quality impairments we've identified here, the files from the EOS-1D Mark III are as we described them earlier in the article: the best we've seen from a digital SLR."

http://www.robgalbraith.com/bins/multi_pag...cid=7-8738-8908


"One can also quibble as to whether the Canon 5D has been the best yet (my opinion), but there will be no argument that with the MKIII Canon has now exceeded what has come before, and has set a new benchmark for low noise."

http://luminous-landscape.com/reviews/came...III-Field.shtml

As to the second quote I know that noise alone isn't everything, but take the comment in broader context.

I'll post others if I can, but those were the two that first came to mind.

There have been some discussions in some of the European magazines that the image quality is not as good as the 1Ds mkII. or 1Ds for that matter. While there is no doubt the 1DmkIII looks like a decent camera, the reviews you have linked to do not give a side by side test and would really like to see such a comparison before making judgement.

True (Where art thou DPReview!?!) but I, personally, tend to take more from reviews by people who have used the gear out in the field on real assignments. I would like to see a structured comparison between the III and the 5D though.

There's no substitute for real comparisons of real images taken with flawless methodology and technique.

When one considers the susrprisingly large variation of lens performance amongst the same lens models and even variation between sensors in the same camera model, then it's absolutely essential to have comparison shots of the same scene taken with impeccable technique, which not only means comparisons using the same lens and exactly the same FoV, but the same lighting, the same focussing, the same exposure and f stop and, of course, tripod, mirror lock-up and remote release.

Before the Rob Galbraith forum closed, I was involved in a lengthy discussion comparing the image quality and resolution of the 5D with that of the 1Ds2. The discussion was heated and polarised with some photographers claiming the 5D actually produced better images. Others claimed there was no significantly relevant difference and Jack Flesher who owned a 1Ds2 but had only borrowed a 5D for a few shots, claimed there were subtletlies in the 1Ds2 images which could only be revealed through expert printing and that such differences could not be revealed on the internet.

As far as I was concerned, the thread was a complete debacle. There were few camparisons of images demonstrated and possibly none that were rigorous. Most arguments seemed to revolve around personal prejudice and bias. The matter was unresolved in my view, and then the thread was subsequently closed for no good reason.

Michael's 'Measuring Megabytes' DVD provided some comparisons between the 5D and 1Ds2 that appeared to indicate that the 1Ds2 was capable of recording slightly finer detail (as one would expect), but those comparisons were also flawed because the focussing was not exactly the same with each camera. (Sorry! Michael ).

This is the first time that I've come across the claim that the 1Ds2 produces less satisfying skin tones than the 5D. It might well be the case. If it is, then someone please demonstrate it. Put your mouth where your money is.

I want to know if this characeristic, if it exists, is a peculiarity of an uninformed public familiar with grainy film that it is still (subconsciously) using as a reference point, or whether there is a 'real' plasticky effect that this relatively unbiased and particularly objective photographer can discern.

Funny how memories are so different about recent history...

Ray, if I am recalling the thread over on RG correctly, another poster proved with image crops that the 1Ds2 revealed more detail than the 5D. I claimed the detail difference shown -- mostly detail in distant trees as I recall -- would be revealed in a GOOD print. I don't recall using the word professional, though I did offer to have anybody stop by my studio and we would do the comparison and make prints on my system. FWIW, nobody took me up on that. As I recall, you were one of the lone voices that claimed it was an insignificant difference IN YOUR OPINION. Micheal, Bill and Charlie showed it again in their mega-comparison, but again in your opinion they failed. In my opinion they did a stellar job with as near to ideal capture conditions as possible. So I do agree with you when you claim "most arguments seemed to revolve around personal prejudice and bias" ...
~~~

Now that that's out of the way, back to the topics at hand. First, I've never heard of the skin-tone issue either and never had them with my 1Ds2 -- sounds like processing issues Edmund. Color was not as good out of the box for skin unless you had a good profile or calibration, but that is a separate factor. As for direct comparisons, Uwe and I did an INFORMAL shoot with the 5D, 1D3 and new Fuji. You can DL the raw files and process to your heart's content here: http://www.outbackphoto.com/reviews/equipm.../1d3_5D_S5.html. The image on the 5D is slightly larger as is the sensor, so this required us to use a different zoom setting on the 24-105 with each camera, a choice of lens and usage factor that will eliminate the comparison as flawed from the purists in the group, so if this is you, read no further...

My short conclusion regarding the 1D3 and 5D is that they are essentially equal as respects image detail. With my current processing which is refined and includes a custom calibration for the 5D, and not at all practiced with the 1D3, the 5D showed more shadow detail. IMO with perfected processing, each is essentially going to be the same on detail, color and DR. Live view is a very cool feature -- with the magnification, you can live focus on the LCD just like a view camera. The added benefit is the mirror is already up for taking the image -- so yes, we finally have one-button mirror up, though indirectly. Speed is stunning, and practically a joke for my limited needs, but sports shooters are sure to relish it -- it is essentially 10MP-per-frame movie capture. The 5D is lighter and smaller than the 1D3, though the 1D3 is notably lighter than previous 1 cameras. Ergos are great with either if you are used to and already like Canon ergos. The 1D3 has added some bracketing refinements, so now it's easier to shoot a series for HDR blending with a single shutter press, another nifty feature. Individual lens autofocus calibration is a huge new feature, and the promise of no more sensor dust could be huge if it really works. Personally, I don't need the speed, don't have to have live preview and am not bothered by having to manually execute a series of exposures when I need an HDR set, AND my 5D's are already paid for, so for now I see little reason for me to upgrade. If I were buying new, the 1D3 would probably get the nod for its added features and 1-series durability.

Cheers,

Jack,
The extra detail in the distant trees appeared in some other sample images on LL in the very thread where I pointed out the focussing discrepancies in relation to the 'measuring megabytes' tests.

I have no argument that the 1Ds2 can provide more detail than the 5D. The issue revolves around the relevance of that difference and the apparent reluctance or incapacity of seasoned photographers to demonstrate such differences.

Okay, so we agree and can let this one go...


Agreed. And I'll add that "relevance" is in the eye of the beholder...


If you are referring to me, I offered a direct print comparison and nobody took me up on it, so any "reluctance or incapacity" was not on my end...

Cheers,

Jack,

I explained my position on the 1DsII in detail to Canon Inc's Ohara-san. He had an interesting comment: "We know that in the end the quality of the image depends on the quality of the pixels". I would hope therefore that history in terms of the original 1DII and 1DsII will not repeat itself. As for my ability to process my Raw imagery , I guess it's probably equivalent to that of everyone else here who has spent a few years doing it. Of course, Adobe doesn't send me freebies anymore, so I'm still stuck with CS2, but then as you know I've been increasingly forced to use C1 of late



Edmund

Enigmatic — on both your part and Ohara-san's. ;-)

Nill
~~
www.toulme.net

True. I remember that. It would have been interesting if someone had sent you a couple of portraits for print comparisons.

However, you will recall my skepticism regarding your argument that such differences can only be revealed on a print, are invisible on a monitor at any degree of enlargement and cannot be communicated over the net.

Cheers!

Thanks for making these files available Jack! The difference in detail indeed is minimal. Reminds me a bit of the difference between D2X and 5D in detail. Apart from the detail of the 5D I like it to be a bit more crisp than the 1DIII file. Would a sharper lens such as the 24-70 yield a bit more detail in the 1DIII do you think?

Kind regards,
Bart

You know, there comes a point where all these pixel-peeping concerns get very tedious. I'm losing interest fast.

Show the images. Sorry to shout, but I'm not interested in endless waffle.

I am not sure the 24-70 would give a different result than the 24-105 but I am curious about how the two would look using one of canon's stellar L optic --- like the 85mm 1.2 or 300/400/2.8 L IS.

Hi Mark: The problem with using a good L prime is the different sensor crops make the image coverage different and difficult to compare directly -- hence our use of at least an L zoom

Bart: To be more specific to your question, and I think Mark has answered it, the 24-105 is probably as sharp as the 24-70 at f8 where we shot -- the bigger advantage to the 24-70 is it has less distortion. The 1D3 does have a smaller pixel pitch, so in a raw resolution comparison with a sharp-enough lens it should show more detail at the pixel level, but not necessarily at the full image level.

Cheers,

"Show the images" ??? Uh, I gave you the link to download them for yourself earlier in the thread!

I totally appreciate that point --- just like with our scientific work there is always someone who wants to be critical

Indeed -- clearly holds true in the camera gear testing world too!

,

I think Rob G has some addendums to his initial enthusiasm

I have been shooting with the ds1mkII since it came out and ONLY shoot models (fashion/beauty) I have shot over 200.000 images (na d a new shutter at 134.000) with the ds1 and never once had a feeling the skin texture looked plastiky..until I intentionally retouched the hell out of them.
I have to day I like my leaf 65 better on skin..but I thought that had to do with it being 16bit instead of 12. I still use my ds1 as a back up...or when I am sick and tired of the shitty focus on the MF camera....and I show a mixture of ds1 and leaf 65 to my clients....they don't even notice the difference and even used the ds1 images for billboards (up to 48ft in length).

I guess is depens a lot on your post production.

Did you ever shoot a 5D on the line and compared it with the 1Ds pics? I guess it's currently between 5D and 1Ds for flawless image quality? The 1DIII should have image quality as good as or even better than out of the 1Ds since the amount of photons is about the same and even are larger, so more sensitive, but it appears not to me. Although the 1DIII files look almost as detailed as the ones out of the 5D, the 5D files look a bit more real to me. Guess the only way to find out is to shoot both cameras from a tripod with a prime set to infinity to rule out any focus variability. Thanks all for your takes on this topic.
Kind regards,
Bart

Sorry! I missed that. Must be going blind with all this pixel peeping.

Trying to open those files has been a frustrating experience, especially considering I'm on a 56k dial-up connection. The latest ACR 3.7 for CS2 does not appear to support the 1D3. I can no longer use the beta version of CS3. Haven't yet bought the upgrade.

I decided I might as well uninstall CS3 since it's no longer of use. Unfortunately doing so affected the operation of Bridge in CS2. I tried repairing the installation using the original CS2 software, which worked, but brought me back to version 9.00, so I had to download the 9.02 CS2 update and Camera Raw v3.7.

So several hours later, I have both raw files, I've successfully uninstalled CS3 and CS2 is back to normal, but I can't open the 1D3 file.

Who was it who said life is not meant to be easy.

Unfortunately, that's no guarantee for anything being in focus at all.

It's better to take pictures in a controlled situation, e.g. a studio, and take test pictures to check where the field of focus is.

Download the DNG converter and use that instead.

Well, shooting at infinity at least tells you about how the pixels compare. Not that hard to point a camera towards anything within the infinity range.
Shooting in a studio can only be reliable if the lens is focussed, then taped to secure that it won't change focus, and then mounted on the other camera. (DPreview.com has several comparison pics that aren't focussed at the very same point, so, sort of pointless.) This won't give the same image on a cropped and a full size sensor, but will at least give you an idea of how the sensors compare focussed at the very same point.

It's the issue with doing gear tests for the shooting public... No way to please everybody because everybody has their own ideas about what needs to be done to make it a valid test, yet no concensus on what those test parameters should be is ever reached. It's why I recommend everybody tests for themselves and personally don't try to share anything approaching scientific anymore. I simply shoot an easily repeatable "real world" scene with a range of detail and lots of bright highlights and deep shadows, and toss a Macbeth color checker in for reference. I still share my impressions but allow everybody the opportunity to draw their own conclusions from my results if they feel so motivated. So far this approach seems to create less turmoil...

,

And how do you propose to make this a reliable, repeatable test?

Are you aware that "infinity" on the lens doesn't necessarily mean that the point of focus will be the same from hour to hour, depending on e.g. temperature?


Shooting test shots for resolution can only be reliable if the lens is focused. Period. It has nothing to do with whether you do it in a studio or not, but in a studio, it's easier to control than outdoors.

I think the issue here, Jani, is that the closer the test target the greater will be the significance of any degree of misfocussing. Focussing at infinity using autofocus is probably going to allow greater accuracy with less hassle.

On the other hand, distant objects are not necessarily good targets for comparing resolution, due to often lower contrast.

Another issue which might complicate matters is the fact, apparently, that all lenses do not necessarily exhibit equal performance at all focussed distances. Lens 'A' might be betterthan lens 'B' at infinity whereas lens 'B' might be better than lens 'A' at a close distance of 2-3 metres, at the same f stop.

Having tested a few of my own lenses, I'm acutely aware that the slightest degree of misfocussing can ruin the results, particularly at wide apertures. For this reason I prefer to use test charts with progressively narrower lines and manually focus if necessary. On some part of the chart, and at an appropriate distance from the chart, the lines will be spaced so that they will cause aliasing and color aberrations which are clearly visible through the viewfinder.

When these artifact are at their most obvious, this represents the most accurate focussing I know of. Turn the focussing ring by the slightest degree and you can see such colorful aberrations diminish and/or disappear entirely.

It's also interesting that sometimes when using autofocus on such a chart (Norman Koren's is ideal) the autofocus will appear to be 'spot on' the first time the shutter is half depressed, slightly off-focus the second time and back on focus the third time.

Indeed, shooting at a fixed focuspoint on the lens gives greater accuracy and less hassle. If one doesn't wait for several hours (?) to mount the lens on both of the two camera bodies, chances that the focus will be exactly the same, are at least very high. The second reason to use infinity is because most lenses are made to work best at infinity. But even mounting a cheap close-up lens upon the used lens wouldn't ruin the test, since the test isn't about testing the lens, but the bodies/ sensors. Such a test could show which one of the sensors shows the most crisp pixel/picture without sharpening, and if the 10mp sensor will yield as much resolution and crispness if uprezzed to 12mp. Personally I think not, but who knows. Until now I haven't seen this happen yet. At least, with the same brand. I know of many D200 users stating it yields the same detail as a D2X, but if you read through the tests and look at what you are actually seeing, 12mp will give you a visible advantage over 10mp. Obviously this won't be seen in print at double truck size, but it will if enlarged for advertising purposes.
From what I have seen, a 5D 12mp does show some more resolution than a D2X 12mp both with the right lenses. But it's little, about as little as the difference between 1DIII and 5D. This means that unhappy D2X users could easily choose for the 1DIII and keep their current resolution, that is, if the camera focusses equally well

That is unless you need to perform microadjustments to the focus of the camera, because of sample variation ...

And how can you know that without checking the focus manually, anyway?

These debates, while informative to a degree, are akin to arguing over whether one woman, painting,photo or exotic car is more of less beautiful than the other. Beauty, the 'ideal' skin tone ( 'ideal' or photo-realistic?), etc are all in the eye of the artist/client.

The best bet is, ideally, is to see/take side-x-side ISO Y shots from each body, same lens, etc etc (good luck) and then make up your own mind which one you like best -- or best suits the desires of your client(s).

That being said, the 'strength' of an AA filter will have an effect on the degree of skin detail and subtle shade differences recorded.

Personally I think AA filters have been made far, far too robust of late. I mean, just how big an issue is moire going to be for 9/10ths of us? I feel like having a bumper sticker made up that reads "USM sucks".

Also, don't kid yourself. The greater a camera's noise-free ISO capability, the greater the in-camera signal manipulation that is done to produce those images -- manipulation you have NO control over aside from the ISO selection. Sometimes that manipulation makes an image, sometimes it breaks it, sometimes you just need to live with it.

In short, until they come up with a low-cost (telephoto-capable) camera system that ideally replicates the human eye and our in-brain image processing capabilties, photography will remain an art of compromises. There ain't no such thing as a free lunch.

I think you have over-estimated the qualities and capabilities of the human perceptive system.

The following quip applies equally well to that:

I can find no rational explanation for the perception that skin tones look more plasticky with the 1Ds2. It's been established that the 1Ds2 provides marginally more detail than the 5D. That marginally greater detail should also apply to human skin or any texture whether animate or inanimate.

Perhaps there's some reaction between certain brands of female make-up lotions and the 1Ds2 AA filter .

Moire is more of an issue for those who use high quality primes at their sharpest apertures. Landscape shots at f11 and f16 are less of a problem.

Wow! A lot of discussion since I've been away. But I have to say, I am so totally and thoroughly speechless by the comment that "The 1Ds2 and 1D2 are not good for people". I guess I'll have to throw away three years worth of images for clients.

That is truly the craziest thing I've ever heard Plastic people? Edmund, perhaps you are responding to the smoothness of tone and lack of grain. The only Canon I've owned that was a bit of a challenge rendering good skin tones was the original 1D. Even the 1Ds could be tough, because, in many ways, it has a lot less latitude for error that the Mark II and Mark III series. To me, it's all in the quality of the RAW pocessor and the "operator" of the software. If anything, the 1Ds2 often has "too much" detail on someone's skin. I find I usually have to retouch most of the pores, lines etc. It's almost like shooting someone with Velvia versus Provia. Less detail is often more flattering on people.

As for the 1Ds2 "providing marginally more detail" than a 5D. . . try A LOT more detail. I just finished doing work on images that were enlarged to 30" on the longest side, 400dpi, cropped, from the 5D. I've had to do similar things with 1Ds2 images. I found the 5D images wanting and falling apart long before my 1Ds2 images have under similar circumstances. If you check out the DPReview on the 5D, you'll find a healthy increase in resolution of the 1Ds2 over the 5D. Moire is a huge issue when you shoot fashion. It's one reason many fashion clients held out with film for so long. I think Canon answered with the stronger AA filter. If you look at the resolution charts, an older 1Ds actually almost out-resolves the 5D, but with moire at the extinction.

Basically, the cameras are designed for different markets. Give me any 1Dxxx over anything else, any day.

Nemo

Since I don't own a 1Ds2, all my comments on the differences between the 5D and the 1Ds2 result from images posted on the net and comments from people who own both cameras.

I got the impression from highly enlarged images on my monitor from Michael's "Measuring Megabytes' DVD and other images from the net, which represent really huge prints, greater than 30", that the differences were marginal.

If you disagree, show me some 100% (or even 200%) comparison crops that are easily downloadable.

For anyone who is interested, I downloaded Jack and Uwe's 5D and 1DMK3 raw files, processed them in DPP, then upressed each to 24 by 36 and printed a similar crop from the files. Peering at the prints from inside a foot, the 5D print clearly shows more detail to my eye, also reflecting the difference I saw onscreen. From normal viewing distance, the detail difference between the two prints is neglible or non existent. The colour looked quite close, but to my eyes, the 1DMK3 is slightly richer with greater depth, probably a result of the 14 bit vs 12 bit.

I owned both cameras, the 5D and 1Dsmk2 (just sold the later). In my opinon at low isos (e.g. 100) the 1Dsmk2 gives slightly more detail when the comparisons are done with a decent lens and careful shooting on a tripod but honestly there was not a huge difference. The 5D starts to shine at higher ISOs. So for 35mm sized sensors, i still think the 1Dsmk2 is king for resolution potential when shooting at the low isos. How much real world difference this makes is subjective.

Looks like there are significant focusing problems on the 1DMIII:
http://www.robgalbraith.com/bins/content_p...cid=7-8740-9006