Michael, 1Ds3 test image results yet?

[billh]

Hi Michael,

Apparently the 1Ds3 is beginning to be delivered to people. Would you mind sharing your impressions of the image quality (and AF performance if possible)?

Thank you!

Bill

“In all, the cameras compared were a Hasselblad H2 with Phase One P45+ back, a Hasselblad H3D-39 II, Sinar Hy6 with e-Motion 75 , and a Canon 1Ds MKII and a 1Ds MKIII. Whew.
We did still-life set-ups as well as a model shoot – testing all ISO settings, resolution, noise, frame rates and long exposures. It will take a week or so for me to collate the results, analyze them with Craig and his assistants, and prepare the results for publication. Watch for them. It should be fascinating.”

[michael]

It will be a while till I can write and publish results as I have a week of teaching and traveling coming up. But, here's the short version about the 1Ds MKIII.

It was very close to the three MF backs in terms of resolution. All three backs showed very fine detail that simply wasn't visible on the IIIs. The reason presumably being because of the AA filter. But I caution that it was only very fine detail that disappeared, and unlikely to be seen except by the most critical eye, and then only at 100% on screen or a very large print.

Of course at 21MP the files were smaller than the 33MP and 39MP files, and so again a bit more limiting in absolute print size possible.

Colour reproduction was very good; nothing to comment on.

Grain was far superior at ISO 400 (and above) to the MF backs, as would be expected. But, the real surprise was that compared to the 1Ds MKII noise at all speeds was comparable. Not better, but comparable. So even with the smaller photo sites Canon has been able to match their previous standard. Quite an achievement.

This is all from memory and therefore subject to change when I actually sit down and collate the results. (I had an all afternoon session at Studio One last week, processing the files and analyzing them with one of their senior capture techs, but the results are still in note form).

Michael

[billh]

Thank you very much for responding. You have given me a sense of what you saw compared to the backs, but do you remember what differences you saw between the III and the II?

Bill

[michael]

Definitely a crisper more defined image. Combined with the additional new features, a definite upgrade and improvement.

Michael

[ashdavid]

Definitely a crisper more defined image. Combined with the additional new features, a definite upgrade and improvement.

Michael
[a href=\"index.php?act=findpost&pid=157242\"][{POST_SNAPBACK}][/a]

Thanks Michael, the ability to produce sharp images was one of the features I was most concerned about.  

[MarkKay]

These are the kind of comments I have been waiting for.  From the limited examples and comparisions I have seen, I had a similar impression.   Looking forward to seeing the results in the future article....

It will be a while till I can write and publish results as I have a week of teaching and traveling coming up. But, here's the short version about the 1Ds MKIII.

It was very close to the three MF backs in terms of resolution. All three backs showed very fine detail that simply wasn't visible on the IIIs. The reason presumably being because of the AA filter. But I caution that it was only very fine detail that disappeared, and unlikely to be seen except by the most critical eye, and then only at 100% on screen or a very large print.

Of course at 21MP the files were smaller than the 33MP and 39MP files, and so again a bit more limiting in absolute print size possible.

Colour reproduction was very good; nothing to comment on.

Grain was far superior at ISO 400 (and above) to the MF backs, as would be expected. But, the real surprise was that compared to the 1Ds MKII noise at all speeds was comparable. Not better, but comparable. So even with the smaller photo sites Canon has been able to match their previous standard. Quite an achievement.

This is all from memory and therefore subject to change when I actually sit down and collate the results. (I had an all afternoon session at Studio One last week, processing the files and analyzing them with one of their senior capture techs, but the results are still in note form).

Michael
[a href=\"index.php?act=findpost&pid=157228\"][{POST_SNAPBACK}][/a]

[billh]

Here are three comparison images - I sharpened them and the rank seemed quite clear to me.

http://forums.dpreview.com/forums/readflat...thread=25835108

Hi All,

Here are full-res files from the Canon 1DsMKIII, Phase One P21+, and Mamiya ZD. All files were shot in RAW and processed to 16bit Tiffs. Then files were converted to jpg format for uploading to web. Camera mirrors were locked up for each exposure. No sharpening or noise reduction was applied.
Sincerely,

Ted

1DsMKIII:
Canon DPP v3.2
50MM f1.4
1/25 sec
f11
(ISO 100)

Phase One P21+:
Capture One v3.7.7
Mamiya 80mm f2.8
1/30 sec
f11
(ISO 100)

Mamiya ZD:
Mamiya Digital Photo Studio v1.3.0
Mamiya 80mm f2.8
1/15 sec
f11
(ISO 50)

http://www.tpedersen.com/1DsMKIII_P21_ZD/

[Christopher]

1DsMKIII:
Canon DPP v3.2
50MM f1.4
1/25 sec
f11
(ISO 100)

Phase One P21+:
Capture One v3.7.7
Mamiya 80mm f2.8
1/30 sec
f11
(ISO 100)

Mamiya ZD:
Mamiya Digital Photo Studio v1.3.0
Mamiya 80mm f2.8
1/15 sec
f11
(ISO 50)

[a href=\"index.php?act=findpost&pid=157451\"][{POST_SNAPBACK}][/a]

Thanks for the test but I think there are some major problems. First of all the Canon 50/1.4 at f11 isn't good at all. I would never use it at that range. Second view is much wider from the Canon so it does not suprise me at all that the P21 shows more detail, I mean it has a much smaler space to cover.

Still thanks for the images.

[billh]

Thanks for the test but I think there are some major problems. First of all the Canon 50/1.4 at f11 isn't good at all. I would never use it at that range. Second view is much wider from the Canon so it does not suprise me at all that the P21 shows more detail, I mean it has a much smaler space to cover.

Still thanks for the images.
[{POST_SNAPBACK}][/a]

Hi Christopher,

They are not my images. Ted Pederson posted them. I agree about the Canon 50mm lens - both about the lens and the aperture, but then again, I don’t know how good a Mamiya 80mm lens is ......I assume he chose them because they were both “normal” focal lengths for their respective formats. I am really curious to see what Michael says/shows when he has time to post his report. Since I use a 1Ds2, I’m particularly curious to see a comparison between it and the 1Ds3. It will also be interesting to see the difference in the final output files (sharpened, etc.) from the 1Ds3 and medium format cameras. I am also going to try a D300 Tuesday (I can’t find a D3) to see how the AF servo compares to the 1D2.

There was a post the other day mentioning his 1Ds3 was “hunting” between background and foreground -
[a href=\"http://forums.dpreview.com/forums/readflat.asp?forum=1032&thread=25831465]http://forums.dpreview.com/forums/readflat...thread=25831465[/url]

“The 2nd difference is more of an issue, and something I don't like at all. Because the Mark III has a larger and finer AF area, it is much more likely to pick up on some foreground item (perhaps a branch or a cable) than the major item you are interested in.
I ran into a very nasty "continual focus attempt and fail" problem with a number of shots. This happened to me with telephoto lenses. The camera can see something in the foreground, tries and focus on it but fails, so goes to the background and focuses as expected - but then sees something in the foreground, so tries to focus on the foreground and fails... and repeats this continually. Very strange, and very different to every other camera I've used. I tried the same photo with a Mark II and Mark III and got very different results - the Mark II would lock on to what I was expecting, the Mark III would keep hunting. I suspect this behavior would go away when you swap to a smaller AF area, and there is even a Custom Function to stop continual focus attempts. But not something I particularly like.”

[Mort54]

There's a lot to quibble with in this comparison, but still, the ranking seems obvious - Phase the best, then the ZD, and then the 1DsIII. Keep in mind that the P21+ is 18 MP vs the 1DsIII at 21 MP, and the ZD at 22 MP. The lack of an AA filter clearly helps the Phase (I don't know if the ZD had an AA filter). I do agree with others that the results are quite close, and the 1DsIII is certainly the most user friendly of the bunch. Tonality looks very similar on all three, and I don't really see any more shadow detail on the MF backs, which is an area they are known to excel in. If I was going for maximum image quality, I'd get the Phase. If I was going for maximum convenience, I'd get the Canon.

[MarkKay]

I am not sure how you rank the ZD second.  There may be some camera shake but as is I think this image is clearly 3rd

There's a lot to quibble with in this comparison, but still, the ranking seems obvious - Phase the best, then the ZD, and then the 1DsIII. Keep in mind that the P21+ is 18 MP vs the 1DsIII at 21 MP, and the ZD at 22 MP. The lack of an AA filter clearly helps the Phase (I don't know if the ZD had an AA filter). I do agree with others that the results are quite close, and the 1DsIII is certainly the most user friendly of the bunch. Tonality looks very similar on all three, and I don't really see any more shadow detail on the MF backs, which is an area they are known to excel in. If I was going for maximum image quality, I'd get the Phase. If I was going for maximum convenience, I'd get the Canon.
[a href=\"index.php?act=findpost&pid=157661\"][{POST_SNAPBACK}][/a]

[Gary Yeowell]

In that test to my eyes the Phase is the clear winner, the Canon a close ish second and the Mamiya a very clear third. If i had to chose one from this test, the Mamiya would not even be a consideration, regarding the Canon and Phase the choice is much harder. On purely ergonomics, ease of use and features, the Canon walks it, but absolute picture quality goes to the Phase without a question of doubt. For me i went with the 1DS3 as a travel shooter, i also love the ease of using T/S lenses but i do still miss my P20 on my 'V' Blad which in terms of colour differentation is still very hard to beat. As an example, look at the grass in the above test, this in my opinion is where the Canon fails every time, 'foliage', it's just crap! big blob of one colour green and no detail...... then look at the Phase, detail of each blade of grass and lovely subtle shades of green just like film.

[httivals]

I agree that the Canon is a close 2nd.  The Mamiya is way out of the running, and the moire is horrible on it.  For some reason, the P21 has substantially less morie than the Mamiya. . .  I disagree entirely about the greens.  I prefer the Canon greens to the P21 greens, in this image.  The P21 greens are much too yellow, given the overcast conditions.  I think there may be contamination from infrared; maybe the P21 doesn't have good infrared filtration (like the Leica M8)? . . .  This comparison is the first one that has made me reevaluate buying a 1DsIII.  I'd rather wait for a 5DII for the weight and size savings, but there's no doubt that after seeing this comparison my lust for a Mamiya AFD and a ZD back (or P25 back) is entirely cured.

[Gary Yeowell]

The greens in my opinion are always crap out of a Canon compared to my previous P20, they always render them as a blob of the same colour green without any shading whatever. I say this not to bash Canon, I have after all just got a 1DS3, but i still don't like the way it renders foliage. (Lovely at everything else though)

[cescx]

It is necessary to recognize, that canon has made a great effort, and that the results are excellent, very next to MF, and surely, superior in some conditions of shoting, where the speed of reaction, the IS and other technological advantages, prioritize by ahead of the resolution.

But it can that most positive, is the reaction of the MF manufacturers, when seeing itself reached about a product infinitely economic, hopes that he is positive for us and our current account.

[Quentin]

I'd rather wait for a 5DII for the weight and size savings, but there's no doubt that after seeing this comparison my lust for a Mamiya AFD and a ZD back (or P25 back) is entirely cured.
[a href=\"index.php?act=findpost&pid=157745\"][{POST_SNAPBACK}][/a]

To base buying decisions on the basis of a "test" like this is about as absurd as it gets.

Quentin

[CatOne]

To base buying decisions on the basis of a "test" like this is about as absurd as it gets.

Quentin
[a href=\"index.php?act=findpost&pid=157903\"][{POST_SNAPBACK}][/a]

Agreed.  It's one set of crappy pictures out a window in flat lighting of a non-challenging subject, with random lenses of varying quality. Sheesh.

[Ray]

Agreed.  It's one set of crappy pictures out a window in flat lighting of a non-challenging subject, with random lenses of varying quality. Sheesh.
[a href=\"index.php?act=findpost&pid=157925\"][{POST_SNAPBACK}][/a]

... and the test was clearly biased in favour of the two MF backs, but particularly the P21, by failing to match the FoV of all cameras in at least one dimension. Having cropped the 1Ds3 image to the same FoV as the P21 image I find the file size is reduced to around 43-44mb. Dividing by 3, I get an equivalent pixel count for the 1Ds3 of just 14.46mp.

The comparison is thus between a 17.23mp MF sensor with an aspect ratio of 4:3 and a FF 35mm sensor with aspect ratio of 3:2 cropped to 4:3 (which would make it 17.8mp) then reduced further in size to 14.47mp because FoV was not matched in any respect.

In addition, there's also a mismatch of DoF. The 1Ds3 should have been using f8 and all cameras should have been focussed on the same spot.

[Quentin]

I think even the most careful comparison is always going to be suspect given so many variables, but at least we can say that this test, with all due respect, is not a careful test,  although the 1Ds III is clearly going to be an excellent camera.  Once you start comparing different formats at these resolutions, factors like depth of field, lens quality, camera shake / vibration, diffraction limits etc are probably more important than the back / chip itself.  In other words, you end up with a system, format and software, not a chip, comparison.

[dseelig]

I shoot action assignment work, a few portraits and some landscapes. I can afford one camera and the zd is just too limited. Yes I would love the better wide angle but I make more money shooting action and for what I do if one camera is it it has to be a canon. The thing is what you do with a camera makes it a better choice for each particular person. I wish I could afford the new rollie leaf camera I cannot . I could get a zd but for me a canon is a better choice there are no absolutes here, what is good for me will not be good for everyone. Enjoy what you have and do not worry about what is best, just what is best for you, by the way I have a leica M8 for my walk around camera and love it as well. David

I think even the most careful comparison is always going to be suspect given so many variables, but at least we can say that this test, with all due respect, is not a careful test,  although the 1Ds III is clearly going to be an excellent camera.  Once you start comparing different formats at these resolutions, factors like depth of field, lens quality, camera shake / vibration, diffraction limits etc are probably more important than the back / chip itself.  In other words, you end up with a system, format and software, not a chip, comparison.
[a href=\"index.php?act=findpost&pid=157959\"][{POST_SNAPBACK}][/a]