dont know if you saw the added crops on the first page. I created a square image and uprezed all shots to the size of the square. I actualy posted some crops of that gravel, and you will not see much of a big difference IMO
I think the lack of a big difference is the surprise and certainly shines a more positive light on the 1ds3. I don't know about the mamiya lens but the Canon f1.4 is only fair and yet still holds it's own.
As for microcontrast take the files in to photoshop and do a little fine sharpening. The 1ds3 seems as if it might have a little additional detail available compared to the 1ds2
glad to see the 1Ds3 holding it's own against the MF cameras! What chip size are those cameras, do they have physically larger chips than the Canon? If they do, I'd say that the Canon is doing quite well! I never thought that the Canon 50 was super sharp, (I wouldn't call is soft though) but I guess there are probably sharper lenses available for the MF bodies as well. Looks like a good real world comparison :-)
Glad to see comparisons with no sharpening. I never sharpen mine till I'm at my print resolution. Might be interesting to see an 11x14" or 12x18" print ready file, sharpened and ready to go, to compare them as prints instead of digital files! :-)
mark1958 wrote:
The newer digital backs have almost twice as many pixels as the p21
that would make a nice comparison, see what type of difference there is for a mediumish sized print, like a 12x18" print, done with one of the newer backs and the 1Ds3
One point of reference for me, please. I know the 50/1.4 well. But I know nothing about the Mamiya 80/2.8. Is the Mamiya lens considered to be a good lens?
dcmiller wrote:
I really don't like these comparisons of "unsharpened" files.
When I provide comparison images, I always supply them with zero processing though I usually explain in the disclaimer that I am providing them this way so that each individual viewing the files can apply their own processing routines. Lord knows my CS2 game needs improvement.
To the group:
In CS2, the file sizes are recorded as:
1Ds mkIII = 5616 x 3744
ZD = 5328 x 4000
P21 = 4906 x 3682
Maybe I assumed too much, but I figgered the p21 would be at least 21 mp. From this list, doesn't it stand to reason the 1Ds mkIII would have more resolution all other factors being equal?
Another queston. Are the ZD & P21 backs crop sensors? Or rather what cam was used for the DBs?
Just did bit of research on my own. The P21+ is a 1.3x crop sensor according to the Phase One website explaining in part the narrower field. If I were considering MF, the P25+ would be my starting point. The ZD is full frame. For the 1Ds mkIII to be so much wider than the ZD, I am guessing that the lenses were not equivalent.
rfkiii wrote:
Maybe I assumed too much, but I figgered the p21 would be at least 21 mp. From this list, doesn't it stand to reason the 1Ds mkIII would have more resolution all other factors being equal?
Another queston. Are the ZD & P21 backs crop sensors? Or rather what cam was used for the DBs?
The P21 is only 18 megapixel and has a 44x33mm sensor. The ZD has a 22MP 48x36mm sensor.
The ZD is probably the weakest performing MFDB on the market. It says something that even the bottom-of-the-range Phase One back beat the Canon here, with less resolution. It would be more interesting to see a popular 22 MP MFDB back like the P25 up against the Canon. I expect the Canon would be beaten by a wider margin.
Of course the 33MP and 39MP backs are in a different league.
Won't someone shoot a model full body and then a macro of the head/hand (nail polish model style)...
That way we can have the best lenses, focused right up the the model, with the same exact lighting and then we can pixel peep around the eyes, and judge the colors of the makeup and skin etc..
I mean really, in the end if your shooting weddings you'll need a faster camera. If your shooting landscapes you'll buy what you can afford (large format if possible), and if your shooting fashion you'll want the best camera for the price/quality.. Which is where I think the whole 1Ds3/MF tests should be targeted towards..
I should add - it's remarkable that the Ds3 is even getting close with the NR requird, AA filter, etc. Stick a Leica or Zeiss lens on the Ds3 and the gap would narrow much further (micro contrast, etc).
I guess the next question would be where does the cell count on the 35mm format run out of steam w.r.t to noise and NR killing detail?
At what MP count are we going to end up with a FF D2Xs equivalent -- great pics - up to a midest ISO level?
At what MP count (or has that been reached already) are we going to sit back and go "... hang on a sec, who the @#$%^ needs a _____MP camera anyway...?"
httivals wrote:
Who would buy a Mamiya ZD after seeing this comparison? Even if there are user errors/flaws, there will always be some in practical use.
Since some asked about the comparability of the Mamiya 80/2.8 against the Canon EF.
I think it is on a similar level with a similar center resolution as the EF 50/1.4 and so of good use for this check.I had the Mamiya lens attached to Canon DSLR's with a Zörkendörfer rig, so I saw them side by side projecting to the same sensor.
The Canon is a Gauss/Planar type closely inspired by the same specification Contax lens.
The Mamiya also seems to be a Gauss type with 6 elements / 5 groups, instead of 7 elements / 6 groups.
(Just thought: Would be interesting to have this same lens on all three camera systems to be able to judge just one parameter: the sensor. Hopefully someone will do this while I reflect on skipping one Canon camera generation this time.)
tom in mpls wrote:
One point of reference for me, please. I know the 50/1.4 well. But I know nothing about the Mamiya 80/2.8. Is the Mamiya lens considered to be a good lens?
Yes, it's a "plastic fantastic". An inexpensive standard prime with good glass.
The 1Ds3 shot is the best sample I've seen so far. It's hard to compare the dynamic range; this is one are where I think the MF backs will still have a big advantage, plus cleaner shadow detail.
With regards to the ZD back, its perceived performance is highly dependent of the RAW converter. Raw Developer sounds to be the favored RAW editor amongst ZD owners.
The Mamiya 80mm F2.8 AF lens isn't Mamiya's best lens. Probably no better or any worse than a Canon 50mm F1.4. They sell on Ebay for ~$200.
John.. I also mentioned in my earlier post in this thread that I think the dynamic range is still going to remain much better on the MF backs. I agree with you that this image does not give a good test of that. In my experience with the 1DsmkII and 5D and the leaf aptus back, the leaf back gives much better DR when doing most landscapes. I cannot imagine the 1DsmkIII is much different than the 1DsmkII in this regard. On the other hand, I was impressed with the resolution with the 1DsmkIII. I would like to see the best glass tested as well.. and agree the 80mm vs 50mm are probably reasonably fair and comparable. I think Michael R's comparisons are going to be quite informative when he posts them on luminous-landscape. Of course if not soon, I will do some of my own comparisons..Next few weeks will prove to be interesting and generate a lot of discussion.
John Black wrote:
The 1Ds3 shot is the best sample I've seen so far. It's hard to compare the dynamic range; this is one are where I think the MF backs will still have a big advantage, plus cleaner shadow detail.
With regards to the ZD back, its perceived performance is highly dependent of the RAW converter. Raw Developer sounds to be the favored RAW editor amongst ZD owners.
The Mamiya 80mm F2.8 AF lens isn't Mamiya's best lens. Probably no better or any worse than a Canon 50mm F1.4. They sell on Ebay for ~$200.
I will say that one of my first thoughts on using the 1DIII was that Canon was getting closer to MFDB: Better, "cleaner" colors out of the camera, more graceful going to white or black (Significantly less blocking up of high DR images). I hope the 1DsIII is the same.
I felt the 1DsII lost some of the magic of the 1Ds. I'm someone who liked the 5D files better than the 1DsII files. The 1DIII files are clearly superior to 5D files, especially before processing.
The ZD files are very nice at high EV. Significantly sharper out of the camera than anything I've seen from Canon, including the 1DIII.
I will also say to that I was pleasantly surprised with the completeness of the 645ZD kit. Didn't feel like a work in progress at all. Charged the battery, put in a CF card and started taking pics. After reading some past evaluations of MFDB, I expected first use of the back would be like starting a nuclear reactor for the first time. But it was easy.
Hubsands ZD review is right on, IMO.
When Mamiya improves noise one stop, or increases throughput, they will satisfy a significant market share.
I tested a 1Ds2 against a Sinar 22mp back a number of years ago and in terms of resolution it was a very close run thing - this test is not a suprise. 22mp MF digital is not were it's at as far as resolution is concerned - if this is what you need from your camera then you need to look at 33 and 39mp backs which comfortably outperform these 22mp sensors.
I have an 80mm Mamiya lens in front of a P45+. It's clear that the lens is outresolved by the sensor at infinity - a lot of detail pops out after sharpening. Or else, maybe my back's focus plane is set off.
Even so, I would be surprised if there were any contest between this back and the new Canon.
