Edward C. Nemergut Photography

Canon 1Ds Mark II Experience Report

Last Update 10.1.05

Preface and Philosophy

Let me begin this report by saying that I do not plan to expound upon the numerous technical specifications of any camera. Cameras are tools—tools designed for one simple purpose: to capture and record light. I have never understood the pointless rambling that goes on in various blogs, forums, and chat rooms where miscellaneous individuals debate the merits of various cameras. Time is better spent developing one’s craft than memorizing the specifications of this year’s camera. I once read, “When Ernest Hemingway and F. Scott Fitzgerald met at a café in Paris in the 1923, they probably didn’t discuss typewriters.” I quite agree.

As such, I plan to make this report practical. What works for me? What doesn’t? How does the camera help me to convey my photographic vision? In addition, I plan to keep this report ongoing. Indeed, to write a report after any finite period of time (let alone a few days) would wrongly imply that I know everything about the camera and my opinion will never change. This is false—I’ve been around long enough to know that my opinion changes quite often. My favorite feature today might drive me crazy in a few thousand frames. I may still learn something new about the camera a year from now. Anyway: watch this space—I’ll add and edit as experience molds my practice.

Feel and Finish

Taking the camera out of the box, the first thing I noted about the 1Ds Mark II is the camera’s weight. I should preface this statement by saying that for the past 5 years I have used (almost exclusively) a Mamiya 645AF system. For those of you that might not be familiar with the Mamiya, the 645AF is an auto-focus, medium format camera and, at 3.0 pounds (according to my scale), is quite heavy in it’s own right. Indeed, I’ve been mocked innumerable times by photographer friends who can’t believe I hike with “that thing.” Nevertheless, at 3.5 pounds (without a lens) the 1Ds is heavier. (For the sake of further comparison, Canon’s new 20D weighs just over 1.5 pounds).

I know this may seem a silly thing to point out; however, this was totally unexpected. When I decided to make “the switch” from medium format film to 35mm digital, I never expected that I would be moving to a heavier system. I used a Canon 1n (almost) 10 years ago and I’m quite sure it was much lighter.

Despite its weight, the 1Ds Mark II is extremely well balanced and fits neatly into my hand. The fit and finish of the camera is extraordinary (think of the difference between the “feel” of new $50,000 BMW and used 1985 Chrysler Le Baron). The kit includes a neck strap; however, a wrist strap (which, as I understand it, had been included with the original 1Ds) is notably absent.

Menus and Operation

For the most part, the menus are logically arranged and relatively easy to use. I did not use the original 1Ds; however, it is my understanding that things are similarly set. That said, I’ve owned a Canon 10D for almost two years and have to say that I much prefer the 10D menu-selection system. With the 10D, to change menu item or adjust a custom function, I hit the menu button and use the rear dial to make my selection. After the selection’s made, I hit a small “enter” button in the center of rear-dial. With the 1Ds, one must hold down the “menu” button while turning the dial to reach the appropriate menu. Then, one must now hold to down the “select” button while again turning the dial. Selections are made only after one releases the “select” button and pushes “menu” again. All in all, it’s not terribly complicated and I’m certain that 95% of users will have it mastered in less than 15 minutes. Nevertheless, it does seem needlessly complicated. Perhaps Steve Jobs has spoiled me with the easy-to-navigate iPod.

Start-up

In the past, photographers have complained of (relatively) long "start-up" times with previous digital SLR's. My (now old) 10D took several seconds to "boot-up" when it was first turned on. Although the 2 or 3 seconds that it would take the camera "wake" seems, to some readers, a nominal delay, I can say with certainty that, in many situations a 2-3 second wait results in a missed photograph.

With the 1Ds Mark II (and the 20D, as I understand) Canon has solved this problem. Canon says that start up times range from 200-300 msec (that's 0.2-0.3 seconds). I haven't wasted time trying to test these specs; however, I can say, with certainty, that after turning on the camera, it's ready to take a picture by the time I raise the viewfinder to my eye.

Speed

My 1Ds Mark II manual tells me that I can shoot at 4 frames per second for up to 11 frames saving Raw files and up to 32 frames saving JPEGs. I haven't timed the camera, but near as I can tell this seems fairly accurate. Once I’ve filled the camera's memory buffer with my 11 Raw files, I can continue to shoot about 1 frame every second until the memory card is full. Those wanting to shoot faster should look at the Canon 1D Mark II--at 8.5 frames per second, you could almost call it digital video camera. (Well, not really, but you get my point).

For me, 4 frames per second is plenty fast. Nevertheless, the only fault I have with this speed is the fact that it is totally unadjustable. That is to say, I'm not sure why the user cannot adjust this speed to 1 or 2 or 3 frames per second. For example, in some situations--say, in the studio--I might want to shoot at 2 frames per second, but for a longer period of time. On the other hand, in other situations--say, photographing wildlife--I might desire the maximum speed of 4 frames per second. I'm not sure why Canon has not added this (apparently simple) software upgrade to their high-end digital cameras.

Now, it wouldn’t be fair for me not to note that there is no currently available digital camera in the world with totally adjustable shooting speeds. This is an artifact of the film-camera days where it was easier to construct a motor that only had to turn at one or two speeds. Nevertheless, in a digital camera, where there is no film to wind, having the shutter fire slower than its maximum speed shouldn’t be too complicated. Perhaps we can look forward to this in the Mark III.

I've received notes from several readers pointing out that the shooting speed is adjustable via Personal Function (not to be confused with Custom Function) 19, the Eos Viewer Utility, and a home PC. While I admit that I did not realize this when I wrote the above, it does seem to be unnecessarily difficult. Although I appreciate the fact that the speed is adjustable, it is a bit of a pain to hook the camera up to a computer everytime I want to change the speed. Can Canon make this a Custom Function in the future?

Initial Evaluation of Image Quality

One of the first questions I was eager to answer was: How will the image quality compare with the medium-format film I’ve shot for the past 5 (or so) years? Although this question seems to ignite a terrific emotional response among a number of pro-film and pro-digital photographers, I have to say that I am more or less apathetic to any particular process. I’m interested in results—what gets me a better photo at the end of the day. As such, I arranged a simple test.

Both of the following images were taken from a tripod mounted camera using mirror lock-up and a cable release. The film used was Fujichrome Velvia 100f. It was processed at a professional lab and scanned at 3200 dpi using my Minolta Dimage Scan Multi Pro. Canon Raw images were processed with Adobe Camera Raw (version 2.4-beta).

The Results...

Canon 1Ds Mark II with Canon EF 50/1.4 @ f/8. ISO 100

Mamiya 645 AF with Mamiya 80/2.8 @ f/8. Fujichrome Velvia 100f

Note the slight difference in the aspect ratio between the 35mm 1Ds Mark II and the medium-format Mamiya 645. Also, note the increased dynamic range offered by the digital 1Ds Mark II and the higher contrast and saturation of Velvia 100f.

Examine 100x200 pixel crops of the tower in the upper right-hand corner of the image. Both files have absolutely no sharpening of any kind. These represent minimally processed images


Canon 1Ds Mark II	Velvia 100f

Here, the Canon file is up-rezed to approximate the Mamiya file size.


Canon 1Ds Mark II, upsized	Velvia 100f

Wow! I was totally shocked when I first looked that these images on my monitor. The 1Ds Mark II solidly trounces the film-based Mamiya. Note the “pair” of diagonal steel support crossbeams near the bottom clearly resolved by the Canon. The Mamiya is only resolving a single bar (see below). Incredible.


Canon 1Ds Mark II, upsized	Velvia 100f

Now before everyone start criticizing me for not getting a drum scan, these on-screen images mirror what I can see on the light box using a 12x loupe! The Mamiya is only resolving a single bar!

That said, I do not print “minimally processed images” to put on my wall. I was curious about seeing how the Canon would compare with Mamiya after I had applied a degree of capture and output sharpening to both images. Below are 100x200 pixel crops of the bushes, downspout, and concrete foundation in the middle right of the image.


Canon 1Ds Mark II	Velvia 100f

Here again, the Canon file is up-rezed to approximate the Mamiya file size.


Canon 1Ds Mark II, upsized	Velvia 100f

Again, the Canon flat out-resolves the film-based Mamiya. Note the leaves of the bushes and the detail in the exposed concrete foundation. Also, note the increased noise (film grain) in the sharpened film image.

To compensate for this increased noise, I went back to the original image and used NeatImage to reduce the film grain before sharpening again. The same 100x200 pixel crops are shown.


Canon 1Ds Mark II, upsized	Velvia 100f

The grain-reduced image clearly has less noise with very little (if any) loss in overall resolution; however, it is neither as clean nor as sharp as the Canon 1Ds Mark II.

Although I’m not going to display all the results here, I ran similar comparisons using different lenses at different apertures. The end-result was always the same—the Canon clearly out-classes the Mamiya, and beats film. In addition, I compared Astia 100f shot in my studio to the 1Ds Mark II and again the film-based Mamiya was repeatedly trounced by the digital Canon. Finally, I printed a number of different images (portraits, landscapes, etc...) using my Epson Stylus 2200. It’s impossible to display the images here; however, I can clearly say that at any size greater than 8x10, the image produced by the digital Canon is clearly better.

Preliminary conclusion: the Canon 1Ds Mark II is a fantastic machine and my Mamiya system is sold.

I will continue to add to this report with new observations as I learn more about the 1Ds Mark II. For now, I’m extremely pleased with the Canon 1Ds Mark II and can’t wait to learn more about it…

Answers to comments received via e-mail

Since posting this review I have received a great deal of e-mail. Some comments/questions and my response:

This is not a fair comparison. The appropriate test is to compare the Canon 1Ds with drum-scanned film.
While I am certain that a drum scan conducted by an professional lab (such as West Coast Imaging) would have yielded sharper results, I believe this comment is invalid for two reasons:
- Of greatest significance, when I look at the images on a lightbox with a 3.5, 8x, and 12x loupe, I see very little difference between the image produced by the Minolta scanner and the image I see through the loupe! I figure that if I can't see much difference with a 12x loupe, the drum scanner probably can't either. Obviously, there is no way for me to post online what I can see with my own eyes; however, I welcome anyone to to conduct a similar test. Send me the results, I love to see them! Again, based upon what I can see with a loupe, I agree that the drum scanner would likely produce a better image, but it would still far short of the image produced by the Canon.
- I'm not sure how relavent it is (at least for me) to compare drum-scanned film as drum scans are extremely expensive and it is not my habit to routinely get one. I find that, 95% of the time, the file produce by the Minolta from 120 film is good enough for exhibition and sale of prints sized up to 20x24. Therefore, for me, (and most photographers) the comparison above is valid because it is an accurate reflection of my normal workflow.
The comparison is not fair because you did not use Velvia (or Provia 100f or Astia 100f or whatever)...
- While I agree that each of the films above may produce a "better" image for variable reasons (i.e. Velvia has more accuance that the Velvia 100f (at the expense of having more grain), Astia 100f has less grain than Velvia 100f (at the expense of less accuance), etc...) I think it is very difficult to argue that any film is going to produce a better overall image and that any such difference is going to "make up" the difference with the Canon. All the films above are excellent ISO 100 E-6 films--films most people consider to be the best of what's currently available. The Canon is clearly better (no pun intended) and what type of film I use to make a comparison probably doesn't make much difference!
This comparison only shows that the 1Ds Mark II is better than the Mamiya 645 AF. The Contax (or Pentax or Rollie or Hasselblad or whatever) would have bettered the Canon. The Mamiya lenses are known to be inferior.
- While I agree that, technically speaking, I should restrict my conclusions to the Mamiya 645 AF system as I did not test every medium-format system available, I think it is valid to extend the conclusion to medium format in general. I have owned a number of other medium-format systems in the past and can personally attest to the quality of the Mamiya lenses. Granted, there are lenses that stand out in particular: the Mamiya 645 AF 210/4 and the Contax 645AF 120 Makro (for example) are excellent lenses. But, overall, I believe it is hard to say that, one lens system is always better.
  That said, the test above was carried out with a $300 Canon consumer lens (the 50 f/1.4)--not Canon's best lens by a long shot. Even if a $3000 Zeiss lens can outperform the Mamiya and equal the Canon, you are now talking about spending 10x the amount of money for equal quality--not a good deal, in my opinion. Furthermore, I have done comparisons using Canon "L" glass (the 35 f/1.4L, for example) which produce even sharper results than the 50 f/1.4.
Why not wait for the Mamiya ZD? The ZD or any of the currently available medium format digital backs will better the Canon...
- This is a very interesting question. First, I should say that I agree completely: Medium format digital is better than 35 mm digital, just as 120/220 film was better than 35 mm film back in the "old days." That said, I made my decision partially based on the fact that I do not expect Medium Format digital to be around in 5-10 years.
  Moore's Law is brutal. In two years, Canon's next camera will equal the MF backs of today. Of course, there may be a new MF back by then too--but maybe not: Canon has sales of copiers, printers, fax machines, scanners, and a billion consumer digicams to pay for R&D while Sinar, Phase-One, Mamiya, and Leaf do not. Canon owns its own CCD and CMOS factories. Phase-One and Leaf (and Mamiya) have had to out-source their CCD manufacture. This may seem insignificant, but it is not--the technology industry is different than the traditional photographic industry. Ask anyone in the software industry how hard it is to compete with a massive company like Microsoft or Dell that have the R&D, the cash, and the factories to make things happen. Granted, MF back demand outpaces supply right now (as Reichmann has pointed out) but what does that really mean? On one hand this is great for Leaf and Phase-One--demand outpacing supply means higher prices. But, on the other hand, it is an indicator that MF backs can't win: a supply problem is another indicator that they lack the infrastructure needed to compete. They may have a better product, but if they don't have any to sell, they are not making the cash they need to build next year's chip.
  Business philosophy aside, a medium format back costs about $20,000 more than a 1Ds right now. (Rumors have it that the Mamiya ZD will change that, but for now we don't have official pricing on the ZD). Even if I had $20,000, I'm not sure medium format's edge in quality is necessarily cost effective. In the meantime, I'm much better off to spend the money I save by not getting a MF back traveling to India and Africa! In ten years, the fact that I was able to take the trip will be more important than whether the photo has 12 or 14 bits or 16 or 22 megapixels!
The differences observed can be entirely attributed to different depths of field. You cannot compare an 80mm lens to a 50mm lens, etc...
This is an inappropriate comparison. The appropriate compaison is an 11x14 (or 16x20 or 20x24, etc...) print from both cameras.

Questions?

E-mail me.

Other Reviews

Navigation

Photgraphy Home

All photography on this website is best viewed on a calibrated monitor (gamma 2.2) and at least 1024x768 resolution. If your monitor is calibrated correctly, each of the 16 boxes below (from black to white) should be clearly visible.

ecn@nemergut.com