Canon EOS 1Ds
A Preliminary Field Report
This is the fifth and final part of a multi-part field test of a pre-production Canon EOS 1Ds. If you have not yet read Part One, the introduction to the piece, please do so first as the entire review will only make sense once you've seen all of the parts in the correct order.
There will be a segment in The Luminous Landscape Video Journal's next issue (Issue #6, scheduled for release in November) showing the camera being used on a landscape and nature shoot and an analysis of prints produced with the 1Ds.
Days 6 & 7— Monday, September 30 & Tuesday, October 1
Though I have split this report into multiple parts to avoid slow loading pages, nevertheless you're going to find that this page is slow unless you have a broadband connection. But, since I need to display many test examples a short wait is the price you'll have to pay.
Jumping to Conclusions
During the last 2 days that I had the 1Ds for testing I was able to spend them shooting landscapes in the Bruce Peninsula and Muskoka regions of north central Ontario. I was very fortunate to have been able to have photographer Thomas Knoll accompany me on this shoot. Thomas works with a Canon D60, and so we were able to compare usage of these two cameras as well as images.
Canon EOS 1Ds with Canon 70-200mm f/2.8L IS. 85mm @ ISO 200
Thomas was also an excellent companion to have along in evaluating images from the 1Ds as he is the author of Photoshop. Thomas wrote the first version of Photoshop himself and has been a member of the Photoshop development team at Adobe ever since, working on enhancements and additional features. Needless to say, Thomas knows a thing or two about digital imaging, and as a passionate photographer since childhood contributed real-world photographic experience as well as his considerable technical expertise.
I mention this because it was very useful to have a second set of experienced eyes, along with Video Journal director Chris Sanderson — who also is an experienced photographer, and who added a third perspective to our evaluations.
I want you to know this because it illustrates the perils and pitfalls of testing, even for those with considerable experience. Thomas, Chris and I have between us about 75 years of experience as photographers, and some pretty good background and expertise in digital imaging. Yet it was only in the last hour of the last day that we realized that we had been mislead in one of the critical aspects of our image evaluation.
Here's what happened.
We had done a lot of our discussion and evaluation of the 1Ds' image quality, especially resolution, based on the photograph of the Toronto skyline that I had shot the first day with the 1Ds, 35mm film, and in medium format on a Pentax 645. You can read about this in section one of this report.
We had looked at various parts of this image, both in large prints and on-screen, comparing and discussing how each system had rendered detail, grain / noise, colour and so forth. We were especially taken by the way in which the 1Ds appeared to reproduce detail in the texture of the red building. (Relative size cropped enlargements from each of the three camera are shown below. These are the same frames found in Part One.)
This texture was clearly visible in the print from the 1Ds, while the prints from the 645 and 35mm showed the building to be almost uniformly solid red. Since the texture in the 1Ds print looked like vertical siding, and since I know the building has vertical siding, the natural conclusion was that the 1Ds was providing higher resolution.
Canon 1Ds Crop — Actual Pixels @ 72ppi
Canon 1V / Provia 100F Crop — Actual Pixels @ 72ppi
Pentax 645 with 200mm f/4 Lens on Provia 100F — Actual Pixels @ 72 PPI
What you'll see is that the the way that the 1Ds reproduces this wall appears to be quite a bit "crisper" on screen. On large prints, frankly, it clearly appears to show higher resolution than medium format.
This turns out to not be so. After some very careful last-minute scrutiny (just to be sure) we realized (it was Thomas who first saw it) that the apparent higher resolution of the 1Ds was in fact caused by a form of artifacting. Somehow the frequency of the line pattern on the building was "beating" with that of the chip, and so what we were seeing was a sort of doubling of the texture width with this particular subject. This texture width spacing was just enough to be resolved by the printer, while the correct texture width which was actually visible on screen in the 645 scan was too fine to be resolved by the printer. A fluke — but it really threw off our judgment.
In short, whereas we had initially decided that the 1Ds was superior to 645 format scanned Provia 100F in terms of resolution, our final conclusion is that while it's very close, medium format still has a slight edge in this area. Not huge, but there. Needless to say 35mm film was left in the dust.
But, having said that, don't you jump to conclusions. Resolution is just one of the factors to be considered when evaluating image quality. In terms of noise / grain, the 1Ds is vastly superior to film, and this plays a huge role in subjective image evaluation. Read my final conclusion at the end of this report before deciding anything for yourself.
Long Exposures and Noise Reduction
Like the Canon D30 (and unlike the D60) the Canon 1Ds uses dark frame noise reduction. This can be turned on and off as desired by the user. When turned on, after any exposure of more than 1/15 sec the camera takes a second blank exposure of the same length and then "subtracts" it from the actual image to eliminate identical noise that appears in each. This can work quite effectively, but of course it means that exposure times are doubled.
Canon EOS 1Ds with Canon 15mm f/2.8 Rectilinear Fisheye lens @ f/3.5. ISO 400.
This photograph was taken with the Canon 15mm f/2.8 rectilinear fisheye lens. Incidentally, I used a program called ImageAlign to remove the barrel distortion that is a standard byproduct of using this lens. (A review of ImageAlign is coming up on the site shortly).
The exposure was for 1 minute and 13 second. I was fortunately able to work with the RAW file using an as-yet unreleased RAW converter that is able to open 12 bit RAW files produced by the 1Ds. (I don't yet have Canon's latest software for the 1Ds). On the original file one can see just a few bright stars. But working in 16 bit mode (the file is actually in 12 bits) I added 3 stops of exposure compensation. This was sufficient to nicely bring out the Milky Way. Yet, the amount of noise that this added is quite acceptable. There is also some minor banding visible. (The ambient temperature was about 18C (65F). At lower temperatures banding and hot pixels would be significantly reduced).
100% Crop — 73 second exposure
100% Crop — 268 second exposure
The 100% crops above are from the slightly over one minute exposure that's reproduced full-frame and also from another frame taken at the same time, also at ISO 400, but this time for about four and half minutes. Of course the star trails in that frame are much elongated due to the longer exposure. In this frame you can start to see "hot pixels". These too can be fixed with software.
I can't really give you a definitive low light comparison with the D60 because after a few exposures clouds started to roll in and it was time for bed. But I think it's fair to say that the 1Ds can do a reasonable job in this area. Once I have a 1Ds on a longer term basis I plan on doing some additional tests which will be reported on these pages.
Odds & Ends
Here are some thoughts on a variety of small topics related to the 1Ds' features and functions. These are in no particular order.
The ability to voice annotate files is very handy. Now, since the Canon 1D and 1Ds have a microphone built in, how about adding some circuits to permit voice control? I can't think of anything handier when working in a fast paced situation than being able to train the camera to recognize certain commands such as "spot meter", "aperture priority", "decrease exposure two stops", etc, etc. The camera when at eye level is already being held close to ones mouth, and the additional firmware and processing power to permit this shouldn't be too onerous. Be careful though. Don't let any other manufacturer hear of this idea.
The camera has 3 frame autoexposure bracketing, as do many film based as well as digital cameras these days. That's fine, but when shooting with a digital camera what I'd like to be able to do is shoot just two frames, say about 1.5 stops apart. This will give me extra highlight and shadow detail and be just what's needed for exposure compositing. Not having that third redundant frame will save a lot of space on disk.
Kodak announced that one of the features of their new DCS Pro 14n is that there is a sensor built in that informs the camera if its being held horizontally or vertically, and which rotates the images according. Very neat. Canon should incorporate this feature as well. It could easily be done by determining which of the shutter releases was pressed.
Photograph by Thomas Knoll
Chris & Michael Discussing the 1Ds
So, here we are at the end of 6 days of testing the 1Ds. What are my conclusions?
Firstly, a few things need to be said. This was a pre-production camera. A few firmware features were not implemented in the version that I was using. This includes image magnification on the built-in colour LCD. This will of course be part of the final firmware on shipping cameras. I have evaluated image quality based on what I've seen with this sample camera, but it's usually the case that full production cameras are somewhat better than these early samples. I also have not had a chance to evaluate a number of secondary items like the Firewire interface and the final RAW conversion software,
Having said all that, I feel that the 6 days that I had to use the camera have given me a pretty good idea of what its capabilities (and shortcomings) are. When I reviewed the Canon EOS 1D last year I was taken with its solidity and ruggedness. The 1Ds is identical in this regard, and in terms of handling. This is a camera that I feel I could take on any assignment and it would survive whatever I put it though as well as any camera could. Also, since I am a long-time Canon user I feel totally at home with the controls and layout. Someone coming from another system might take a while to really become familiar with the highly complex but ultimately usable interface presented.
Image quality is of course what everyone is most interested in. Without descending into a detailed analysis (which I've already done in the 5 pages presented thus far) the easiest way for me to summarize is to say that the 1Ds produces the best combination of resolution, colour accuracy and low noise that I've yet seen in a digital camera. (You'd have been surprised if I'd said otherwise, right?)
What about a comparison with both 35mm film and medium format? I'm afraid that film has definitively lost the battle. The 1Ds's full-frame 11MP CMOS sensor produces a 32MB file — as big as a typical scan. But this file is sharper and more noise free than any scan I have ever seen, including drum scans. There simply isn't a contest any longer.
The 1Ds also fares very well against medium format. Is it sharper than 645? No, not quite, but really very close. When you add in the extremely low noise of the images compared to scanned film, and add in all of the cost and workflow advantages of shooting digital over shooting and scanning film, in my opinion the 1Ds is to be preferred. I'll gladly take the huge reduction in noise (grain) over slightly lower overall resolution any day of the week. Thomas and Chris basically have concurred with this finding after reviewing many sample images that we shot together.
Since I'm primarily a medium format landscape and nature photographer this presents me with a quandary. When I get a 1Ds (and my order is already placed, with deposit), what do I do with my medium format gear? What do I take with me on a shoot from now on? Should I sell my MF equipment?
I haven't made up my mind yet, but when I think about the advantages of shooting with my wide range of 35mm lens, including tilt/shift, high speed and IS lenses, as well as several very high quality zooms and super-telephotos, I'm hard pressed to see the advantages of staying with medium format film any longer. Of course medium format digital backs as they become more widely available and reasonably priced may well shift the balance the other way. We'll see.
Then there's the matter of price. As this is being written in early October the $8,999 price mentioned by Canon U.S. in their Photokina press release has gone by the boards. The camera definitely will not be priced that high. What will the final street price be? That's anyone's guess at this point, but my sense is that after the shock of Kodak's 14 Megapixel full-frame DCS Pro 14n, and it's $4,000 street price, Canon will become competitive in its pricing. Will it be that low? Likely not, as the 1Ds is a much more sophisticated camera than the Nikon-based 14n. We should all know soon.
Finally, I would like to thank Canon Canada for the opportunity of testing this camera. At the time it was the only one in the country and I know that many thousands of readers around the world join me in thanking them for this early opportunity for there to be an independent evaluation of the EOS 1Ds.
OK Canon — what's next?
Issue #6 of The Luminous Landscape Video Journal contains a field evaluation of the Canon EOS 1Ds along with a discussion about this camera with Thomas Knoll, the original author of Photoshop and a leading expert on digital imaging. A featured interview with Thomas will also appear in a future issue of The Journal. Find out more.
Update: October 21, 2002
Rob Galbraith, a respected photojournalist, has just published a review of a pre-production Canon EOS 1Ds. While his camera had some unfortunate functional problems which mine didn't, nevertheless his impressions of the 1Ds paralell mine in many respects. He writes, "At its best, the preproduction EOS-1Ds body in hand here generates photos that match or exceed the level of clarity of the best 6x6 Hasselblad and Mamiya 6x7 prints I've ever made."
While some have expressed concern with my comparison of 1Ds output with that of scanned film and inkjet prints (which I can't agree with — but that's another story), Rob has done his comparison with "fibre-based and RC custom prints from TMax 100 B&W vs both colour and black...", as well.
He writes, "The smallest landscape detail in EOS-1Ds frames holds up at or beyond the level of the traditional darkroom prints, even when comparing at the equivalent of a 16 x 20 inch enlargement. I'm talking about fine, smooth, photographic detail, free from sharpening-induced pixelation or other digital oddities."