Jump to content
Drew

Rob Galbraith's final findings on the Canon 1D MarkIII series AF

Recommended Posts

I'm not trying to start another "Canon vs Nikon" flame war ^_^ I'm just trying to offer my real-world experience with both cameras side-by-side. I'll try and pull up some test shots to demonstrate this, but in the mean time, to explain what I meant by "detail" vs. "resolution"...

 

As Eric said, resolution is pretty easily quantifiable, and "detail" seems more subjective. What I mean from my experience, is that if I'm shooting a bear photo (I happened to be testing these bodies in Alaska) looking at RAW/NEF files, the files coming from the 1D3 had clearly defined hairs with clear detail. On the D3, that level of definition wasn't there - I mean, the bear had hair obviously, but you couldn't count the individual hairs on it. It could be corrected somewhat with post-processing, but you can't bring back what wasn't there to start with. Like Drew said above, the files from the 1D3 seem to be more crisp and sharp.

 

The D3 files are bigger - it's a higher resolution camera. The 1D3 files seem more detailed/crisp/sharp. As I pointed to earlier, it's the same thing that Galbraith mentioned - the 1D3 produces a more detailed image (not higher resolution). To put it in terms that underwater photographers can relate to, it's the difference between seeing the teeth on a tiger shark as it's about to bite your dome port, and being able to count the serrations on the edges of those teeth :)

 

As for the AF tests, my guess is that they wanted as equal settings as possible - the 3D dynamic focus is kind of like letting the camera select the AF point (a typical dynamic AF setup). When I tested the D3, I disliked the "3D dynamic area tracking wonder focus" for the same reason that I dislike allowing the camera to select the focus point on Canons - it misses the focus more often than I would and is predisposed to jump off the subject if anything comes in the path of it.

 

Hope this helps shed some light on what I meant earlier.

Share this post


Link to post
Share on other sites

Brian

Yes please put some pics together to illustrate your point. The resolution chart pics on DPreview show the D3 having higher resolution with a 'softer' look. Whereas the 1D3 has crispness but moire shows up on lines earlier, thus showing lower resolution. I suspect the AA filter is making the D3 go softer. And isn't that a role reversal? I remember when the 1D2 came out and everyone was complaining how soft it was because of the AA filter. So Canon listened and corrected it with the 1D3. And now the D3 has the same affliction apparently. From the limited exposure of 2 days with the D3 shooting gannets, dolphins and whales, I'd say the files can take more sharpening in post like the 1D2 files could way back. The higher resolution should give more 'detail'. Still what Craig is saying is that once you have to literally split hairs over the image, then it's down to a preference of looks and not a technical issue.

I think we've established that the Canon vs Nikon battle is over. Ok so it'll never be over but it's nice to think it is.

Share this post


Link to post
Share on other sites
Still what Craig is saying is that once you have to literally split hairs over the image, then it's down to a preference of looks and not a technical issue.

 

I think that the word is NUANCE - that's what we are discussing. At this level of gear there are a lot more factors to consider than simply image quality which is at an all time high on these cameras.

 

But to throw a spanner in the works I'm using a Leica M8 as my topsides camera as much as I can these days - superlative optics, lightweight and superb files (although as always there is web controversy over it). One thing is for certain though - it can't be housed - so discussions as to its merits here are irrelevant and pointless so no arguing with this one!

 

At the end of the day many people seem to forget that cameras are tools and if they deliver images fit for the purpose that they were shot then they are the appropriate tools for the job. Discussions of nuance can be interesting but IMHO they often devolve into unwinable arguments, rather than an acceptance that most equipment is capable of exceedingly good results.

Share this post


Link to post
Share on other sites
Why doesn't Nikon make a 14/2.8?

 

They do http://imaging.nikon.com/products/imaging/...f_28d/index.htm :)

 

I mean, the bear had hair obviously

I liked that :B):

 

 

I find the bigger problem with the 1DmkIII and underwater photography is its sensor size falling in between FF and DX more than anything else.... so Drew, you thought the nikon/canon battle is over but what about the sensor size one? :)

 

I would also like to see real world improvements on using no AA filters in our cameras as I do not think moire is very subject to happen underwater and taking off the filters would be positive.

Share this post


Link to post
Share on other sites
But to throw a spanner in the works I'm using a Leica M8 .... it can't be housed -

As they say, Paul... Bag it! Ewa Marine has a bag for it. Good for 33ft too :) Slap that Tri-Elmar 16-24 on and you have a semi decent wide angle.

I too am looking into the M8 but will wait for the 2nd hand market to drop. I can't stand throwing $10k on a camera system I'll use for very specific purposes like landscape and parties

 

I find the bigger problem with the 1DmkIII and underwater photography is its sensor size falling in between FF and DX more than anything else.... so Drew, you thought the nikon/canon battle is over but what about the sensor size one? :)

Well discussion of crop vs full sensors has been done ad nausuem. However, I particularly like the 1.3 sensor because it's in between the 2. You get a bit of crop benefits and a bit of FF benefits. I like the 15fe look on the 1.3 crop, curvaceous yet not so round :B): Corners on the 1.3 is better on the wider lenses... 16-35, 14 even the Sigma 12-24 (well if you get the right domeport config). It's a distinctive look. Horses for courses and all that.

Share this post


Link to post
Share on other sites
curvaceous yet not so round :) C

Yes we hear that's how you like your women

Share this post


Link to post
Share on other sites
As they say, Paul... Bag it! Ewa Marine has a bag for it. Good for 33ft too :) Slap that Tri-Elmar 16-24 on and you have a semi decent wide angle.

I too am looking into the M8 but will wait for the 2nd hand market to drop. I can't stand throwing $10k on a camera system I'll use for very specific purposes like landscape and parties

 

 

Well discussion of crop vs full sensors has been done ad nausuem. However, I particularly like the 1.3 sensor because it's in between the 2. You get a bit of crop benefits and a bit of FF benefits. I like the 15fe look on the 1.3 crop, curvaceous yet not so round :) Corners on the 1.3 is better on the wider lenses... 16-35, 14 even the Sigma 12-24 (well if you get the right domeport config). It's a distinctive look. Horses for courses and all that.

 

Ahh Yes a Ewa bag - didn't spring to mind I have to admit :B): On the other hand, what other digital camera can you use a 70 year old lens on with complete compatability? I've tried a pre-WWII lens and whilst clearly not up to modern standards, it was surprisingly usable! There are also lenses such as the Cosins/Voightlander 12mm which I'd like to try (16mm equivalent FoV and VERY small). Here's a shot taken on the M* if you are interested (NT underwater): http://www.l-camera-forum.com/leica-forum/...r-lighting.html

 

I too find the 1.3 crop a workable compromise - I don't care for smaller crops myself - its a personal thing - but since I shoot a lot of full aperture material I like to us either FF or 1.3x. To my mind 1.5/6x seem to lose some of the focus differential of the 'larger' formats

Share this post


Link to post
Share on other sites

If there were such thing as detail unique from resolution in a digital system then someone would be able to measure it. Quantization and linearity in digital cameras are nearly ideal so there's little room for differences between those two terms. If the subject were lenses then I could see the point considering the subjective nature between contrast and detail in an analog system. Again, I think lenses play a role in these subjective impressions. I think others covered this well; it's mostly a matter of slight differences in the final presentation. Nikon's AA filter is stronger than Canon's and its lenses are different. The final result is bound to look different to some degree.

 

Craig,

Why would you want to remove these filters?

Because I *think* that the additional elements underwater (water, ports) cause enough softening so long as the pixel pitch is sufficiently high. Lenses are already challenged and, in macro anyway, the system is often diffraction limited. On a 1Ds3, f/16 diffraction will blur out any potential moire. That may be true even at f/11.

 

The D200HR example is exactly what I'm thinking. I contacted the companies that do that sort of mod and inquired about a D3 version. They mainly cater to Canon because most people interested like all the filters removed for astronomy photography. With its past leadership in noise control, that market is totally dominated by Canon right now. Still, it may be a viable mod once the 24MP version hits. I'm not sure the D3/D700/5D pixel size is small enough.

 

As far as the UV-IR filters go, there are two types. The type that DSLR camera makers use are the dyed ones. Dyed UV-IR filters cause red sensitivity to be poor compared to green and blue and that hurts us underwater. The interference type filters don't have that problem but they have varying performance with angle of incidence and are more expensive. A retrofit of the alternate filter type substantially improves the natural white balance of the sensor and reduces the need for magenta filtration. Canon users can get this done but Nikon users are currently out in the cold. :)

Share this post


Link to post
Share on other sites

Reading this thread makes my head hurt.

 

I've been quite happy with both my 1D and 1Ds MarkII and MarkIII's, and don't seem to have a problem with AF on either MkIII body (with current firmware, which seems to have corrected the early low-contrast AF issues). From what I've seen, I think Eric is quite happy with his, too, and gets fabulous images with them. OTOH, when I look at the stuff Alex and the rest of you top-notch Nikon shooters get (including a lot of which I own in print), I'm stunned at the quality of stuff you're producing.

 

Bottom line is that I can't imagine that any of us are suffering ANY disadvantage in real-world applications with either brand at this level of equipment. The shortcomings with the most impact on results remain BEHIND the viewfinder, not inside the camera body. I'm not saying analysis of the differences is without merit (or just measurebating), but - ESPECIALLY when it comes to uw use - are Galbraith's comments about the nuances of one output versus another REALLY that pertinent to what you guys can get with these bodies?!

 

P.S. FWIW, I, too, like the 1.3 crop (not to mention the blazing fps speed) of the 1D bodies in a lot of applications. While I only use the non-"s" bodies about 10% of the time, usually when I'm trying to capture lots of frames to catch fast-happening action, several of my very best shots (speaking of bears' hairs!) have come from the 1D bodies. That has emphasized to me that it's about far more than just resolution/megapixels (even if they're very high quality megapixels).

Edited by bmyates

Share this post


Link to post
Share on other sites

Well, this has gotten interesting :)

 

I don't see why there can't be detail unique from resolution - pixel pitch and other factors all influence this (as does AA filter, lens, etc). It may be measurable - I don't know if a measurement exists for it, but I'm basing my statements (which have been agreed with by others) based on what I can see with my eyes. Craig clearly knows more about the technical details of this stuff than I do - I just know what works for me and makes the images I need.

 

You can take beautiful photos with either system - I never intended this to be a Canon vs Nikon thing (although it feels like some are treating it that way). There are lots of photographers who are better than I am shooting with both systems. I don't have a loyalty either direction - my employer provides whatever gear I ask for. I personally used to shoot Nikon, went over to Canon, and now am shooting a mixed bag. I mentioned lens wishes earlier. I with Canon would make a better speedlight system too. I'm kinda hoping for big things from the new 5D - if not, a D700 will probably end up in my kit for a lot of reasons.

 

They're all just tools to get the job done - that's what it really comes down to in the end.

 

On the other note, I agree with Bruce - I like the 1.3 crop for a lot of applications. I shoot a lot of sports, and the extra reach is nice. Shooting full-frame, I need to carry longer glass or use extenders. Plus I like the 1.3 crop with a 15mm underwater (sort of like using a teleconvertor on the 15mm with a 5D).

Share this post


Link to post
Share on other sites
Reading this thread makes my head hurt.

Hey Bruce, i think this thread's been pretty mellow. :)

Actually I think these discussions are great; most of what I know about digital cameras and how they work started here on Wetpixel.

Edited by loftus

Share this post


Link to post
Share on other sites
Actually I think these discussions are great; most of what I know about digital cameras and how they work started here on Wetpixel.

 

Me, too! But some types of learning are more difficult/painful than others. Sorting through all these Wetpixel discussions of resolution, pixel size and pitch, detail and the like to try to figure out what's really real and how much is just opinion causes at least a few neurons to misfire in my wee little brain! :)

Share this post


Link to post
Share on other sites
I don't see why there can't be detail unique from resolution - pixel pitch and other factors all influence this (as does AA filter, lens, etc).

I think in the context of the entire camera system there can be. You must agree on what those terms mean or you may end up arguing over language and nothing else (as often happens). :)

 

It is the case that photographers speak of acutance distinct from resolution. In the context of this discussion, differences in acutance are directly related to the AA filter strength. These can be largely eliminated through sharpening though an image will always be better if it's details aren't lost in the first place.

 

If there's a difference between "detail" and "resolution" between DSLR bodies and it's not acutance, then I'm totally at a loss as to what it could be. It must be analog function and a body has very little analog image processing it it. There is the AA filter (acutance), the UV-IR cut filter, the bayer filter and the microlens array(s). What remains is the light box.

 

It is not my intention to contribute to yet another brand flame session either. I simply believe that opinions on which of the bodies is better boil down to personal preference, familiarity, lens differences, testing methods, etc. Once the 24MP Nikon comes out, I think underwater shooters will choose systems based what lenses and ergonomic features they prefer.

Share this post


Link to post
Share on other sites

Ok, forget pixel numbers and sizes, resolution and detail, and all the other techincal terms for a minute and reflect; what influences the capturing device more than anything else - answer is the lens of course. Above water, lens characteristics are a well known and oft discussed nuance of image creation, but underwater we have several problems which cause us to work in specific ways.

 

Firstly we have to use dome ports to work with wide-angle lenses. These are, in essence, simple lenses and they impart characteristics, even if accurately aligned to the images created by the lens being used. In simple terms, the use of a dome port effectively limits the aperture which we can use underwater fairly severely unless the images being shot have no corner detail in them. As a practical approach I'd suggest that f/5.6 is a viable limit above which fewer images can or are taken (using larger apertures).

 

Macro is a different story with the flat planar glass mostly causing chromatic problems - again most noticable in the corners and substantially correctable in software. But again, macro work usually requires small apertures and diffraction limitation is an inherent problem with aiming for high depth of field. Mostly we probably shoot at f/11 and smaller.

 

There are of course exceptions but I suspect I'm generally correct in my comments.

 

So discussing nuances of sensor design when we are using optics under significant constraints seems perhaps a little errr, well, not pointless, but maybe unimportant in many ways. I find that my Canons (1DSs and 5D) can and do show signs of lens imperfections above water. Underwater they show the compromises detailed above. I am yet to be convinced that I need to upgrade my cameras and housings for the vast majority of my images. Others will disagree and will be perfectly correct to do so for particular types of work. In fact I've taken another route above water by adapting several Leica lenses for my Canons and I find these can be extremely capable. But since I can produce perfectly acceptable 24" x 16" prints from my existing cameras, I'm not sure that I can see the reasons to upgrade and even with more pixels, faster AF, bigger buffers and all the other advantages of the latest generation of equipment I'm not enticed. I'm working on a book on fish at the moment and my biggest problem is stalking the little b*****s, something which NO camera system will help with!

Share this post


Link to post
Share on other sites

Can anyone tell me what advantages a hardware low pass filter (AA filter) has over performing frequency filtering in software. I would have thought that it would be advantageous to perform the filtering in software using wavelets, Fourier transforms or even simple Gaussian filtering on the raw image rather than permanently attenuating spatial frequencies before they are sensed. Then you could have a setting in the camera that would alter the default AA filter strength in the camera just like we do for contrast and saturation. Maybe I'm miss understanding the purpose of the AA filter?

 

Cheers,

 

Mike

Share this post


Link to post
Share on other sites

You cannot perform anti-aliasing in software. It must be done prior to the sensor. Once high frequency information has aliased to a lower frequency it cannot be differentiated from data that might already be there.

Share this post


Link to post
Share on other sites
Once high frequency information has aliased to a lower frequency it cannot be differentiated from data that might already be there.

 

Doh! Yes, of course. I guess I never really thought that one through.

Share this post


Link to post
Share on other sites
Rob Galbraith just updated his 1DMk3 AF issue findings. Seems there isn't much change despite all the work. Still only action shots apply and fast moving ones like billfish and tuna would really stress the AF.

 

http://robgalbraith.com/bins/multi_page.as...8740-9068-10086

 

Rob notes, in his very thorough analysis, some photographers have enjoyed significant AF performance since the 1DMKIII fix. Gratefully, it did make a big difference on my camera body. Not that I shoot much of the long-lens sport he does, but I do shoot equestrian action because my daughter rides hunter/jump and the camera was never critically sharp before. After the fix I did a round of kiteboarding, see http://stephenfrink.blogspot.com/2009/05/k...ttempts-at.html. I shot both 5DII and 1DMKIII, and the combination of 10-frames per second and stellar AF acquisition made the 1DMKIII clearly the go-to camera body. Longest lens I used was 70-200 F-4, but it was really a terrific combination for this kind of action.

Share this post


Link to post
Share on other sites

Stephen

I think it's also a lot to do with expectations. When I had my 1D3, I expected it to outperform my 1D2 in terms of AF tracking at 10fps, like Rob and other sport shooters did too. I dumped mine a few months after I couldn't get a good keeper rate (Blue dot) and kept my 1D2 series.

I'm surprised you kept the 1D3 since it wasn't critically sharp on the ratio of captures.

I wonder if the 1Ds is also affected.

Share this post


Link to post
Share on other sites
Stephen

I think it's also a lot to do with expectations. When I had my 1D3, I expected it to outperform my 1D2 in terms of AF tracking at 10fps, like Rob and other sport shooters did too. I dumped mine a few months after I couldn't get a good keeper rate (Blue dot) and kept my 1D2 series.

I'm surprised you kept the 1D3 since it wasn't critically sharp on the ratio of captures.

I wonder if the 1Ds is also affected.

 

Drew - I never had an issue with my 1DsMKIII, and from the very beginning have been pleased with its AF performance. Much better than my 1DsMKII in fact, especially with my 100mm macro UW.

 

As for the 1DMKIII, I did have a nagging regret that I sold my 1DMKIIN with so few actuations, as that remains a very nice camera. But I love the combination of the crop + frame rate of the MKIII and high res of 1DsMKIII BOTH fitting in same housing. That was a stronger determinant than a few soft sports shots, for what I do for a living.

 

Now that the focus is dialed in better on the MKIII, and I still have the synergy of two very different cameras able to fit into the same housing, I'm very pleased.

Share this post


Link to post
Share on other sites

Maybe the limit of the AF processors right now is about 9 fps with a mirror system. I mean even the D3 can't AF track past 9 fps in DX mode. If the AF processor on the Ds3 series isn't the problem, then it has to be the 10fps. Can 1 or 2 fps extra make that much of a difference in accuracy?

Share this post


Link to post
Share on other sites

If it is a processor limit, it should be an easy thing to test - run the same subjects at maximum frame rate limited to 7/8/9/10 fps, and look for a significant crossover point where percentage in focus drops off sharply.

Share this post


Link to post
Share on other sites

Sponsors

Advertisements



×
×
  • Create New...