brainiac wrote:
It's still for sale, although I'm considering hanging on to it until the 5DmkII price gets rebated. The 1 series AF and speed is, ahem... enjoyable, and I don't get the impression so far that the 5DmkII is going to improve much on the 1Ds3's ISO 12800 performance:

Please hang on to it... at least until you have the 5D2. So you can provide proper ISO comparisons between the two cams... We can't trust Dpreview and the likes to do it properly, can we now?

scott f wrote:
brainiac wrote:
John Power wrote:
Contrary to what Dasrocket opines, I suspect most will notice very little difference, if any, between a 16 x 20 from either camera of a similarly exposed scene. I say this because it is true when comparing my 30D to my 1DSMK2 and that is about double the megapixels.

Whether you see the extra detail in print is also highly dependent on your printing process, printer, medium, and print size.

Not to be argumentative, but what is the point of going for more megapixels, if there is little to no difference on print? What are we gaining besides bigger files to handle and store?


Not much. You can take and print great shots with a 3 megapixel camera. It is for each to judge whether she needs or wants to pay for that facility. But the 5D2 has lots of other tricks, and great overall image quality, even if it is too sharp for you. And luckily they included a 5 and 10 megapixel raw mode, so pick a size and, dare I say it, have eosfun.

Daan B wrote:
brainiac wrote:
It's still for sale, although I'm considering hanging on to it until the 5DmkII price gets rebated. The 1 series AF and speed is, ahem... enjoyable, and I don't get the impression so far that the 5DmkII is going to improve much on the 1Ds3's ISO 12800 performance:

Please hang on to it... at least until you have the 5D2. So you can provide proper ISO comparisons between the two cams... We can't trust Dpreview and the likes to do it properly, can we now? :D


Depending on the result of that comparison I might have to publish them after selling the 1Ds3 ;-)

Great post... everyone thinking of a 5D should read this.

brainiac wrote:
Daan B wrote:
brainiac wrote:
It's still for sale, although I'm considering hanging on to it until the 5DmkII price gets rebated. The 1 series AF and speed is, ahem... enjoyable, and I don't get the impression so far that the 5DmkII is going to improve much on the 1Ds3's ISO 12800 performance:

Please hang on to it... at least until you have the 5D2. So you can provide proper ISO comparisons between the two cams... We can't trust Dpreview and the likes to do it properly, can we now?


Depending on the result of that comparison I might have to publish them after selling the 1Ds3 ;-)

Hmmm, am I sensing a little doubt wether the 1Ds3 will hold its ground within the high ISO arena?

If you need the higher native resolution for large prints or extensive cropping then I think it's very relevant to compare whether the mkII is as sharp and noise free at it's native resolution as the 5D is at it's. Yes the print size is larger but if you need that print size then are you getting as good quality as you were getting at the lower print size that originally fulfilled your needs? Hence I think that the DPReview comparisons are valid, not as a like for like comparison of course, but to see whether a mkII is as good at doing 21 megapixels as the mkI was at doing 12.8!

> Hmmm, am I sensing a little doubt wether the 1Ds3 will hold its ground within the high ISO arena?

Well, I'm not sure what you mean there. Canon are explicitly stating that the 5D2 sensor is newer and better, so we should expect the 5D2 to beat the 1Ds3 at iso 12800. On the other hand, high iso 5D2 samples I have seen online haven't been very impressive. I would say that they look like badly handled 1Ds3 files. Here's an iso 6400 comparison with the D700:
http://www.flickr.com/photos/erikfive/2868125472/sizes/o/

In that example the blotchy colour noise in the 5D2 file looks pretty bad. Has it been resized down to 12 megapixels to match the D700 file? We don't know, but we do know is that at iso 12800 the D700 and downsized, NR'ed 1Ds3 comparison looks like this:

Beni wrote:
If you need the higher native resolution for large prints or extensive cropping then I think it's very relevant to compare whether the mkII is as sharp and noise free at it's native resolution as the 5D is at it's. Yes the print size is larger but if you need that print size then are you getting as good quality as you were getting at the lower print size that originally fulfilled your needs? Hence I think that the DPReview comparisons are valid, not as a like for like comparison of course, but to see whether a mkII is as good at doing 21 megapixels as the mkI was at doing 12.8!

I take your point, but they don't present the information like that. Here's what DPReview actually says:

"A 21 megapixel camera - even a full frame 21 megapixel camera - is always going to struggle more at higher sensitivities than one with a less ambitious resolution, so it comes as a pleasant surprise to see that the EOS-1Ds Mark III does an excellent job at up to ISO 800, and (from a noise point of view) more or less matches the EOS 5D right the way through the range. Noise (in particular chroma noise) at the highest settings is slightly lower than it was on the EOS-1Ds Mark II, but this may well be down to a slightly more aggressive approach to NR. Of course the Nikon D3, which has the advantage of far larger pixels, produces noticeably less noisy results at higher ISO settings (without any serious impact on detail) - and keeps on going all the way to an unprecedented ISO 25,600.

In truth the Mark III isn't likely to be the first port of call for anyone prioritizing very high ISO performance any more than it would be the first choice for a sports photographer; that's not what it's designed for. As a studio camera it offers far, far more versatility than a medium format camera/back (most of which max out at ISO 400) but there's a limit to what you can expect at ISO 1600 and 3200. It certainly seems that Canon is applying more luminance NR (and perhaps less chroma NR) at the highest ISO settings than Nikon does with the D3."

They have singularly failed to notice that if your goal was a 16x20 inch print in very low light requiring iso 12800, then a 1Ds3 will do about as good a job as a D3, and a better job than a 5D. They have failed to see this because they have based their opinion on comparing 100% crops at differing magnifications, as their noise page shows.

brainiac wrote:
> Hmmm, am I sensing a little doubt wether the 1Ds3 will hold its ground within the high ISO arena?

Well, I'm not sure what you mean there. Canon are explicitly stating that the 5D2 sensor is newer and better, so we should expect the 5D2 to beat the 1Ds3 at iso 12800. On the other hand, high iso 5D2 samples I have seen online haven't been very impressive. I would say that they look like badly handled 1Ds3 files. Here's an iso 6400 comparison with the D700:

In that example the blotchy colour noise in the 5D2 file looks pretty bad. Has it been resized down to 12 megapixels to match the D700 file? We don't know, but we do know is that at iso 12800 the D700 and downsized, NR'ed 1Ds3 comparison looks like this:

So do I believe Canon that this sensor is better than the 1Ds3 one? Well, yes - they wouldn't say it if it wasn't true, but so far the difference looks like being marginal, to say the least. The 5D's iso bump over the 1Ds3 looks like being a bit of not very good in camera NR, and the presence of those iso settings, obviating the need to use exposure compensation and raw. That last limitation is fairly academic if the in camera jpegs at very high speed don't look very nice, as you will want to shoot raw at high speed anyway, to have a chance of rescuing the picture from the kind of blotchy colour noise we see in the flickr example.

I agree with you... The posted high 5D2 samples I have seen so far appear to have been (heavily) processed by in-camera NR. And it doesn't look very pretty to me. They have tried to even out flat surfaces with little detail and OTOH tried to maintain as much detail as possible. Overall the files look very soft... Could this have been a side effect of the in-camera NR?

But this is all done with JPEG's. I wonder how the RAW files will look. Will the in-camera NR affect the RAW files or not? And if so, how will that look in comparison to 1Ds3 RAW files? Since both cams use basically the same sensor, I don't expect differences in noise levels in RAW files to be beyond marginal though. But only time will tell

brainiac wrote:
digitalbug30d wrote:
I find it pointless to compare a 12.8 MP camera to a 21.1 MP camera...
this is like comparing a 100 hp car to a 200 hp car and expect the 100 hp car to do the same performance as the 200...apples and oranges

It's actually a lot more like comparing a 100 hp car to a 130 hp car, in a 70 mph speed limit zone.

no, it's like comparing a 100hp light car to a 200hp heavy car. horsepower to weight ratios...

too many fall prey to the numbers game. the megapixel race is alive and well.

Daan B wrote:
I agree with you... The posted high 5D2 samples I have seen so far appear to have been (heavily) processed by in-camera NR. And it doesn't look very pretty to me...
But this is all done with JPEG's. I wonder how the RAW files will look. Will the in-camera NR affect the RAW files or not? And if so, how will that look in comparison to 1Ds3 RAW files? Since both cams use basically the same sensor, I don't expect differences in noise levels in RAW files to be beyond marginal though. But only time will tell

Well, I correctly predicted, in November last year, that the 1Ds3 would have a comparable noise performance (providing you want to make pictures, not crops) to the D3. I am going to be bold and make another prediction here: the 5D2 iso 12800 noise performance, optimally processed from raw, will look pretty much the same as that from the 1Ds3. Why?
(1) because Canon likely made some effort with the 1Ds3 sensor, so tweaks are unlikely to radically transform its performance, and
(2) sensor design and fabrication is a very expensive business in which costs are acutely sensitive to volume, so Canon would be foolish not to aim to use its latest sensor in more than one model at different price points, if it can get away with it, and
(3) Canon has stated that the 5D2 sensor is derived from the 1Ds3 design, and
(4) no detailed explanation has been offered of how the 5D2 sensor differs from that in the 1Ds3, and
(5) because 5D2 high iso performance is looking awfully familiar to me - it looks exactly like what i would get if I shot my 1Ds3 and didn't optimally process the files.

So basically the 5D2 is a 1Ds3, without all the fancy stuff, and Canon's claim to have enhanced the sensor is just lipstick on a supermodel. The fact that it does 3.9 fps suggests that the 1Ds3 could have done 7.8 fps if they had wanted to offer that, and that as we all know, these cameras are not the best that they can be in every feature, but are tailored to distinguish themselves from each other in the way that the marketing department wants them to. For instance, the 1Ds3 could have had a 6400 and 12800 iso setting, but Canon knows that pixel peepers and DPReview, in their ignorance, would have been scornful due to the lack of good in-camera NR.

simonella_viru wrote:
brainiac wrote:
digitalbug30d wrote:
I find it pointless to compare a 12.8 MP camera to a 21.1 MP camera...
this is like comparing a 100 hp car to a 200 hp car and expect the 100 hp car to do the same performance as the 200...apples and oranges

It's actually a lot more like comparing a 100 hp car to a 130 hp car, in a 70 mph speed limit zone.

no, it's like comparing a 100hp light car to a 200hp heavy car. horsepower to weight ratios...

too many fall prey to the numbers game. the megapixel race is alive and well.
5D = 128 hp car with aerodynamics and weight such that it tops out at 125 mph and takes 8 seconds to do 0-60 mph.
5DII = 211 hp car with aero and weight such that it tops out at 150 mph but still takes 8 seconds to do 0-60.
Speed limit for 8x12 prints is 60 mph.
For 40x60 prints and shooting landscape with great lens and no situational limits, there is no speed limit.

OK, I've definitely fallen prey to the numbers game.

The best reasons to buy the 5DII would be everything besides the number of pixels, but even this "modest" increase in MP will be a positive, never a negative.

simonella_viru wrote:
brainiac wrote:
digitalbug30d wrote:
I find it pointless to compare a 12.8 MP camera to a 21.1 MP camera...
this is like comparing a 100 hp car to a 200 hp car and expect the 100 hp car to do the same performance as the 200...apples and oranges

It's actually a lot more like comparing a 100 hp car to a 130 hp car, in a 70 mph speed limit zone.

no, it's like comparing a 100hp light car to a 200hp heavy car. horsepower to weight ratios...


It's actually a lot more like sitting and masturbating in your car.

simonella_viru wrote:
brainiac wrote:
digitalbug30d wrote:
I find it pointless to compare a 12.8 MP camera to a 21.1 MP camera...
this is like comparing a 100 hp car to a 200 hp car and expect the 100 hp car to do the same performance as the 200...apples and oranges

It's actually a lot more like comparing a 100 hp car to a 130 hp car, in a 70 mph speed limit zone.

no, it's like comparing a 100hp light car to a 200hp heavy car. horsepower to weight ratios...

too many fall prey to the numbers game. the megapixel race is alive and well.

Resolution is proportional to the square root of the number of pixels. You would be right if you were talking about flatbed scanners.

cogitech wrote:
simonella_viru wrote:
brainiac wrote:
digitalbug30d wrote:
I find it pointless to compare a 12.8 MP camera to a 21.1 MP camera...
this is like comparing a 100 hp car to a 200 hp car and expect the 100 hp car to do the same performance as the 200...apples and oranges

It's actually a lot more like comparing a 100 hp car to a 130 hp car, in a 70 mph speed limit zone.

no, it's like comparing a 100hp light car to a 200hp heavy car. horsepower to weight ratios...


It's actually a lot more like sitting and masturbating in your car.


genius

It's very hard to get conclusion on two different digital cameras, especially when they are different in pixel density (magnify ratio). I've seen reviews by comparing Canon and Nikon DSLR even with different lenses. A reasonable comparison would be using adapter to test the same lens on two system with same pixel density. Although the adapter may still affect the result, but this is probably the best way to go.

since you've made that pictures with 17-40mm on a 5D, the loss of a little sharpness is acceptable(from what I see in your pictures).

gabimaster wrote:
since you've made that pictures with 17-40mm on a 5D, the loss of a little sharpness is acceptable(from what I see in your pictures).

The image was made with a 1Ds3, at f11. Whether the corners are acceptable is a matter of personal taste. I am perfectly happy with them, but serious landscape shooters mightn't be.

brainiac wrote:
Daan B wrote:
I agree with you... The posted high 5D2 samples I have seen so far appear to have been (heavily) processed by in-camera NR. And it doesn't look very pretty to me...
But this is all done with JPEG's. I wonder how the RAW files will look. Will the in-camera NR affect the RAW files or not? And if so, how will that look in comparison to 1Ds3 RAW files? Since both cams use basically the same sensor, I don't expect differences in noise levels in RAW files to be beyond marginal though. But only time will tell

Well, I correctly predicted, in November last year, that the 1Ds3 would have a comparable noise performance (providing you want to make pictures, not crops) to the D3. I am going to be bold and make another prediction here: the 5D2 iso 12800 noise performance, optimally processed from raw, will look pretty much the same as that from the 1Ds3. Why?
(1) because Canon likely made some effort with the 1Ds3 sensor, so tweaks are unlikely to radically transform its performance, and
(2) sensor design and fabrication is a very expensive business in which costs are acutely sensitive to volume, so Canon would be foolish not to aim to use its latest sensor in more than one model at different price points, if it can get away with it, and
(3) Canon has stated that the 5D2 sensor is derived from the 1Ds3 design, and
(4) no detailed explanation has been offered of how the 5D2 sensor differs from that in the 1Ds3, and
(5) because 5D2 high iso performance is looking awfully familiar to me - it looks exactly like what i would get if I shot my 1Ds3 and didn't optimally process the files.

So basically the 5D2 is a 1Ds3, without all the fancy stuff, and Canon's claim to have enhanced the sensor is just lipstick on a supermodel. The fact that it does 3.9 fps suggests that the 1Ds3 could have done 7.8 fps if they had wanted to offer that, and that as we all know, these cameras are not the best that they can be in every feature, but are tailored to distinguish themselves from each other in the way that the marketing department wants them to. For instance, the 1Ds3 could have had a 6400 and 12800 iso setting, but Canon knows that pixel peepers and DPReview, in their ignorance, would have been scornful due to the lack of good in-camera NR.

Pretty obvious conclusion... If Canon gave it to us all the way, what will there be left to sell us next year? At least they know for sure that 5D2 owners will jump in the air when the 5D3 finally gets its 9 cross point AF + 6 assist points next time around ... Only to hear shortly thereafter that Nikon has introduced a 300 cross point eye controlled AF sensor... And the story continues... And the poor folks who are trapped inside these marketing fairytales, don't seem to realize that IQ has virtually remained the same all along... I am not buying it anymore (both the cams and the marketing BS)

phuang3 wrote:
It's very hard to get conclusion on two different digital cameras, especially when they are different in pixel density (magnify ratio). I've seen reviews by comparing Canon and Nikon DSLR even with different lenses. A reasonable comparison would be using adapter to test the same lens on two system with same pixel density. Although the adapter may still affect the result, but this is probably the best way to go.

Yes. That's why the example below was shot with the very same lens, via adaptor, on a D700 and 1Ds3 at the same aperture and shutter speed. The idea was to ensure that the same amount of light was hitting the sensor regardless of camera settings. The Nikon was set to iso 12800 (H1?) and the Canon to iso 3200 (H). Optimal raw processing and noise reduction were applied to each, and the 1Ds3 file was downrezzed to the dimensions of the D700 file. While the noise patterns were originally very different, with the Nikon having a little more lumi noise and the Canon having a lot more colour noise, after noise reduction the files are surprisingly similar.

Brainiac.....you are a breeze of fresh air
I am going to the office of DPR and slap those guys a arround...!

Daan B wrote:
And the poor folks who are trapped inside these marketing fairytales, don't seem to realize that IQ has virtually remained the same all along...

I was with you until this line. 5 years ago I could not shoot iso 12800 in very low light. Now I do it at every job. Image quality has been transformed, and I, for one, am grateful.

brainiac wrote:
Daan B wrote:
And the poor folks who are trapped inside these marketing fairytales, don't seem to realize that IQ has virtually remained the same all along...

I was with you until this line. 5 years ago I could not shoot iso 12800 in very low light. Now I do it at every job. Image quality has been transformed, and I, for one, am grateful.

When you compare cameras of any manufacturer according to the same specs, the differences in IQ will only be marginal at best. In the end, all cameras share the same basic design and technology. THAT was my point

brainiac wrote:
scott f wrote:
brainiac wrote:
John Power wrote:
Contrary to what Dasrocket opines, I suspect most will notice very little difference, if any, between a 16 x 20 from either camera of a similarly exposed scene. I say this because it is true when comparing my 30D to my 1DSMK2 and that is about double the megapixels.

Whether you see the extra detail in print is also highly dependent on your printing process, printer, medium, and print size.

Not to be argumentative, but what is the point of going for more megapixels, if there is little to no difference on print? What are we gaining besides bigger files to handle and store?


Not much. You can take and print great shots with a 3 megapixel camera. It is for each to judge whether she needs or wants to pay for that facility. But the 5D2 has lots of other tricks, and great overall image quality, even if it is too sharp for you. And luckily they included a 5 and 10 megapixel raw mode, so pick a size and, dare I say it, have eosfun.

Not sure what you mean by "too sharp"? I haven't seen an image yet from it to suggest it has great image quality yet. Every high ISO image I've seen is mediocre at best, which isn't surprising as I don't think Canon has launched a camera yet with a portfolio of excellent samples.
I guess I'm just playing the devil's advocate in all of this.

Umm, that doesn't work.

Uprezzing creates (read: fabricates, interpolates, whatever) pixels that weren't there before. A straight 21MP file from the 5D2 has actual data in all of those pixels. You can't compare the two.