Canon EOS 5DS Production Models Out in the Wild

I agree with those who question the relevance of these images. The choice of lenses are, at best, peculiar and the quality of their images rather poor (to be fair, my view is limited to the few I bothered to open). It puzzles me a bit that Canon is handing out these early cameras to people like this. One would think that the people first in line were chosen to act as good publicity. I cannot see how this can be regarded as good publicity.

So, I guess we have to wait for our own delivery and judge then ...
 
Upvote 0
For a given sensor size, focal length, and shutter speed, the mean magnitude of motion blur due to camera shake will be the same regardless of the pixel density or size. However, the smaller and more closely spaced those pixels, the greater the sensor's ability to DETECT such movement during the exposure. This should be plainly obvious to anyone who has ever used a digital camera.

Diffraction behaves the same way: with higher pixel densities, all else being equal, the size of the Airy disk is the same but the ability to SEE the disk across multiple pixels is increased. What was once too small to be observable due to the size of the pixels, is now detectable.
 
Upvote 0
privatebydesign said:
Bghead8che said:
I'm sorry but this is some serious misinformation and is the same hyperbole surrounding the D800 and how you would "HAVE" to use a tripod, double your shutter speed, and take 10 handheld shots for 1 one you planned on keeping. The whole thing has been repeatedly debunked by actual owners.

For pixel level sharpness more pixels demand more stability.

Pixel level sharpness with more pixels implies greater magnification/output size.

For same size output stability is pixel density agnostic, a 5D MkIII and 5DSR will display the same camera shake at the same sized output.

You are right for a same sized output scenario, you are wrong for a pixel level test.

Exactly. But...most RAW converters have a one-button loupe and/or zoom-to-100%, making that pixel level test an easy way to judge sharpness.
 
Upvote 0
lichtmalen said:
IglooEater said:
@lightmalen couldn't agree more- I would love to see some *meaningful* tests and comparisons if ever you feel like publishing them!

I probably will, I should also offer an English translation on my otherwise german homepage ;)
I used GoogleTranslate. It seems to work okay most of the time (Well enough for me to understand what people are trying to convey)
 
Upvote 0
seanature said:
The Canon reps said repeatedly that this is a camera that is designed for tripods.

And the same reps say that the 7D2, of same photosite density, is a hand-held all the way camera ;) ;D ::).

All the talk about how the 5Ds would be impossible to use hand held.... ::).

(that said, of course, to take full 50MP advantage you need slightly higher shutter speeds on average than for a 5D3 just as 5D3 needed slightly higher than for the 10D)

So personally, I wouldn't be so worried and afraid about the pixel count and tripods and so on.

(OTOH, this seems like a camera designed by a team of MBAs. All it offers is the, currently for a short time, highest MP count. It's not really the ideal landscape/IQ camera unless you are in a MP are all scenario, as the sensor is behind at low ISO. Personally I'd rather shoot a 36MP Exmor than this for top landscape or other IQ. And then the RAW buffer is hideous and since the MBA types were afraid of any hint to not induce people into needing both a 5Ds AND a 7D2 they crippled the crop mode to not work in RAW! So you don't get an OK and fantastic RAW buffer as with D810 for full and cropped RAW, but a poor and poor RAW buffer and you are locked into 5fps and dealing with giant, storage and card space wasting files even when shooting distant birds, etc. And then the movies don't include any basic usability features nor even special video AF so the video usability is much worse than any old model that can use ML or the ones that have the special AF for video. And then the movie quality won't match a 5D3 shot in ML RAW (granted that mode is a bit of a pain due to mega files and so on) and it doesn't sound like it will match the video quality of regular modes of other stuff coming out. It just sounds like a camera designed by some MBA who is protecting this and that and milking this and that. I bet a lot will keep 5D3 (5Ds ain't free, far from it, a nice trip or two or lens or some bills paid might be more enticing), wait on 5D4 or go elsewhere. Although for some it will be very cool.)
 
Upvote 0
Also keep in mind that early Canon samples tend to stink. For whatever reason, the cameras usually end up going in the first pre-release week or two to those who have trouble taking any in focus, non-blurred due to shake or motion, exposed properly, weird settings (in cam jpgs usually with sharpness down and NR turned radically high) etc. etc. shots. I never got it, since it seems like poor marketing, but I guess they prefer handing them out to buddies and good words over nice samples out in the wild. Who knows. But never trust the early samples for overall image quality (other than if you get a RAW you can use masked area to test for banding and DR, but that is a side matter) so don't ever be much worried by them.

(I should also add that I haven't bothered to look these specific samples over since I'm not interested in this body so maybe they are actually fine or maybe not.)
 
Upvote 0
LetTheRightLensIn said:
seanature said:
The Canon reps said repeatedly that this is a camera that is designed for tripods.

And the same reps say that the 7D2, of same photosite density, is a hand-held all the way camera ;) ;D ::).

All the talk about how the 5Ds would be impossible to use hand held.... ::).

(that said, of course, to take full 50MP advantage you need slightly higher shutter speeds on average than for a 5D3 just as 5D3 needed slightly higher than for the 10D)

So personally, I wouldn't be so worried and afraid about the pixel count and tripods and so on.

(OTOH, this seems like a camera designed by a team of MBAs. All it offers is the, currently for a short time, highest MP count. It's not really the ideal landscape/IQ camera unless you are in a MP are all scenario, as the sensor is behind at low ISO. Personally I'd rather shoot a 36MP Exmor than this for top landscape or other IQ. And then the RAW buffer is hideous and since the MBA types were afraid of any hint to not induce people into needing both a 5Ds AND a 7D2 they crippled the crop mode to not work in RAW! So you don't get an OK and fantastic RAW buffer as with D810 for full and cropped RAW, but a poor and poor RAW buffer and you are locked into 5fps and dealing with giant, storage and card space wasting files even when shooting distant birds, etc. And then the movies don't include any basic usability features nor even special video AF so the video usability is much worse than any old model that can use ML or the ones that have the special AF for video. And then the movie quality won't match a 5D3 shot in ML RAW (granted that mode is a bit of a pain due to mega files and so on) and it doesn't sound like it will match the video quality of regular modes of other stuff coming out. It just sounds like a camera designed by some MBA who is protecting this and that and milking this and that. I bet a lot will keep 5D3 (5Ds ain't free, far from it, a nice trip or two or lens or some bills paid might be more enticing), wait on 5D4 or go elsewhere. Although for some it will be very cool.)
This was very informative. Maybe this information was available elsewhere but I just noticed (apart from the fact I had already read that fps is not increased in crop mode...) So this is crippled intentionally for landscape and studio users and many 7DII users report AF problems. So it seems that there are no camera solutions when we are FL limited at least for birds... I am sticking with my 5D3s for now...
 
Upvote 0
LetTheRightLensIn said:
Personally I'd rather shoot a 36MP Exmor than this for top landscape or other IQ. And then the RAW buffer is hideous (...and...) the movie quality won't match a 5D3

Sounds like it's not the right camera for you! Your needs should be much better met by the 5D3 replacement.

On the other hand, I never shoot bursts, or use crop mode, I rarely go above ISO 1600, and I prefer dedicated video cameras. I do, however, bump into the limitations of current Canon cameras for fine detail in large prints. All the things that you list as limitations are immaterial to me.

Don't worry though, I'm sure the next Canon full frame will be geared towards you.
 
Upvote 0
tron said:
Maybe this information was available elsewhere but I just noticed (apart from the fact I had already read that fps is not increased in crop mode...) So this is crippled intentionally for landscape and studio users and many 7DII users report AF problems. So it seems that there are no camera solutions when we are FL limited at least for birds... I am sticking with my 5D3s for now...

What a silly thing to say. If a camera's fps is limited by the processor then it can increase the fps in crop mode, if the fps are limited by the mirror and shutter mechanicals crop mode won't give a fps change, which is more "crippled"? I'd argure ff cameras that can increase fps in crop mode are "crippled" by their manufacturers cheapening down on the processors.

But yes, Canon have a target market for the 5DS that they think will be best served by the feature set, don't be disappointed if that doesn't include you or you can't devise a workaround to force it into your personal need, inevitably it wouldn't live up to your expectations anyway. The cost of ff 10fps mirror boxes and shutters is not insignificant and there is no reason for the actual target market to pay for them because you are too cheap to buy a crop camera for the jobs it is best suited to.
 
Upvote 0
privatebydesign said:
tron said:
Maybe this information was available elsewhere but I just noticed (apart from the fact I had already read that fps is not increased in crop mode...) So this is crippled intentionally for landscape and studio users and many 7DII users report AF problems. So it seems that there are no camera solutions when we are FL limited at least for birds... I am sticking with my 5D3s for now...

What a silly thing to say. If a camera's fps is limited by the processor then it can increase the fps in crop mode, if the fps are limited by the mirror and shutter mechanicals crop mode won't give a fps change, which is more "crippled"? I'd argure ff cameras that can increase fps in crop mode are "crippled" by their manufacturers cheapening down on the processors.

But yes, Canon have a target market for the 5DS that they think will be best served by the feature set, don't be disappointed if that doesn't include you or you can't devise a workaround to force it into your personal need, inevitably it wouldn't live up to your expectations anyway. The cost of ff 10fps mirror boxes and shutters is not insignificant and there is no reason for the actual target market to pay for them because you are too cheap to buy a crop camera for the jobs it is best suited to.
I guess you are at "insulting mode" as usual. Or may be even more "Canon fan mode" than myself.
The fact that you prefer to attack instead of mere expressing your opinion could indicate a psychologically troubled person.

I believe there are some Nikon FF cameras that increase a little fps when shooting in APS-C mode so how silly of me to think that Canon could do it too. By the way I did not ask for double frame rate. Even 5D3's 6fps could be something. I also didn't ask for reports of 7DII problems in AF or locking. So I was thinking of possible alternate solutions in order to kill (ok shoot) two birds with one stone (OK camera).
 
Upvote 0
Eldar said:
I agree with those who question the relevance of these images. The choice of lenses are, at best, peculiar and the quality of their images rather poor (to be fair, my view is limited to the few I bothered to open). It puzzles me a bit that Canon is handing out these early cameras to people like this. One would think that the people first in line were chosen to act as good publicity. I cannot see how this can be regarded as good publicity.

So, I guess we have to wait for our own delivery and judge then ...
Canon has already produced their official marcomms and promo videos so they are now putting the camera(s) into the hands of people with a wide viewership regardless of skill level. They obviously have their viewership and anybody who is skilled enough to identify when a camera not near it's performance limits does not really need to input from such reviewers.

Take a look at TN for example. He has a huge following on YT, but unlike some reviewers he is also working professional and has produced good work. He got a review 5Ds-R last month and has been posting a set of interesting comparisons vs the D810 and 5D-III for studio and landscape photography. His impressions are generally positive considering how cynical and negative I thought he was in past months. While I don't agree with everything TN has to say, I fully understand him getting an early copy in order to share his impressions of the camera to his wide (global) audience.
 
Upvote 0
tron said:
This was very informative. Maybe this information was available elsewhere but I just noticed (apart from the fact I had already read that fps is not increased in crop mode...) So this is crippled intentionally for landscape and studio users and many 7DII users report AF problems. So it seems that there are no camera solutions when we are FL limited at least for birds... I am sticking with my 5D3s for now...
No, it is not "crippled" intentionally to fit a profile - instead they defined the camera's niche by catering to its strengths. The software is limiting the ISO sensor output - And I'd make an educated guess that's primarily because of the image processing size. The camera can't apply as much NR and in-camera correction at the max output and still maintain a "Canon acceptable" frame rate for the buffer size - considering its not using C-Fast or UHS-III (as far as I can tell). There are only so many mbps you can push through the pipe at UHS-I class 3 (30mbps to be precise) where as a UHS III class 1 can achieve 220mpbs. C-Fast is as fast as many SATA SSD disks (up to 6gbps with current tech), because that's what it is only with a more rugged interface connector. But its rare, and C-Fast has the commensurate price to show it. By not using either of those formats, they limit themselves in A) data transfer rate from device to card and B) the amount of time for in-camera processing before it must pump the bits to the storage device to clear space for the next frame.

The sensor is nearly the same pixel density as the 7DmII, but covers almost 62% more surface area. Without dumping a much more powerful processor (or a pair of them) into the frame, they were never going to be able to hit the same processing performance - especially at very high ISO - as the 7DmII. The algorithms eat a *lot* of processing, and increasing the processing area (surface area pixel count) is squared in terms of processing costs. So its not 2*T, its T^2... and a bit, and that assumes no increase in pixel color depth.

So rather than let very noisy images get out in the wild straight from the camera, (or have substantially longer times for writeback, and commensurately larger files) they chose to cap ISO at 12800, although going over 6400 is probably not wise. They actually do the same with all their cameras given the technology of when they're made. With current technology, you're not going to get very many more MP than what the 5Ds/5Ds R/7DmII have because of the limitations in physics. Some things *can* be done... but they're esoteric in nature and deal with complex computational algorithms or interesting adjustments to chip manufacture such as Back-Side Illumination. You either interpolate a proper image using calculation, or you increase photoreceptor size/efficiency. The physics (and physicists) are not going to budge, unless someone develops warp drive. And even then... probably not in our lifetimes.

The 7DmII issues are likely limited to a few bad bodies that made it past QC, a few users who haven't compensated for the higher pixel density, and a few users who didn't bother to calibrate their lenses. Remember, "a few" when you're talking about several thousand produced bodies can easily number a few dozen per thousand. And even a few of them, posting on every site they can, can make a camera seem like a real problem.

Does the 7DmII body have an inherent design issue? No. I had a "problem" 7DmII body, AF/Soft focus across the board. I noticed the issue right away and after some consternation exchanged it within two weeks. The second body has been perfect and is beyond my expectations for an APS-C sensor.

By the way, the 5DmIII sensor is one of the best sensors produced by Canon. Other than some features, you are certainly not sacrificing anything to stick with it. If you get a 5Ds/5Ds R (or a 7DmII), you're going to be disappointed at 100 percent crops cause there will be noticeably more noise on the 5Ds at anything over ISO 400 - and over ISO 1600 on the 7DmII - depending on the camera settings and in-camera NR. The 5DmIII's pixel density is far less than most APS-C cameras, and is certainly less than a 5Ds/5Ds R/7DmII. That means the 5DmIII will have less noise, all things being equal - and if noise is your image quality determiner, the 5DmIII will win every single time. That part is physics at play. Outside of construction and build quality, everything else falls to the user and their experience with the tools at hand.

Just my opinion. Your mileage may vary.
 
Upvote 0
@pvalpha I am satisfied with my 5D3s quality. I was thinking about 5DSR as a partial alternative to 7DII due to its issues (and it could serve as a FF camera at the same time). Otherwise I am not Megapixel maniac more like mega iso maniac which in this case 5D3 suits me.

It is just for cases when I am Focal Length Limited (birds) that I started thinking differently (and the camera I would get would not make me sell my 2 5D3s....)
 
Upvote 0
privatebydesign said:
pvalpha said:
The sensor is nearly the same pixel density as the 7DmII, but covers almost 62% more surface area.

No, the 5DSR sensor is 864mm², the 7D MkII sensor is 329mm². That means the 5DSR covers 262% the area of the 7D MkII or 162% more, not just 62% more.
Forgot the 1. I had 1.62 when I did the math. :/ Should have just googled it.
 
Upvote 0