Final firmware Canon EOS 5DSR real-world samples posted

Canon DPP........

It would be interesting to see a couple of the images processed in DPP to see if the native software deals with the shadow lifting any differently/better(?) than ACR........or supply the raws to allow a couple of the users to have a go at it.

Thanks, MIke

Jun 24, 2015permalink

siggo

Agree. I think DPP should always include some images processed using the camera maker's software in addition to OOC JPEGS.

Jul 15, 2015*permalink

(unknown member)

I downloaded IR's RAW still tests this afternoon and checking them out in DPP4 on my 2560x1440 Dell am IMPRESSED. I can see very fine grain with their own settings in deep shadows at 1600,( behind the bottles etc) and they are at least a third stop underexposed. The noise performance is as good as Nikon D8xx series machines
IMHO. I am impressed, but you do need to be very careful about any sharpening at all, as at this magnification the results become very easily visible earlier, and not in a good way. I am impressed.

Jun 24, 2015permalink

mikedodd44

They seem to have got rid of the old problem of banding but if you look at the highest resolution version of the aurora image there seems to be a lot of 'blotchyness' in the dark sky. Not sure this could be called noise as the blotches are much bigger than this and don't think it is compression artifacts either.

Jun 24, 2015permalink

ttran88

Hope you brought a second camera to have better personal files for keeps.

Jun 24, 2015permalink

PhilJeffrey

I think the aurora pics nicely point up the strengths and weaknesses of the camera - it's really only going to shine in well-lit scenes loaded with subject detail where the megapixels trump the noise. In lower detail dark scenes, well it's a Canon chip with small pixels so us Canon shooters should not be shocked about that noise. (I own a 7D2 amongst other Canons). ETTR to mitigate it somewhat, but a Sony chip-based camera and probably X-trans will produce a more workable image under these circumstances. Shame it's not possible to compare it to a 7R II for the same aurora image.

Jun 24, 2015permalink

zigi_S

I can see a galaxy(don't know the name) on the up right side.

Jun 24, 2015permalink

(unknown member)

Scarily big, you see the problem with that copy of the 16-35mm II!! not a lens I would choose to use. The car interior deliberately underexposed by a stop, was already at least a stop underexposed due to the meter relying on the highlights outside, so to be objective (rather than objectionable!) you need to meter directly off a midtone inside the car , or point an incident light meter from inside the car itself towards the lens position (with the photographer stood there blocking with your shadow, more light (assistant necessary!)

I note that the lens at all apertures cannot evenly cover the sensor, so it would be nice to gat one that can and do some more please! Love the back of the preacher's car shot!

Jun 24, 2015permalink

Collie Camp

Well, they did not use the 16-35II, they used the new 16-35 / 4L. Reviewed as a very very good lens, but this copy seems decentered.

Jun 25, 2015permalink

fedway

I'm from the Seattle area too. Where did you go to get those Northern Lights shots?

Jun 24, 2015permalink

edu T

Nothing personal, but I wonder why every now and then do some people seem to be unwilling to read earlier comments *before* commenting themselves?
a) for the sake of redundancy; b) short attention span; c) little belief in others having the same idea or question first; d) . . . ?

Jun 25, 2015permalink

fedway

Ah found it farther down. I'm too busy to read all the comments. Perhaps, you're retired, unemployed or independently wealthy. I'm none of the above.

Jun 25, 2015permalink

I see impossible. Whatever that means...

Jun 24, 2015permalink

watson076

I see a lot of negative criticism regarding these photos. While I myself have moved on from Canon to m43rds, I do feel like defending Canon based on these samples for a few reasons:

1) ACR is noisy. It will bring out noise in the cleanest sensor even at base ISO. So I really don't know how fair it is to judge noise.

2) ACR colors are artificial. I know you can tweak them to taste, but I quit using Lightroom after a year with the software. My biggest gripe is the rendering of colors which look good for some applications, but for many things (including skin tones and blues etc) the colors typically don't look attractive. I think Lightroom's workflow is fantastic, but their overall image conversion is just so-so.

3) These images are post processed to taste, so do they represent the camera or the photographer ? Personally I prefer to see OOC JPEGs (or RAWs converted from the manufacturers own software) to get an idea of what they envisioned the sensor to produce.

Jun 24, 2015*permalink

joelakeland

I see plenty of hot pixels, white ones.

Jun 24, 2015permalink

mpgxsvcd

Nothing in these pictures make me want to run out and buy this camera. The lens wasn't great for these images and the noise was distracting. It wasn't the photographer' fault. It was the equipment's limitations this time.

Jun 24, 2015permalink

JeanPierre Thibaudeau

What is this? A full frame camera with all sample images shot between 100-800 ISO?? Nothing above 800! When DPR reviews a 4/3 sensor camera, the ISO is pushed beyond ISO 6400 on their samples.

Does Canon need help to hide their high ISO performance?

We want to see samples at ISO 12000 and beyond for the new 5D full frame cameras.

Jun 24, 2015permalink

mpgxsvcd

It doesn't go beyond ISO 12,800 and even that ISO is already in its expanded range.

Jun 24, 2015permalink

JeanPierre Thibaudeau

So it's even worst than I thought, then.

Jun 24, 2015permalink

nandbytes

the native ISO is 6400. That's kind of a trade off you are making for 50mp.

Jun 24, 2015permalink

BeaniePic

Jean Pierre who's "We". I for one don't!

Jun 26, 2015permalink

JeanPierre Thibaudeau

I do understand why you do not want to see the high ISO samples. I am scared too.

Jun 26, 2015permalink

J A C S

It seems that the default black point of ACR with this camera (blacks too black) triggered shadow pulling in almost every shot. What if one just adjusts the black level instead?

Jun 24, 2015permalink

Johannes Zander

Nice pictures...
just don't blow 'em up to much and you won't see any noice.

Jun 24, 2015permalink

https://www.flickr.com/photos/33448355@N07/sets/72157655014407345

at ISO 100-200? congrats!

Jun 24, 2015permalink

Zorak

Seriously, that's bad. Even considering post process, result on this first picture is fugly, that's only around iso 1600...

Jun 24, 2015permalink

Digimat

Guys...you need to look at the conversion settings, that first iso800 shot is +1.1EV, +12 Shadows, +13 Blacks, + Tone Curve. Thats at least 2 Stops lifted (ISO3200) and we all know the canon sensors dont like to be pushed.

Seems to be about the same Noise you get out of a 7D II, which is..well...kind of okay but not terribly good. However, you can use quite aggressive Noise Reduction and still retain a lot of Detail for common print sizes, because you have so many Pixels to play with.

So...you will have a lot of Resolution up to about ISO800, maybe equivalent Resolution up to ISO3200 compared to a 5D3, and a bit better than average Canon APC-S Res @ISO6400.

Jun 24, 2015*permalink

J A C S

You know that Canon sensors do not like to be pushed even at ISO 800?

It should be the same as your 7DII at pixel level.

Jun 24, 2015permalink

Digimat

i dont have a 7DII, but yes, thats exactly what i wrote, isn´t it? ;)

Jun 24, 2015permalink

Menneisyys

The ISO400 shot of the same subject, done with similar lighting (SAM_023_ISO400) uses dialed-down shadows (-16) and only +.25 Exposure. Only blacks have been increased, which shouldn't have an effect on, say, the colored sky's noise. This results in an effective ISO of about 400 - and not more.

Nevertheless, the resulting image is still noisy. Of course it's a lot better than the pretty much useless first shot you've referred to - but still noisy.

Jun 24, 2015permalink

Andrea Visconti

"pretty much useless first shot"? Wow! your usability standards are quite high! What do you do with your photos?

Jun 24, 2015permalink

Menneisyys

Andrea, just compare the noise level to, say, these shots:

These are OOC JPEG's shot with my Fuji X-E1, at base ISO, with 30s, last summer in a particularly dark night (Finnish Summer nights aren't dark; this one was dark because of the heavy clouds). They show little noise.

Of course, it's apples-to-oranges (APS-C vs. FF; ISO200 vs. ISO800; 16 Mpixels vs. 54 Mpixels) but it's the end result that counts. And my shots are a lot cleaner than those of the Canon.

Jun 24, 2015permalink

J A C S

The ratio of detail vs. noise in those Fuji shots is disturbingly low, given the fact that SS did not have to be limited.

Jun 24, 2015permalink

Andrea Visconti

Menneisyys, you're also comparing a OOC JPEG (that has some degree of noise reduction applied) with a directly converted RAW with exposition pushed up and no noise reduction applied (at least, I suppose it's not since it's not specified). The noise in that photo is almost only luminance noise that is quite easy to get rid of just in lightroom.

Anyway my point was that the photo looks to me still perfectly usable for any use I can think of, not that has particularly low noise.

Jun 24, 2015permalink

capanikon

Wow ... such low noise at ISO 800.

Jun 24, 2015permalink

futile32

Resolution and Detail are very good as one might expect.

Jun 24, 2015permalink

andra1978

you can try shots any flowers... with canon 5DSR on flowers is more fine detaily compare with sony senzors.... for example

Jun 24, 2015permalink

garyknrd

I do not know much about these cameras. But the 800 ISO shot looks like my 7D noise. I would not buy one. Just me I guess.

Jun 24, 2015permalink

thx1138

You should have gone to spec savers. The 5DsR is better than the 5D III at least to ISO 3200 and downsampled it blows it away for detail, has less shadow noise and no banding. It's a class above the 7D II in noise and of course detail.

Jun 24, 2015*permalink

If this "blows away" a 5D III, then how bad was this camera??
I myself think this 5Ds is technically obsolete before it hits the shelves.

It's a good weather landscape camera, but even therefore the dynamic range is not state of the art.

Jun 24, 2015*permalink

From a layman's point of view:

I'm beginning to think that Canon hasn't figured out the technology yet to get lower shadow noise from their sensors, or that the only technology that is "usable(?)" is owned by sony & canon won't cough up the dough to get licensed to use it. Something's goin' on that makes no sense, to me anyway. I'm still hopin' that Canon does have a clue & is gonna show some progress in the very near future.............but the hopes are fading.

Otherwise, very nice photos. Resolution is very nice, but I'd rather have 24M images & better shadows.

Jun 24, 2015*permalink

ProfHankD

Canon's using an older fab and sensor design, but the Magic Lantern folks have figured out how to use alternating lines of different ISOs to get near-Sony DR from at least some Canons. I'm sure they'll do it ASAP IF it is technically possible....

Jun 24, 2015permalink

thx1138

Before you comment like this do some research, the 5DsR is much better than the 5D III. And no it's not as good as the Sony sensors, but it's much closer than you might expect. Sure you can't or shouldn't push shadows more than 3 stops, but comparative analysis to D810 show the 5DsR to be doing admirably given the older sensor tech.

Jun 24, 2015permalink

Zdman

As it happens the enhanced Dynamic Range of the sony sensors is only at ISO's below 800. If you took a similar shot with the 810 you'd probably get only slightly better noise because of the lower pixel count.

Jun 24, 2015permalink

badi

Mike, Canon is using an older tech (mainly larger scale for electornics) as it is quite expensive to switch the process. It's not really a matter of "they don't know how to", it's just a matter of time and money.
Probably a little while they'll continue with current tech, and at some point will decide either to switch their manufacturing to more modern standards, or to buy sensors from others companies.

Jun 24, 2015permalink

http://tinyurl.com/CanonNoise

thx1138......Your "observations" could be debated (better than the 5dIII), but this is probably not the place. I'll keep an eye on the Canon Full-Frame forum, as a spectator, to see if there's a "rabbit in their hat"

Jun 24, 2015permalink

Prof Hank & Badi.........thanks for 'splainin that. I don't pay enough attention to what's goin' on behind the scenes, just focus on the results I see here and elsewhere.

Danke, Mike

Jun 24, 2015permalink

So ISO 800 is the maximum usable and even this is crazy noisy?
What a shame... Canon seeing impossible

Jun 24, 2015permalink

Samuel Spencer

Not so fast, read the ACR conversions in the descriptions. It was a dark scene, and parts of the image have been lifted. After adjustments the dark parts are more comparable to a higher ISO than 800.

Jun 24, 2015permalink

I could see that. But even if it's pushed to like 1600 or 3200 it's still bad. I mean maximum ISO of this camera is 6400 right? So I would guess the maximum usable ISO is one stop below the native range would be 3200, but more likely 800-1600 is acceptable as it looks like. So that's comparable to the performance of a 2011 APS-C camera with 24 megapixel (NEX-7) with twice the sensor size but same pixel size.
In other words by today means, it's quite an expensive good weather camera...

Jun 24, 2015*permalink

My four year old Nex 5n doesn't look that noisy adjusted like that in Lightroom...
Look at the doll's heads in the iso640 car-interior shot...

Jun 24, 2015*permalink

thx1138

You guys are seriously delusional and clearly have no clue. Download some files and compare it to D810 and 5D III even at ISO 6400 and then talk crap.

Jun 24, 2015permalink

JohnTh

@Eric Hensel: Sooo... your Nex has 50 MP?
I think that's pretty unfair to do such a biased comparison.

Jun 24, 2015permalink

Delusional? It's there for all to see.

Jun 24, 2015permalink

EOSHD

ISO 800 on the aurora shot not looking too nice is it? Looks as noisy as the Nikon D750 is at 6400!!

Jun 24, 2015permalink

I do not understand: EOS HD is a member of DPR? Because if yes, then how I should read this post? From which perspective, as long as we compare a 24 Mp camera with a 50 Mp camera, and the latter is noisy? I understand the emancipation of Nikon/Sony adepts around, but I have difficulties in understanding this sort of comparisons... Sony A7s should be compared as well. It will provide a surprise in high iso noise. Don't you think?

Jun 24, 2015permalink

SETI

Can you please use different lens? Can we see studio portraiture? Thanks

Jun 23, 2015permalink

graybalanced

So, where in North America were you able to see the Northern Lights? DPReview is in Seattle now, right? Was it in that region?

Jun 23, 2015*permalink

Samuel Spencer

Yep, north of the city in a place called Camano Island.

Jun 24, 2015permalink

Technically North America includes Canada.

Jun 24, 2015permalink

graybalanced

Thank you Samuel! Found it on Google Maps. Wow, that's definitely further south than the aurora is typically seen.

HowaboutRAW - If DPreview is able to get aurora photos I was curious how far south they were. Since they are based in the USA I guessed the pics were not taken in Canada and that turned out to be true. Therefore, even though it is a fact that North America includes Canada, in this context, that fact was not very important.

Jun 24, 2015permalink

graybalanced:

My point is that the contiguous US states are not the entirety of North America.

Indeed, I've seen faint northern lights in Minneapolis.

Also: Mexico is considered part of North America too.

Jun 24, 2015permalink

@HowaboutRAW: and there was Pangaea, once upon a time...

Jun 24, 2015permalink

Is Southern Oregon contiguous enough?
https://scontent-sjc2-1.xx.fbcdn.net/hphotos-xpt1/t31.0-8/q90/s2048x2048/11312819_902389303132894_8231258476074054307_o.jpg

Jun 24, 2015permalink

The Silver:

And what does billions of years ago have to do with anything?

Jun 24, 2015permalink

EH,

Yes, but why ask?

Jun 24, 2015permalink

Due to very few Leica articles, I'm improvising. Sometimes this includes geography.

Jun 24, 2015permalink

HaR: "Yes, but why ask?" It was a joke, son.
Folks seem surprised the Aurora was visible south of Canada

Jun 24, 2015permalink

graybalanced

HowaboutRAW - Not sure why you're bugging me about the North America thing. The only, and sole, reason I mentioned NA at all, is because it was a verbatim quote from the article itself. If it's that important to drag Canada and Mexico into it, it looks like you have an issue with the author of the article.

Jun 24, 2015permalink

graybalanced,

But the article clear that the lights were unusually far south in North America.

Jun 25, 2015permalink

steelhead3

Nice getting the northern lights...I did see some grain in the sky in a few pics...is that out of the ordinary?

Jun 23, 2015permalink

Samuel Spencer

That's due to processing. Check the descriptions in the images for specific adjustments. The edges are going to show it more than center due to vignetting correction.

Jun 23, 2015permalink

vignetting correction makes a ISO 800 shot so noisy? That's the problem of those Canon sensors, no dynamic range...

Jun 24, 2015permalink

Mick Benjamins

Grain is what you get on film, this is noise.

Jun 24, 2015permalink