Pages: [1]   Go Down

Author Topic: Canon 5DsR High ISO testing and mini review  (Read 6203 times)

tesfoto

  • Full Member
  • ***
  • Offline Offline
  • Posts: 145
Canon 5DsR High ISO testing and mini review
« on: June 06, 2015, 03:27:03 am »

Hey all LL

I had the chance of testing a pre production model of Canon 5DsR up against my 5DIII.

Testing can be done in many ways, I tested the way I would use the cameras in real life, and this is my findings: (I only shot RAW not any jpgs - sorry but I can't post any files, you have to take my words on this)

Low ISO / Sharpness:

The 5DsR is much sharper than my 5DIII, this is no surprise and the difference is like applying a sharpening to the 5DIII of 100/1/3 in Photoshop.
But it seems like the 5DsR has a much better contrast and color depth, it kind of has a more 3D look compared to the 5DIII - for me this is very good news.
Noice is very well controlled, and you can pull the shadows a lot and still have a clean file (see later)

High ISO:

Much to my surprise the 5DsR is as good as the 5DIII up to ISO 6400, perhaps even a little better. I did not test 12.500.
I can be little hard to judge as the files have different sizes, but developing the 5DsR to the size of the 5DIII give a really good impression.
I would happily shoot 3200-6400 ISO in hand held low light situation with my 5DIII, and will do the same for the 5DsR - as a reference I would no go to 1600 on my 5DII

Night shot:

I did some long exposure testing (city) and som high ISO too - and the 5DsR are on pair with the 5dIII. However the 5DsR have a different color look (more warm) and I kind of like the 5DIII better.
Noice is very well controlled, I could pull the exposure in Lightroom up to 4 stops and still have a usable file. The 5DIII I could pull about 2 stops. Very clean shadows, but pulling 4 stops shows that is is still a Canon camera (magenta/green cast).

Color:

Much more color contrast and color dept in the 5DsR, however on long exposure the 5DsR seem to have different look.

Highlights:

The highlights are about the same, perhaps the 5DIII is a tiny bit better of controlling blown out highlights (lamps at night)

Moire:

To my surprise, it was kind of difficult to get some moire in the 5DsR. I had to shoot different fabrics / patterns several times to finally get moire (shot with 24TS) - it was actually easier to get moire on my 5DIII

Hand held:

I did not play around with the new 5DsR settings, but the 5DsR shutter is much more soft. However I can shoot handheld (24-70 at 70) and get consistant sharp images at 1/45s with my 5DIII, I had to go for 1/90 for the 5DsR, so that is a setback.

Other Findings:

The 5DsR appears to be a slower Camera, not as responsive or fast shooting as the 5DIII - you can feel the much bigger file sizes.
Lightroom is much slower in generating previews, making adjustment etc. that will many be fixed, but I am thinking about a new computer (currently using MacMini3GH i7 and 16GB Ram)
New great feature, a leveling arrow in the finder is great and sensitive - this is really great.


All in all a great camera as the 5DIII with much better image quality - I have on order, and the rumors tells me that Canon ship on June 16 in my Country (Europe)

Cheers, TES

Logged

Bart_van_der_Wolf

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 8914
Re: Canon 5DsR High ISO testing and mini review
« Reply #1 on: June 06, 2015, 04:06:55 am »

Hey all LL

I had the chance of testing a pre production model of Canon 5DsR up against my 5DIII.

Hi,

You're privileged, and lucky. Thanks for sharing your findings.

Quote
Testing can be done in many ways, I tested the way I would use the cameras in real life, and this is my findings: (I only shot RAW not any jpgs - sorry but I can't post any files, you have to take my words on this)

Low ISO / Sharpness:

The 5DsR is much sharper than my 5DIII, this is no surprise and the difference is like applying a sharpening to the 5DIII of 100/1/3 in Photoshop.
But it seems like the 5DsR has a much better contrast and color depth, it kind of has a more 3D look compared to the 5DIII - for me this is very good news.

Yes, I've been predicting this for a while already, the higher sampling density will result in a significantly higher MTF response. The color depth can also be partly caused by the supposedly new Color processing and maybe CFA filters.

Quote
High ISO:

Much to my surprise the 5DsR is as good as the 5DIII up to ISO 6400, perhaps even a little better. I did not test 12.500.
I can be little hard to judge as the files have different sizes, but developing the 5DsR to the size of the 5DIII give a really good impression.
I would happily shoot 3200-6400 ISO in hand held low light situation with my 5DIII, and will do the same for the 5DsR - as a reference I would no go to 1600 on my 5DII

You may get even better results by sticking to a maximum of approx. ISO 1600, and then pushing the exposure level further in the Raw converter. That way you also gain more highlight headroom because you still underexpose but do not amplify the signal into clipping. The only drawback is that the humbnail images that are embedded in the file will be relatively dark, but the Signal to Noise ratio will likely be better. Noise reduction algorithms may also be better at reducing noise on less amplified images.

Quote
Moire:

To my surprise, it was kind of difficult to get some moire in the 5DsR. I had to shoot different fabrics / patterns several times to finally get moire (shot with 24TS) - it was actually easier to get moire on my 5DIII

Sensitivity to developing moiré artifacts is reduced by the denser sampling of the 5DS R. The 5DS will be even better at avoiding it, and is probably better suited for fashion/fabric/product/architecture shots. The difference in resolution and MTF may be very small in practice, also because diffraction will already begin to visually soften the images at f/4.5 and gets more prominent at narrower apertures. Diffraction will function as a sort of Optical Low-Pass Filter (OLPF or AA-filter).

Cheers,
Bart
Logged
== If you do what you did, you'll get what you got. ==

tesfoto

  • Full Member
  • ***
  • Offline Offline
  • Posts: 145
Re: Canon 5DsR High ISO testing and mini review
« Reply #2 on: June 06, 2015, 05:43:54 am »

I forgot about diffraction, and I am not going into the more theoretical discussions, but this was my findings

Diffraction:

I tried several aperatures trying to get moire with the 5DsR, and I got the strongest moire at f8 the same shot at f11 the moire was still there but a little less, at f16 the moire was gone. At f16 the 24TS on the 5DsR was still sharper than the 24TS on my 5DIII at any aperature.

I am not sure if moire is telling something about diffraction, but I guess there might be a connection.

I photograps a lot of Architecture, and I will happily shoot at f11 on the 24TS both on the 5DIII and 5DsR - If I get moire (time will tell) I can go to f16 and still have fairly sharp images without moire. But this need to be proven in real life situations. I am not too worried about moire in the 5DsR and as an architectural photographer I will get this camera and not the 5Ds.

Cheers

TES
Logged

tesfoto

  • Full Member
  • ***
  • Offline Offline
  • Posts: 145
Re: Canon 5DsR High ISO testing and mini review
« Reply #3 on: June 06, 2015, 05:51:20 am »

Regarding Noise:

I find that the 5DIII produces great luminous noise at 6400 ISO, almost like film grain and I like this look a lot.
The 5DsR produces the same kind of noise and has a beautiful look at 6400 ISO too.

Cheers, TES
Logged

tesfoto

  • Full Member
  • ***
  • Offline Offline
  • Posts: 145
Re: Canon 5DsR High ISO testing and mini review
« Reply #4 on: June 06, 2015, 05:55:44 am »

BTW

Just confirming that the 5DsR body is identical to the 5DIII. Same battery, same motor grip, RRS L-bracket will fit.

Cheers, TES
Logged

Bart_van_der_Wolf

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 8914
Re: Canon 5DsR High ISO testing and mini review
« Reply #5 on: June 06, 2015, 06:37:18 am »

I forgot about diffraction, and I am not going into the more theoretical discussions, but this was my findings

Diffraction:

I tried several aperatures trying to get moire with the 5DsR, and I got the strongest moire at f8 the same shot at f11 the moire was still there but a little less, at f16 the moire was gone. At f16 the 24TS on the 5DsR was still sharper than the 24TS on my 5DIII at any aperature.

I am not sure if moire is telling something about diffraction, but I guess there might be a connection.

Hi,

There is a connection, they are directly related.

The TS-E 24mm II, which I also use (and like a lot), has the highest resolution (in a limited DOF) at approx. f/4.5 - f/5 at its center. Due to the more limited DOF at that aperture range, only those structures in the plane of best focus will run a risk of showing moiré and false color and mazing artifacts. This will be more obvious when you use a Rawconverter that extracts more resolution from your Raw files, like Capture One Pro (which since version 8.3 has preliminary support for the 5DS and 5DS R). The extreme corners, especially when shifted a lot (I try to not exceed 7 mm) keep improving until f/16, but then by that time the center will be going down in resolution. At approx. f/11 there is a good balance between center and corner performance, but then diffraction will probably already kill most potential aliasing artifacts, even on the 5DS R with its Low-Pass Cancellation filter.

Quote
I photograps a lot of Architecture, and I will happily shoot at f11 on the 24TS both on the 5DIII and 5DsR - If I get moire (time will tell) I can go to f16 and still have fairly sharp images without moire. But this need to be proven in real life situations. I am not too worried about moire in the 5DsR and as an architectural photographer I will get this camera and not the 5Ds.

If you only shoot architecture at narrower apertures, the 5DS R should do fine. However, when shooting at wider apertures (which is often possible due to the availability of tilt), you may encounter issues that take a lot of effort to suppress/repair in post-processing, if at all possible. In those cases (especially when using a tripod) it pays off to shoot an additional image at a narrow aperture in addition to the actual shot with a wider aperture. At f/14, the 5DS/5DS R sensor will be totally diffraction limited at the higherst spatial frequencies, and aliasing due to sampling pitch will be totally eliminated for the most important (green and red) wavelengths. Then the diffraction limited/filtered image can be used to locally repair problem areas. When shooting without a tripod, the 5DS will already prevent most issues due to its OLPF, and saves a lot of time in post-processing.

Cheers,
Bart
« Last Edit: June 06, 2015, 06:42:19 am by BartvanderWolf »
Logged
== If you do what you did, you'll get what you got. ==

Bart_van_der_Wolf

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 8914
Re: Canon 5DsR High ISO testing and mini review
« Reply #6 on: June 06, 2015, 06:44:56 am »

BTW

Just confirming that the 5DsR body is identical to the 5DIII. Same battery, same motor grip, RRS L-bracket will fit.

Yes, that's helpful for those who use such L-plates, no new investment required at that end.

Cheers,
Bart
Logged
== If you do what you did, you'll get what you got. ==

Khurram

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 488
Re: Canon 5DsR High ISO testing and mini review
« Reply #7 on: June 07, 2015, 03:17:26 pm »

BTW

Just confirming that the 5DsR body is identical to the 5DIII. Same battery, same motor grip, RRS L-bracket will fit.

Cheers, TES

So you didn't have any issues with the cable release???  I read from another reviewer that some of the cable slots may not be available on the 5DsR when using the 5DIII L-Bracket and pre-ordered the new L Bracket.  however, if there is no issues accessing the cable release with the older bracket, i'd prefer to avoid the additional cost.
Logged
----
[url=http://www.pbase.com/kssphotog

tesfoto

  • Full Member
  • ***
  • Offline Offline
  • Posts: 145
Re: Canon 5DsR High ISO testing and mini review
« Reply #8 on: June 07, 2015, 05:20:16 pm »

So you didn't have any issues with the cable release???  I read from another reviewer that some of the cable slots may not be available on the 5DsR when using the 5DIII L-Bracket and pre-ordered the new L Bracket.  however, if there is no issues accessing the cable release with the older bracket, i'd prefer to avoid the additional cost.


Sorry, I did not test with the cable release. I was not aware of this issue, well I might just pre-order the new RRS L-bracket as well.



Logged
Pages: [1]   Go Up