DPReview Adds EOS 5DS R To Studio Test Scene

[neuroanatomist]

Iso100: look at the Color wheel/red patch next to the playing cards - imagine that seeping into something you want to show to a client. The same with the color checker, most obvious in the 2/2 blue patch. Or the green below it.

This is probably some strange effect of the beta software that was used to convert the raw. If you download the raw file, it looks just like every other camera.

It's an artefect of demosaicing & AB stretching from random LSBs.
ACR doesn't have mandatory denoising parameters set yet while DPP already knows what to do. Transforming from L to AB makes things less obvious, but more intrusive and harder to fixthough. :-[

Speaking of bias...you were quite happy to bash the camera's IQ, presumably knowing at the time that the issue was the half-baked RAW conversion software. Interesting....

If you're in the job market, you should see if DxO is hiring. You appear to have the necessary qualifications…

 

[Lawliet]

Iso100: look at the Color wheel/red patch next to the playing cards - imagine that seeping into something you want to show to a client. The same with the color checker, most obvious in the 2/2 blue patch. Or the green below it.

This is probably some strange effect of the beta software that was used to convert the raw. If you download the raw file, it looks just like every other camera.

It's an artefect of demosaicing & AB stretching from random LSBs.
ACR doesn't have mandatory denoising parameters set yet while DPP already knows what to do. Transforming from L to AB makes things less obvious, but more intrusive and harder to fixthough. :-[

Speaking of bias...you were quite happy to bash the camera's IQ, presumably knowing at the time that the issue was the half-baked RAW conversion software. Interesting....

If you're in the job market, you should see if DxO is hiring. You appear to have the necessary qualifications…

It either speaks of your bias or inability to read to the last sentence.

 

[PureClassA]

Downloaded RAW and opened in LR. Looks good. Color Moire in black and white pencil sketches (played with the one on the left) , but Moire brush got rid of it nicely.

 

[PureClassA]

And yes, the 85 1.8 is very dated, but it's still a darn good performer. It is a bit frustrating though that these other camera bodies are being shot with much newer and far more expensive glass. If the more appropriately comparable $2000 85L really wasn't ANY better then no one would be buying it...

 

[zlatko]

In the Studio Test Scenes, Can someone explain to me why DPreview is using the 85mm f/1.8 from 1992 which was designed for Film cameras on a 50mp 5DSR while the Nikon D810 studio scene gets the crazy good and expensive Nikkor 1.4G which is one of in not Nikon's very best prime lenses with barely any CA and excellent contrast?

The reason is ... let's see ... um ... oh, yeah ... WTF!

 

[sanj]

I checked on Brian's website. It is not amongst Canon's sharpest lenses.

 

[Eldar]

What lens should they have used, and why?

They seem to be having issues with off brand, so that would rule out a zeiss otus
and of course rule out at least some cross brand consistency.

In my view they should have used an Otus. Then you could have the same top performing lens on both Nikon, and Canon. That would make their comparison tool much more interesting. I do not understand why they should have issues with 3-party lenses, if that is what makes your comparison valid.

 

[Mt Spokane Photography]

I checked on Brian's website. It is not amongst Canon's sharpest lenses.

Other than LOCA's, the 85mm is sharp enough, and at f/5.6, the corners and edges are virtually equal to the center sharpness. That's a very important item that is often overlooked. CA's are low as well.

Just be happy they did not use the 50mm f/1.4, it has the center sharpness, but otherwise it suffers.

It would be interesting to have them use a Zeiss lens that was similar for Canon, Nikon, and Sony, but they have a huge database of images using that lens, so its probably better to keep on using it for comparisons sake.

 

[sanj]

I checked on Brian's website. It is not amongst Canon's sharpest lenses.

Other than LOCA's, the 85mm is sharp enough, and at f/5.6, the corners and edges are virtually equal to the center sharpness. That's a very important item that is often overlooked. CA's are low as well.

Just be happy they did not use the 50mm f/1.4, it has the center sharpness, but otherwise it suffers.

It would be interesting to have them use a Zeiss lens that was similar for Canon, Nikon, and Sony, but they have a huge database of images using that lens, so its probably better to keep on using it for comparisons sake.

You right. I checked only at full open. But most lenses fine at 5.6. No?

 

[exquisitor]

And yes, the 85 1.8 is very dated, but it's still a darn good performer. It is a bit frustrating though that these other camera bodies are being shot with much newer and far more expensive glass. If the more appropriately comparable $2000 85L really wasn't ANY better then no one would be buying it...

I also wondered how 85L II can be worse than 85/1.8. According to TDP the 85L II looks a bit better in the corners than 85/1.8 at f/5.6. http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=397&Camera=453&Sample=0&FLI=0&API=6&LensComp=106&CameraComp=453&SampleComp=0&FLIComp=0&APIComp=4 At least not worse... Of course there is sample variation, but I would expect the much newer design like 85L II (2006) to deliver better results than old design for film (1992), especially in terms of contrast given new advances in coatings and optimizing it for digital.
Anyway, IMO the comparison is still valuable and telling a lot about the performance, which is great...

 

[Valvebounce]

Hi Dilbert.
That is part of the human condition,* mostly we will complain about stuff to anyone and everyone except the one person that can actually change what we are complaining about! ;D
* It seems to be a very British condition but happens to some extent everywhere I have been!

Cheers, Graham.

In the Studio Test Scenes, Can someone explain to me why DPreview is using the 85mm f/1.8 from 1992 which was designed for Film cameras on a 50mp 5DSR while the Nikon D810 studio scene gets the crazy good and expensive Nikkor 1.4G which is one of in not Nikon's very best prime lenses with barely any CA and excellent contrast?

Why don't you post a comment on dpreview and ask that question there rather than here?

 

[zlatko]

What lens should they have used, and why?

85L or 100L, because they are better lenses.

 

[privatebydesign]

I have always hated it when people have proclaimed a sensor size smaller as a next size up "killer", ie, the latest Nikon DX or the Fuji X100T is a FF killer.

I have always baulked at people who have said any 135 format sensor is a medium format killer too, but looking closely at the test images it seems there is a much bigger difference between the Phase One IQ180 and the Pentax 645Z than there is between the 645Z and the 5DSR, which I find remarkable.

Obviously the best MF digital is still going to best the 5DSR with ease, but for those contemplating a 645Z (and the associated lenses), depending on your shooting style and subject matter, the 5DSR deserves a serious look, particularly when you throw in the Canon range and quality of lenses.

I think as the software and processing skill level increases on these 5DSR files we are going to have a bit of a base level realignment.

Personally I still dn't have a need or desire for 50MP, I'd love an excuse for one but I just don't have it, but if these IQ improvements are carried through to the 1DX MkII at 25ish MP I will be one very happy photographer.

 

[ritholtz]

What lens should they have used, and why?

85L or 100L, because they are better lenses.

DPR says it is the sharpest prime with FL between 50-85mm. Can't believe Canon yet to come up with better prime than 25 year old one. Sharpest Canon prime between 50-85mm FL is a nice discussion to start.
From DPR:
Essentially, we try to shoot a camera w/ the best available on-brand 50-85mm prime. For Canon, that's the 85/1.8.

 

[raptor3x]

What lens should they have used, and why?

85L or 100L, because they are better lenses.

Neither of those lenses are significantly sharper than the 85 1.8 at f/5.6.

 

[privatebydesign]

What lens should they have used, and why?

85L or 100L, because they are better lenses.

Neither of those lenses are significantly sharper than the 85 1.8 at f/5.6.

The 100L is noticeably better in the corners, certainly for contrast (which is a core aspect of resolution), and in the center it doesn't introduce the false colours the 85 f1.8 does so an accurate resolution measurement is much easier.

In truth the 100L would be a much more appropriate lens to be doing these tests with.

 

[zlatko]

What lens should they have used, and why?

85L or 100L, because they are better lenses.

Neither of those lenses are significantly sharper than the 85 1.8 at f/5.6.

The 100L is noticeably better in the corners, certainly for contrast (which is a core aspect of resolution), and in the center it doesn't introduce the false colours the 85 f1.8 does so an accurate resolution measurement is much easier.

In truth the 100L would be a much more appropriate lens to be doing these tests with.

Thank you. I was just about the write the same thing. IQ tests on the-digital-picture show the 85L is slightly better in the corners, and the 100L is significantly better in the corners. The 85/1.8 is a fine lens, but there is a difference.

 

[zim]

What lens should they have used, and why?

85L or 100L, because they are better lenses.

Neither of those lenses are significantly sharper than the 85 1.8 at f/5.6.

The 100L is noticeably better in the corners, certainly for contrast (which is a core aspect of resolution), and in the center it doesn't introduce the false colours the 85 f1.8 does so an accurate resolution measurement is much easier.

In truth the 100L would be a much more appropriate lens to be doing these tests with.

Thank you. I was just about the write the same thing. IQ tests on the-digital-picture show the 85L is slightly better in the corners, and the 100L is significantly better in the corners. The 85/1.8 is a fine lens, but there is a difference.

Thanks for all the replies, very interesting, the 85L has it then as the 100 is outwith the range they use which is a pity.
As an aside I didn't realise that one of the great strengths of the 85L was actually sharpness at higher Fs, all talk seems to revolve around wide open creamy goodness. It must be a fabulous lens to use!