It depends on the location on the chart (as some already noted). For example the 5DM4 gives less color artefact than the "mighty" D810, which is quite good.

https://www.dpreview.com/reviews/image-comparison?attr18=lowlight&attr13_0=canon_eos5dmkiv&attr13_1=sony_a7rii&attr13_2=nikon_d810&attr13_3=pentax_k1&attr15_0=raw&attr15_1=raw&attr15_2=raw&attr15_3=raw&attr16_0=25600&attr16_1=25600&attr16_2=25600&attr16_3=25600&attr126_3=normal&attr171_1=off&attr171_3=off&normalization=compare&widget=1&x=-0.88061848256023&y=0.8314884601902539

At least I can't see the big differences from before where Canon always had a lot of banding and color noise where Nikon/Sony/Pentax still remained clear.

Wow grats it slightly beats a D810, a two year old camera with more pixels to deal with!

Didn't you mean to say the Sony gives a more lively jaundiced look?

My idea was that that the human eye (brain) is specialised in reading faces so this would be the best place to judge image differences, since modern sensors are so close.

@stepanfo "When stopped down, lenses used in the studio shot comparison are equally sharp. "
Pls prove your statement.

A quick look into DXO (although I don't like DXO very much) shows the opposite. For even more emphasis on price I selected the Otus for as example:
https://www.dxomark.com/Lenses/Compare/Side-by-side/Canon-EF85mm-f-1.8-USM-on-Canon-EOS-5DS-R-versus-85mm-F1.4-EX-DG-HSM-Canon-on-Canon-EOS-5DS-R-versus-Zeiss-Carl-Zeiss-Apo-Planar-T-Star-Otus-85mm-F14-ZE-Canon-on-Canon-EOS-5DS-R__241_1009_322_1009_1383_1009

Emfor: Click measurements, sharpness, field map, select aperture 5.6 - all are the same - excellent.

@Step Well, the field map colors of DXO are not very accurate, are they?
Does this green mean 77% or 99% ? The scale ends at 80%.

I'm interested in a lens comparison at 50mp like photozone.de started. Unfortunately those lenses used by DPR have been testet by photozone in the past at 20mp (not all of them have been tested there). It is no surprise that they perform similar (but not equal) at 20mp.

Quite the authoritative post and I love the alpha male proclamation at the end! Flex for me! Now bare your teeth and say "GRRRRRRR!" *click* (reviews image) oooooohhh, intimidating! You MUST be right if you're that emphatic!

Depending on the specifics of your work.
Autofocus is probably worse than in any modern amateur camera. RAWs falling apart, if only slightly touch the slider "Shadows". It still works, and works well - but there is nothing "awesome"

Not just that but also this is most likely the default picture profile and not the dedicated portrait picture profile.

Its a beta version of ACR unfortunately.

Adobe modified the Adobe Standard profile in order to make it more independent. For Canon, that means dramatically less saturation, starting with 80D I think.

'Camera Standard' profile should get you back to Canon land. You can try this with some 1DX Mark II files. See what happens with the reds when you go from Adobe Standard -> Camera Standard. "Booom" as Fro Knows says...

Many Sony APS-C and FF mirrorless are all compressed RAW with exception of A7II and A7RII if you updated it to latest firmware. I rather stick to Canon or so who does uncompressed raw file.

nathto: Not quite. Canon _only_ offers compressed RAW. But this is actually a good thing, because Canon RAW is compressed losslessly, which means that no data is lost in conversion and back.

What the hell: https://www.dxomark.com/Lenses/Compare/Side-by-side/Nikkor-AF-S-NIKKOR-85mm-f14G-on-Nikon-D800E-versus-Sony-FE-Carl-Zeiss-Sonnar-T-STAR-55mm-F18-on-Sony-A7R-versus-Canon-EF85mm-f-1.8-USM-on-Canon-EOS-5DS-R__388_814_1252_917_241_1009

Even on a 50MP sensor the canon is last in sharpness... At f/5.6 the canon is still last. It's really hard to make this comparison fair.

Lets mount the Otus line lens for Canon 5D MK iv and the results will be significantly different!

Current comparison is real eyewash. Honestly why dpreview stuff do such nonsense?

DPR staff, could please comment on this? Posting such a high-profile studio comparison shots taken with vastly different era / quality lenses does not seem that fair.

Is the reason that Canon lacks any better lens for suitable focal length (85mm)?

@emfor - thanks for the correction. I had 1.8 on the brain after both Canon cameras used the Canon 1.8.

To me, using the Sigma 50mm Art would have made sense. It's available in both Canon and Nikon mount, and for Sony via adapter. At that point, we're comparing sensor resolution, AA (or not) filters, and CFA for each of the cameras, which is how it should be.

@ogl They have to be sharper in some aspects since they omit AA filter (sony and nikon, don't know pentax AA?). However they pay with pronounced moire: https://www.dpreview.com/news/9234421352/canon-eos-5d-mark-iv-added-to-studio-scene-comparison?comment=4552972155

Certainly not better.

Yeah.. If you would check different parts of the test chart you would see that the result vary. That's because the test is done incorrectly, or the test chart had suffered some malformations in time.

You mean.. Sigma Quattro is sharpest out of all cameras on the test...

That's the magic of going AA less. AA-less brigade will still claim AA filter is a bad idea.

UGLY - blue and yellow banding.

Wow - 2500 ISO. Not all that useful

Meant no disrespect. That number just seems low for a FF camera. Happy to use my 5D mk ii in that range but it does come down to taste.

@Yossarian: You are comparing at 100%, choose the same size. The D810 is very close (which is very likely due to the lacking AA filter. Do some NR to get the same level of detail and they will be identical in real life). The Sony is still the best package in my opinion due to the dual gain sensor tech. It has other drawbacks compared to the 5div, though.

The D810 is barely worse than any of the comparable cameras at 12800 and below.

Plus if you aren't using the comp setting you are doing it wrong.

On top of that, the D810 has the cleanest files at low ISO than any other camera in these tests, excluding pixel shift from the Pentax K-1.

But never mind, let's complain that the D810 is a bit noisy at 12800, an ISO you probably use less than 10% of the time. :)

(Let's ignore that the D810 is also two years old)

By the way, moire hasn't been a problem with my D810 or 5DSR ever. I prefer the D810 because it has less clean up needed in post as the D810 is quite a lot better than the 5DSR at all ISOs.

Compared to the D810, A7R II and even the 5DSR the IV is lacking, it's an odd camera. Overpriced too, but the 5D III has been overpriced for a few years compared to the competition so this is not a surprise.

It doesn't matter whatever setting you choose, the D810 is noisier and the moiré is poor. Its closer with the Sony but moiré is a big issue here too.

Do you guys like moiré artefacts in your photos?

I don't particularly see that making observations on poor moiré makes me a fanboy but that's your call.

I realise that it is not specific to these sensors. However, in the context of this discussion, it is relevant to comment on it. Or should we just ignore it. If so, perhaps we should ignore the gain in sharpening?

@duxx High iso noise improved: https://www.dpreview.com/reviews/image-comparison?attr18=daylight&attr13_0=canon_eos5dmkiv&attr13_1=canon_eos5dmkiii&attr13_2=nikon_d810&attr13_3=sony_a7rii&attr15_0=raw&attr15_1=raw&attr15_2=raw&attr15_3=raw&attr16_0=12800&attr16_1=12800&attr16_2=12800&attr16_3=12800&attr171_3=off&normalization=full&widget=1&x=-0.27606038820992096&y=-0.7289629100408049

5DIV has much less noise in high ISO, especially chroma noise. 5DIV noise is much finer. Actually 5DIV seems to be better than 6D in ISO 12800 and above.

Doesn't have the moiré artefacts of the D810 or the Sony. And looks like better noise reduction at higher resolution.

@Duxx

Looks like Canon have added 8mps and reduced the noise. That is impressive.

Sensor looks better than D810 at higher ISOs and looks at least on par with sony at these higher settings. Does not suffer the moiré issues that both these sensors clearly have.

Will be looking at other factors such as DR and ISO invariance when these come out.

@Philou Strange enough, in your example I can still see some color artifacts for a6000 (with AA) in the lower left corner (left of the child's face) and on the right side (the jacket). Is it an ACR bug?

Saying the 5D IV looks much better than the D810 and A7R II is crazy to me. They are very close and considering the A7R II and D810 have substantially more resolution, you can add more noise reduction to make up for any perceived improvement on the 5D IV files noise wise.

That said, the 5D IV is barely better than the D810 until you push past ISO 6400. This is just another dated sensor from Canon.

I would much rather have a 5DSR than the 5D IV so that I have a huge megapixel count to clean up with NR or just switch to a D810 or wait for D810's successor.

@drive So it *is* better if you push past ISO 6400? But it is dated?
Nice ὀξύμωρος

The D810 doesn't appear to hold up very well in this comparison and suffers from moiré artefacts.

https://www.dpreview.com/reviews/image-comparison?attr18=daylight&attr13_0=canon_eos5dmkiv&attr13_1=canon_eos5dmkiii&attr13_2=nikon_d810&attr13_3=sony_a7rii&attr15_0=raw&attr15_1=raw&attr15_2=raw&attr15_3=raw&attr16_0=100&attr16_1=100&attr16_2=3200&attr16_3=100&attr171_3=off&normalization=full&widget=1&x=-0.6178278566250529&y=0.3980383649422197

By the way, the crisscross patterns in the D810 and Sony photos is not additional detail..... Yet alone the colours

@ photogeek

Is it really so bad? http://briansmith.com/sony-a7-a7r-a7s-lens-guide/

If you are into wide angle, or the 50mm focal length, then the Sony FE lens line up is the best in the industry.

If you are into telephoto it is the worst in the industry by far.

A7R2 runs circles, while MarkIV takes real photos...

And R2 takes unreal photos then? :-)

Can i have some of what you're smoking?

Actually, the Sony gives more detail and has more accuracy. The Canon has more vivid Colors . Canon for colors, Sony for black and white ?? I thought a camera should be good at both...
Anyway not that much difference but again like somebody else asked for, why not'mentioning the lenses being used for the test ? Sharpness also dépends on the lens, doesn't it ?

@Timbukto - Likewise, I have nothing against the AA filter. Viewing sharpness and noise are very subjective. This can be verified by reading the various posts on these two subjects. The fact that people adjust sharpening by points is further proof. Other photo content such as color and drama will trump these two every time.

Viewing sharpness is asking for an audiophiles review of a stereo system. A simpler task is to rate the legibility of fine text, in which case the MP is the overwhelming factor beyond AA filter.

I would love to mount my Sigma 24-35 Art and 100-400 II on this new bad boy!

Price ? Are you kidding ? More than 4000 euros in Europe, that's a shame !