Sony a7R III added to studio scene comparison

>> Because DPR use different lens
Really? And DXO claims both lenses have roughly equal sharpness what raises a question that may be it is not due to different lenses but is the reflection of inaccurate focus or generally poor sharpness of A7RIII unless you use pixel shift.

11 months ago*permalink

darrinlingle

when doing these tests across different brand cameras, they should use third party lenses (different mounts) for a more accurate comparison.

11 months agopermalink

Peter von Reichenberg

Is it only me seeing the CA in A7RIII...? And bit sharper/cleaner picture of D850? Well...

11 months agopermalink

If you bothered to read any other posts you'd have the answer to your question. 90% of comments are about the CA.

11 months agopermalink

Peter von Reichenberg

I have dived in last five reactions only...

11 months agopermalink

nah

11 months ago*permalink

DPReview007

My pre-order for the A7R III went in a couple days ago!!!

Not having to waste hours (days actually) on AFMA is worth it alone!

Add to that:
- more consistent focusing as the PDAF is actually on the sensor
- not missing shots due to eye-AF
- no low-pass filter
- higher megapixel count

I don't even need this comparison tool vs. the 5D IV. It will give me better results on average.

Fingers crossed it works as advertised.

Still hoping Canon will come out with their FF mirrorless system soon. Meantime, Sony gets a chance :)

11 months agopermalink

Andrew2468

My 2p / 2cents worth...
The studio scene tool is more about lenses than sensors and electronics. How about some additional new tests aimed at the latter. E.g.
1. How resistant to electro-magnetic interference is the sensor/electronics. E.g. does the DR reduce if someone standing next to you is talking on their phone when you fire the shutter? And if you have to shoot near heavy electrical gear?
2. Temperature and humidity. E.g. is image quality the same shooting in the arctic as the tropics. How well does the sensor cope?
3. I think I read light hitting the sensor obliquely can be an issue. Can evenness be tested, preferably without a lens, e.g. direct light beams at the sensor in varying places and at varying angles and analyse the numbers.

Perhaps something like the above might reveal there are real-world differences between makes and models that we don't know about.

11 months agopermalink

Ladies and gentlemen, sdaniella just had a baby. LOL

11 months ago*permalink

fft2000

Wow, great questions! Finally someone thinking instead of just trolling.
1) Absolutely! The Sonies aren't protected, in fact most cameras aren't protected. Look here, you can clearly see the interference:
https://www.dpreview.com/reviews/image-comparison/fullscreen?attr18=daylight&attr13_0=sony_a7riii&attr13_1=nikon_d850&attr13_2=canon_eos5dmkiv&attr13_3=olympus_em1ii&attr15_0=jpeg&attr15_1=jpeg&attr15_2=jpeg&attr15_3=jpeg&attr16_0=100&attr16_1=100&attr16_2=100&attr16_3=200&attr126_3=2&attr171_3=1&normalization=full&widget=1&x=0.279307649659362&y=-0.7455291340042918
And as you can see it's the Olympus that does best in that regard. I think they make great use of the wireless antenna, they measure the external microwave background and use the antenna to counter and erase (hope it's the correct term, I am no native English ;)) the waves.

(continued in next post - stupid char. limit)

11 months agopermalink

fft2000

But there are solutions, e.g. you could buy one of those stickers available for your phone to protect you against microwaves. It will for sure improve your image quality! If you still think it's not up to your standard wrap your camera in tin foil but leave the front lens open - use wired mesh (e.g. metallic mosquito net) for that.

2) The sensor does well, only the IBIS will work worse. Mostly because YOU (the photographer) is shivering from the cold. But IQ in general will be much better as there is less radiation going on (more heat, more radiation) - and for that look at 1)

3) Those tests ARE done and for photographers that request ABSOLUTE image quality lens manufacturers came up with SHIFT/TILT lenses so that photogs can fine tune the angle the light hits the sensor.

11 months agopermalink

Rick Knepper

What program are folks using to view the A7r III RAWs?

11 months agopermalink

Angrymagpie

And I'm sitting here looking at stunning photos in the photo books made by photography maestros, wondering how the IQ police in DPReview's comment section would say about these images

11 months agopermalink

mjbauer

They are consumed with the piece on the iPhone X movie, they'll get here soon

11 months agopermalink

@angrymagpie
Most of the best photos ever taken were taken with a four element uncoated lens or similar.

But this is a gear review site. Complaining about an emphasis on testing and image quality is like taking a golf club to a tennis court and telling everybody they're playing the wrong game.

11 months agopermalink

Angrymagpie,
so you are sitting there enjoying yourself thinking 'why not go and write a stupid ass comment on DPR?'

11 months agopermalink

sandy b

From a Nikon user, they look stunning. As do the D850. Well done Sony.

11 months agopermalink

Angrymagpie

I understand that this is a gear review website. But I wasn't expecting this place to be populated with image quality fundamentalists who seem consumed by pixel peeping level quality at the cost of everything else

11 months ago*permalink

Lea5

@Angrymagpie. Wot? Where have you been the last years? This is THE place #1 of all brickwallshooters, pixel wankers and blackframers of the entire planet.

11 months agopermalink

Angrymagpie

@Lea5 the If only there's a more photography oriented forum with comparable following as the DPReview
I'm pretty sure that the forum has plenty of good photographers. Would be great if the site could shine more spotlight on them from time to time
It seems that many people here would benefit from some regular reminder that IQ is perhaps one of the least important aspects in photography

11 months agopermalink

Lea5,
that's a rare amount of gibberish there. You sound as if Basher and Matsui (fictional characters from Ocean 12) were in a front-end collision with sdaniella. LOL

11 months agopermalink

>> image quality fundamentalists who seem consumed by pixel peeping level quality at the cost of everything else

Where exactly did you see "at the cost of everything else"? I think you are imagining the problem that does not exist outside of your own head and are generally confusing things in your head. I did not see here anybody but you saying what you claim. So it must be just you who are on that line of thinking.

Do not forget that if I say I love good meat (e.g. excellent picture quality) that does not mean that I do not like salads (e.g. artistry not directly related to picture quality). You are just being downright silly here and the only one here who takes sides by refusing to acknowledge separate importance of both.

11 months agopermalink

@Angrymagpie, I'm what you'd call a rare case (is that rare?) of photographer and gear head in equal measures.

I've been a pro (with a very promising career) and I admit to love gear and pixel peeping.

But there's a catch: I make myself very aware to never let the gear head in me win.

What you see on this website (and thus complaining) is when the gear side wins on people.

Either a lot of people here have forgotten what photography is all about in favor of gear obsession, or they've never been a photographer in the first place.

"people here would benefit from some regular reminder that IQ is perhaps one of the least important aspects in photography"

Personally, I do that quite often, talking about the business aspect of pro photography, or inviting people to care less about charts, and to remember even 12 yo cameras are good enough when skill is employed.

11 months agopermalink

Cont.

Know what happens when someone does that here? Either sound of crickets, or it doesn't take long until we get back to the usual DR chart talk.

People want to keep their brick wall tests, DR talk and gear of the year obsession.

Discussing actual photography (or photography market) here rarely leads to anything.

No wonder there's next to zero pros left in DPR forums.

Wanna see a popular post? Buy the most recent camera, mess up your exposure and lift by 5 stops. That's popular.

It's will never be "buy a 10 yo camera, which is good enough, and go out shooting".

11 months agopermalink

@marcio_napoli
>> Either a lot of people here have forgotten what photography is all about in favor of gear obsession, or they've never been a photographer in the first place.

It is almost as silly as saying that if you are discussing the technical side of cars you are not a good driver. Discussion of the technical side of cars does not stop you being a good driver. You are jumping to all the wrong conclusions drawing cause-effect relationship of the wrong kind.

Let me tell you this. Like most good drivers like their gear and learn the technical side of it so do many photographers. While you can make good photos with crappy equipment it is rather silly to do that if you have a choice. But you are free to make silly choices.

11 months ago*permalink

@marcio_napoli
>> Buy the most recent camera, mess up your exposure and lift by 5 stops. That's popular.

You are the only one who suggests that. I did not read anybody else here having such silly ideas as you have. This really says more about you than about those 'others' you claimed you talked about.

It seems to me from your conclusions you did not get the point of "lifting by 5 points" as you ridiculously assume it is for wrong exposure while it really is not.

11 months ago*permalink

miksto, I don't know how to say it, but are you really talking about DPR?

Are you sure we're talking about the same place? Where have you been the last 10 years??

I can't even begin to take your comments seriously, as it's painfully obvious you're defending this place for whatever reasons. Maybe somehow offended by my comments?

If offended or not, it doesn't matter. To be honest, not here to make friends.

What matters is that we all know what this place has become, don't come with nonsense like " I did not read anybody else here having such silly ideas as you have."

You know full well what I'm saying.

Post a thoughtful discussion about photography market, technique, marketing, lighting, or even about gear itself (in more thoughtful ways than usual), you get sound of crickets.

Post a "here's a lens cap on shot, at ISO 100k or shadows lifted by a billion stops", you reach 150 posts in 3, 2, 1...

Please let's not be naive here, this is DPR we're talking about.

11 months agopermalink

@marcio_napoli
Learn to read. I said nothing about the place and your problem and your confusion equally have nothing to do with the place. You will not be happy anywhere if you assume if someone likes to know about cars they are a no good driver or that a good driver should not have interest in horse powers.

You just continue being silly attacking the place and people while your problem has nothing to do with it and is your own internal problem. May be you are the kind of photographer who is no good in technical side but that is wrong to start assuming everybody is like you. You must be too self conceited to think that way.

>> Post a "here's a lens cap on shot, at ISO 100k or shadows lifted by a billion stops", you reach 150 posts in 3, 2, 1...

And what? Ignoring the fact that this ridicule comes only out of you and in this form nobody else is bragging like that but you. Still if you think ignorance on ability to lift shadow is a sign of anything good then you let yourself down.

11 months agopermalink

@marcio_napoli
>> Post a thoughtful discussion about photography market, technique, marketing, lighting, or even about gear itself

You must be totally self conceited if you believe you are the only one to judge what is thoughtful as you seem to presume you know what is. May be yours are not really as thoughtful as you like to think and you are just fooling yourself because it makes you feel better about yourself? Or are you the only one who is right and almost a lost Messiah, unrecognized genius who's judgements are so ahead (or behind) of time that you better keep quiet and un-understood by your contemporaries? Or may be you have a small circle of similarly self-conceited ahead/behind of time geniuses? Any of this touches a cord in your soul and makes you happy about yourself? Or you are just happy from making others look bad as that makes you feel above them?

11 months agopermalink

I'm almost laughing, *almost* because it's not even funny, it's just silly.

Look pal, you say I'm judging? And what exactly are you doing about me, if not judging me on many different levels? Even worse: you judge not knowing a single thing about me.

I've been a pro with a very promising career, I did go the extra mile, I've put the extra effort. I did not buy awesome pro gear just to shoot brick walls and discuss DR on forums.

I'd be glad to link my portfolio here if that helps you chill down.

Did you read the part I'm also a gear head? I have nothing against gear. What I have something against (if you wanna call it that way) is obsession with the best gear to achieve the smallest goals, something that you see happening a lot here.

Everything I said has solid basis. They're not just random free attacks. And know what? Criticism is good, it makes one evolve.

Everyone knows what DPR forums is all about. Don't come with this BS saying I'm acting superior, or similar nonsense.

11 months agopermalink

>> Everything I said has solid basis.

That is an easily verifiable lie. I can immediately see a few points above that are of no solid basis and are instead very shaky and easily dis-proven opinions but I reckon it is not worth my time digging that hole. Nor is it worth yours if you are a pro. Enjoy yourself and be easier thinking you have the only truth.

11 months agopermalink

@marcio_napoli
"If offended or not, it doesn't matter. To be honest, not here to make friends."

Honestly, I don't know what you are here to do then. Certainly giving your lengthy, unwanted and dubiously accurate opinions to people that don't care to hear them seem like a total waste of time.

11 months agopermalink

Chris Tsorotes

Just wait for the A7r IV should be out in 6 months

11 months agopermalink

Why would it be?

It took the A7rII two and a half years to be replaced. The A7II was released 3 years ago and has not yet been replaced.

The A6000 has also not been replaced - after almost four years. The A6300 is two years old and the A6500 a year. None of these cameras has been replaced. Why would an A7rIV be out in six months?

Here's a tip for you - if you want to make a joke - pick something with a kernel of truth in it.

11 months agopermalink

lumberjack63

In o ne way you are right: While Canon and Nikon are sleeping, concentrating on DSLR and all the things, they are used to do, Sony is working hard, dictating the pace ... ;-)

11 months agopermalink

Rick Knepper

lumberjack63, you have to recognize the kernel of truth i.e. Sony releases new models more quickly than other brands. My problem with the quip being that it is old and worn out. Actually, I'd like to see Sony release an A7r IV in 3 months with a healthy increase in MPs.

11 months agopermalink

Photography Matters

That camera may be the A9r in 6-8 months.

11 months agopermalink

Chris Tsorotes

Maybe it's just me, but Sony doesn't have a good track record with long product life cycles and long term support.

11 months agopermalink

It's just you. You keep talking about things you clearly know nothing about. You can put the battery from the A99II into Sony's first DSLR from 2006. In fact from 2006 up till 2017 they have used the same battery in every single DSLR and every mirrorless - the FM500 and the FW50. The only deviation is with the new Z battery that debuted in the A9. The first camera introduced since the A9 also uses that battery, so they will obviously use that going forward.

They took over an obsolete mount (Minolta A-mount) and have produced 20+ bodies for it and 30+ lenses including a 500mm f4. They are on to their second generation adaptors for A-mount to e-mount, to allow A-mount users to migrate across with their lenses and have even produced an adaptor with a motor in it to power screw-drive a-mount lenses from 1985.

The A99II has a 42mp sensor and a pro spec. It works with lenses created 32 years ago by another manufacturer. If you think that's a bad record of support you're frankly trolling.

11 months ago*permalink

Chris Tsorotes,
its sounds like you bought a camera back in 2010 and haven't used the site until today.

11 months agopermalink

Rick Knepper

@Photography Matters

I haven't seen a hint of rumor of an A9r since July on SAR. Where are you hearing this?

11 months agopermalink

ozturert

Rubberdials, consistency in battery has actually been a negative thing :) Sony even had to give 2 batteries with A7RII and A7SII.

11 months agopermalink

hjs_koeln

@Rick Knepper & Photography Matters
Re. higher MPs: Yeah, I was hoping for more MPs as well. There were rather authentic sounding rumors about something new with approx. 70MPs on SAR (BEFORE the A9 came out). Ever since the a9 was released, no new rumors about higher MPs that I´m aware of.

Re. a9: Now with the A7r III having most of the new tech of the A9 I´m wondering about the purpose of the A9. Three possible scenarios:
a) a technology demonstrator type camera whose tech is trickling down into the a7-series
b) a singular model for sport shooters and journalists, periodically updated
c) the beginning of a new series (a9, a9s, a9r) to ring in the frontal assault on Canikon, with having Sony released cameras around the Canikon buyer base so far.

IF a new series, higher MPs may well be reserved for an a9r. If not, we will see higher MPs in a a7r IV (which may take a while).

And for dreaming: Imagine 70-80MPs & pixel-shift mode. I´d probably bankrupt myself for that.

11 months agopermalink

LiangMing

I am glad the Canon 5D MK4 is holding up IQ at ISO 800-6400, compared to the new competition from Sony and Nikon.

11 months agopermalink

It's only year old. I think that is the very least you could expect.

11 months agopermalink

The A7RIII consistently looks OOF compared to the D850 and A7RII.
Sometimes it almost looks as blurry as the 5D4 with its low pass filter.

11 months agopermalink

Repeating the same nonsense in multiple posts won't make it more right.

11 months agopermalink

https://www.dpreview.com/reviews/image-comparison/fullscreen?attr18=daylight&attr13_0=sony_a7riii&attr13_1=nikon_d850&attr13_2=canon_eos5dmkiv&attr13_3=sony_a7rii&attr15_0=raw&attr15_1=raw&attr15_2=raw&attr15_3=raw&attr16_0=100&attr16_1=100&attr16_2=100&attr16_3=100&attr126_0=1&attr171_3=1&normalization=full&widget=1&x=0.12944393801276208&y=-0.7417836124943415

It's consistent across the image. The A7RIII is soft.

11 months agopermalink

11 months agopermalink

@9vII
It's not a lens test. Try to ignore the differences in optics. The widget demonstrates sensor performance.

11 months agopermalink

If it's a lens problem they should get a better lens and do it again.
Everyone has always used this to compare basic sharpness across different sensors, and it was a reasonably good comparison until now.
DPReview is just destroying the value of their tool using that lens.

11 months agopermalink

9VIII,
you seemed to think this is a sensor problem, that is incorrect.

11 months agopermalink

>> Try to ignore the differences in optics

All good unless you consider that according to DXO the lenses used for A7RII and A7RIII are roughly equally as sharp. And if DXO is right then it means one of two:
1) In-camera sharpness (whatever that means from sensor to other processing that takes place) of A7RIII is poor unless you use pixel shift.
2) The A7RIII sample focusing is not spot on.

11 months agopermalink

skillet01

The a7riii JPEG's in the studio scene seem to have about twice the file size as the a7rii JPEG's, at all ISO settings. Are both JPEG Quality settings set to Extra Fine?

11 months ago*permalink

DPR: probably should have switched lenses when reviewing a lower profile camera...

11 months agopermalink

Mateus1

To be honest D850 shows a liitle more details and images are sharper but I would buy A7r III anyway.

11 months ago*permalink

ahrensjt

Agree on both points. I found the D850 to have better detail/color/contrast... but A7Riii is close enough, and the other features would pull me there (Eye Autofocus, AF tracking in video, among others).

11 months agopermalink

Sharpness is lens based. Not the best lens in use these days (switched to Sony 85 from Zeiss 55).

11 months agopermalink

The A7RII looks sharper than the A7RIII.
Either there is something wrong with the comparison or there is something wrong with Sony's new sensor.

11 months agopermalink

There is something wrong with the comparison. II used the Zeiss 55 1.8, a high end lens. III uses the Sony 85 1.8, a consumer lens.

Why the change, you ask? DPR uses first party 85s for image quality tests by default, but Sony didn't have a standard 85 until very recently. ALL prior FF E-mount testing had been done with the 55. This changes with the III.

Because ALL brands are tested with different lenses, it is impossible to test the resolving power of one brand's sensor to another. Because of that, these tests are for noise and color only.

11 months agopermalink

https://www.dpreview.com/reviews/image-comparison/fullscreen?attr18=daylight&attr13_0=sony_a7riii&attr13_1=nikon_d850&attr13_2=nikon_d750&attr13_3=nikon_d500&attr15_0=raw&attr15_1=raw&attr15_2=raw&attr15_3=raw&attr16_0=6400&attr16_1=6400&attr16_2=6400&attr16_3=6400&attr126_0=1&attr171_3=1&normalization=full&widget=1&x=0&y=0.07071072883657764

@BlueBomber
Succinctly explained.

11 months agopermalink

xPhoenix

Interesting. I would take the sharpness differences with a grain of salt. I sincerely doubt the D850 is sharper. As for noise, when viewed at full size, the Sony seems close to the D850. Difference is probably negligible. What I still find interesting is that the D500 looks cleaner at high ISO (when viewing 1:1). Also, if you're interested in the cleanest shots at high ISO, the D750 is still king, unless you want to drop $6500 on a D5.

11 months agopermalink

thx1138

The D500 is not cleaner at high ISO, I own the D500 and 5D4 and 1DX and they both are much better than the D500 above ISO 1600. The 5D4 and 1DX are as good as the A7RII/III and D850 at high ISO.

11 months agopermalink

xPhoenix

In the link I posted above, I compared the a7riii to the D850 and D500. The D500 is actually cleaner. It's clearly visible. I'm talking about 1:1 viewing. If you downsize the image from the a7riii or D850, then it's a different story, but pixel for pixel, the D500 is cleaner.

11 months agopermalink

TMax1980

Due to variances in lenses, and possibly external factors, it's difficult to gleen much from the studio scene comparisons. ... The problem is that ARW files are compromised when processed in ACR (same for Olympus). That the Oly studio scenes of Oly multi-shot photos were taken with a Panaleica lense and processed in ACR disqualifies the test.

Another factor regarding multi-shot photography is that it's critical for the camera to be locked down and not subject to the slightest amount of vibration. Photo studios in repurposed warehouse spaces with wooden floors are often prone to vibrations--footsteps, outside traffic, etc.

What lenses are used for comparing one camera to another?

11 months ago*permalink

Schellenursli

Does anyone know if Pixel Shift Mode can be combined with bracketing?

11 months agopermalink

Rialta

No, I read on Petapixel today is cannot be used with bracketing.

11 months agopermalink

secondclaw

It does not. Bracketing is disabled in this mode. At least that's what those with the camera have reported.

11 months agopermalink

Michael Ma

What's up with camera manufacturers removing their AA filters when it's obvious they still need it? I would rather not have false color in my images at the cost of losing some detail that I'll have to discard anyway. Camera manufacturers need a reality check on what's more important.

11 months agopermalink

iwannabesedated

Since 2012, I've been using a D800E, followed by a D810 (in 2014), and more recently a D500, and the lack of an AA filter has never been an issue for me. I suspect the same is true for the vast majority of D800E, D810 and D500 owners. On the other hand, the improved sharpness has been a true boon. I think Nikon made an excellent decision in leaving the AA filter out of the D850. Regards.

11 months ago*permalink

jhinkey

Likely because with the high resolution sensors the large majority of their users no longer need an AA filter to reduce false color. True that for some applications it is still an issue, but those are not the majority of users. Plus, it costs Nikon less for not having it on the sensor cover glass.

11 months agopermalink

Lightright

That's why you have pixel shift feature to compliment the loss of AA filters with higher resolution sensors coming out. You are only limited to shooting landscapes though which I think is a fair compromise since moire is more noticeable with still subjects having very little to no moving elements on the scene at all. Pixel shift eliminates moire while increasing the resolution and perceived sharpness of your photos at the same time.

11 months agopermalink

AA filter don't play nice with pixelshift...

11 months agopermalink

Felipe Rodríguez

I've also used (and enjoyed!) a D810 and now an A7RII, both without AA filter, and I've never experienced any trouble because of that (well, very few times a very slight moire, to be honest, but you had to be a real pixel peeper to detect it). Quite the opposite, I've noticed a clear improvement regarding sharpness. So, no AA for me, please.

11 months agopermalink

I've had moire on my A6000 and A6300s, which have an AA filter.

11 months agopermalink

thx1138

Owning a 5DsR almost never see issues due to lack of AA filter. Very occasionally in the feathers of parrots I’ll see moire, but for landscape, architecture etc it’s a huge non-issue.

Also as the sensor res increases you tend to push interference effects to smaller and smaller scales, the pattern needs to have pitch similar to that of the sensor (taking into account lens magnification of course). Something that might have caused moire on a 20MP camera at a set distance, will not cause it or it’ll be less noticeable on a 50MP camera, as now the pattern scales are not commensurate. Of course a finer pattern will cause moire on the 50MP camera, but in the real world that gets harder unless you are deliberately shooting repetitive patterns like fly screens or very fine high thread count fabrics say. It’s highly variable but in general it’s not something you have to worry about as the resolution increases.

11 months agopermalink

Lightright

For me, Pentax seems to have a better implementation of the pixel shift in K-1 than Sony's A7R III. The red thread on the spool on K1 appears to be far more distinguishable than A7R III but it looks less saturated compared to Sony's punchier and contrasty version. Other than that, both sensors performs spectacularly with the kind of lens used as the other factor for differences in comparison.
https://www.dpreview.com/reviews/image-comparison/fullscreen?attr18=daylight&attr13_0=sony_a7riii&attr13_1=sony_a7riii&attr13_2=pentax_k1&attr13_3=pentax_k1&attr15_0=raw&attr15_1=raw&attr15_2=raw&attr15_3=raw&attr16_0=100&attr16_1=100&attr16_2=100&attr16_3=100&attr126_0=1&attr126_1=2&attr126_2=1&attr126_3=2&normalization=compare&widget=1&x=-0.306752400436686&y=0.5199794149784049

11 months ago*permalink

Stu 5

Where is the A7R III pixel shift sample to compare because it does not show in the dropdown box.

11 months agopermalink

Lightright

Just click the link I provided.

11 months agopermalink

Stu 5

Yes totally agree with your observations. I notice on the playing cards there appears to be halos on some of the lines. Looks almost over sharpened.

11 months agopermalink

Sony's pixel shift image was processed in Sony's software, not ACR (like everything else), so it's hard to compare accurately. I'm sure they'll update when ACR becomes compatible with pixel shift.

11 months agopermalink

caravan

Hats off to Sony again ! Sony looks like the clear winner by a lot in the scene comparison.

11 months agopermalink

The A7RIII looks like either the lens or the sensor is having severe problems picking up detail. Even the A7RII looks better across virtually the entire scene.

11 months agopermalink

TOF guy

To me the recent batch of cameras are so close that high ISO performance is not what would decide me towards one brand or another

11 months agopermalink

bleachfix

I was just thinking the same. There is very little difference even when pixel peeping.

11 months agopermalink

https://www.dpreview.com/reviews/image-comparison/fullscreen?attr18=lowlight&attr13_0=sony_a7riii&attr13_1=sony_a9&attr13_2=sony_a7rii&attr13_3=sony_a7_ii&attr15_0=raw&attr15_1=raw&attr15_2=raw&attr15_3=raw&attr16_0=25600&attr16_1=25600&attr16_2=25600&attr16_3=25600&attr171_1=1&attr171_2=1&attr171_3=1&normalization=compare&widget=1&x=0.047130276343759506&y=-1.0934781164906688

Pretty much the same with APS-C. Sensors are hitting a hard limit with current tech. And Sony recently entered a new generation of sensors (A7R II, A6300), so who knows if anything else can be wrung out of them. The A7R II sensor has already been massaged to give it a slight high ISO advantage (and minor DR, depending on the source) in the III.

11 months agopermalink

coudet

Massive improvement at high ISO. Seems Sony's claims of reduced read noise are true.

11 months agopermalink

snapa

Especially if you compare the A7rIII to a few Canons & Nikons! It looks like Sony is pulling ahead of the pack very quickly.

11 months agopermalink

rhlpetrus

Really? To me, in RAW, the Sony and D850 look the same at 6400 and 12800, the highest I'd dare to go. Jpegs are not a good basis for comparisons, since, if you are going to do NR, dedicated softwares do better than in-camera.

11 months agopermalink

jhinkey

Hardly "massive" between the A7RII and A7RIII. It's better, but not by that much.

11 months agopermalink

Around 2/3 stop at 25,600.

11 months agopermalink

Sergey Borachev

Why doesn't DPR just use the same physical lens when possible, e.g. The same Canon 85mm lens mounted on the Sony camera as well as on the Canon camera?

11 months ago*permalink

brendon1000

Unfortunately it's not an ideal situation. The Sony canon adapter has some tolerances which can affect IQ.

One option is to use a sigma lens that is available on most mounts but that too has issues with sample variations depending.

11 months agopermalink

Sergey Borachev

Keep returning until.you get a good one. I mean DPR can find a good adapter, can't it? And also a good one for Nikon lenses. Shouldn't be too hard for DPR, I think,.esp since that can make its test results much more reliable or meaningful than tests results that used different lenses. And no more arguments any more about lenses affecting test results in future, for tests between FF cameras from these 3 top camera sellers.

Cheap food for all the apologists' complaints each time, but will the camera makers allow DPR to do that? No where to hide any more.

11 months ago*permalink

brendon1000

I am no expert but from what I have read a canon lens on Sony will never match entirely the performance of the same lens on canon. An adapted solution is generally a compromise in quite a few ways and IQ is one such way.

I can't find any sort of data to support my claims but the corner performance in general is the one most affected on adapted lenses used on Sony.

11 months agopermalink

manifest destiny

@brendon1000: the 28mm Zeiss Otus for Canon mount, adapted via Metabones onto the a7Rii gives corners so sharp I could use 'em as a saw!

11 months agopermalink

aftab

DPR knows better, but maybe they have changed the lens from 55mm (a7RII) to 85mm (a7RIII) to keep the playing field even as 85mm/1.8 has been used for both Canon and Nikon. This has introduced some unwanted consequences. a7RII RAW appears sharper than a7RIII RAW. Some are wondering if IBIS or focus was an issue. I don't understand how IBIS could be an issue if this feature wasn't used (DPR can clarify if it was used). I don't think focus was an issue, I don't expect DPR to be lazy with their tests. And then there is the question of CA. DPR mentioned earlier that Sony's 85mm/1.8 produces 'awful lot' of CA. Then why a7RIII's RAW (standard mode) doesn't show them but JPEG does? Interestingly, DPR hasn't said anything to clarify these issues. I know someone once said, If you don’t understand my silence, you won’t understand my words. :)

11 months agopermalink

Steven R. Rochlin

Yes, DPR does what they can and i commend their efforts. Agree the a7RII looks sharper than the a7RIII. Especially in the corners, so yes lens/focus issue is my bet here.

11 months ago*permalink

PWPhotography

FE 55/1.8 is in higher class than FE 85/1.8. Maybe in future DPR should also use the lenses from the same manufacturer such as Sigma 85/1.4 Art in the lab test. Sigma should release Art-series lenses to FE mounts next year. Rumor is 35/1.4 Art will be the first released to FE mount that originally planned in this year but now push to Q1 next year.

Also personally I prefer Imaging-Resource test scene over DPR's lab scene that shows more details.

11 months ago*permalink

jhinkey

Really these scene tests are only good for judging noise characteristics.
Since they use difference lenses in some cases, sharpness just cannot be directly compared between these high resolution sensors where the lens characteristics dominate the image sharpness.

11 months agopermalink

I expect DPR are looking into it, because the difference in sharpness between the A7R Mk 2 and 3 is so obvious.
Could the 85 be that much softer than the 55? Comparing to the 5D4 images I am wondering if there is a weak AA filter, because that is what it looks like.

11 months agopermalink

tangbunna

purple CA with pixel shift

11 months agopermalink

sensors doesn't produce CA, lenses do it...

11 months agopermalink

Richard Murdey

@Anulu

maybe so but the CA with pixel shift enabled is indeed much worse - but only at the border .. very strage

11 months agopermalink

Different software. ACR doesn't support pixel shift yet.

11 months agopermalink

If it was misfocused like many people here said, then the image would not be sharp in the pixel shift feature shot, would it? Didn't some of you also said the Arii was misfocused in its test?

11 months ago*permalink

Eric Hensel

If you just look at RAW high-ISO noise patterns -they are much sharper with the Nikon.
Baked in sharpening?
https://www.dpreview.com/reviews/image-comparison/fullscreen?attr18=lowlight&attr13_0=sony_a7riii&attr13_1=nikon_d850&attr13_2=canon_eos5dmkiv&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&attr126_0=1&attr171_3=1&normalization=compare&widget=1&x=1.0432432432432432&y=0.5022343158721924

11 months agopermalink

bartnoel

yes

11 months ago*permalink

rhlpetrus

Cameras' files respond differently to same sharpening settings and profiles. But I agree the D850's raw conversions look a bit oversharpened.

These camera are so close as to make IQ not an issue regarding choices. Body abilities, functions, how they will be used, and lenses available will be much more relevant to determine which is best for you. Pixelpeeping is not a good way to judge these cameras. (Still, some seem to see "huge" differences. Maybe in the eyes of beholders.)

11 months ago*permalink

Daiken

Man, that lens on the A7R III ruins this comparison. It's either really soft or out of focus. I know sharpness isn't something to look at for a sensor/camera comparison but it's really hard to ignore. I think the test should be redone with a better lens.

11 months agopermalink

Agree, but not better lens, just need to be properly focused, this lens is very-very sharp

11 months agopermalink

@Anulu
It's not that sharp actually, for its focal length. https://www.lensrentals.com/blog/2017/06/sharpness-tests-of-the-sony-fe-85mm-f1-8-g/

The 55/1.8 is better - as is the Nikon 85/1.8. I think if DPR is determined to use the Sony 85/1.8 they need to re-shoot the 7rII photos with it, especially as it is DPR highest rated camera in the semi-pro full frame category and as such is a reference point for users of many brands.

The widget is showing the A7rIII sensor as performing better than the A7rII one but the reduced accutance is confusing people.

11 months ago*permalink

@RubberDials
It's sharp, sharper than GM and Batis and sharper than 55 1.8 to. A tried all of them and i own fe 85mm 1.8
https://www.dxomark.com/Lenses/Compare/Side-by-side/Sony-FE-85mm-F18-on-Sony-A7R-II-versus-Sony-FE-85mm-F14-GM-on-Sony-A7R-II-versus-Zeiss-Carl-Zeiss-Batis-85mm-F18-Sony-FE-on-Sony-A7R-II__1818_1035_1680_1035_1536_1035

11 months agopermalink

@I'll take Olaf over your opinion if you don't mind. :)

Did you actually look at Roger's curves? And 10 copies tested on an optical bench beats 1 lens tested on Imatest at god knows what distance by DXOmark.

11 months agopermalink

Doesn't need to look charts :D Because i using it and used the other lenses to. I look my photos not charts :D

11 months agopermalink

@Anulu
And yet you felt the need to link to DXOmark... :)

Back in the real world, the MTF curves of ten samples tested on an optical bench trumps your statement about the one lens you own, presented without any evidence.

You won't really find anyone who doesn't think their 85 is the sharpest lens ever created because they tend to shoot portraits with it - which is one of the most undemanding subjects you can shoot if you want to test the sharpness of a lens - large structures of smoothly transitioning tone. You only need good contrast at 10p/mm for that.

11 months agopermalink

You doesn't even know what am i shooting... :D
I'm felt to need link dxo, because to show you, not just i'm saying fe85 is sharp :)
But you are trully ridicolous, if you think i will take your opinion and mft charts over my REAL experience, and Real mean using them in the real world, for photography, for work... real doesn't mean looking charts and websites on the internet as you do :D
Take my advice : Rent those lenses take up your ass, leave your internet at home, and take pictures ! Don't let be fool you by websites, because pictures born in the real world :)
Anyway the mentioned lenses are all stellar, all of them ! The differences are negligible, but this test shoot is clearly oof, as many other folks already mentioned on this forum, maybe this happened again :)
https://www.dpreview.com/news/8576711309/now-we-know-sony-a9-is-sharper-than-we-thought
And sorry for my bad english, it's not my primary

11 months ago*permalink

Gon0S

@Anulu
So, if experts like Dpreview reviewers are, one more time, fooled by manual focusing on a SONY body, maybe the error is somewhat in the SONY device and not the reviewers, isnt'it ?
Why do they managed to 'nail' it with Nikon, Canon, FUji, Olympus, Pentax, Hasselblad, Leica, Panasonic, etc... but SONY ?

11 months agopermalink

@Gon0S
I don't know what and how are they doing , but i never had an issue with Manual Focusing in a Sony body

11 months agopermalink

Burnt to a Crisp

Great to see Sony putting Nikon to shame! And off-tripod, Sony's formidable IBIS should make the difference even greater.

Canon isn't even in the running...

11 months agopermalink

Jetfly

Why? Have you ever heard about IS?

11 months agopermalink

rhlpetrus

I don't know where these brand fans see these differences. Well, that what's DPR is about ...

11 months agopermalink

@Jetfly

Have you ever heard of IS + IBIS? Yes, they can combine. ;)

11 months agopermalink

Gon0S

LOL, the shame is for all the guys like you using hyperbolic wording for meaningless differences...

You must really have never used anything else than a SONY Alpha to claim Nikon latest bodies are put to shame.
But as your account is not older than a few weeks, I guess you have to show the world that using a high spec SONY device give you the ultimate power to tame other brand users... The funny part is that all your claims change nothing to your skills. And throwing big words to the face of other brand users will never give you better skills.

So, why not go back to video game forums and battle again between xbox fanatics versus sony fanboys ? THIS is your real playground.

11 months agopermalink

Funny, since I used Nikon for years, and switched to Sony to alleviate issues I was having with them (which reviewers are just now starting to notice). No, it hasn't changed my skills, but it has allowed me to have full confidence in the camera, and to continue shooting where I would otherwise be chimping. And feel free to check my account registration date. :)

11 months agopermalink

Gon0S

@BluebomberTurbo
Does @Burnt to a Crisp be your war account ?

About chimping, still seeing loads of MILC users 'chimping' at theirs pics afterwards...
At best, you chimp directly in the EVF. At worst, you chimp using the back lcd screen. And in the end you sort your keepers on the +24" monitor...

11 months agopermalink

Life recorder

Two things.

1) the D850 appears to be a better camera with higher output IQ.
2) if you don’t believe #1 then you are agreeing these comparisons are of little value because too many things can affect results.

The truth is there are a dozen things more important when considering a camera....unless you are a gear head who thinks pixels viewed at 200% are more important than content...and the D850 output is clearly better.

11 months agopermalink

You are right in a sense, because there are so many variables, however: I recently purchased a landscape photography book (Looking For The Light, Peter Latham, New Zealand). He uses a Nikon D810 and a Mamiya 645 / Phase One P65, and mostly primes.
I am guessing a bit, and while shots from the D810 look good, the level of detail from the P65, even when reproduced in a book, is truly outstanding - I haven't seen anything this good before (detail wise).
If you print and are good at it, there appear to be significant benefits from high resolution (and good lenses no doubt).

11 months ago*permalink

Clayton1985

@Life recorder... I don't fully agree or disagree with #1 or #2. If you do consider the variables then you can get value out of these comparisons without needing to declare a camera to be clearly better.

11 months agopermalink

"the D850 output is clearly better."
No, it isn't
You doesn't even looked the test shoots right ? :)
https://www.dpreview.com/reviews/image-comparison/fullscreen?attr18=lowlight&attr13_0=sony_a7riii&attr13_1=nikon_d850&attr13_2=canon_eos5dsr&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&attr126_0=1&attr171_3=1&normalization=full&widget=1&x=-0.24609778317643485&y=-0.9019021739130434

11 months agopermalink

Mais78

Looks like the standard Raw settings for the A7riii hide noise in the shadows by closing them more than the other cameras. Once you open them to the same level you will get same amount of noise of mkii and d850?

11 months agopermalink

@Mais78

That's how Adobe chose to profile the camera. And as we've learned, Adobe sucks at profiling. Not Sony trying to hide noise.

11 months agopermalink

https://www.dpreview.com/reviews/image-comparison/fullscreen?attr18=daylight&attr13_0=sony_a7riii&attr13_1=nikon_d850&attr13_2=canon_eos5dmkiv&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&attr126_0=1&attr171_3=1&normalization=full&widget=1&x=-0.24211485870556068&y=-0.8611136260751471

The D850 is clearly sharper.

11 months agopermalink

Mais78

@BlueBomber I doubt it is Adobe as this camera is not yet supported by them

11 months agopermalink

cantor

Looking at the wood cut or the paper bill, I would say the Nikon D850 is significantly sharper than the A7riii at ISO 100 (raw). That surprised me. Is the image in focus?

11 months agopermalink

Softer lens than previously used.

11 months agopermalink

ozturert

Rubberdials, I think Canon 100mm f2.8 L IS has 4 ax IS, others are 2 but in telephoto lens-based solutions are better than IBIS (probably worse than ibis+lens based solutions). That's why Olympus delayed 300mm f4 Pro to add IS.

11 months agopermalink

stevo23

Which lens? What if DPR uses an Otha for every FF and APS test?

11 months agopermalink

MyReality

As long as the same lens is used for all tests, it does not make a difference. If I measure two distances in inches, there is no problem, but if I measure one in inches and the other in centimeters, there will be a problem.

11 months agopermalink

stevo23

@MyReality - Exactly. They should use the Zeiss Otus 85mm at f/5.6 and thereby eliminate all potential variables other than the sensor. (Even though using the same one would ensure consistency, why not use the best every time?)

11 months agopermalink

https://www.dpreview.com/reviews/image-comparison/fullscreen?attr18=lowlight&attr13_0=sony_a7riii&attr13_1=nikon_d850&attr13_2=sony_a7rii&attr13_3=sony_a7sii&attr15_0=raw&attr15_1=raw&attr15_2=raw&attr15_3=raw&attr16_0=25600&attr16_1=25600&attr16_2=25600&attr16_3=25600&attr171_2=1&attr171_3=1&normalization=compare&widget=1&x=-0.16847859095050105&y=-0.9019021739130434

@stevo23

The adapters add another level of inconsistency. They can tilt the focal plane depending on tolerances, and add slight misfocus to the top or bottom of the scene. Then again, it seems like even that would be better than using the overly soft and colorful Sony 85/1.8...

11 months agopermalink

stevo23

@BlueBomber - I think it would be fairly solid, especially if you get a good one for each mount. It's just beyond me how anyone is supposed to take these as useful when they introduce yet another variable such as the FE 85 1.8.

11 months agopermalink

dbm305

There are lots of comments here comparing sharpness of small details between similarly resolving cameras. DPR maybe you should have a message telling people what to look for, and what aspects of the scene are likely down to lens and shooting, and what may actually be sensor performance.

11 months agopermalink

Mike Sandman

Yes, that would be helpful.

Looking closely at these comparisons, one thing is clear. They all have superb image quality, and pixel shifting brings them close to the Phase One benchmark. Whether it's Nikon 850, Sony a7Rii and iii, Pentax K1, or Canon Mark IV, the IQ differences are small, most notably in noise.

Image quality is not much of a differentiator for this class of camera - you have to look really hard to see differences. What you should choose depends on what you're looking for -- mirrorless technology, ruggedness, weather sealing, dynamic range, range of lenses and accessories, size/weight... to each his own. Whatever criteria you prefer, there's an amazing option.

11 months ago*permalink

TrojMacReady

Everything as expected (mostly similar to the A7RII, minus variations due to probably the lens used), except this:

Seems like a nice shadow noise improvement still, creeping closer towards the A7S(II) in this area as well (the last still imaging stronghold of that sensor).

The difference is even much more pronounced at ISO 51200 and its top ISO (102k), but here the addition of the usual RAW NR muddies the waters of most of these sensors.

11 months ago*permalink

I don't know if it's the lenses or, as some people suggested, the IBIS is lowering the sharpness, but there is definitely something wrong going on here. I expected to see a better IQ from the D850, but the IQ from the A7R III is strangely soft. Even if compared to the A7R II.

11 months agopermalink

Just curious, why did you expect 42mp sensor out-resolves 46mp sensor? Any particular reason?

11 months ago*permalink

I don't. I think you misunderstood what I wrote. I expected to see a better image quality from the D850 than the A7R III.

11 months agopermalink

Ah got it, sorry.

11 months agopermalink

No worries!

11 months agopermalink

Different lens. No IBIS on tripod tests. And the test is for noise and color reproduction, not sharpness.

11 months agopermalink

That's a good point.

11 months agopermalink

Arca45Swiss

Are those green leaves in the test scene Cannabis ?

11 months ago*permalink

Mike Sandman

dPr staff: I assume that comparing JPEGS is a less "accurate" way then looking at RAW files to compare cameras. But there's a question about RAW, too:

Why is the a7Riii RAW file 82MB while the a7Rii is 41.4 MB? Are you using uncompressed RAW for the a7Riii and the default compressed RAW for the older version? Doesn't that introduce some uncertainty in comparing these two Sony cameras?

11 months ago*permalink

aftab

Something curious happening with a7R III's JPEG conversion. Look at four corners and toggle between RAW and JPEG. JPEG seems to have loads of CA while RAW doesn't. As if camera's JPEG is creating CA which didn't exist in RAW. How is this possible? So, maybe those are not CA produced by the lens, rather some sort of color artifact? Also, why did DPR change lenses between a7RII and a7RIII? 55mm used with a7RII seems to have better corner sharpness and than 85mm used with a7RIII.

11 months agopermalink

seb_t

Well actually, I've downloaded the RAW and checked by myself (using CaptureOne), and indeed the RAW do *not* have AC in the corners while the JPEG does... weird.
But on another hand, what I've also checked is a comparison of the full JPEG from ACR available for download (which does match the RAW - so no mismatch) with the preview JPEGs used in the web tool (retrieved using network traces). For example: https://2.img-dpreview.com/files/p/S300x200T6x9~image_comparison/studio/faa391a74ead43429b9b7866cacd94e2.acr.jpg
... and it appears that the preview web JPEG are a bit softer than the ACR JPEG they are extracted too. I'm not sure but it looks like a stronger JPEG compression used by the webtool for the A7r3 preview thumbnails. The difference with the original JPEG from ACR is very slight but it does make a visual difference, especially since the preview thumbnails from the other cameras does not present such difference.

11 months agopermalink

seb_t

A picture is worth a thousand words:
https://i.cubeupload.com/zVBtxM.jpg

Above: an extract from the A7R3's "JPEG ACR" made available for download
Below: the same extract as used in the DPR web preview tool (URL in my post above)
Merged JPEG save as quality 12 and uploaded in cubeupload which does not add extra recompression. The difference is subtle, yet the JPEGs used for preview in the case of A7R3 RAW looks more "jpeg compressed"

11 months agopermalink

Looks like the older A7RII is sharper unless you use pixel shift on the new A7RIII. With pixel shift it is kind of on a par with the older A7RII. Also the newer A7RIII seems to have some funny color shift compared to other cameras.
While expecting improvement the studio comparison seems to show the older one is better.

11 months agopermalink

sai1

A7RIII is clearly sharper on my screen than the A7RII with RAW selected

11 months agopermalink

ClaytonTurner

Not on mine and I'm viewing it on a 5K screen. The rii is clearly sharper.

Its also been noted that 2 different lenses were used.. which probably is responsible for the differences.

11 months ago*permalink

sai1

Still we should see the same camera in the top spot. ;-)

11 months agopermalink

>> A7RIII is clearly sharper on my screen than the A7RII with RAW selected

Odd. I was also checking with RAW and the old is way sharper WITHOUT pixel shift. With pixel shift it is just on a par. I have a 4K monitor.

>> Still we should see the same camera in the top spot. ;-)

Looks like Nikon gets it.
Still I am surprised the new Sony is worse than the old Sony. Something is fishy.

11 months agopermalink

sai1

Hm had to check again. You're right A7Rii looks sharper, must have mixed that up somehow. Probably not 100% in focus.

11 months agopermalink

The LENS on the II is sharper than the III. What's between the lens and sensor (sensor cover, microlenses, CFA layer) should be optically the same between cameras, and have the same potential for sharpness.

11 months agopermalink

G Sciorio

Wonderful options available for photographers. Loads of frustration for camera geeks.

11 months agopermalink

johnperrydk

Exactly

11 months agopermalink

sbu

Lots of anger from the sony fanboys in this thread, lol -
Now they will not be able to enjoy photography anymore knowing d850
has a promille better image quality.

11 months agopermalink

https://www.dpreview.com/reviews/image-comparison/fullscreen?attr18=lowlight&attr13_0=sony_a7riii&attr13_1=nikon_d850&attr13_2=sony_a7riii&attr13_3=nikon_d850&attr15_0=raw&attr15_1=raw&attr15_2=raw&attr15_3=raw&attr16_0=6400&attr16_1=6400&attr16_2=12800&attr16_3=12800&attr126_0=1&attr171_3=1&normalization=full&widget=1&x=0.14467051320983895&y=0.561141304347826

Uh, try looking at the test first?

11 months agopermalink

Dietrich Reinhardt

At 100 ISO, looks to me that D850 out-resolves the competition. Also, can't say that I see much difference between the two Sony's.

11 months agopermalink

mick232

I doubt the D850 outresolves any camera with pixel shift.

11 months agopermalink

Magnar W

The results might well be a competition between lenses, more than the different sensors!

11 months agopermalink

cna

Pretty clear to me ...
In RAW, the D850 out-resolves all 3 other cameras at any ISO.
I do not understand why the Sony a7rIII is softer than the Sony a7rII (lens ?)

11 months agopermalink

@cna Glad it wasn't just me.
(RAW) D850 looks best, A7Rii looks noticeably sharper than Mkiii, which is probably the lens.
Can DPR shed some light on this please? (Apologies if you already have...)

11 months agopermalink

feraudy

I have the impression that the D850 jpegs have more sharpening applied to them.

11 months agopermalink

cna
"In RAW, the D850 out-resolves all 3 other cameras at any ISO."
Look it again this time with opened eyes :)
https://www.dpreview.com/reviews/image-comparison/fullscreen?attr18=lowlight&attr13_0=sony_a7riii&attr13_1=nikon_d850&attr13_2=canon_eos5dsr&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&attr126_0=1&attr171_3=1&normalization=full&widget=1&x=-0.06067415730337077&y=-0.8054347826086956

11 months agopermalink

tgchan

THANK YOU/.

11 months agopermalink

NicoPPC

Well, every moderns cameras do the job vey well. Maybe the 5D4 a little better in high ISO. Fuji X-T2/20, D810, even 5D3 doing well..

Nothing one would really notice on real-world pictures...

DPR: There is not any sharper lens for Fuji ?

11 months ago*permalink

"5D4 a little better in high ISO"
Better than what ?

11 months agopermalink

vscd

Than the A7R3

11 months agopermalink

Patlezinc

I agree. A7RIII more noisy at high ISO (« high » is not adapted, it starts at 800 iso)

And we can see some magenta fringes all over the frame (lens issue?)

11 months agopermalink

https://www.dpreview.com/reviews/image-comparison/fullscreen?attr18=daylight&attr13_0=sony_a7riii&attr13_1=canon_eos5dmkiii&attr13_2=canon_eos5dmkiv&attr13_3=fujifilm_xt20&attr15_0=raw&attr15_1=raw&attr15_2=raw&attr15_3=raw&attr16_0=12800&attr16_1=12800&attr16_2=12800&attr16_3=12800&attr126_0=1&attr171_3=1&normalization=full&widget=1&x=-0.6339714372531146&y=-0.5993662290629244

Haha, You guys are joking right ? Look it again :)
https://www.dpreview.com/reviews/image-comparison/fullscreen?attr18=lowlight&attr13_0=sony_a7riii&attr13_1=nikon_d850&attr13_2=canon_eos5dmkiv&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&attr126_0=1&attr171_3=1&normalization=full&widget=1&x=-0.06067415730337077&y=-0.8054347826086956

11 months ago*permalink

NicoPPC

@Anulu. Yes.

11 months agopermalink

randomguy

You guys are looking at full size, if you are going to compare noise at least view it at the same size for all cameras (hit the comp or print button top right).

11 months agopermalink

vscd

@ISO12800 even the "old" Canon 5D Mark III holds up well. No big difference to newer sensors...

11 months agopermalink

@NicoPPC
Look it again with the right settings. Lower resolution and bright scene hide the real noise level.
https://www.dpreview.com/reviews/image-comparison/fullscreen?attr18=lowlight&attr13_0=sony_a7riii&attr13_1=canon_eos5dmkiii&attr13_2=canon_eos5dmkiv&attr13_3=fujifilm_xt20&attr15_0=raw&attr15_1=raw&attr15_2=raw&attr15_3=raw&attr16_0=12800&attr16_1=12800&attr16_2=12800&attr16_3=12800&attr126_0=1&attr171_3=1&normalization=compare&widget=1&x=-0.6339714372531146&y=-0.5993662290629244

11 months agopermalink

Yup, gotta use the low light setting. Unless you're shooting ISO 12,800 in broad daylight...

11 months agopermalink

vscd

@Anulu. Even if you're right... that's RAW and nobody ships the final picture without getting rid of the chroma-noise. DXO alone would leave the Sony and the Canon quite equal...

11 months agopermalink

IMHO, the samples taken in studio scene with A7R III are misfocused.
Try to compare with A7R II side by side and you will find every image soft focused from iso 100 up to 102400...
I think DP testers set the (wrong) focus on ISO 100 and then set the camera to MF for any other iso.
It is not a lens issue because the softness is on both corner to corner and the center area.
Considering this, the studio scene test is pretty useless.
And... Pixel Shift on iso100 only? :/

11 months ago*permalink

Livio Spallone

It doesn't seem possible that the new Sony is worse of the old model, they should shot the scene again.

11 months agopermalink

I think you are right.
I was blaming the lens, but after another look, the differences in sharpness are too much, surely.

11 months agopermalink

https://www.dpreview.com/samples/4348798595/sony-fe-85mm-f1-8-sample-gallery

It's so strange to see so many pro users here that ignore all of this...
Someone says the 85mm lens is full of CA, but they should know when the CA comes out too...
...misfocusing

11 months agopermalink

aftab

85/1.8 produces 'awful lot' of CA according to DPR.

11 months agopermalink

dbm305

Visible differences is sharpness are all about the lens and shooting. These scenes do not test sharpness or resolution when the resolutions are so similar. nor do they test CA. At most they test noise. The exception is pixel shift, where the difference is great enough to be visible is a scene like this.

11 months agopermalink

@dbm305: You are possibly right, but I don't think it unreasonable for viewers to expect a studio test scene with lots of fine detail examples to show relative sharpness between different cameras.
As it is, the A7Rii is outperforming the Mkiii for rendered detail - Is it the lens? or something else like an AA filter, processing, IBIS - the list goes on.
A lot of us would like to know whether there is a problem, or as you suggest, the studio test scene is worthless for assessing detail.

11 months agopermalink

dbm305

It’s not impossible that there could be subtle differences between sensors with the same or similar nos of pixels, it when they both have no AA filter and IBIS is turned off like here I think if there were a visible difference between them one or both would be defective. Sharpness of image at low ISO on raw files just isn’t a point of difference that makes for choice between cameras with similar resolutions.
I agree though that DPR should say something. Posting studio comparisons with visible sharpness differentials like this is bound to raise questions in the minds of viewers. BTW, one significant difference which will explain part of it is the the 85mm lens used on the r3 will have less DOF than the 55mm one used on the R2, and at these distances that’s a huge effect compared to any sensor effect.

11 months agopermalink

Not so sure it was misfocused, the pixel shift features yields very sharp details.

11 months agopermalink

They definetely had to test it with the 55mm lens.
CA could vary also because if you choose a 85mm you have to place the camera further than on 55mm, and this change the angle the studio lights reflect on the lens.
But again, I am pretty sure we have a misfocus problem
To the ones who states 85mm has a very bad CA... are you really sure it is normal at f5.6? :)

11 months agopermalink

About the Pixel Shift shot, the camera could have been refocused for that shot, Sony Imaging Edge Software could have corrected CA, etc...

AND

I don't see electronic first-curtain shutter feature enabled on R III, as it was on R II

11 months agopermalink

I seriously doubt they refocused. Even for some weird reason that they do, what is the odd of the multiple shots were all in focus, but the single shot was not?
If you look at the jpeg image with some sharpening applied, the Sony file is almost as sharp as the Nikon's. In fact it, looks sharper than the Sony raw image itself.
But anyway, we're only guessing here, right? :)

11 months ago*permalink

maljo@inreach.com

Comparing jpgs is silly. Default contrast and sharpening varies among the manufacturers and greatly influences the results.
Nikon’s default settings handily beat the competition but that is about all one can say.

11 months agopermalink

Max Iso

Did you set it to RAW?

11 months agopermalink

danny006

The jpegs have more detail and less noise compared to A7RII. At higher ISO it does even better than the D850 in terms of detail. The 5DS is not performing good, the details are gone at medium and high ISO.

11 months agopermalink

Derrick Louw

When one looks at ultimate image quality, medium and high ISO should not form part of the equation.

11 months agopermalink