Sony SLT-A77 studio comparison samples

Now maybe people will stop picking on the 'ageing' sensor in the new PEN cameras as this NEW Sony sensor has apparently sunk to that level.
Like the PEN cameras, lets all look at the positives of these cameras and not just the high ISO performance.

Cheers

Sep 28, 2011permalink

mediman30

There's no point of convincing people to like this camera if this is not for them. I have tried a pre-production copy of the A77 last Saturday at an event sponsored by Sony UK and all I can say is it's a totally different experience. I have an a700 that I have been using for the last four years but the experience I had with the A77 after using it for only a day was completely different - in a VERY positive way. I don't need to mention all the features you get with this camera compared to the more expensive ones out there but there's only one thing I hope people will do when they see this camera in their local camera shops. JUST TRY IT!

Christmas gift for myself - check.

Sep 27, 2011permalink

Hubertus Bigend

The test shot is clearly out of focus.

Look at the image centre, the focus target, the playing card (which is a substantial amount behind the focus target) and the wall – I'm quite sure the image was focused by AF, and the AF missed the focus target in the centre and hit focus either on the playing card or even on the wall.

Stopped down to f/9 with an APS-C sensor, it's of course harder to notice than, say, with a 35mm full frame SLR at f/8, like the Nikon D3 (which shows everything perfectly in focus except for the playing card, which is massively out of focus).

Sep 27, 2011permalink

Camera Mount Clamp

That is a nice camera.

Sep 27, 2011permalink

harold1968

ISO is merely good, not excellent, however dynamic range and colour resolving is awesome (see D X O M A R K).
Arguablly real photographers with fast glass don't need to shot over ISO 800, although 1600 and 3200 can be useful.
The current obession with high ISO performance should not deter real photographers, this is an outstanding camera.
As M9 sales show, it is preferable to have detailed base ISO results that trash every other FF camera on the market then excellent really high ISO performance.

Sep 27, 2011permalink

eprahm

Very well said for those of us who are trying to be practical and not hung up in theory world. Will be my next camera within a month or so.

Sep 27, 2011permalink

eprahm

Very well said for those of us who are trying to be practical and not hung up in theory world. Will be my next camera within a month or so.

Sep 27, 2011permalink

+1!!! Having read all the above, I have a feeling, that nobody shoot at the ISO lower than 1600... If U prefer 7D with less noise but bad DR and colours, it's Your choice, but it doesn't mean, that 7D is better camera...

Sep 27, 2011permalink

onlooker

> As M9 sales show

Are you serious? I suspect if all Sony managed was M9-level sales, heads would roll. Sony is not a boutique operation.

And your definition of a "real photographer" as one who does not shoot above ISO 800 is laughable.

Sep 28, 2011permalink

Steve oliphant

Wow what these test are so different from image resource . com, the sony is at the bottom end of the test there.There seems to be money changing hands here,the canon was better at 1600,but the nikon clearly was the best of them at 1600 iso. I use a Canon 7D and it's very good, but i do work at a camera store and i have done my owen test, and the nikon seemed to be a bit better in the red's for sure, the 5D shots here are clearly out of focus, nice eh, and i know that this camera will crush all of them buy two stops.

Sep 27, 2011permalink

bavarius6

The JPEGS just don't hold up past 3200, details smeared away. Unfortunately for Sony everybody wants massive ISO's and this sensor just can't deliver, translucent mirror or no translucent mirror. Looks 2 stops noisier than 16mp sensor so taking SLT into account it's still 1 3/4 stops noisier. Like 24mp FF in D3x this sensor needs capping at 6400 max IMO.

Sep 27, 2011permalink

digby dart

The sensor is clearly straining under the weight of that translucent mirror and all those pixels, low shutter lag live view comes at a price. That said, seeing what the sensor sees is a revelation in photography, in old terms its seeing what the film sees which is fundamentally a mighty step forward in the pursuit of photographic excellence.

Perhaps the poorer high iso performance will be offset by the improved perception of final outcome that low shutter lag live view gives, only time will tell. Some of the a55 shots I've seen with better glass have been stellar.

Sep 27, 2011permalink

David G Peterson

The jpeg's look pretty good the raw's should get better with a more optimized raw converter. Another thing that concerns me is the need for really good glass for this and the Nex-7 with their 24mp APS-c sensor's....I guess you'll have to go primes or Zeiss $$

Sep 27, 2011permalink

ThomasH_always

Sony missed the memo again: "We want better pixels, not more pixels."

Both D7000 and 7D look much better in magnification (thus crop in real life) and of course, the 5D as the full size camera is still a stop or more ahead.

Note that Canon announced an Oct 26 EOS event, thus there might be something new from them, changing this situation even more.

Sep 27, 2011permalink

RedFox88

No, that is for a new printer not any camera.

Sep 27, 2011permalink

sensibill

There's no evidence 'it's the pixels'. The A33 with lower MP looks just about as bad, and there's the fact that a 100% crop of the 24MP is going to be a much tighter crop of a given scene. It could also be that Sony just didn't do as good a job with this sensor, and of course there's the fact that it's losing 1/2 a stop to that pellicle (which is not worth the sacrifice in light, IMO).

Sep 27, 2011permalink

Dustinash

Speak for yourself. I think this is exactly what I am looking for, if only it shot 4k video. but 60P? Hells ya. and 24 MP is very useful to some of us

Sep 28, 2011permalink

anagram4wander2

DPReview should be ashamed of itself.

The JPGs are similar across all the default cameras.

The A77 RAW conversion is using an Alpha version of ACR.

Unless Sony has some magic to make the JPGs look so much different.

This is not a level playing field.

Bye DPReview - That sucked.

Sep 27, 2011permalink

darkref

bye!

Sep 27, 2011permalink

treepop

people get so worked up :(

Sep 27, 2011permalink

Lars Rehm

It's a Beta-Version. We use those pretty frequently on many of our reviews as, very often, there is not yet any official ACR support for the cameras when we are testing them. We also use our own processing as detailed on our RAW sample pages. Images are processed in ACR with all sharpening and NR turned off. Sharpening is then added in Photoshop via an unsharp mask.

Sep 27, 2011permalink

anagram4wander2
"Bye DPReview - That sucked"
Not only DPR then, go to DxO, in SNR measurements put on screen mode and put A77 against D7000, nikon is whole step better than sony which is pretty much what I see in DPR's samples

Sep 27, 2011permalink

http://www.dpreview.com/galleries/5639653565/photos/1428222/studiocomp

Comon anagram4, the RAW converter version, type, brand or whatever makes only a slight difference to noise appearance if any. Sony just has a very good JPEG engine in the camera. Blaming DPREVIEW for this is a cop-out. You should be ashamed of yourself!

Sep 28, 2011permalink

Eleson

So, on the stamp, is the chin rastering done with lines or crosses?

Sep 27, 2011permalink

JPMontez

That is curious indeed!... Pentax 645D seems to backup A77's. Afterall more resolution seems to have its benefits...

Sep 27, 2011permalink

Eleson

so here are the ISO 1600 comparison

http://www.dpreview.com/galleries/5639653565/photos/1430052/studiocomp1600

Smooth noise VS details.
Nikon really have noise under control, but where are the details???
I like the 645D shoot :)

Sep 28, 2011permalink

djec

genuine question: why are the file sizes (raw) not larger relative to the other lower MP cameras?

Sep 27, 2011permalink

JimWarp

Lots of noise as JPEG or RAW (even with ISO 200), lots of CA, soft, strange colors. A definite buy ...or not.

Sep 27, 2011permalink

JPMontez

What do you think of these other RAW conversions made from the A77 samples with IDC and Capture One? Quite better than the ones coming from ACR, no?

http://forums.dpreview.com/forums/read.asp?forum=1037&message=39450219

Sep 27, 2011permalink

http://forums.dpreview.com/forums/read.asp?forum=1037&message=39453100

R I P A 7 7 V I V A 5 N ? : )

Sep 27, 2011permalink

jlamesa

Maybe guys, you all have to read that thread:

Sep 27, 2011permalink

Thanks! Very interesting!

Sep 27, 2011permalink

Menneisyys

Wowz... the 5N is exactly one(!) stop better than the A77... the A77 is indeed pretty much awful at high ISO's.

Sep 27, 2011permalink

As you may expect with higher pixel density but perhaps it should be not as much as we are seeing here

Sep 27, 2011permalink

RedFox88

What about all you that kept repeating the myth that smaller pixels doesn't equate to more noise? Certainly looks like that myth has been shattered with the noisy a77.

Sep 27, 2011permalink

sensibill

There's no myth. The size of the pixels and density of this sensor is still well ahead of compact sensors that can still turn out _relatively_ low read noise. The MP isn't the culprit here, it's the pellicle mirror sapping 1/3 of the incoming light.

Sep 27, 2011permalink

thx1138

Stop looking at pixel level noise it tells you nothing about different cameras with different resolution. A 100% view on the Sony A77 corresponds to a massive enlargement, a print several feet long and you would not stand 3" to examine a real print of that size. The ONLY fair way to compare the two cameras is to equalize their pixel counts, and usually you would downrez the A77 to 5N size. If so you would find most of that stop difference disappeared, but the A77 file might still have more detail.

I would also wait for full RAW support from ACR rather than look at terrible jpgs.

Sep 27, 2011permalink

JPMontez

When looking at dpreview's RAW samples at ISO 3200, Sony A77 looks indeed much noisier when compared to Canon 7D. But when you look at DxOMark measurements, noise difference at this ISO level is pretty much neglectible. What could justify this?...

Sep 27, 2011permalink

DxO renders RAWs differently than ACR I think

Sep 27, 2011permalink

Either DxO Mark can't be trusted or Sony still has to make a camera worthy of its sensor. (Well, maybe NEX 5n is it.)

Sep 27, 2011permalink

Aslo yo have to choose screen option for DxO graphs for better comparison of DPR samples to DxO results

Sep 27, 2011permalink

also lol

Sep 27, 2011permalink

ck3

AFAIK DXO measures noise in the midtones. If you look at the samples here, you'll notice that the 7D has quite a bit darker shadows than the A77 (whatever the reason for that may be). The darker shadows in turn help to conceal the noise as you will notice, when you manipulate the A77 samples accordingly.
Nevertheless the shadow noise of the A77 isn't something to brag about...

Sep 27, 2011permalink

Subjectively, the A77 images look much noisier than the 7D and this is where DxO misses the mark - pun intended. Their noise measuring instrument obviously operates on different parameters than the human eye. Pity because photography is all about what the eye sees.

Sep 27, 2011permalink

TrojMacReady

There's something going on with the Sony RAW files ever since the introduction of the A35. The A35 has the exact same sensor as the A55, but looks quite bad in comparison. The same blotchyness is present in the A77 RAW files as seen here. In fact, the A77 downsampled to A35 looks pretty similar, noise wise.
But both their DXO measurements don't translate into the visual performance one would expect and to me that's a first. The A35 unsurprisingly showed the same results in DXOMark as the A55, yet visual performance is a non sequitur. Likewise for the A77. Will be interesting to find out what 's going on here.

Sep 27, 2011permalink

Hannu108

" Their noise measuring instrument obviously operates on different parameters than the human eye. Pity because photography is all about what the eye sees."

So true.

Sep 27, 2011permalink

It was interesting to see that the panasonic GH2 with its even smaller pixels (on a smaller sensor) delivers more detail at ISO 800 and 1600. To me the new kid from sony looks like a 15yo who is almost to strong to walk.

Sep 27, 2011permalink

It's strange, but i see the opposite... Choose ISO 1600 and look at the playing cards, for example... And compare colours with that of 7D )) Did U still think, that Canon's sensor better with such colours. Ha-ha ))

Sep 27, 2011permalink

Low Budget Dave

Are you looking at the same samples I am looking at? The GH2 colors are compressed and the image grain past 800 shows clear evidence of over-sharpening.

Sep 27, 2011permalink

@hippo84: the playing card at 1600 JPEG seems to be going A77's way but raise it to 3200 and things are even. Turn it to raw and the A77's rendering gets dismantled.

Thanks for pointing out the card. It's funny but you are right. From the other areas I checked, the A77 was lagging behind. But however I also noticed that the other cameras were off focus here.

Sep 27, 2011permalink

Hey! was talking about the raw-files!

Sep 27, 2011permalink

And even looking at the jepg there is more fine detail in the GH2's 1600 ISO file than there is in the sony's 800 ISO file. outch!

Sep 27, 2011permalink

Button Pusher

@hippo84

It is pretty well known that the playing card should not be used as the ultimate reference point in this studio shot comparison due to focusing differences. Nice try though.

Sep 27, 2011permalink

Downsize the samples guys !!! downsize the samples !!! it will make the a77's samples looks as clean as tears of child, it will make disappear all this nasty noise, correct the colors and make comeback all lost details and it will be better than any other aps-c camera , just downsize them !!!
I am just kidding :)

Sep 27, 2011permalink

but...I'm serious...Anyway - GOOD JOKE!

it's myth, downsizing the samples NEVER lower the noise.
and it's absolutely idiocy to buy 24 MP camera, shoot in RAW and downsize every picture....

Sep 27, 2011permalink

TrojMacReady

Not a myth. There is a reason people in these forums state small pictures posted in forums doesn't tell us much about noise performance. I can show you 0.5 MP image from my A500 at ISO 12800 with no noise. But believe me, it's there at 12MP. ;)

Sep 27, 2011permalink

Frederik Paul

Why should someone downsize when I lose the extra MP then? You could simply use only 18 or 16 MP from the start.

Sep 27, 2011permalink

Yes...IIt's my mistake. Downsize from 12 till 0.5 MP - really works. :))) Or from 24 to 1-2 MP. But I don't think that we need to do it...

Sep 27, 2011permalink

Hannu108

@Frederik Paul, You'll gain little bit sharpness...

Sep 27, 2011permalink

putomax

that's not true, half the izeit lowers noise by half ;-)

Sep 27, 2011permalink

http://www.dxomark.com/index.php/Cameras/Compare-Camera-Sensors/Compare-cameras-side-by-side/%28appareil1%29/734|0/%28brand%29/Sony/%28appareil2%29/619|0/%28brand2%29/Canon - any questions? Aren't excellent DR and color depth more important at the majority of shooting conditions?

Sep 27, 2011permalink

what am I doing with excellent DR but no detail that I get from sonys crappy (except 24 f2.0) lenses? makes no sense at all..

Sep 27, 2011permalink

ryansholl

Well I'll be damned, Sony actually made a good lens, and it's a 24 F2? What am I doing with all this other AF Zeiss glass?!

Sep 27, 2011permalink

recycle.

Sep 27, 2011permalink

have a look - A77 ar ISO200 is worse than K-r at ISO800....

Sep 27, 2011permalink

I've never seen before so high level of noise at ISO200-400 in RAW. ISO800-6400 are really BAD.
It's worse than 10 MP CCD APS-C sensor from Sony.

Sep 27, 2011permalink

Menneisyys

Yup, I even see speckles of noise at ISO100 on the surface of the blue watch...

Sep 27, 2011permalink

A77 has HUGE amount of noise in RAW from ISO200 and above.
K-5, 5D Mk II and D7000 are much better.

Sep 27, 2011permalink

Low Budget Dave

I think there are people here out to get Sony. When I look at the samples, I see it differently. The Nikion shows smoothing (or blur) especially at the edges, and the Pentax is almost identical. The only things different from Pentax are the crop and the color curves.

Sep 27, 2011permalink

it seems to me we saw different photos...
I say about RAW.

Sep 27, 2011permalink

http://forum.getdpi.com/forum/showthread.php?t=30519

I say about noise in RAW. Not about resolution and DOF.
24 is more than 16. Always.

Sep 27, 2011permalink

victor china

the A77 is about a stop behind at the pixel level, if you normalize it it is about half a stop behind... and that's probably the light loss of the translucent mirror.

I expect the NEX-7 to be at the same level of the 16MPX sensor when size is normalized... meaning that the increase in resolution is allowing real gain of resolution at base iso, and no real loss at high isos...

Sep 27, 2011permalink

Lea5

Seems to me a great camera regarding to the marvellous images in your link.

Sep 27, 2011permalink

jlamesa

When you all get tired peeping pixels, see some real life photos shot witn A77

Sep 27, 2011permalink

zstan

Pity nobody appreciates the AF speed and dynamic range of the a77...

Sep 27, 2011permalink

slncezgsi

I have not seen any comparison that would shot the dynamic range of A77. The DPreview studio shots does not really allow to test that by naked eye.

Sep 27, 2011permalink

http://forums.dpreview.com/forums/read.asp?forum=1000&message=39451949

Take a look at DXO test. A77 gets 78 and 7d's score is only 66. Being a little worse at high ISO, A77's sensor has great color depth and DR.

Sep 27, 2011permalink

Debankur Mukherjee

The Image tests are out. Fitting 24 MP in a APS-C Sensor has made the sensor more noisy in RAW mode and in JPEG mode the sharpness is lost due to high Noise reduction applied. This sensor needs even better in-camera post processing to give better results in high ISO.

Sep 27, 2011permalink

sensibill

Smaller, more dense sensors like m43 going from 12MP to 16MP weren't this affected, so it's either the pellicle or not Sony's best sensor - not necessarily the MP.

Sep 27, 2011permalink

exp1orer

Panasonic G3 is even better than this Sony A77

Sep 27, 2011permalink

onlooker

Based on these RAW images, if I was going to get an APS-C camera today, I would get the 7D.

Sep 27, 2011permalink

http://www.dpreview.com/galleries/8017913541/photos/676011/aimg_0049_1

I replaced the D7000 with the NEX C3 in the Studio Comparison quartet and I'm leaning toward this NEX for 3 reasons. 1) I still don't have a mirror-less one in my collection. 2) It equals or even slightly edges the 7D in raw. 3) I already have the 7D.

Sep 27, 2011permalink

RubberDials

The 7D hasn't got any in-body IS - shoot a fast prime below 1/160 and camera shake will have more of an impact on the final image detail than noise.

Sep 27, 2011permalink

macky patalinghug

taken with at 1/20 with a 35mm f/2

Sep 27, 2011permalink

RubberDials

Now post the one with the IS on. Nikon and Canon users can't test this unfortunately. On a Sony cam take a shot with IS on and one with it off. Up to 1/160 there is a difference in critical sharpness. Nice shot btw.

Sep 27, 2011permalink

hikenhi

Put this SLT-A77 against the Pentax K-5, RAW, ISO 3200. Any Questions?

Sep 27, 2011permalink

rishi o'

Put the SLT-A77 against the 1/3 cost 5n at ISO 3200.

Sep 27, 2011permalink

TrojMacReady

Why would we put it against a camera that already applies NR in RAW at ISO 3200?

Sep 27, 2011permalink

I replaced D7000 w/ NEX C3 before I moved the cursor (in 3200 ISO) raw to the Greek looking face then to that Margaret Thatcher looking woman then to that chick in the Kodak frame and observed that: the 5Dmk2 was on top; followed by NEX C3, then the 7D then the A77. It was only on the Thatcher-like face that I had some difficulty choosing between the 7D and the A77.

I did the same thing in 3200 JPEG and found that in Thatcher-like face the Canons were better. In the Kodak chick I saw a a close duel between the 7D and NEX C3 for here the Sony appeared cleaner but the Canon seemed to render the eyes better. The best rendering over all was from the 5Dmk2 while the A77's capture was at the bottom looking like it used a watercolor filter.

From these quick glances I can say that: 1. The Canons here make better JPEG than Sony. 2. The current APSC sensor technology still can't beat the 2nd generation FF sensor. 3. In terms of raw, the Sony 16MP sensor is currently the best APSC sensor.

Sep 27, 2011permalink

itev

Mr Sony : I want the Nex-5N sensor in the a77 body !

Sep 27, 2011permalink

mjkerpan

I want to see the tests again when RAW support is better. ACR's support for the A77 is SUPER preliminary and it shows: in low-contrast areas the JPEG samples of the same area...

Any way, once the processing workflow bugs are worked out, this looks like it will be a great camera for getting lots of detail at low-to-moderate ISO.

Sep 27, 2011permalink

BozemanMark

The NEX-5N looked very good compared to just about everything at every ISO. Put that sensor in a NEX-7 body and Sony won't be able to make enough of them.

Sep 27, 2011permalink

Yep. I also like like that Sony 16MP on my Canon :)

Sep 27, 2011permalink

Mssimo

What lens does dpreview use? How good/sharp is it?

Sep 27, 2011permalink

Sony 50mm f/1.4 @ f/9

Sep 27, 2011permalink

Midnighter

While I know its needed for DOF f9 is beyond the diffraction point of 24MP APS-C (1.5 factor). So, in effect, what you are saying is the shot is diffraction limited.

Sep 27, 2011permalink

At a third of a stop beyond f/8? Nothing that would be visible to the eye. In that case, all of dpr's test shots would be diffraction limited, since they are mostly shot at this or eqv. apertures. The GH2 photo is shot at f/6.3.

Sep 27, 2011permalink

Midnighter

Diffraction hits f8 at 21MP for APS-C x1.5. The GH2 has a pixel pitch of 3.6, even smaller than the a77 sensor so its further below f8. However my guess is the GH2 would be diffraction free at f6.3.

Sep 27, 2011permalink

I enlarged the GH2 photo at ISO800 to SLT77 size, and it showed similar noise but much more detail. I'm sure the SLT77 is a nice camera, but maybe it's time for Sony to think about lenses that can actually handle all the megapixels without weighing twice times as much as the camera?

Sep 27, 2011permalink

thx1138

Yes this is a good way to compare different cameras with different resolution, or you could have down rezzed the Sony to GH2 size. The 7D is very unforgiving on glass, Sony will be worse.

Sep 27, 2011permalink

Sosua

Which lens are you referring to that weights twice as much as an A77?

Sep 27, 2011permalink

Sosua

Which lens are you referring to that weights twice as much as an A77?

Sep 27, 2011permalink