DxOmark says: A99 new full-frame sensor is Sony’s best yet.
DxOmark (Click here) just published their full Sony A99 sensor test. And it scores right in the middle between the Canon and Nikon Full Frame. The only real downside is the Low Light performance which is less good compared to Canon and Nikons score. But if you shoot at Low ISO than the A99 is on par with the best Nikon FF and far better than the Canon. DxO writes: “The use of Sony’s SLT technology in its top-of-the-line full-frame shooter delivers both pros and cons. The obvious downside is low-light performance compared to its closest competitors. The upside to the technology, of course is fast performance with full-time live view and continuous autofocus, which makes the A99 one of the best choices among semi-pro cameras for shooting video.”
It’s a good result for the A99 and at leats for me color depth and dynamic range are far more important than the image quality at 6400 ISO 
I suspect that the Sony RX1 will score an even better results because of the lack of the transparent mirror. And this may another a good reason why a NEX FF makes sense…don’t you think?
A99 and RX1 links:
A99 Best price in US is at Bigvalue ($2,600 on eBay here).
A99 body Amazon, Adorama, BHphoto, Jessops, Wex UK, Wex Germany, Sonystore DE-UK-FR-IT-ES-NL-SE-FI-CH and more links at Alphacameradeals.
A99 with kit lens links Alphacameradeals.
RX1 Amazon, Adorama, BHphoto, Wex UK ,Wex Germany and more links at Alphacameradeals.
kalpurush :)
5 months ago |Well, a99 is a fantastic camera- no doubt about that. Though, I wish if the price was a bit lower, like $2,199 or so.
SonyCyberPunk
5 months ago |What?!?
Look at the differences in DXO scores. Even though many DXO scores matter little in the real world, I have to just hang my head really low when I see this data. Poor Sony FF shooters.
What a fail.
kalpurush :)
5 months ago |@ SonyCyberPunk
Photography is not all about high ISO noise – there’s a lot more than that!
Fire1
5 months ago |Damn, The high ISO score is wrong.
I really wished the A1 is coming. One FF will not help Sony gain in the FF market.
Olaf
5 months ago |This shows that its time that the mirror is “upflipable”. Dr and Color Depth us same as the d600
Olaf
5 months ago |Typo. The sony camera mirror should be upflipable
Brendon
5 months ago |If the mirror can move then it would be a regular SLR and not an SLT.
Sony has already stated they won’t make any more SLRs so I guess that’s not going to happen.
Can’t say I am thrilled at the lower performance but I have already accepted that I am going to stay with A mount and so I will work around the limitations and concentrate more on the pros.
I still feel weird being unable to shoot quickly in Live view when I use Canon/Nikon DSLRs.
Vivek
5 months ago |That Nikon D600 really shows what can be done with the 24MP Sony sensor.
admin
5 months ago |LOL. But the RX1 may beat the Nikon…and you will write “That Sony really shows what can be done on a Nikon camera with a Sony sensor”
Joe
5 months ago |I doubt the RX1 does as well in the Sports/Low-Light category, which is what killed the A99 in comparison to the Nikons. I wonder about their value for the A99 as it basically says the average ISO for what they term “excellent image quality” is around 1500 for the A99, where the A900 was 1400 and the Nikons are near 3000. I can tell you now, the difference between the A99 and A900 is probably closer to a half stop above 400. Eye test, not lab test. But that’s just me.
Lofote
5 months ago |The RX1 does not have a translucent mirror, so low-light IS better than the A99.
Joe
5 months ago |You must have missed the “in comparison to the Nikons” part of my statement.
Vivek
5 months ago |That is an easy one, Admin.
1. No color shifts/casts in D600 (Rx1 has those).
2. True 14 bit depth of RAW files from the D600 ( fake 14 bit from Rx1).
3. A built in VF in D600 ( the dumb OVF of zony is pretty expensive and is not included).
Do you have any others to compare?
mict
5 months ago |You are definitely wrong. Good optical pentaprism made of optical grade glass costs much more then a tiny oled screen.
Vlad
5 months ago |Some RX1s have those. And why “fake” 14bit?
Reason
5 months ago |Vlad, please link us to your work- I am curious to see the images from someone who is so adamant that they have 14 bit RAW. 12 bit just isn’t enough right? And what about the majority of clients who only receive 8-bit files? What are they smoking?
fredrik
5 months ago |rx1?? ahahah no… its a fake camera for child… go 1dx if you want serious results
Sam
5 months ago |This sensor is behind the Nikon implementation in every area. Sony are making impressive strides but even they realize. that SLT isn’t the future. Posts like this will look bad when Sony abandon that avenue and people talking up mediocre performance relative to the competition now go back and talk about how much better the new kit is.
If you shoot many systems, as I do, you’ll note that people who are switching today (most notably Pentax shooters who can’t wait any longer for full-frame) are moving to Nikon, not Sony. I believe Sony will only be truly competitive when they ditch SLT and can beat or equal SLR’s on every front.
St.
5 months ago |You can see what change there is only from removing the AA filter of D800E vs. D800.
It’s not a surprise that SLT will take lots of light and the results will be even worse.
A99 is wonderful camera – I love it, but noise wise can’t beat the DSLR at the moment.
I also agree that Sony are doing the right thing, leaving back the SLT technology.
Can’t wait for the FF NEX.
Lofote
5 months ago |Sony is doing it actually right, by doing it differently. Video is unbeatable in the Sony and thats what their strongest horse is. And their smaller SLT easily kill DSLRs in fps speed. They are not doing the same things as Nikon and Canon anymore, and why should they. They do it right by thinking different and building a different customer base with SLTs and NEXs.
Kaigee
5 months ago |you are right, following people footstep only will be “another” one of them, it not mean they can be the best. if want to be the best need to have his own way to fight!!!
Vlad
5 months ago |How is the sensor behind the Nikon in every area? You are telling me that 0.2 in DR and 0.1 in Color depth is some noticeable difference?
It is really funny to read unfounded generalized statements like this:
“you’ll note that people who are switching today (most notably Pentax shooters who can’t wait any longer for full-frame) are moving to Nikon, not Sony.”
Joseph
5 months ago |Now this really makes you wonder about the DXO testing. Other reviews clearly show the A99 on par with competitors and better than the D800 in low light and it appears DXO is saying otherwise. Also, DXO scores the Canon 5D MKIII lowly in low light performance, but some hands on tests and other reviews show that it equals, and in some cases, exceeds the D600 in high ISO performance. This would make it seem that the A99 is no better than the A900 and A850 in low light performance. Too many sites, publishers and testers placing their “blessings” on what they perceive as lowlight performance and image quality. O know that the A99 is perfectly usable at the higher ISO values from experience and don’t really put a lot into the so called lab and theoretical tests. I make my conclusions based on prints and customer satisfaction. Still, it makes you wonder if someone spilled coffee on the lab equipment when they were testing.
Jack
5 months ago |Yes, I’m no expert, but that wide range of scores for ISO seems questionable.
JiM
3 months ago |I’ve always said ISO score from DxO is useless. Mind they ranked A350 with noisy CCD sensor that was useless over ISO 400 to be on par with A700 where I used ISO 1600 for A3 size prints. And of course A700 is still better noisewise from A77 even if DxO says it is not. Most people using raw format use Adobe software that works a lot better than DxO and gives at lkeast 1-2 stop advantage in shadows which DxO spoils turning black in to green
Joe
5 months ago |I have both the A99 and A900 and the A99 blows the A900 out of the water in low light, not only in retained detail and noise, but the noise in black areas is much cleaner and the noise that is present is more film-like.
quicky
5 months ago |What you guys are comparing is the improvement of the raw converter used by the A99 compared to the A900. Obviously the noise reduction algorythm are much better on the A99.
DxO mark assess the real noise of the sensor (sensor + mirror for A99) on the raw file before the raw conversion. AFAIK, they are the only testers to do so. All other reviewers are mesuring the RGB pictures, so after raw conversion thus including noise reduction filters.
DxO mark is testing the sensor only
Others are testing the sensor + a raw converter (the one of the camera usually)
2 differents approaches that tell you different things.
(BTW, the raw converter of the A900 wasn’t really great, so it is no surprise to see big noise improvement on the A99 after raw conversion, Sony software engineers have gone a long way since 2008)
Vlad
5 months ago |This is why:
“An SNR value of 30dB means excellent image quality. Thus low-light ISO is the highest ISO setting for a camera that allows it to achieve an SNR of 30dB while keeping a good dynamic range of 9 EVs and a color depth of 18bits.”
Once you hit lower than one of those criteria, no matter which, your score goes down.
I am still surprised how people are bitching about ISO scores, when they probably won’t use them at all. Why not complain about Canon’s crappy DR?
Mike
5 months ago |Why we don’t complain about Canon crappy DR? Because we don’t own Canon cameras so we don’t care.
You may have not noticed yet, but most of these complaints are not about which vendor is better. We are pointing out a Sony design decision forced upon us that doesn’t make a lot of sense for many stills shooters. In fact, it almost nullified 4 years of sensor development.
Vlad
5 months ago |My point about Canon is that people will buy it. Same like they will buy the Sony. I would like to see how many people will actually base their purchase decision on DXOmark’s score. You are just overreacting like most here. You may not have noticed, but nobody is forcing anything on you. You are free to switch to another company. I also don’t see how the amazing DR and Color depth have been nullified? Seems like ISO has become the reference for everything now. Seriously, what ISOs do you use?
Rob
5 months ago |Sony can make an Alph FF too – don’t you think that makes even more sense?
RurikII
5 months ago |Wow… ISO is so low… Almost full stop from D600. Lowest FF score for last 10 years?
It looks that mirror they use in 99 is a bit different to one used in 77, takes out more light. 77 loses around 2/3 of a stop.
Cody
5 months ago |The low-light score doesn’t seem right. The mirror eats about 1/3 stop, so I was expecting a low-light score around 2400. This 1555 result implies that the A99 has a full-stop disadvantage to the D600, and that definitely just isn’t true comparing my results to D600 shots I’ve seen. It also implies that the APS-C D7000 only has a 1/3 stop disadvantage vs the A99, and I don’t see that, either (not even close).
I have a lot of respect for DxO, but something is fishy here.
Joe
5 months ago |I agree. From my apples-to-apples experience with the A900, the A99 is at least a half stop better than the A900 above 400. Same environments, same subject, same lens, same metering.
Mike1
5 months ago |I’m not that sure. I also got a A850. High ISO was always a weakness. If A99 is only half stop above A900 (or A850), then the ISO score is about right. The ISO performance of my Pentax K5 is better than the A850.
JiM
3 months ago |Your Pentax can be better in ISO score only if you use jpg out of camera or compare in-camera denoised K5 “raw” with real raw from A900 using robot. If you use Adobe raw software A900/A850 are way better than K-5 or any other APS-C camera on the market. There’s at least 1 stop advantage.
lavorkybel
5 months ago |That high iso score is weird. A99 received higher high iso score on dpreview than D800 a 5D mkIII
RickJ
5 months ago |Why are you trying to push a FF nex so hard in all your posts? It’ll be a niche product with limited appeal. Attaching huge FF lens onto a small, slow auto focusing device is counter productive.. “look at this full frame picture… that’s… out of focus..” By the time they revamp a NEX to full frame.. it’ll be essentially an uncomfortable square version of a SLR that’s a tad smaller.. have fun holding a 200mm G lens with a goofy square NEX body attached to it. It’ll look ridiculous
Don Cox
5 months ago |I regularly use a 200mm lens on a NEX, and the pictures are not out of focus.
There is no problem in holding the lens with the NEX attached to it. Your concerns are misplaced.
Reason
5 months ago |Professional shooters use cameras for 10+ hours, my events/weddings I am shooting for 10-12 hours straight. I guarantee you that the tiny NEX body will be a strain to fingers and left-hand because of improper balance. the Battery grip on my A850 helps counterweight the lens and reduce fatigue on my wrists.
NEX body IMO is not ideal. Size really matters, I know everybody hates that, but I think a camera that molds the shape of your hand, (your hand is bigger than an NEX) is much more comfortable. I only wish the A900/850/99 were LIGHTER- weight is the only reason those larger cameras can become fatiguing, not their size.
Maxwell
5 months ago |From samples I have seen I would guess maybe 1/2 stop between
A99 and D600 in high ISO. And the A99 high ISO noise pattern is
way better then A900. More grainy, looks better for me.
Yes, a FF NEX is very welcome.
JiM
3 months ago |If you use Adobe raw software A900 keeps still up to date and there’s little if any difference in picture quality between A900 and A99.
However jpg engine was improved in A99.
If Sony would gave new software to A900 many people wouldn’t even look at A99.
That’s why I decided to buy second hand A900 instead of A99.
Glass pentaprism of A900 is way better than EVF for me.
Amateur Sony Shooter
5 months ago |I firmly believe DXO got ISO score wrong. All other major review sites, as well as my own experience (I had both A900 and A99 for a while before I sold A900) all indicating A99 is easily 1-1.5 stop better than A900 in high ISO. I also took some testshot of D600 and it’s only half stop cleaner at most, while I prefer A99′s sharper image at the same ISO level. Something is wrong there with DXO’s ISO scoring system.
JiM
3 months ago |As I have written above, no difference in raw quality between A900 and A99 at least in Auto ISO range. From ISo3200 up A99 is bit better but this is no longer quality acceptable for professional work. So for pro photographer using the lowest ISO possible there’s hardly any advantage in buying A99. I think it’s rather an expensive toy lacking OVF and some really nice functions available in A900 like ZONE balancing and craw+small jpg recording. For me there would be much better to get improved software to A900 than buying A99, because A900 sensor is good enough.
Think how nice it would be if Sony give us “A950″ with A99 sensor, OVF, tilting screen and simple LV on it.
Kursad Sezgin
5 months ago |How Sony Made a Loyal Alpha Customer Switch His Mind:
“I will certainly buy A99″ > These were my words before it hit the market.
“I am planning to buy A99″ > These were my words after seing the price tag.
“I will think twice before buying A99″ > These are my words after seing all SLT’s price going down but A99 staying the same.
” I will wait a couple of months to buy A99″ > These were my words after seing Sony roadmap switching to NEX development rather than A-Mount lens improvements.
“I will not buy A99″ > These are my words after seing the ISO achievement of A99.
OK. Tests or DxO scores may not be that important. Actually to some extent, they are not enough to build decision mindset alone. Most important thing is the user perception, the complacency it delivers. The ease of use, even getting used to the menu is important for the next investment.
However, there are very basic but very important missing points in Sony’s strategy. The roadmap is switching from one to another within 1-2 monts. The price tag is more than A99 delivers. Why should I switch from A57 to A99? To gain 1 stop advantage? 1 Evs Dynamic Range gain? Or 1.5 bits of color depth advantage? Does these all three sum up to $2.300 price difference? From my point of view, no! I will not go that far to compare A57 to A99 of course. A99 is better in means of build quality and photo output for sure. But in my calculator, the difference does not equal to $2.300. Most important point here is the change in Sony roadmap. They declared that they will not introduce any FF A99 contenders in 2013 but rather chose a path to develop a FF NEX model. What that brings? Well, we were expecting a development in lens models, right? CZ 50mm is on the way. 70200G will have a successor at last. Let’s not forget 300mm, too. But hey, Sony is still trying to sell rebranded Tamrons and Minolta’s! Why not bring SSM to all these models? I own CZ85 1.4 and CZ2470. Both decent lenses, right? But hey, they don’t have weather sealing! I won’t be able to shoot in damp weather even if I buy A99. What about this?
Sorry Sony… you are pissing your loyal customers big time!
JustSoldMySLT
5 months ago |+1
Abit77
5 months ago |+1
maggy
5 months ago |+1
John Maverick
5 months ago |It’s funny but you really don’t sound like a very loyal SONY customer.
tjf
5 months ago |a99 – noise generator
dynax
5 months ago |I think ISO score is completely WRONG… let’s check comparatives on DPreview for example… what’s happened DxO? Why the same D600 Sony sensor score 1400 points less in A99!?
Blackriver
5 months ago |I’m sorry but to score the A99 that low in low light performance is an epic fail. Real world observations put it MUCH higher than that.
Bill
5 months ago |Strage ISO results. If you look at the trio of the A57, A580 and Nikon D5100 the scores are 785, 1183 and 1121. The two cameras without the SLT mirror being about the same and the A57 being 1/2 stop lower, which is the same as the loss due to the mirror. There is no Sony without a mirror in the new sensor but the difference between the A99 and D600 results, 2980 vs. 1555 implies a 0.9 stop difference due to the mirror. The A99 as mentioned by others only scores slightly ahead of the old A900. I see no reason Sony would need a mirror with more loss than on the A57, maybe there is some differences in the color filters, or something else accounting for the additional 0.4 stop difference. Or maybe the scores are wrong. Going to the site, the A99 is listed but the scores are not being displayed.
Beka
5 months ago |I think it has something to do with pdaf pixels on the sensor. nex 5R’s iso score is 910 and 5N-s score is 1080 so as we can assume AF pixels on sensor decreases low light performance 1.2 times. if we look at a57 and d5100 or a580 we’ll see that mirror loses half stop light so divides high iso performance by 1.5 so logiically that a99′s performance should be 1.5*1.2=1.8 times lower than d600. 1555*1.8=2800 which is pretty close to d600′s score and the difference may be caused by different technologies of nikon and sony to receive information from sensor.
I might be wrong this is just an opinion. sorry for my English!
skeptical
5 months ago |This is the most intelligent remark in this thread. Indeed the only rational explanation here (unless DxO made a major error, which I find unlikely), is that the AF points on the sensor are eating up additional ISO sensitivity. Considering that the A99 has a rather large number of these (102), it could be a significant factor. It really calls into question the wisdom of combining two AF technologies that are both detrimental to ISO performance. The result is a full-frame camera marginally better than APS-C in low light performance.
Mno
5 months ago |Marginal? My a99 is so much cleaner than than the a77 that I sold my a77.
(I would have preferred to keep both but I can’t afford that!)
Tanglefoot
5 months ago |I assume that the missing pixels, where the on sensor AF points are, have to be interpolated from the data from the surrounding pixels. I wonder if this could have messed with the results at all? Just a thought…
Beka
5 months ago |as far as I’m concerned pixels aren’t missing on a99′s sensor. Some pixels on sensor are divided into two parts and one part receives light from one side of lens while the other part gets light from other side of lens. This separate information is used for focusing, but for final image they use the sum of these two parts. That is, pixels aren’t missing but in fact each part receives limited portion of photons
so together they don’t get as much light as they could if weren’t divided.
once again, there is a chance that I am wrong though I believe I’m not
danny
5 months ago |DxOMArk rating are very inconsistent. From what I’ve seen the A99 is about half a stop worse than the D600 and 5DMarkIII.
Steve
5 months ago |I try not to be a fanboy and I’m not willing to trash any and all negative reviews or comments because it makes my camera or camera maker look bad, and I don’t think DxO is part of some big conspiracy to make the a99 and Sony look bad.
I’m willing to bet that DxO noticed how poor of an ISO score the test came back with and tested it several time again just to make sure…
I’m a sills photographer and yes I know that the video types don’t care about anything but video and I know that some of the Sony video fanboys think the a99 walks on water but, no way in hell will I ever buy full-frame camera that scores as bad as the a99 is scoring at the price Sony expect me to pay for this camera.
Vlad
5 months ago |If you are really basing your buying decision on DXO scores, there isn’t much to be said…
The Lotus Eater
5 months ago |Sorry, but DXO messed up big time here. The ISO score doesn’t tie in with real world results or previous tests on other SLT cameras where the 0.5 stop light loss resulted in a comparable drop in ISO score.
Milt
5 months ago |Where’s Steve McQueen? He was the one that claimed to have the math that said the ISO score would be around 2k.
anthony
5 months ago |i got to agree with teh iso score aspect being off. ive used d800 and my a99 taht i now own and its just shy of d800 from the pics ive taken thats its pretty negligable.
mt
5 months ago |Ok, most A99 users don’t agree with that weird iso score. So let’s sue that DxO motherfuckers!!
Rob
5 months ago |The A99 and slt technology are all stopgap measures. Sony now has the capability to make mirrorless Alpha cameras (both APSC and FF). I say, get on with it Sony. Sales of the A99 will die off in a very short term, and you’d better be able to pull a rabbit out of the hat, as not having a decent FF option available – again – won’t get you any switchers, and lose you some more of your fan base (and, why abuse people who support you?).
ferrera
5 months ago |A900 – ISO scores 1431
A99 – ISO scores 1555
But the real ISO performance of A99 is obviously much better ….
QueenB
5 months ago |Sony A77 – ISO score 801
Sony NEX7 – ISO score 1016 (remove mirror, 25% better)
Sony A99 – ISO score 1555
Nikon D600 – ISO score 2980 (remove mirror, 100% better)???
Denis
5 months ago |These things cannot be measured in percents. The right formula is ln₂(X₁-X₂) (except for the color depth, that is already an extent of the power of 2)
Denis
5 months ago |Oops, log₂, of course, not ln₂
JiM
3 months ago |jpg quality is really better, raw quality is very similar for both. There’s no big difference in raw quality between any odl or new 24Mp camera. Where newer cameras have less noise, they loose color gain. The best camera noisewise is still good old 12Mp Nikon D700.
laz
5 months ago |Guys, be careful that DXO offers two set of measurement, one for printing quality and the other for screening quality, check both out
oskar
5 months ago |I read somewhere that the 0.5 stop light loss assumption is kind of a misunderstanding. The relative effect of the translucent mirror is actually worse the higher the iso. So the same mirror that causes a 0.3 light loss at say iso 800 could cause a 0.9 light loss at 1600. (numbers not correct here, just trying to make a point). Could this be the explanation for those results?
Vlad
5 months ago |Why is the A99 listed as Preview still?
Kaigee
5 months ago |haha, you are right!?! why???
Henkbenk
5 months ago |Come on, this is the difference:
http://www.dpreview.com/reviews/sony-alpha-slt-a99/21
Poke
5 months ago |The overall score is excellent. The iso score is a bit disappointing but also somewhat misleading. I would never think for one second that my 850 could keep up with my 99 in low light conditions. I mean not even close. I would hate to print anything shot at 1600 with my 850 but won’t hesitate to shoot at 6400 and print it with my 99. However, if shooting in good light I wouldn’t hesitate to use my 850 because I still feel the 850 can hold its own. But these scores tell a whole different story. If shooting in good light my 850 does not hold a candle to the 99 and should stay home. If shooting in low light they would appear very equivalent to each other despite the fact I know this not to be true as far as appearance goes. How do you think Canon feels. Their 5DM3 gets blown away by every new Nikon and only beats the 99 in low light. In normal light these scores would indicate that Sony one ups the Canon. However, Canon shooters probably feel differently about that. however, it takes make me wonder if I shoot a lot of action/sports in low light should I use another camera besides the 99. If I would go by just the scores I would not even consider the MK3. I would probably go right to the 600. However, I think I would be smart and really look at the MK3 as a viable option. However, I will stay content in the near time shooting at 6400 and 8000 with excellent results with my 99. Oh, by the way did I mention I cannot do nearly that well with the 850?
JiM
3 months ago |I’m really amazed how many of you compare not raw quality but jpg engines. At least up to ISO3200 raw data from A900 is enough to prepare high quality A3 size prints.
JiM
3 months ago |I forgot to tell – using Adobe software. Dxo rawer is not even comparable with it, loosing colour information in shadows. All I can say – buy yourself A850 or A900, use Adobe raw software, you won’t need new camera. Leave DxO away.
BigP
5 months ago |Low-Light-Iso 1550 is a misleading number
Look at the live samples on dpreview.com see ISO 3200/6400 with A99 vs A900 vs D600 vs 5Dmark3. At higher iso A99 is very good and if some measuring algorithm DxOmarku to measure the “only” 1555 then quietly let. All the time it’s just a statement but the reality is different.
Worse would be the opposite.
If Low-Light-ISO 2500 and measured the contrary, the samples appeared as of 1600: o) So this is pretty irrelevant.
Joseph
5 months ago |Well explained. I have never been one to judge based on the lab measures. I have shot the same venue with both the 5D MKIII and A99, both excellent results. The SLT technology isn’t bad at all, and the EVF remains as a plus now. For those that own the A99, keep shooting and enjoy. We know the excellent high ISO performance it offers. For those interested, I can’t tell you what to do, but I don’t see any reason that you would be disappointed.
Varn
5 months ago |Lucky I use my A99 to shoot real world objects and not measurement charts, hence the results didn’t bother me at all.
Simon
5 months ago |To put things in perspective: I’m still selling stock photos taken with a 4MP P&S from many years back. Do you think the graphic designers and other ‘customers’ who are purchasing those images care about the finer details of the tool I used?
Vivek
5 months ago |Absolutely not!
When coming to buying a camera, I would choose the one that gives the best bang for the buck. That is the point here.
mt
5 months ago |Are DxO tests JPEG or RAW? Beacuse if they are testing JPEGs with NR OFF then maybe this ISO score have sense. On Dpreview ISO test with NR OFF A99 is the worst -> http://i47.tinypic.com/mievz9.jpg
Mike
5 months ago |4 years of sensor development nullified by a piece of plastic wrap…
Alan
5 months ago |Dun take the test too seriously, as you see, 5D mark III didn’t score so well as them but why still so many ppl like it and getting it? Because it is a good camera, because many ppl see the result in real life, from picture, from video, how many ppl is look through a test of marks to get it? rarely… i do have sony a99, it is a great camera, it make me become more and more loving it day by day after different shooting in picture and video. I very proud of showing my result taken with a99 with my friends and family. it is really incredible! my family member use nikon, we taking a same shot in very dark night, the result come out, i won, brigther, noise lesser, everything just look impressive. i duno how it score so low in ISO, but alot review had prove that it really give an incredible shot at night, and i have this a99 i very understand it as well.
Anyway, test out yourself for ur equipment! if it is nice to you, then it is nice. Different equipment different use. Buy the one suit you the most. a99 really fast and more correct in auto focus although the mirror make some quality loss, so i have no complain about the mirror, only if the mirror can flipable of course is the thing im waiting for
Aramiss
5 months ago |Henkbenk:
Yes , you have given us a very good comparison link
http://www.dpreview.com/reviews/sony-alpha-slt-a99/21
For my eyes the vinner in iso the Canon, almost identical the second Sony. The third is in iso (but with the best resolution of course) the D800.
And there is something wrong with the files of the D600. Very washed like soft. Two days ago I tested the D800, D600 with my Eos 1Ds M3 and I felt the same. D800 and 1dsM3 are identical in high iso, with D800 super resolution and wide DR and with high quality thick pictures of the Canon. And D600 was slim, soft and boring in competition with the two others.
I also tested for one week the Sony Rx-1 and quality of the files is winner. Above the 1DsM3 and D800. I m waiting for what DXO will show us about Rx-1 : )
davy793
5 months ago |There are lies, damned lies and statistics.
Forget the lies and go with your eyes.
Merry Christmas to you and my A99
Joel Benford
5 months ago |A while back I did my own high ISO comparison of full frame cameras I consider buying, at ISO 6400 only — they all do 3200 well enough for me and none do 12800, so that’s where the action is. I took the “Raw Martini” shots from dpreview and put them through Lightroom 4.3. I carefully equalized the white balance, the black and white points, the mid-tones, and the output size.
My personal ranking for the noise was, in descending order:
1. D600
2. D800 = 5d3*
4. A99 = RX1**
6. 5d2
[* 5d3 dydnamic range ain't as good as its noise, but the noise is my main concern. **The RX1 seems to overstate it's ISO, and brightening the picture up to match the others makes noise much more obvious.]
For what it’s worth, I agree with other posters here that the A99 feels more like half a stop behind the D600 under these conditions. I think less of its noise is in the mid-tones, so it’s less offensive. It looks better than it measures.
Even more subjectively… I wouldn’t pick between these cameras on high ISO noise, they’re just not that far apart. I’d pick on AF speed/accuracy, ability to choose the AF point you want, and available lenses. E.g. if the 5d3 gets a wide open 50/1.4 in focus at EV5 and the d600 needs stopping down to f/2 to be reliable, the 5d3 has better noise in practice…
[I own a NEX-6 BTW, and before that I had a Contax film system. Enjoy deciding what my biases are.]
nikaei
5 months ago |“”"I suspect that the Sony RX1 will score an even better results because of the lack of the transparent mirror. And this may another a good reason why a NEX FF makes sense…don’t you think?”"”
maybe the refresh line in alpha apsc (3x,5x,6x) that will happend this year will had the new up/down mirror mode…maybe not.
but fear not, sony will add it sooner rather than later
Cooking
5 months ago |So, Nikon cook their RAWs a little more… That scores worse in my book. DXO? I may as well tell you it is as reliable as politicians.
Future m43 user
5 months ago |There’s no way it’s a full stop between A99 and D600. At most it’s 1/2 stop, so high iso score for A99 should be about 2000.
Joseph
5 months ago |It’s not even half a stop behind the D600. This should finally show people how unreliable DXO ratings are. It’s no different than when using their RAW editing software compared to Lightroom, Bibble and Aperture. You get different results sometimes with different software. What this does state is that Sony shooters should stay clear of DXO editing software and opt for one of the others for better results since this is the best they can get according to them. I know with certainty that real world results and other real world reviewers clearly show that the A99 is equal to its competitors in high ISO noise control and in some cases superior with standard image quality.
Guess that’s why I never gave DXO any of my money. Glad I never did.
JiM
3 months ago |The same I have said years ago. For Sony cameras you have to use Adobe raw software and that’s all. Thanks to that I don’t afraid to make large prints (A3 size) out of A700 and A900 in whole ISO range. Even ISo 6400 of these cameras retains good colour detail with really nice grain visible only when you put your nose to the picture.
michael
5 months ago |DxO finally released the measurements. Consistent with http://www.sonyalpharumors.com/dxomark-the-sony-a580-is-much-better-than-the-sony-a55/, but IMO don’t reflect well human perception of noise and DR. Just compare the NEX-7 measurements with the A99 – above ISO 6400, the NEX-7 outperforms A99 (independent of measurements for screen or print). Owning both cameras, I cannot find such a behavior in my RAWs.