Dxomark: A55, A33, Nex

[OneGuyKs]

I disagree I think the newer APS-C cmos sensors are much better than the older 24mp CMOS one FF or not. Resizing does not help a lot here..the sensor just has more noise.
Owning a 10mp CCD model from Sony I can assure you it does not equal the 6mp one "Printed" for low light high ISO. The 14mp CCD is even worse...again resizing here won't help you.

I will take dxomark over your conjectures. The scores are based on noise vs signal (from RAW, not jpegs), so they have to be printed equal size. They have tested the cameras and applied the same test to all brands in lab environment. You have not. I will take their words over yours.

[Greg Beetham]

Huh? The dxomark scorers are calculated at *EQUAL SIZE* (8 MP). That's a fact. I am stating facts. You are not.

It's meaningless, you don't use a 24MP camera at 8MP, or downsize all your 24MP images too 8MP every time you take a shot. What about the 6MP camera? do you then up-res it to the arbitary 8MP figure (wherever they got that from, plucked it out of thin air probably...gee 8MP is a nice round figure, lets go with that).
Greg

[mike2008]

It's not meaningless at all. You do have to engage your brain a little bit though. Easy said then done, I know..

"Sensor Overall Score is normalized for a defined printing scenario—8Mpix printed on 8”x12” (20cmx30cm) at 300dpi resolution. Any other normalization, even with higher resolution, would lead to the same ranking, given that any camera that could not deliver the chosen resolution would be eliminated from the comparison."

[deathvalleydave]

Regarding the A850/A900's high scores for low light, keep in mind that these are RAW not jpeg scores, so we all know how much better the A850/A900 perform in the noise category when shooting RAW vs. jpeg. Yes, the new APS-C cameras are better for low light jpegs, but DXO scores sensors all by themself, not jpeg signal processing.

Plus, remeber how DK says the A850/A900 are very capable for high iso work...

[stevecim]

I have both the A350 and A550 and going by DxoMarks results the A350 should out preform the A700 at High ISO, but it does not even get close to the A700 at ISO 800 and higher. I don't think their test mean much compared to real word results . I would be more interested DK opinion on each camera since he has used nearly every Sony Alpha body.

[bfitzgerald]

I have both the A350 and A550 and going by DxoMarks results the A350 should out preform the A700 at High ISO, but it does not even get close to the A700 at ISO 800 and higher. I don't think their test mean much compared to real word results . I would be more interested DK opinion on each camera since he has used nearly every Sony Alpha body.

Exactly there isn't a hope in hell the A350 does better in low light than the A700. Heck some folks were scared to even hit ISO 800. I tried everything to match the 6mp CCD from the 5d, matching exposures, allowing for slightly less ISO sensitivity. It was simply not possible you could downsize, upsize do what you wanted it just didn't give as good prints for high ISO. ISO 1600 was pushing it on the 10mp CCD, 3200 was not really very good either. Nice enough below that mind.

This simply highlights how odd and frankly illogical DxO's scoring is. Folks even said the A700 was better than the A900 resized down, there is no question the newer 12-16/18mp CMOS sensors are better for high ISO.

[OneGuyKs]

Huh? The dxomark scorers are calculated at *EQUAL SIZE* (8 MP). That's a fact. I am stating facts. You are not.

It's meaningless, you don't use a 24MP camera at 8MP, or downsize all your 24MP images too 8MP every time you take a shot. What about the 6MP camera? do you then up-res it to the arbitary 8MP figure (wherever they got that from, plucked it out of thin air probably...gee 8MP is a nice round figure, lets go with that).
Greg

There is nothing "meaningless" about it. The scores are calculated at equal print. And that's how it shoud be. Dxomark got it 100% right.

[OneGuyKs]

I have both the A350 and A550 and going by DxoMarks results the A350 should out preform the A700 at High ISO, but it does not even get close to the A700 at ISO 800 and higher. I don't think their test mean much compared to real word results . I would be more interested DK opinion on each camera since he has used nearly every Sony Alpha body.

Exactly there isn't a hope in hell the A350 does better in low light than the A700. Heck some folks were scared to even hit ISO 800. I tried everything to match the 6mp CCD from the 5d, matching exposures, allowing for slightly less ISO sensitivity. It was simply not possible you could downsize, upsize do what you wanted it just didn't give as good prints for high ISO. ISO 1600 was pushing it on the 10mp CCD, 3200 was not really very good either. Nice enough below that mind.

This simply highlights how odd and frankly illogical DxO's scoring is. Folks even said the A700 was better than the A900 resized down, there is no question the newer 12-16/18mp CMOS sensors are better for high ISO.

Maybe he is doing something crappy with his A350 RAWs. A700 looks like a doggie-doo in IR jpegs too though. I will take dxomark words over his.

[bfitzgerald]

I don't have an A350 but I have processed raw files from it and the A700. There is no doubt at all that the A700 is superior at high ISO. Regarding jpegs Sony have been behind rivals for some time now, I don't believe there is an Alpha on the market that has what I would consider "good" jpegs esp at higher ISO levels.

You can put a Panzer tank through the holes in DxO..if you want to take their word (which in the context of a real world user is dubious and pretty useless in many respects) fine knock yourself out. You'll be hunting for a long time to find anyone who feels the A350 is great at high ISO or as good as the A700. The new CMOS sensors absolutely murder some of these older CCD ones, and older CMOS ones. DxO does not reflect that..so let's call it out and drive a few more Panzers in the holes. Academics might like it..everyone else will wonder what all the fuss is about.

[Greg Beetham]

There is nothing "meaningless" about it. The scores are calculated at equal print. And that's how it shoud be. Dxomark got it 100% right.

It's a theoretical print not an actual print, we've been through this before, it doesn't have much if any relevance to anything useful regarding the resolving potential of the individual sensor at low light & high ISO.
Sony I'm sure is doing something different with the newer larger MP sensors and noise, they might be doing something along the lines of Nikon with noise binning or they might have invented a system of their own that actually works.
Greg

[David Kilpatrick]

The problem with all this is that when we tested the A350, for example, ACR 6.x or LR 3.x didn't exist and Sony had not yet employed DxO to produce in-camera firmware/software - which they apparently now do. DxO might perhaps favour Sony since DxO is possibly responsible for some of the embedded software in Sony cameras from mid-2009 onwards. This a core part of DxO business, and at the time Sony was known to be working with DxO their cameras made a quantum leap in quality - both raw files and in-camera JPEGs.

Barry is living in the past to some degree, and the real world results from the 2009-2010 Alphas are pretty exceptional. The Alpha 350 (2008) is stuck with 2008 in-camera routines but still has really very beautiful colour rendering (after the KM 7D and especially the 5D, one of the best DSLRs for the look of its colour). But the A350 is not stuck with 2008 processing any more - sadly, dozens of websites including my own reports naturally reflect what was possible in the weeks after its release.

Here is a dark (1/15th at f/4.5) CZ 16-80mm ISO 800 clip from the A350 at 100% processed using normal 25/25/25 sharpening and NR values in ACR 6.3, but boosted by one full EV (+1.0) in exposure, therefore actually equal to ISO 1600. Honestly, it's not so much worse than new models. The big change has been how Adobe and other raw processors treat .ARW files, and Sony using new routines of similar quality in-camera - just like Nikon has done for years.

[bfitzgerald]

ACR might have improved (I remember the cries of defence about NR raw as well back then) but the facts remain. The old 14mp CCD is now used in the A290/A390 and with no significant improvements. There is still more noise there even if ACR has improved (and it has a lot)

It's a bit like saying your 5 year old car isn't so bad because the cost of a 3 year old one has gone down too..it's all relative. It's not living in the past, if ACR is better all round then it is for ALL cameras including the newer models. I've no problems with CCD's as it happens but ..if the A350 is better for high ISO than the A700..let's see it then.

You could argue the 14mp CCD is nice tonality wise..on the other hand Sony stuck with it despite everyone else going CMOS (no doubt due to a warehouse full of them) and deliver significantly better high ISO as a result. If you're a low light shooter you'll want to pass on the A350. ACR made my A200 better at high ISO (colour noise was the big problem) trouble is it also made my Km5d better as well. Guess which is still better? Answers on a postcard please

[alphaomega]

I am not in any way an expert, but I do own A700, A350 and A550 cameras. I am impressed by the quality of A350 images; colour density, DR and also noise performance. I mistakenly took some ISO800 images forgetting to dial in ISO200 for conditions. These sailed through Alamy and frankly I was impressed. They were developed from RAW. Please do not under estimate what Sony's A350 can deliver. It is short of functions compared with the A700 and slow, but delivers excellent results. I am afraid I am not "on messsage here" but I fail to understand why Sony's Jpeg quality is always slated. I do not use it often, but my wife and son do. I find Sony's Jpegs fine. If they are substandard then Sony's competitors must deliver absolutely fantastic results "straight out of the can". Sorry that I cannot go into technical details. I just take pictures, develop them, upload to Alamy and sell. Just a retired businessman enjoying taking photos and thinking that Sony make good cameras. No need for me to switch to Pentax or Nikon to enjoy Sony sensor quality.

[David Kilpatrick]

Barry, the new cameras are not necessarily better. The Canon 60D I am using this week is FAR more noisy than the A350 at settings from 100 to 800, and only gains at higher settings. Even the A55 actually is no better than the A350 at common settings like 100-400, both have visible noise (and both are better than any current Canon, which might surprise people, and deeply upset Canon fans, but it's actually true - now Canon is entirely 18 megapixel in the 550D/60D/7D range, the image quality is not the same as the quality they earned their reputation with).

The A55 absolutely slaughters almost anything else. I'm off for a week next week abroad, and this time, the A55 and NEX-5 will be all that goes with me. No A900. The sheer accuracy of focus from the A55 - far superior to any of the previous models on routine subjects - makes its 16 MP images a fair match for the 24 MP A900. The viewfinder is not ideal, but in the past I was willing to put up with Leica III viewfinders to get Leica quality output from film, and now I will put up with an EVF to get this quality.

A 580 also beckons but the A55 has GPS and I want this for travel. It will make captioning of open landscape shots much easier.

David

[pakodominguez]

I don't have an A350 but I have processed raw files from it and the A700. There is no doubt at all that the A700 is superior at high ISO. Regarding jpegs Sony have been behind rivals for some time now, I don't believe there is an Alpha on the market that has what I would consider "good" jpegs esp at higher ISO levels.

Just curious: the A350 and A700 raw files you did process were from a sort of "scientific"/controlled test? or just random pics? I don't know about the A350, but the A700, well exposed and with high quality lens is hard to beat -that sensor have a special quality and it is a shame the A700 didn't sell better.

In other hand, The Maxxum 5D 6MP sensor was the last (and better) generation of that sensor and I find nortmal that you could get a really good image quality from it (the opposite happened with the Nikon D70, with the "same" sensor, that had a less than average image quality IMHO) specially noisewise. But in my experience, it wasn't until I used the A700 that I could find a real difference between an average lens and a good quality lens, probably because the higher resolution, probably because better technology... Your A200 have the same A100 10MP sensor (I guess improved) but image quality on that series was not au pair with the actual technology.

The NEX JPGs are better than the A850. I do shot raw anyway in both cameras.

Regards

Pako