Jump to content

I think DXO labs should reconduct its A7S test


sanveer
 Share

Recommended Posts

These tests of the A7S show how the internal processing on the A7S in Jpeg (and presumably video), is rather complex. Comparing RAW images for noise and dynamic range among other things may not be suitable. I feel the colours have been muted slightly, but that has somehow pushed the Dynamic Range across the various ISO settings. I feel DXO labs did not conduct their tests accurately. I could be wrong. It's jusr my opinion.

http://m.techradar.com/reviews/cameras-and-camcorders/cameras/digital-slrs-hybrids/sony-a7s-1255921/review?src=rss&attr=all

Link to comment
Share on other sites

EOSHD Pro Color 5 for Sony cameras EOSHD Z LOG for Nikon CamerasEOSHD C-LOG and Film Profiles for All Canon DSLRs

DXO has rarely been wrong with their measurements - Sony's probably doing some pretty clever tricks on the image processing side.

 

I completely agree. I like the DXO lab testing. But, I feel, that not only is the Sony sensor superb, even their algorithms are great, especially for doing what Neat Noise or Neat Video does, within the camera itself. After looking at the various videos, I would even think its in the league of something like Cinnafilm's Dark Energy (I am obviously pushing it).

 

I would like to see someone test the ISO 409600 range, and then try and clean it in post. I am curious as to the level of detail on that one. 

Link to comment
Share on other sites

DxO is completely irrelevant for video shooters. Unless you are shooting Magic Lantern Raw. Yes, it might give a suggestion of the capabilities, but it all depends on the internal processing of the raw data, so it can get either worse or better than the DxO results.

 

Also, I'd take the Techradar results with a grain of salt. I often see results in their graphs that are completely different from other lab tests.

 

Like this one. Look at GH4 dynamic range in raw, I wish it was that good! No way it has between 13-11 stops over the whole iso 100 - 25.600 range.

 

Sony_7S_TIFF_DR_03-900-90.jpg

 

DxO looks much more reliably to me if you look at the DR measurements: http://www.dxomark.com/Cameras/Panasonic/Lumix-DMC-GH4---Measurements

Link to comment
Share on other sites

These tests of the A7S show how the internal processing on the A7S in Jpeg (and presumably video), is rather complex. Comparing RAW images for noise and dynamic range among other things may not be suitable. I feel the colours have been muted slightly, but that has somehow pushed the Dynamic Range across the various ISO settings. I feel DXO labs did not conduct their tests accurately. I could be wrong. It's jusr my opinion.

http://m.techradar.com/reviews/cameras-and-camcorders/cameras/digital-slrs-hybrids/sony-a7s-1255921/review?src=rss&attr=all

I don't think I understand why you think they made mistakes for the a7S, or somehow the tests are wrong only for the a7S.  If you don't do the same tests for all cameras, then you don't have anything.

 

Michael

Link to comment
Share on other sites

DxO is completely irrelevant for video shooters. Unless you are shooting Magic Lantern Raw. Yes, it might give a suggestion of the capabilities, but it all depends on the internal processing of the raw data, so it can get either worse or better than the DxO results.

 

Also, I'd take the Techradar results with a grain of salt. I often see results in their graphs that are completely different from other lab tests.

 

Like this one. Look at GH4 dynamic range in raw, I wish it was that good! No way it has between 13-11 stops over the whole iso 100 - 25.600 range.

 

Sony_7S_TIFF_DR_03-900-90.jpg

 

DxO looks much more reliably to me if you look at the DR measurements: http://www.dxomark.com/Cameras/Panasonic/Lumix-DMC-GH4---Measurements

I about fell off my chair when I saw this Tech Radar chart with the GH4 being up there.  Based on other graphs I found, it looks like it is blowing every other camera out of the water, including the D600, D800, 5DMk3.  I wish it was true, but I'm not believing this.

 

Michael

Link to comment
Share on other sites

I don't think I understand why you think they made mistakes for the a7S, or somehow the tests are wrong only for the a7S.  If you don't do the same tests for all cameras, then you don't have anything.

 

Michael

 

Well they did test it with a "pre-release" firmware and only afterwards Sony came forwards with the release firmware that should have 15.3 stops in RAW. So there's atleast one mistake already.

 

And how do they always get Nikon/Sony made sensor cameras to test so easily but never Fuji? I think they're mainly for marketing for certain firms.

Link to comment
Share on other sites

Guest Ebrahim Saadawi

 A day after their test release, Sony released a firmware claiming increased dynamic range in raw to 15.3, which means it was less before the release. I would like DxO to comment whether they tested it after or before the firmware upgrade, and whether there's a change at all or not. 

It would be great if they can clear it up with a simple comment, 

so we would know if Sony's claims are false, or if the tests are incorrect. The comment asking this on their page seems to be one of the few unanswered.

Link to comment
Share on other sites

A day after their test release, Sony released a firmware claiming increased dynamic range in raw to 15.3, which means it was less before the release. I would like DxO to comment whether they tested it after or before the firmware upgrade, and whether there's a change at all or not.

It would be great if they can clear it up with a simple comment,

so we would know if Sony's claims are false, or if the tests are incorrect. The comment asking this on their page seems to be one of the few unanswered.


Aaaaah, ok. Interesting. I didn't know that. Maybe DXO Labs should answer it and quickly. Otherwise both Sony and DXO Labs seems to be overstating things a little.
Link to comment
Share on other sites

  • Administrators

DxO is completely irrelevant for video shooters. Unless you are shooting Magic Lantern Raw. Yes, it might give a suggestion of the capabilities, but it all depends on the internal processing of the raw data, so it can get either worse or better than the DxO results.

 

Also, I'd take the Techradar results with a grain of salt. I often see results in their graphs that are completely different from other lab tests.

 

Like this one. Look at GH4 dynamic range in raw, I wish it was that good! No way it has between 13-11 stops over the whole iso 100 - 25.600 range.

 

Sony_7S_TIFF_DR_03-900-90.jpg

 

DxO looks much more reliably to me if you look at the DR measurements: http://www.dxomark.com/Cameras/Panasonic/Lumix-DMC-GH4---Measurements

 

TechRadar say they use equipment and software supplied by DxO to do the testing. You'd think after a result like that they would stop and check their methods. Clearly some kind of operator error!

Link to comment
Share on other sites

Well they did test it with a "pre-release" firmware and only afterwards Sony came forwards with the release firmware that should have 15.3 stops in RAW. So there's atleast one mistake already.

 

And how do they always get Nikon/Sony made sensor cameras to test so easily but never Fuji? I think they're mainly for marketing for certain firms.

I'm guessing Fujifilm's X-Trans sensor crashes their software. :P

Link to comment
Share on other sites

TechRadar say they use equipment and software supplied by DxO to do the testing. You'd think after a result like that they would stop and check their methods. Clearly some kind of operator error!

It's not the first time I see a weird graph like that on TechRadar. I never valued their tests much personally...

 

For example:

 

Nikon_D800_TIFF_SNR-580-100.JPG

Look at the Nikon D700 graph. So you should shoot at ISO 3200 instead of 1600? the signal to noise ratio is suddenly as good as at ISO 100? - Everybody who has ever used this camera knows this isn't true.

 

Canon_EOS_70D_RAW_DR-580-100.JPG

The EOS 60D has a just under 10 stops of dynamic range from ISO 100 to 3200? Not true.

 

 A day after their test release, Sony released a firmware claiming increased dynamic range in raw to 15.3, which means it was less before the release. I would like DxO to comment whether they tested it after or before the firmware upgrade, and whether there's a change at all or not.

 

I don't believe a firmware upgrade can give a sensor suddenly two stops of extra dynamic range. The RAW image data (like DXO tests) is really up to the hardware Maybe with S-Log you can squeeze out more dynamic range cause of image processing, but DXO tests will never show these kind of results because they test raw sensor data.

 

 

I'm guessing Fujifilm's X-Trans sensor crashes their software. :P

 

DXO never tested a X-Trans sensor either. Understandable since those camera's work different - you can even see this in Adobe Camera Raw or Lightroom. So it probably doesn't work with their test methods. Same goes for Sigma Foveon chips. Never seen DXO-like tests of those. Just the conventional bayer sensor cameras.

Link to comment
Share on other sites


I don't believe a firmware upgrade can give a sensor suddenly two stops of extra dynamic range. The RAW image data (like DXO tests) is really up to the hardware Maybe with S-Log you can squeeze out more dynamic range cause of image processing, but DXO tests will never show these kind of results because they test raw sensor data.

Right - unless the previous firmware was significantly faulty.  S-Log shouldn't affect DR results either.  What can affect results is noise reduction because it lowers the apparent noise threshold in the darker tones.  But that is misleading since it reduces resolution/detail as well.

Link to comment
Share on other sites

By the way: the '15.3 stops firmware', is the lauch firmware: V1.0, which comes with all final production camera's. DXO usually tests production camera's, if not, they note it is a pre production model. In case of the A7S they don't, so I'd assume they tested with firmware V1.0.

 

Also interesting is the fact that Sony actually gives a real number for dynamic range. Camera companies never do that. Ever seen Canon, Nikon, Pentax or whatever advertise with the amount of stops of DR a camera has? I haven't. It's just the cinema camera companies that do this (RED, ARRI, BMD).

Link to comment
Share on other sites

By the way: the '15.3 stops firmware', is the lauch firmware: V1.0, which comes with all final production camera's. DXO usually tests production camera's, if not, they note it is a pre production model. In case of the A7S they don't, so I'd assume they tested with firmware V1.0.

 

Also interesting is the fact that Sony actually gives a real number for dynamic range. Camera companies never do that. Ever seen Canon, Nikon, Pentax or whatever advertise with the amount of stops of DR a camera has? I haven't. It's just the cinema camera companies that do this (RED, ARRI, BMD).

Part of the reason is that you have to establish some standard threshold for noise in the darker tones.  And this threshold is somewhat subjective.  I did a lot of DR testing using Imatest and the software gave multiple results for different noise thresholds.  It is up to the user to decide which threshold is appropriate.  This is one reason why you should be VERY careful about drawing conclusions about DR when the results come from different sources.  One of the big values of DxO testing is that the testing schemes are very consistent.  But as mentioned previously, those results can only be considered strong hits about what to expect when shooting video from the same sensor.Not only is there variations in processing of the raw data to consider, but the fact that video uses an electronic shutter may have influence (more noise) as well.

Link to comment
Share on other sites

  • Administrators

I don't understand why the hell tech radar are converting everything to TIFF rather than testing the raw files, because the software supports the raw files from all the leading cameras that DXOMark themselves have tested.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

  • EOSHD Pro Color 5 for All Sony cameras
    EOSHD C-LOG and Film Profiles for All Canon DSLRs
    EOSHD Dynamic Range Enhancer for H.264/H.265
×
×
  • Create New...