Home · Register · Search · View Winners · Software · Hosting · Software · Join Upload & Sell

Moderated by: Fred Miranda
Username   Password

  New fredmiranda.com Mobile Site
  New Feature: SMS Notification alert
  New Feature: Buy & Sell Watchlist
  

FM Forums | Alternative Gear & Lenses | Join Upload & Sell

1       2       3              8      
9
       end
  

Sony A7 and A7R: New Firmware update on March 19th
  
 
akuba
Offline

Upload & Sell: Off
p.9 #1 · Sony A7 and A7R: New Firmware update on March 19th


johntruong wrote:
Also, what do you mean by "16-bit" mode? Are you referring to the RAW? All RAWs from Canon and Nikon are 14-bit, and Sony claims 14-bit RAW recording, but as some have shown on the forums here as well as digilloyd, Sony's RAW files are not true 14-bit.


16-bit mode refers to how many bits are used to store each pixel in memory (or on-disk) by the processing software you are using and during any hand-offs between different pieces of software. In RAW files the image data may be tightly packed (even in an uncompressed file) but in memory each pixel is stored in a native data type for efficient processing. Since all the native datatypes are powers of 2, processing software uses either 8- or 16-bit storage. 16-bit storage uses more memory and processes slower however it it is able to hold all of the original data without loss and additionally gives 2 bits of headroom for intermediate values during processing steps. Using 8-bit processing can easily result in posterization due to the decreased granularity.

The reason many here are skeptical of posterization claims is that no one (to my knowledge) has shown a credible example of it that wasn't a result of either poor choices in the processing workflow or of extreme processing. What constitutes 'extreme' processing is very subjective which is why other replies are asking for a description of your processing steps.



Mar 22, 2014 at 04:46 PM
sebboh
Offline
• • • • • •
Upload & Sell: Off
p.9 #2 · Sony A7 and A7R: New Firmware update on March 19th


akuba wrote:
16-bit mode refers to how many bits are used to store each pixel in memory (or on-disk) by the processing software you are using and during any hand-offs between different pieces of software. In RAW files the image data may be tightly packed (even in an uncompressed file) but in memory each pixel is stored in a native data type for efficient processing. Since all the native datatypes are powers of 2, processing software uses either 8- or 16-bit storage. 16-bit storage uses more memory and processes slower however it it is able to hold all of the original data
...Show more

there are lots of examples of the predilection of the a7(r) to posterize easily (rx1 too). if you look at those examples with the tools that show where sony's algorithm is throwing out data you see that it isn't happening in the sky gradient where the posterization is. so yes it is an issue with the cameras (depending on your subjects of choice), but no it doesn't seem to be due to the lossy raw compression. i'm happy to send people raws from my rx1 (and a7 once i actually take it outside to shoot) to play with where there is obvious posterization with no processing going straight from raw to tiff.




Mar 22, 2014 at 05:03 PM
akuba
Offline

Upload & Sell: Off
p.9 #3 · Sony A7 and A7R: New Firmware update on March 19th


sebboh wrote:
there are lots of examples of the predilection of the a7(r) to posterize easily (rx1 too). if you look at those examples with the tools that show where sony's algorithm is throwing out data you see that it isn't happening in the sky gradient where the posterization is. so yes it is an issue with the cameras (depending on your subjects of choice), but no it doesn't seem to be due to the lossy raw compression. i'm happy to send people raws from my rx1 (and a7 once i actually take it outside to shoot) to play with where there
...Show more

I'm not sure how you are concluding it is not due to Sony's compression. The tools that purport to show loss are only giving half the picture. They show you were data is lost during the 11 to 8 bit (second) step of the compression - if you look at what this step is doing, you would not expect it to be lossy for gradients. The other half of the compression is the first step of 14 to 11 bit mapping - this step is always lossy. The mapping is non-linear so you could be seeing posterization in the sparser range of this mapping, but I haven't seen an example without multi-stop pushing or pulling. Are you by any chance doing ETTR and then trying to pull the sky back a fair bit? I would be interested in seeing an example.



Mar 22, 2014 at 05:22 PM
sebboh
Offline
• • • • • •
Upload & Sell: Off
p.9 #4 · Sony A7 and A7R: New Firmware update on March 19th


akuba wrote:
I'm not sure how you are concluding it is not due to Sony's compression. The tools that purport to show loss are only giving half the picture. They show you were data is lost during the 11 to 8 bit (second) step of the compression - if you look at what this step is doing, you would not expect it to be lossy for gradients. The other half of the compression is the first step of 14 to 11 bit mapping - this step is always lossy. The mapping is non-linear so you could be seeing posterization in the sparser
...Show more

i will pm you a raw from the rx1 of one of the most egregious examples taken out a plane window where there is a large gradient filling most of the frame. if i have time this weekend i'll try to take something similar with the a7.

i pretty much never use ettr with sony cameras, because i don't find it adds enough to merit the increased danger of clipped highlights.




Mar 22, 2014 at 05:49 PM
snapsy
Offline
• • • •
Upload & Sell: On
p.9 #5 · Sony A7 and A7R: New Firmware update on March 19th


akuba wrote:
I'm not sure how you are concluding it is not due to Sony's compression. The tools that purport to show loss are only giving half the picture. They show you were data is lost during the 11 to 8 bit (second) step of the compression - if you look at what this step is doing, you would not expect it to be lossy for gradients. The other half of the compression is the first step of 14 to 11 bit mapping - this step is always lossy. The mapping is non-linear so you could be seeing posterization in the sparser
...Show more

It's a combination of the two. 11-bit channel encoding by itself wont yield posterization, at least not on current generation sensors.

Article



Mar 22, 2014 at 06:00 PM
akuba
Offline

Upload & Sell: Off
p.9 #6 · Sony A7 and A7R: New Firmware update on March 19th


snapsy wrote:
It's a combination of the two. 11-bit channel encoding by itself wont yield posterization, at least not on current generation sensors.


My reading of sebboh's post is that the issue is the classic bands-in-the-sky gradient posterization, not artifacts at high-contrast edge transitions. If it is the former then it is provably not the second step of the compression either by itself or in combination with the first step - provably in that you can verify on the decompression that the 7-bit deltas didn't require a shift during decoding. That leaves as possible culprits either the mapping step standalone, the mapping step interacting with post-processing, or some other piece of Sony's pipeline is coming into play.



Mar 22, 2014 at 08:13 PM
 

Search in Used Dept. 



Orfeaus
Offline

Upload & Sell: Off
p.9 #7 · Sony A7 and A7R: New Firmware update on March 19th


I haven't upgraded my a7's firmware yet; I wanted to get a feel for the kind of experience others have had. A cursory glance at the posts on this thread indicate some people have had problems with the update or have seen meager improvements in the startup times.

Is the update worth running?



Mar 23, 2014 at 04:05 PM
sebboh
Offline
• • • • • •
Upload & Sell: Off
p.9 #8 · Sony A7 and A7R: New Firmware update on March 19th


i didn't have any trouble, but i did it immediately after getting the camera so i can't really judge the improvement in startup time or if anything else is improved.




Mar 23, 2014 at 05:51 PM
wfrank
Offline
• • • •
Upload & Sell: Off
p.9 #9 · Sony A7 and A7R: New Firmware update on March 19th


Ive written about some. A new found one is a little menu change under delete, now you can delete "all images in folder" instead of one-by-one or by date.




Mar 23, 2014 at 06:29 PM
Timmytimetravel
Offline

Upload & Sell: Off
p.9 #10 · Sony A7 and A7R: New Firmware update on March 19th


I updated successfully after 5 times on a mac. 2 mistakes, when it say wait for the LCD screen, it popped up and said restoring something for more than a minute at which point I went, ok I guess that's what they mean, boot the next 4 times I kept doing other things on the mac, swapping windows, on the web etc. Every time I switched window focus, even after 5mins, immediate fail. So your mileage may vary, for me I gave up using the computer for anything else, waited what seemed like 10-15, then all good. That was usb3 port, old usb2 cable lying around.

Anyone had any more luck with 1080p out on the hdmi out on new firmware? I was going to use the a7 as an unattended wide b camera on an event shoot. Going to an external recorder will give me hours (few batteries or dc coupler) instead of 29m per go, however I am mixing with 1080p footage



Mar 27, 2014 at 08:42 PM
rji2goleez
Offline
• • • •
Upload & Sell: On
p.9 #11 · Sony A7 and A7R: New Firmware update on March 19th


Timmytimetravel wrote:
I updated successfully after 5 times on a mac. 2 mistakes, when it say wait for the LCD screen, it popped up and said restoring something for more than a minute at which point I went, ok I guess that's what they mean, boot the next 4 times I kept doing other things on the mac, swapping windows, on the web etc. Every time I switched window focus, even after 5mins, immediate fail. So your mileage may vary, for me I gave up using the computer for anything else, waited what seemed like 10-15, then all good. That was
...Show more

FWIW - I plugged my USB3 cable into my iMac (not a hub) and closed all other applications. Everything worked fine the first time through. I think it's crazy to have to update this way. Firmware should be simply loaded onto the card and popped back into the camera.



Mar 27, 2014 at 09:17 PM
ChiShutter
Offline
• • •
Upload & Sell: Off
p.9 #12 · Sony A7 and A7R: New Firmware update on March 19th


Quick note: in OS X mavericks the update kept failing about a third of the way through for me, until I left the updater app frontmost and refrained from doing anything else on my computer for the duration of the update.

Annoying, but glad the camera isn't bricked as I initially thought. Successfully updated to 1.02.



May 21, 2014 at 08:04 PM
1       2       3              8      
9
       end




FM Forums | Alternative Gear & Lenses | Join Upload & Sell

1       2       3              8      
9
       end
    
 

You are not logged in. Login or Register

Username   Password    Retrive password