Hacked Sony A7S?!

Thanks a lot paulcurtis.

From what you describe, I also think RAW video is not to be expected. Internal 4K, perhaps, with some luck. And slog3+sgamut3.cine, possible but very unlikely.

Then again less ambitious stuff should be feasible: remapping buttons, custom WB working with slog2 (through auto-exposure or temporary change of PP), etc.
 
paulcurtis said:
The whole point (IMHO) is whether the raw stream can be tapped into and there's no sign of that. yet.
Click to expand...
Sure, raw would be nice, but there's a lot of other functionality that can make the camera better. Have you looked in "source code app\com.sony.imaging.app\sony\imaging\app\avi\C0001R.java" ? There's a mind boggling list of parameters with a numerical code for each one. Things like 4K recording and 100Mbps bitrate for XAVC are there. It would seem like a worthwhile endeavor to see if the camera would respond to changes in these parameters.
 
Ralph B said:
Sure, raw would be nice, but there's a lot of other functionality that can make the camera better. Have you looked in "source code app\com.sony.imaging.app\sony\imaging\app\avi\C0001R.java" ? There's a mind boggling list of parameters with a numerical code for each one. Things like 4K recording and 100Mbps bitrate for XAVC are there. It would seem like a worthwhile endeavor to see if the camera would respond to changes in these parameters.
Click to expand...

There is a lot of stuff in there, that's for sure.

It would need a bunch of people quite a long time to get to the point of trying to poke things at the camera of which there's probably a real chance of bricking the camera too.

Fundamentally, i don't know where this stuff is coming from. If people are going to get into the nuts and bolts then it would make sense there was some information about where the code came from. How was it 'decompiled' and from what. It's very unlikely there are string constants in the firmware, so i wonder where they came from. There are some odd misspellings too, so was this all manually added? I wouldn't have thought so but who knows.

If Nabil wanted to move the project forward he should really be reaching out to the community or at least allowing people to contact him. He says he's got an app running, but doesn't really say how. There's a tiny readme with little information in.

He also says that he hasn't released all the source code, so what was the point of a peek at it?

IMHO i suspect it's possible to write an app, get it into the camera and gain some sort of access to the camera hardware, running in an android sandbox. It would be an enormous amount of work to get it to that point based on nothing and also not very clear what the scope is once you have.

Like mulder i really want to believe, but why release it this way? Unless there is some community out there pulling it apart cooperatively?

cheers
Paul
 
Forget 4k internal (even if possible, that's probably a long way from being stable and usable)- does this mean we might see more capable internal apps that we can sideload??

I had dreams of things I liked about Magic Lantern coming over (ghost image, for example) since Sony supported apps on their cameras. But it turned out Sony kept all the good stuff for internally-developed apps only (like the intervolometer), and third parties only get access to remote control style apps. Boo!
I was never really into the Magic Lantern RAW workflow, but many of the usability features I miss on a daily basis.
 
He will release an app today that brings better color then slog2 to the A7S. Lets wait and see what's he on about.
 
Polaroid22 said:
He will release an app today that brings better color then slog2 to the A7S. Lets wait and see what's he on about.
Click to expand...

Fantastical claims aren't very convincing. I'd rather he release an API or something. At first glance this thing sounded plausible (I've been itching to find a way to make certain apps possible), but the more I'm hearing about it, the more it seems like baloney.
 
M. Gilden said:
Fantastical claims aren't very convincing. I'd rather he release an API or something. At first glance this thing sounded plausible (I've been itching to find a way to make certain apps possible), but the more I'm hearing about it, the more it seems like baloney.
Click to expand...
True, like this morning, "today I will release the app".. 4 hours later.. "Nah I will not release the app, cause I am doing the hard work and you aren't.."
Dude really?
And then his Nab-log vs SLOG2. the slog2 looks better then his nablog. but he says better DR and better color. release the damned thing so we can see for ourselves..
 
Haha it seems to be very easy to get people emotional without showing a thing.
So this kid is a little brat who can't do anything but brags about all the things he could do......but he can't do anything except wasting his and others time lol
 
Ok he released the test app. You can download it from here.
https://twitter.com/nabfathi15

According to him: better colors and more DR. I have no access to my A7S this week, but would love to give it a try. Did anybody else try it already and what do you think?
I hope 4K and if possible 10bit are coming soon!
 
Do you really feel confortable installing untested software from such an opaque and unfriendly source? I know I wouldn't. My crystal ball says: if this bricks a camera he's going to say it's the user's fault and call him an idiot.
 
Samuel H said:
Do you really feel confortable installing untested software from such an opaque and unfriendly source? I know I wouldn't. My crystal ball says: if this bricks a camera he's going to say it's the user's fault and call him an idiot.
Click to expand...
Well I didn't see anything that could do harm to the camera. But I did not read all the code. Some users already verified it is working fine. Hopefully some other coders will get in there and help. 4k and 10bit are first on my list :)

I want to see some video's to see what is improved as the screengrab for now only suggests the same color control with different layout. And hopefully 4K or 10bit internal is possible. Or slog2 at lower ISO's would be great. (as A7S II can do it at iso 1600, I was thinking they were saving that one just for the A7S II, so that could be possible)
 
Last edited:
You have to hand it to him, he appears to have achieved quite a bit here but as has been said i would be very wary of installing anything right now simply because of his aggressive attitude. Compare that to the ML team or in fact any open source team.

He's reverse engineered the API that the play mobile apps use. He's not hacked the firmware or anything though. So whatever apps can be developed would rely on Sony generated internal APIs. So you have to ask yourself what would Sony be doing leaving APIs to access 4K on an A7s for example - if they themselves didn't plan on releasing 4K internal on the A7s.

Now it could be that there is access to the uncompressed data stream from the sensor via the API, in which case it's conceivable that the data could be written elsewhere. The NLog app could be either applying a LUT to this stream or perhaps pushing a LUT to the firmware to use. That by itself offers interesting possibilities.

There is definite interest here as there always has been - i want to believe!

I would be more than happy to speak to him but i've not found any way to access and i appear to be blocked on twitter too! Does anyone have contact details?

Oh and i fail to see the use of NLog - Slog2 at lower ISOs would be more useful for actual footage. The cine curves can be messed around with to get this look - perhaps all that the API is doing is applying picture profile settings to an existing cine curve... I hope not, i hope he's got a LUT working there, something really new.

cheers
Paul
 
You must log in or register to reply here.
Back
Top