Ok quick checkin from BABBQ. After this post I'm putting my BT keyboard away and getting drunk, and likely won't return to XDA until Monday or Tuesday evening.
The official announcement was kinda "meh". It's really what's going on behind the scenes that is mattering.
There are multiple devrel people at Samsung, many of whom weren't even aware of each others' existence. They now are, and I have quite a few new contacts. Now, requests from us are probably going to go through multiple chains. Whether this makes a difference, who knows - but it is a potential step forward.
They also had an engineer from LSI that I talked to personally. It was a bit difficult due to the language barrier, but I sort of got the impression that the problems with current Insignal source on our phones aren't phone-specific - some of this **** is broken even on the OrigenBoard itself. Same case for libcamera - it sounds like the current Insignal one is pretty low-quality/broken, and the "new" source may be improved. It won't be directly usable on our devices, but it may represent a significant step forward. e.g. instead of "it's totally and completely broken", it may be "sorta working and needs some tweaking for each camera module" - See, for example, use of the Nexus S libcamera on Aries-family devices - each one has needed device-specific work, but instead of being an insurmountable task, it may be feasible. However, please keep reading regarding maintainer exhaustion...
I also got the impression that future Insignal drops are going to be more aggressively managed by LSI. Again... none of this is official.
November is likely to be another ICS tarball, however, one that is actually working. The engineer I spoke to said that the actual JB differences are minimal. (As in, I know what they are, and in fact we have an implementation of the JB differences in Gerrit already. The problem is the underlying hwcomposer is apparently broken even for OrigenBoard+ICS with the current Insignal source. Functional ICS HWC shouldn't be too hard to JB-ify. (Look at the Nexus S HWC as an example of what needs to be done to go from ICS to JB.)
The new source drop might also allow us to finally fix HDMI.
The problem is - Thanks to the lack of documentation for the past few months, the team of CM maintainers for Exynos are exhausted and far behind on other non-Exynos-specific stuff. So this isn't going to be a magic bullet. It alone probably won't get us M builds.
However the other "behind the scenes" item is that a bunch of Samsung people were here, and they're probably going to be writing trip reports to management. This trip reports are going to likely include the massive amount of pwnsauce Sony dished out. Speaking of that - Espenfjo, xplodwild, and codeworkx, Y U NO FLY HERE? Chainfire, Jerpelea, and Supercurio were all here. U NO GET SMARTWATCH! :P
So Samsung Mobile might be more receptive to community requests in the future. Our contacts are likely to be in a MUCH better position to facilitate things than before.
So it's a step forward... But it's likely to be months before any of us consider a new Exynos device, and our current devices are likely to always be behind others because we're just too far behind as it is. But if Samsung keeps on improving, after a few months users might see definitive improvements.
With all of that said, I'm off to party and feast. You won't see any more from me until Monday evening at the absolute earliest.
Sent from my GT-N8013 using Tapatalk 2
*so much sig updating needed*
My Github profile - Some Android stuff, some AVR stuff
An excellent post on "noobs vs. developers"
A few opinions on kernel development "good practices"
Note: I have chosen not to use XDA's "friends" feature - I will reject all incoming "friend" requests.
<MikeyMike01> Smali is a spawn of hell
<shoman94> ^^^ +!
<Entropy512> gotta be careful not to step on each other's work. :)
<Bumble-Bee> thats true
<jerdog> compeete for donations