Who Is Gp2x-int


Status
Not open for further replies.
'craigix' said:
And she could speak English, and entertained my request to see the Korean fish market.

err.. as it were.
LOL! *literally*
 
Last edited by a moderator:
I wonder how well the pandora team would fare trying to market the pandora in korea. Would there korean be just as bad as GPH's english, it would be funny.
 
Last edited by a moderator:
'WarmFluffyUK' said:
They haven't fixed the hardware, the fix in on a software level, apparently if you use the SDK you won't get any screen tearing.
They had implemented a rotation in SDL. It reduced performance by 400% to 50% of GP2x levels.

They said they were going to undo these changes due to this. I have also heard something about using the 3D hardware to perform this rotation at no performance cost, as Orkie already said. Maybe something has already come of that?
 
Last edited by a moderator:
Peter R said:
They had implemented a rotation in SDL. It reduced performance by 400% to 50% of GP2x levels.

They said they were going to undo these changes due to this. I have also heard something about using the 3D hardware to perform this rotation at no performance cost, as Orkie already said. Maybe something has already come of that?


(oops, I originally thought you meant it ran at 4x GP2X speed at first, I can see from your other thread how I misread you) What were they benchmarking? Abysmal performance explains why they came here asking for a routine, but writing something close to optimal in ARM assembly language isn't difficult if you know the first thing about the architecture. If they can't do that then they shouldn't be manufacturing an embedded device, or need to hire someone who knows the first thing about how to write embedded software.

I'm not going to be using SDL, especially while it's broken (the audio bugs). I know there are others who won't as well (notaz..).. and since none of us used GPH's SDL on GP2X (did they even provide source?) then if everyone moved to someone else's SDL this solution will have to move with it.

People banging the hardware directly aren't going to be very interested in what SDL is doing. If the 3D hardware is being utilized then a concise code example of how to do this might be appreciated, but it introduces its own complications. It means that you now have to have additional buffers for writing to, ones that the 3D subsystem can access, and it takes away from total throughput if you were using 3D in the first place. It also takes additional bus bandwidth. It's not as good as having hardware that actually works correctly, and I'd prefer it if GPH didn't pass it off as such.
 
Last edited by a moderator:
Exophase said:
[quote name='Peter R' post='720597'
I'm not going to be using SDL, especially while it's broken (the audio bugs). I know there are others who won't as well (notaz..).. and since none of us used GPH's SDL on GP2X (did they even provide source?) then if everyone moved to someone else's SDL this solution will have to move with it.
Only thing ive seen do the sw rotation was fce from Orkie, The SDL was done by Puck (most if not all i think) and a version of the source has been provided.

Even if the the tech guys are a different location, im sure they have phones.
 
Last edited by a moderator:
Yeah, saying "use our SDK and all will be well" isn't really good enough. If the GP2X is anything to go by, their SDK will be horribly outdated anyway, and no one will use it. Apart from quasist, but we all know what a fucking nutjob he is.
 
Status
Not open for further replies.
Back
Top