PCSX2 Linux Development

[123_qwe]

Through Phoronix i came across this document http://www.slideshare.net/CassEveritt/ap...r-overhead# where some opengl optimizations are explained.

[gregory]

Actually phoronix is in my RSS reader

Anyway, buffer storage could help us a lots in my opinion. It is on my todo list but I haven't have free time for several monthes.

[123_qwe]

Actually phoronix is in my RSS reader

Anyway, buffer storage could help us a lots in my opinion. It is on my todo list but I haven't have free time for several monthes.

I've seen that on git you have started working on it Hopefully something great will turn out from it.
I've also been trying to learn OGL and its newer functionality and concerning that section of buffer storage i've found that DolphinEmu's code has been great (here for reference http://code.google.com/p/dolphin-emu/sou...Buffer.cpp ). Any reason why you used DYNAMIC_STORAGE_BIT instead of CLIENT_STORAGE_BIT?
Best of luck and thank you for your work Gregory

[gregory]

well for the moment it is slower ! DYNAMIC tell you that you can update the buffer from client. What it mean exactly I'm not sure, maybe I can drop it. CLIENT_STORAGE is a hint to the driver but I didn't see any impact so ...

[krysto]

Now that PCSX2 is on Github, I'm doing some work on the wiki:

https://github.com/ahtruex/pcsx2/wiki

[krysto]

Now that PCSX2 is on Github, I'm doing some work on the wiki:

https://github.com/ahtruex/pcsx2/wiki

On the other hand, it might be confusing having 2 wikis, as there is already http://wiki.pcsx2.net/. Maybe we should just roll any documentation into the original wiki?

[gregory]

Well I never used this wiki (which is a shame I know). Anyway nice work. By the way, any news on thread stuff?

[krysto]

Well I never used this wiki (which is a shame I know). Anyway nice work. By the way, any news on thread stuff?

Yeah currently reading pthread docs, so I know how to convert those functions into std::thread functions.

[krysto]

Though from what I understand, pthread's initialization is separate from it's construction whereas in std::thread everything is rolled into one. It will be a little trickier than just swapping out the functions 1-to-1.

[krysto]

Just a further update, I spent a huge amount of time today reading through the GSThread GSWorker, and GSRasterizer code, I was finally able to really wrap my head around the concept. I know you explained it to me before, but I didn't quite grasp it the first time around, haha.

Cross-platform threading will make a lot of this code unnecessary. I will try and simplify some things if possible.