Dynarec Completed For Psx4all Project.


Nr.56 posted on Oct 3 2006 at 10:33 PM said:
@ x-code
drink a bottle of port wine, like me. its a lot more efficient. :D
@ zodttd: studies are never waste of time ;)
30% speed increase is a steep rise. if interpreter or dynarec, it's the success that counts.


I'll stick to beer, I'm quite sure that Mr.zodttd needs a little time to relax and rethink matters, be it dynarec or something else, the cook is cooking, there is no need to panic :lol: .
 
Last edited by a moderator:
thanks zod and the rest, this parties been fun to be around for :)

hopefully a release we can play with soon? throw us a bone zod :p
Good work, thanks for trying...I'll leave you in the kitchen and join MR x-code for a beer

grats on page 50 everyone !
 
i think the most important thing is that zod has fun with his coding at the moment. he finds out something new everyday which gives him hope and joy. that's what keeps him at work and i'm sure he will try to get that dynarec to perform well someday. good job mate!
 
my post sounds panicky? :D
that was not my intension. in my opinion this emu is a fantastic part of software.
(If it is possible :blink: ) how much speed increase is to anticipate by a static precompiled game file? (how much affect the translation the performance of the unit?) How large would such a file be?
 
Zodttd can you cemment on the movie on youtube? it had ff7 running at 35 fps in game, and something like 25 in battle?
it seems way way better then the last public beta. Was it real?
That would be so great :D
 
why everybody already saying like 'it wont happen' 'no need to wait anymore'?
Zodttd and his team gave us more than we could have expected before they started working on it.
Im sure he will find something to sort things out.
Only by thinking about a dynarec he showed us he is a genius.
Even as genuis as he can, he needs time and maybe greatings, not farewell.

see you :lol:
 
Honestly, even an interpreter, with the things Zodttd now knows about asm and crap, speed can probably be seriously increased, and the power of the other hardware hasn't been tapped into yet (because it is a bit of a betch to do).
 
Nobody talked about abandoning the idea of psx emulation. It is only the general misconception that a fullspeed "every game runs perfect" psx emulator will be possible on the gp2x. That was once again, denied by one of our forum members and I don`t think it`s a bad thing...

Anyway. It`s amazing how far you guys got with psx4all. I hope you will succeed in improving this and find some new ways to speed it up. Who knows, maybe you will find a way to make that dynarec possible soon? Good work ZodTTD, Unai and Tinnus.
 
Jeez you people are depressing... Many games were playable in 0.1, it has improved vastly since, and it's not even finished yet! And unlike you, Z hasn't given up hope...


Damn, 51...
 
yeah like the man said youre dressing people

whos up for pushing this thing to page 60 ?
 
zod you goin to release the gp2x one soon as i will need somethin to do while i wait for my zaurus sl C1000 to come :p i orderd it form pricejapan just now with wireless card and usb cable thingy :D but i think im still gettin a v43 haha :D
 
I think there's plenty of positive reasons to continue this project.

I mean look at bleemcast. That emulator runs many games, albeit many few at full capacity but they've managed to fine tune certain games such as gt2, tekken 3 and that other one to run even better,

Now my understanding of cpu's and the arm commands(or whatever they're called) could only be less if I didn't know the name of the CPU. But as far as I can see the gp2x has more power availble to it. I understand that did have a dedicated 3d chip, but we have more ram, we have the source code for the system.

All that it requires is for everyone involved to invest time, now everyone on this forum is very busy, especially the programmers, so if everyone works together it can be achieved. Stop being so negative and look what been acomplished already.

And please, I am far from a programmer, but the source code release would be the biggest help in the world for the community.

Please feel free to flame
 
not by me! :lol:

you are right, but the calling for the sourcecode is too soon. Let the team develop this emu. i`m sure they will do. :)
 
Yes there was abit of bad news, but 33% guys and girls thats alot! That will be noteciable, I know thats only on one game, but this sounds like overall good news for the emu! Keep it up guys you have my support and many others.
 
¿33% of 7 frames are 2 or 3 frames isn't it? (my english is bad, sorry :( )
May be in 2D the performance is better than in 3D.May be in 2D is 50 % better than before :)
Time will tell.
 
:)

Comment on the FF7 video: It's real. :) Espetero is a beta tester for Unai, I believe. I'll give a quick test of FF7 on psx4gp2x and see what my numbers look like @ 266MHz.

I don't think there's too much to panic about. The dynarec was our great hope at getting things working "fullspeed". Right now there's quite a few 2D games that are running at about 30 FPS it seems, and that's with an interpreter. I have actually modified the interpreter a few times, and the threaded ARM assembly interpreter was fast, but I didnt know ARM assembly as well as I do now. It might be worth going back to. Right now I'm using an interpreter with assigned registers mapped to specific often-used variables.

The ARM Dynarec was a threaded (not multithreaded btw), 32MB translation cache, regmapped / known constant value, etc...I just dont know what's slowing down the dynarec. I haven't given up on it permantely, but since I'm out of idea's for improving the dynarec, and considering its slower than one of the interpreters...I'll wait til an idea comes along. :)

Haven't heard from Unai yet...more news soon...
 
Back
Top