(01-17-2016, 06:31 PM)Dante38490 Wrote: [ -> ] (01-17-2016, 06:10 PM)tambre Wrote: [ -> ]How ignorant can you be? Did you even read the rules or what I (also vlj) have said earlier?
(01-17-2016, 12:51 PM)Dante38490 Wrote: [ -> ]Then I just activated a feature that was disabled, I do not know why.
Then don't do it, if you don't know what you're doing.
(01-17-2016, 12:51 PM)Dante38490 Wrote: [ -> ]And I put something that should be like this, not as it was before.
in the end there is only one hack on PPUInterpreter.cpp
One hack that is not allowed, due to it being "aggressive", as I have explained quite a few times in the past.
I honestly hope it won't turn into the same situation, as it did on Github, where you ignored our feedback, requests, instructions and explanations.
Except that without the "hack" would work much less games, meme blackdeamon also use multiple hack for games that posting ...
Without the hack there would be far fewer games that would do something.
And for the line that are disabled, we should ask one who has disabled this and see if we could not be reactivated and resolved the problem.
Same for the hack "ppuinterpeter.cpp" if something is missing, it should be added to this problem as solved, so there would be no need to hack
Use hack is rather a good thing, which is to tell the developer, if it has "implémentais / was improving" this feature, this game could finally do something.
It tells them what a place focused their effort.
Yes there would be far fewer games that would work, and that isn't necessarily a bad thing. As noted in the rules, minor hacks, such as BlackDaemon's, where he often changes at most a line or couple.
And for the disabled lines, why didn't you ask in first place then? Why uncomment without investigating and finding out why it was commented in the first place?
For the hack in PPU Interpreter, as I have explained already many times before, the trap instruction is to halt processing and exit, due to a fatal problem in the processing. The exception is basically the same what would happen on a real PS3. Instead of simply disabling currently correct behaviour, you should instead find out why the trap instruction is called, and what causes it.
Hacks overall are not a good thing, they're very bad. Minor hacks are good to let developers know what areas to investigate (as noted in the rules). But for most things, we already know right now where to focus the effort.
Well. the hacks will disappear due the progress of the emulator and, dispite of it´s not in the master branch, the fact is that [BLES00952] is running with Dante´s hack.
I think it´s also good for RPCS3, because it makes the emulator more visible and maybe it helps to attract more developers.
Of course the project has rules and Tambre don´t want it to be a mess.
But why not just put Naruto Shippuden: Ultimate Ninja Heroes 2 [BLES00952] *
* = Works with hacks.
And just remove the ´*´ when it start to work with the master branch?
Sasuke vs Itachi boss battle screenshots ( used decrypted save )
http://imgur.com/a/Kl4d8
it renders the game almost properly
wish the emulator had a speedhack or frame skipper or something to increase the fps a little bit so its in somewhat playable framerate
I don't think it would change anything, according to my testing the render thread is able to pull ~20-30 fps ingame (frame time is around 30ms) but the rest of the emu can't keep up.
Likely llvm would help there, unfortunatly it doesn't support the same kind of hack as interpreter 2. If a TW instruction is executed it will still throw (but without a valid exception_ptr).
(01-18-2016, 12:32 AM)vlj Wrote: [ -> ]I don't think it would change anything, according to my testing the render thread is able to pull ~20-30 fps ingame (frame time is around 30ms) but the rest of the emu can't keep up.
Likely llvm would help there, unfortunatly it doesn't support the same kind of hack as interpreter 2. If a TW instruction is executed it will still throw (but without a valid exception_ptr).
what do you mean? you can play it at 20-30 fps? how?
and yeah, llvm helped dragon ball z to play at full speed while it was like 10 fps without it
(01-17-2016, 07:56 PM)ikki84 Wrote: [ -> ]I think it´s also good for RPCS3, because it makes the emulator more visible and maybe it helps to attract more developers.
Of course the project has rules and Tambre don´t want it to be a mess.
No. Developers don't want to work a developer that uses hacks. That's one of the reasons I'd never want to work on PCSX2, for example. Improper hacks/hacks without explanation are considered the worst and will rather keep better developers away.
And yes, there are rules. The rules are there for a reason. And the rules should be followed.
(01-17-2016, 07:56 PM)ikki84 Wrote: [ -> ]But why not just put Naruto Shippuden: Ultimate Ninja Heroes 2 [BLES00952] *
* = Works with hacks.
And just remove the ´*´ when it start to work with the master branch?
No.
Hacks such as disabling the trap instructions are pretty much horrible. A real PS3 would also "crash". The very purpose of the trap instruction is to halt execution if some error occurs and maybe hand over control to the operating system that will do something with the situation (or whatever). The trap instruction/crash or what you want to call it was put there by the developers of the game!
The real solution is to investigate why trap is called in the first place. You will perhaps find that some method returns completely invalid data for some reason. Then you fix that issue and who knows, maybe this game and 100 more will run just fine now.
In even simpler terms: You have a building with a leaking water pipe. You could either turn the water off and fix the pipe, or leave it on and maybe the first floor will still have some water, but the other 9 floors wont. Also the cellar is now flooded.
(01-18-2016, 04:06 PM)ssshadow Wrote: [ -> ]Hacks such as disabling the trap instructions are pretty much horrible. A real PS3 would also "crash". The very purpose of the trap instruction is to halt execution if some error occurs and maybe hand over control to the operating system that will do something with the situation (or whatever). The trap instruction/crash or what you want to call it was put there by the developers of the game!
The real solution is to investigate why trap is called in the first place. You will perhaps find that some method returns completely invalid data for some reason. Then you fix that issue and who knows, maybe this game and 100 more will run just fine now.
In even simpler terms: You have a building with a leaking water pipe. You could either turn the water off and fix the pipe, or leave it on and maybe the first floor will still have some water, but the other 9 floors wont. Also the cellar is now flooded.
Also since the cellar is now flooded, the pressure is building up in the pipe, and other pipes may also burst at any moment.
What's the status of the game on the latest master build?
with build master show this error
Quote:{PPU Thread[0x1] (main_thread)[0x008d6c44]} Exception:
(in file Emu\Cell\PPUInterpreter.cpp:1699, in function ppu_interpreter::TW)
Please report this to the developers.