Why are you making a PS3 emulator?

[ssshadow]

There are more processors than just Cell on the PS3 motherboard. Things like network controllers, audio controllers etc can be ARM, and this is probably what they refer to.

[Hykem]

There's no ARM microprocessor in the PS3. We do have support for ARMv7 in RPCS3, but that was for PS Vita testing on earlier stages.
This will most likely be moved out of RPCS3 into another emulator in the future.

[mushroom]

There's no ARM microprocessor in the PS3. We do have support for ARMv7 in RPCS3, but that was for PS Vita testing on earlier stages.
This will most likely be moved out of RPCS3 into another emulator in the future.

The article I linked states this:

"The issue specifically revolves around an ARM processor inside the PS3 that controls both the clock as well as other base-level tasks, such as putting the PS3 to sleep. It's believed to be similar to the leap year bug that affected original Zunes and rendered them inoperable on New Year's Eve only to resolve itself the following day."

It's not really needed at all to emulate, but nonetheless the article does state that the PS3 does have an ARM processor.

[derpf]

There's no ARM microprocessor in the PS3. We do have support for ARMv7 in RPCS3, but that was for PS Vita testing on earlier stages.
This will most likely be moved out of RPCS3 into another emulator in the future.

The article I linked states this:

"The issue specifically revolves around an ARM processor inside the PS3 that controls both the clock as well as other base-level tasks, such as putting the PS3 to sleep. It's believed to be similar to the leap year bug that affected original Zunes and rendered them inoperable on New Year's Eve only to resolve itself the following day."

It's not really needed at all to emulate, but nonetheless the article does state that the PS3 does have an ARM processor.

And this article has little to no credibility given neither its sources (Sony itself) nor what quickness I did for a Google search turns up anything related to an ARM processor in the PS3.

[androidlover]

What I wanna know about rpcs3 is how the whole project came into existence in the first place. Did dhrpcs3 just start a github project, and others just heard about it somehow? What's the gap between rpcs3 in 2011 and rpcs3 today and in the near future, because rpcs3 has come a long way to run ps3 games already, even if basically one (swap).

Btw, people still think a ps3 emulator will never happen, or will take decades to load/play games, despite the depths rpcs3 has come in a few years.

[ssshadow]

What I wanna know about rpcs3 is how the whole project came into existence in the first place. Did dhrpcs3 just start a github project, and others just heard about it somehow? What's the gap between rpcs3 in 2011 and rpcs3 today and in the near future, because rpcs3 has come a long way to run ps3 games already, even if basically one (swap).

Btw, people still think a ps3 emulator will never happen, or will take decades to load/play games, despite the depths rpcs3 has come in a few years.

This answers some of your qustions: http://www.emunewz.net/forum/showthread.php?tid=159284

Also: people don't know what they are talking about, at all.

[derpf]

Btw, people still think a ps3 emulator will never happen, or will take decades to load/play games, despite the depths rpcs3 has come in a few years.

You should laugh at these people and then play a nice game of Arkedo SWAP or Scogger HD

[androidlover]

But the thing is ... if powerhouses like Haswell i7s @ over 4 GHz aren't predicted to play heavy PS3 games when rpcs3 is capable, why continue if the road ahead may be a dead end and it'a known? As discussed, nearly any modern PS3 game would make an i7 cry with rpcs3 running it ... even in the future processors do not seem to have such power.

Are you guys just hoping cousin and overall speeds will get better along the way? Because you will be severely limited to any gameplay experience, evenn something like Little Big Planet would be hell to emulate on any processors of this generation thus far.

[ssshadow]

But the thing is ... if powerhouses like Haswell i7s @ over 4 GHz aren't predicted to play heavy PS3 games when rpcs3 is capable, why continue if the road ahead may be a dead end and it'a known? As discussed, nearly any modern PS3 game would make an i7 cry with rpcs3 running it ... even in the future processors do not seem to have such power.

Are you guys just hoping cousin and overall speeds will get better along the way? Because you will be severely limited to any gameplay experience, evenn something like Little Big Planet would be hell to emulate on any processors of this generation thus far.

Wait until a good optimized PPU recompiler is done before making that assessment. Even a 5 ghz Haswell i7 can't run pcsx2 (!) in interpreter mode in most games. Interpreters are inherently slow, and according to AlexAltea, the rpcs3 interpreter is not only horribly slow, it's even slower than what it could be.

Eg: For me Nekotekinas SPU recompiler made that Fractal zoomer homebrew 30 (!) times faster (0,4 fps -> 12 fps) without any other performance tweaks.

A stock i7 4770k can now almost get 2 fps in Disgaea 3. If the PPU recompiler gives a boost similar to the best case for the SPU recompiler (maybe, maybe not), 2*30 = 60 fps, and you are done, it's playable.

Of course this is just pointless speculation, because I do not know, and no one else will really know either until it's done. I just want to put it out there that it's not impossible.

Edit: You are still fucked for Final Fantsy XIII, etc, though

[derpf]

if powerhouses like Haswell i7s @ over 4 GHz aren't predicted to play heavy PS3 games when rpcs3 is capable

This is where you stop, realize this is a dumb blanket statement, that nobody has really predicted anything, and that prediction without evidence is stupid.

Great, now that we have that taken care of, we can stop discussing this at all, and instead reflect on the smooth, smooth sounds of Sexy Sax Man.

---

and according to AlexAltea, the rpcs3 interpreter is not only horribly slow, it's even slower than what it could be.

It's not even the interpreter ATM, it's the memory manager, which is horrendously slow. Given that you can have many memory accesses per PPU instruction, it's not really a surprise that it tanks speed so much. It will be optimized in the future, though.

(The PPU interpreter just kind of sucks in general because there's a lot of spaghetti and indirection going on. It's not very cache-friendly at all, either.)