600 Mhz Powerful Enough To

[Yrx]

All Video Games are my passion and to tell you the truth I never owned a Saturn and would like to see a lot of the games I missed out on (Mostly Nights) so much so that I would pay a developer to work on a Saturn project!

If it looked likely, I would donate money towards it as well. I would need to be convinced that it would be possible, though.

 

[fusion_power]

Cool analysis, but you read more into it than I would ever let on. I'm pretty picky about not releasing information that should not be out there yet. I was just having fun with numbers to show the leap in technology we have in our grasp now!

TI's development platform for 3430 runs at 550MHz (this is public), so here is what I did:

ARM9 x 1.1 = ARM11
ARM11 x 2 = Cortex-A8 = ~1200MHz

We all know it has been proven to go much faster but we won't be trying it until the kernel is completely stable so the overclocking can be the only thing blamed for system crashes at higher speeds.

I personally think some optimized code will make a very capable Saturn emulator. It truly is a matter of who would be willing to put all the time in to making it.

Impressive numbers at all. But maybe it's better not to tell the Batteries what they have to power up, it could be, that they give up and quit and run away in fear before you can grab 'em an assemble the System. :lol:

Saturn Emulator? Hm, Saturn Bomberman is great! With Wlan and 10 Pandoras this could be going into insane fun. (I hope some Emulators will support Wlan)
But maybe the Devs should start from Bottom first, so NES, SMS 8Bitters then Super Nintendo, Megadrive etc... Even GBA and SNES will be enough hard work I think. If this stuff is running fine, the "big ones" could come into play. ^^

Oh, I'm sure MAME will be ported. But is it possible, that this Build for Pandora bases at the LATEST MAME Builds? So we can quit to convert the Roms into older Versions and all this complicated and unfunny stuff.
It would be awesome to play RADIANT SILVERGUN for example full Speed at Pandora

 

[Tobriand]

But maybe the Devs should start from Bottom first, so NES, SMS 8Bitters then Super Nintendo, Megadrive etc... Even GBA and SNES will be enough hard work I think. If this stuff is running fine, the "big ones" could come into play. ^^

You know, I actually think we might get PSX, GBA and SNES before anything else, simply because of who will definitely get dev units. Remember: this thing is so powerful, the code shouldn't need to be anything like as optimised as it needed to be for the GP2x or GP32. A dirty port from a C only emu for Windows for most systems will, in theory, run full speed, let alone if it were to have the optimizations for the 2x version included.

If I were to guess the running order, it would be SNES, MD/SegaCD, Psx, Amiga, Mame, AtariST, TG16, NES, MSX, N64 (but only this late because it took them a while to find one that was open source and working), Spectrum, C64, SuperGrafx/Turbo CD/SuperCD Tg16.

THEN about here is where the non-standard ones would start arriving in some order or another. x68000 would actually work on this device, for example, Saturn might be experimented on, or maybe someone'll get 3DO working. DS may turn up, sooner or later, but probably not for the first couple of years of Pandora's life.

 

[1magus]

I believe ya it's just I think if we ever get a DS emulator it will probably be good but will have the same problems that the PSP suffers from (Very buggy, not fully compatible with all games and some games suffer from sound lag) but if someone puts some work into it than why not.

 

[Squidge]

It'll be easier to port PC emus to, sure, but I think the challenge now will still be along the lines of using the least amount of Mhz for a particular emu. Besides, optimising is fun!

 

[Yrx]

32X would be fun as an add-on to an existing Genesis/MD emulator. There were a few really fun games for that.

 

[Tobriand]

I sense a new style of coding competition: release a stock emu's source to devs and challenge them to optimize it for battery life as well as performance. Then see which is using us more power after the deadline. Then merge all compatible optimizations together and get a very low power emu indeed .

 

[fusion_power]

It'll be easier to port PC emus to, sure, but I think the challenge now will still be along the lines of using the least amount of Mhz for a particular emu. Besides, optimising is fun!

Especially if you can save alot MHz with optimized Software. Running an unoptimized Emulator Full Speed at (example) 600MHz is not that fun like running same Emulator optimized at only 300MHz Full Speed. ^_^

Well as long we will HAVE all the "Standard" Emulators, it doesn't really matter which one comes first.

 

[argor]

N64 (but only this late because it took them a while to find one that was open source and working),

i thing i will be porting Mupen64 wen i get a pandora Mupen64 run well at 1 ghz but after some optimised i thing i can get it run well at 600 Mhz B)

 

[cowai]

I sense a new style of coding competition: release a stock emu's source to devs and challenge them to optimize it for battery life as well as performance. Then see which is using us more power after the deadline. Then merge all compatible optimizations together and get a very low power emu indeed .

:wub:

 

[Exophase]

N64 (but only this late because it took them a while to find one that was open source and working),

i thing i will be porting Mupen64 wen i get a pandora Mupen64 run well at 1 ghz but after some optimised i thing i can get it run well at 600 Mhz B)

Who said Pandora would even run at 1GHz, and I hope you don't actually think that a port of an N64 emu would run well at that anyway. Whoever works on N64 emulation has to write an ARM dynarec and not just cookie cutter optimizations, are you up to that?

 

[argor]

N64 (but only this late because it took them a while to find one that was open source and working),

i thing i will be porting Mupen64 wen i get a pandora Mupen64 run well at 1 ghz but after some optimised i thing i can get it run well at 600 Mhz B)

Who said Pandora would even run at 1GHz, and I hope you don't actually think that a port of an N64 emu would run well at that anyway. Whoever works on N64 emulation has to write an ARM dynarec and not just cookie cutter optimizations, are you up to that?

i am not saying that Pandora will run at 1GHz i am just saying Mupen64 can run good at 1GHz as it standst now i have made test to comferm it and try not to inshold me i nou my english is not wery good but i have bin coding emulators for years B) and even a 2 generation emulators coder my mother did teach me a lot about emulators and coding

 

[chris_r]

Well if you've been coding emulators for years you should know that the processor in the panda is nothing like the emulator in your pc. what is capable on a 1ghz x86 is a lot different to what is capable on a 1ghz arm

 

[argor]

Well if you've been coding emulators for years you should know that the processor in the panda is nothing like the emulator in your pc. what is capable on a 1ghz x86 is a lot different to what is capable on a 1ghz arm

you sayd i did test on x86 processor i never sat i did test on x86 processor :rolleyes: you seme to foget that some of osse have resorsere that is not avemble to all

 

[Exophase]

i am not saying that Pandora will run at 1GHz i am just saying Mupen64 can run good at 1GHz as it standst now i have made test to comferm it and try not to inshold me i nou my english is not wery good but i have bin coding emulators for years B) and even a 2 generation emulators coder my mother did teach me a lot about emulators and coding

You just don't get it. Whatever compilation of Mupen64 you ran had an x86 targeted dynarec - whatever port you do will have to use the interpreter. You can't use the dynarec performance to gauge how it'll run with the interpreter, you must have an ARM dynarec of similar quality to do so. Even then 1GHz ARM Cortex-A8 will probably not perform as well as a 1GHz Pentium 3 or Athlon, especially for dynarec output unless you try to reschedule the code to pair better.

I think anyone who has done a lot of emulator coding would know this. What emulation coding have you done anyway?

you sayd i did test on x86 processor i never sat i did test on x86 processor rolleyes.gif you seme to foget that some of osse have resorsere that is not avemble to all wink.gif

Yeah right.. what was this non-x86 1GHz platform that only you have access to, that you ported to but haven't released for... I think you're making this up.

 

[StealthBagel]

Spelling 'us' as 'osse', 'resources' as 'resorsere', and 'available' as 'avemble'... I hope english isn't this guy's first language.

 

[craigix]

I'm not too in to N64 emulation but I do remember that due to the libs the N64 dev kit comes with high level emulation can be used with quite a lot of success on many of the games, I remember playing N64 games on a 300mhz PC with a 3D card, I'm sure a lot of people remember that emu, it ran most of the nintendo produced games.

 

[sbock]

Yes, the emulator was called UltraHLE based on the Glide API und there is also an OpenGL version of it called UltraHLE 2064.
But even I'm no coder, I think it's no easy task to port this not very polished emulator to the Pandora...

 

[elinscheid]

I know that there are people dinking around with a N64 emulator on the PSP, I wonder how the performance is. If it does run at a decent speed then I bet it is heavily optimized. I don't use my PSP for home brew so I wouldn't know.

I can't wait to see the final specs on the Pandora and then compare it to the PSP. Honestly I won't really understand the specifics but I wonder about the strengths of both systems and what they are capable of.

 

[Joshwaan2k]

Yeah I think Corn 64 was another emus that had low system specs. A 500MHZ PC could run most games full speed.

Also with the CPU do you think it will run Mortal Kombat 2 on a MAME emu almost full speed?

I believe that game runs full speed on a 900MHZ PC.