PS Vita

[Ewlibx]

Hey
PS vita was recently hacked so it now can run homebrews. I just thought it would be really cool to play drastic on that device. Any thoughts on you porting or letting someone else port drastic to ps vita?

[kaikun97]

Hey
PS vita was recently hacked so it now can run homebrews. I just thought it would be really cool to play drastic on that device. Any thoughts on you porting or letting someone else port drastic to ps vita?

Does the PS Vita even have a ARMv7 or higher CPU for starters?

[ericbazinga]

Hey
PS vita was recently hacked so it now can run homebrews. I just thought it would be really cool to play drastic on that device. Any thoughts on you porting or letting someone else port drastic to ps vita?

You'd have to talk to Exophase (the creator of DraStic) about it.

[TkSilver]

Does the PS Vita even have a ARMv7 or higher CPU for starters?

It uses a arm cortex a9.... but only 3 of the 4 2.0GHz cores are available to programs 1 is used for the OS and is unavailable. The processor is no where near the issue that the 512MB ram would be. Not to mention the biggest issue which is that only the psp mode has been hacked and that basically means that it really does not matter what specs the vita has, we can not access them anyways.

[Exophase]

It uses a arm cortex a9.... but only 3 of the 4 2.0GHz cores are available to programs 1 is used for the OS and is unavailable. The processor is no where near the issue that the 512MB ram would be. Not to mention the biggest issue which is that only the psp mode has been hacked and that basically means that it really does not matter what specs the vita has, we can not access them anyways.

I don't think the clockspeed is anything like 2GHz, AFAIK it's more like 1GHz at most.

512MB RAM isn't an issue if you don't need to load 512MB and maybe 256MB compressed ROMs. I do all ARM testing on my Pandora which only has a 1GHz Cortex-A8 processor and 512MB of RAM, the RAM isn't a problem (even my older unit w/256MB of RAM was still enough to test). The CPU power is too little to enjoy some games but most are reasonable with some frameskip.

Pandora benefits from having very low OS overhead and screen update overhead, 1GHz Cortex-A8 or Scorpion Android devices fare a lot worse. I don't know anything about homebrew on Vita so I don't know how much overhead there is there, but having three available cores helps a lot for cutting down rendering time and Cortex-A9 is more efficient with a lot of the emulator's run time (not the 2D/3D parts though, NEON is actually a little worse on Cortex-A9 than A8, but having more cores more than makes up for that)

The bigger question is what the OS is like. DraStic uses some Linuxy things, and at the very least needs some kind of mmap-like functionality. I doubt I'd even entertain a port that's a lot of work.

[reeve]

Not to mention the biggest issue which is that only the psp mode has been hacked and that basically means that it really does not matter what specs the vita has, we can not access them anyways.

A native vita hack was recently released. It has been patched already, but hopefully being able to dump and analyze the RAM will lead to new exploits.

[TkSilver]

I don't think the clockspeed is anything like 2GHz, AFAIK it's more like 1GHz at most.

The cores appear to be running around 800mhz durring gameplay but they probably do go higher like the psp being underclocked to 222 to save battery life and only allowed to go to 333 in specific games. The base arm reference design is for up to 2GHz, that is where I got that number.

A native vita hack was recently released. It has been patched already, but hopefully being able to dump and analyze the RAM will lead to new exploits.

If you are talking about the webkit exploit in 2.60 to 3.18 in various forms of success then yes that is an exploit, but it is also far from a CFW (custom firmware) giving kernal access to unsigned code (ie homebrew). For homebrew to be both reasonable to code and accessible to users you really need a hack thst allows for a modified version of the firmware to be run. That firmware should have the ability to be upgraded (to be able to run current games that will have firmware checks if not already) and the ability to run unsigned code or the ability to sign user code (unsigned code is easier with the new keys that sony is using). That exploit is a good first step and hopefully will lead developers to a better understanding of how the firmware works and how the system works. Hopefully that will lead to a usable hack.... bit it is not there yet

[Exophase]

The cores appear to be running around 800mhz durring gameplay but they probably do go higher like the psp being underclocked to 222 to save battery life and only allowed to go to 333 in specific games. The base arm reference design is for up to 2GHz, that is where I got that number.

There are a ton of variables in how an ARM CPU is implemented in a particular piece of silicon. Process node and manufacturer, physical layout decisions like types of transistors used, configuration parameters and so on affect how high a CPU can clock. When ARM says up to 2GHz they mean with the right kind of design optimized for speed (although really it can go higher in the right conditions). And even if an SoC is physically capable of meeting timing requirements for some clock speed the SoC manufacturer or device integrator may limit maximum clock speed to something lower, for example to keep within power/thermal budgets that are important to the rest of the device.

I know when 3DS came out a lot of people were saying that it could clock at 1+GHz because that's what ARM says is the upper end for ARM11s on (now ancient) 65nm processes. In the end it has been long confirmed that it can't exceed 268MHz.

If you look at other 45nm-class quad core Cortex-A9 small mobile devices none of them can run at anything like 2GHz. I can't find a source that says anything other than > 1GHz, so that's probably what we're looking at here. I was a little disappointed when I heard this, hoping for something more like 1.2-1.5GHz, but the truth is that consoles and handhelds haev pretty much never been aggressive with CPU performance, at least outside of vector coprocessors.

[reeve]

If you are talking about the webkit exploit in 2.60 to 3.18 in various forms of success then yes that is an exploit, but it is also far from a CFW (custom firmware) giving kernal access to unsigned code (ie homebrew). For homebrew to be both reasonable to code and accessible to users you really need a hack thst allows for a modified version of the firmware to be run. That firmware should have the ability to be upgraded (to be able to run current games that will have firmware checks if not already) and the ability to run unsigned code or the ability to sign user code (unsigned code is easier with the new keys that sony is using). That exploit is a good first step and hopefully will lead developers to a better understanding of how the firmware works and how the system works. Hopefully that will lead to a usable hack.... bit it is not there yet

I'm not talking about the webkit exploit. I'm talking about the PSM exploit. It is capable of running unsigned code already on firmware up to 3.51. There's a SDK in development and people are already coding homebrew.

[huckleberrypie]

The bigger question is what the OS is like. DraStic uses some Linuxy things, and at the very least needs some kind of mmap-like functionality. I doubt I'd even entertain a port that's a lot of work.

AFAIK it runs on a Unix-like OS, i.e. based on *BSD, but I'm not sure if that would make things feasible enough for Drastic to run on an API standpoint.