atomicthumbs
I am the king of the collectors, with hospital bed
My point being that "revealing" information can be misinterpreted and totally useless, or it can be informativeSquidge said:Your point being?
Last edited by a moderator:
My point being that "revealing" information can be misinterpreted and totally useless, or it can be informativeSquidge said:Your point being?
Rivroner said::huh: NO real answer? :rolleyes:Squidge said:Two different frequencies![]()
We all know the first batch was run at 500Mhz.
Squidge said:Rivroner said::huh: NO real answer? :rolleyes:Squidge said:Two different frequencies![]()
We all know the first batch was run at 500Mhz.
We didn't know that per se (a few numbers were floating around, between 450 and 550MHz is where I'd put it), although it's good to know I guess.
Now we just need to know what you changed it to ;P Not that it really makes that much of a difference, beyond "hey guys, someone can get Pandoras this high" (and the number is probably not higher than the 625MHz we were told anyway)
TI provides its own kernel (look on TI site). TI recommends CodeSourcery 2007 q1-27 IIRC, which looks good.Pickle said:Did you build your own toolchain or did it come with TI's kernel? Are you even using TI kernel? Or do they just provide kernel patches and you built up eveything from scratch?
750Mhz. I've had it running at 800Mhz, but it crashed before it ran for the same amount of time as before.Exophase said:Now we just need to know what you changed it to ;P Not that it really makes that much of a difference, beyond "hey guys, someone can get Pandoras this high" (and the number is probably not higher than the 625MHz we were told anyway)
Squidge said:750Mhz. I've had it running at 800Mhz, but it crashed before it ran for the same amount of time as before.Exophase said:Now we just need to know what you changed it to ;P Not that it really makes that much of a difference, beyond "hey guys, someone can get Pandoras this high" (and the number is probably not higher than the 625MHz we were told anyway)
I'm hoping that the actual Pandora will clock higher and also give more performance at the same Mhz due to differences on how it's been made/components/etc.
Squidge, I'm really confused on all this, and if you could clarify that would be great. I remember that this chip is said to go up to 1 Ghz. Did they mean that it's highest clock was as powerful as 1 Ghz, but actually a different value? Or am I just completely wrong?
-God Ginrai
Squidge said:750Mhz. I've had it running at 800Mhz, but it crashed before it ran for the same amount of time as before.Exophase said:Now we just need to know what you changed it to ;P Not that it really makes that much of a difference, beyond "hey guys, someone can get Pandoras this high" (and the number is probably not higher than the 625MHz we were told anyway)
I'm hoping that the actual Pandora will clock higher and also give more performance at the same Mhz due to differences on how it's been made/components/etc.
Thank you very much for the answer man![]()
Thanks for the pressure Exo![]()
I was under the impression that 1ghz was likely if (and only if) they used the less power-efficient method of production - i.e. the one that uses something like 30-40% more energy for the same tasks. On the assumption that they wanted energy efficiency (which seems reasonable) over super-ultra-whizz-power, then 750MHz is pretty good (seeing as that process is the one quoted to be at ~550MHz in the early docs).God Ginrai said:Squidge, I'm really confused on all this, and if you could clarify that would be great. I remember that this chip is said to go up to 1 Ghz. Did they mean that it's highest clock was as powerful as 1 Ghz, but actually a different value? Or am I just completely wrong?
-God Ginrai
I definitely agree that you shouldn't chance killing your dev board just yet.Squidge said:Don't forget that the board is still running at TI specified voltages. By increasing the voltage we'll probably get higher speeds. However, I don't want to over voltage a dev board and kill it! Lets wait until we have some "spares" before trying to kill them, yes?
TI may have demonstrated a 1 Ghz version to somebody, but we don't know the conditions used for that (or power requirements?)
And that last part is exactly why I'm not too excited about whatever clock speed it is. I'm more impressed with the actual performance Squidge is getting out of this. That's the main indicator of power.Prophet said:I think 750Mhz is quite nice.
That's more than twice the PSP's top speed for example, and I've seen some really amazing emulators on the PSP.
Heck, it's faster than the factory preset clock speed of the eeepc.
I realize you can't make simple 1:1 speed comparisons based on CPU Mhz alone of course, but still...
The current Linux kernel isn't suited to SquidgeSNES at the moment.God Ginrai said:Btw, have you done any work on porting SquidgeSNES over to the Pandora yet? Or are you waiting until you get the LCD working to do that?
Squidge said:The current Linux kernel isn't suited to SquidgeSNES at the moment.God Ginrai said:Btw, have you done any work on porting SquidgeSNES over to the Pandora yet? Or are you waiting until you get the LCD working to do that?
Can you say why or is it under NDA?
I'm curious about how a kernel couldn't be suited for an emulator.
Are you going to run it on the framebuffer instead of X11? I thought yes. But does this mean the pandora will not have X11? Because I don't know how we will switch betwen x11 and framebuffer apps, it's not as easy, I think..
I think you probably hit the nail on the head with that one.Exophase said:Probably just means that it doesn't have some syscalls that he's relying on at the moment...
