Thanks !
The Flash and the PSRAM on this board are both rated at 133 MHz, no worries, didn't messed with those... yet.![:)]( "Smile")
On the other hand, running fully bloated python from Flash with a 10000 loops of floating point and integer operation (with consistent results up to 636 Mhz), generating 1MB of garbage in RAM and PSRAM, with real-time output on USB, I would say that it is more than just testing the cores, but otherwise you are right, valid points.
The QSPI divider clock was automatically set by Pimoroni Micropython (nice surprise, on other versions you have to set it manually or live with it under-clocked at 1:4 ratio ).
The voltage must be set manually up to 1.3V, and for higher voltages you have to unlock a safety register, I don't recommend that unless you know exactly what you are doing ... it can go up to 3.3V.
The chip is B variant with the bigger chip package ... it looks like it can deal well with the heat.
Also there is an overheat protection, un-lockable, but just don't.
Please take the 600+ just as a stretch test, wouldn't use like that... except if it's absolutely necessary, for a short while. You can change the frequency on the run, quite glitch-less.
But at lower frequencies (multiple of max QSPI) it looks pretty workable, so far.
The Flash and the PSRAM on this board are both rated at 133 MHz, no worries, didn't messed with those... yet.
On the other hand, running fully bloated python from Flash with a 10000 loops of floating point and integer operation (with consistent results up to 636 Mhz), generating 1MB of garbage in RAM and PSRAM, with real-time output on USB, I would say that it is more than just testing the cores, but otherwise you are right, valid points.
The QSPI divider clock was automatically set by Pimoroni Micropython (nice surprise, on other versions you have to set it manually or live with it under-clocked at 1:4 ratio ).
The voltage must be set manually up to 1.3V, and for higher voltages you have to unlock a safety register, I don't recommend that unless you know exactly what you are doing ... it can go up to 3.3V.
The chip is B variant with the bigger chip package ... it looks like it can deal well with the heat.
Also there is an overheat protection, un-lockable, but just don't.
Please take the 600+ just as a stretch test, wouldn't use like that... except if it's absolutely necessary, for a short while. You can change the frequency on the run, quite glitch-less.
But at lower frequencies (multiple of max QSPI) it looks pretty workable, so far.
Statistics: Posted by gmx — Mon Sep 02, 2024 12:17 am