How many CPU cycles does the 6809 at normal running frequency made in one second?
Currently, unoptimized DragonPy runs around 55000 cycles/sec.
CPU perfomance...
CPU perfomance...
... too many ideas and too little time ... Related stuff written in Python:
Dragon 32 emulator / PyDC - Python Dragon 32 converter: https://github.com/jedie/DragonPy
DWLOAD server / Dragon-Lib and other stuff: https://github.com/6809
Dragon 32 emulator / PyDC - Python Dragon 32 converter: https://github.com/jedie/DragonPy
DWLOAD server / Dragon-Lib and other stuff: https://github.com/6809
Re: CPU perfomance...
894886.25 (14318180 ÷ 16) - or slightly fewer on some Dragons.
Re: CPU perfomance...
894886.25 when a Dragon is fitted with a 14.31818 MHz crystal. 888625 when a Dragon is fitted with a 14.218 MHz crystal.
Re: CPU perfomance...
Wow. Than i'm far away from that.
But ok, i didn't do any optimizing and have many logging action inserted... there is still a lot air
But ok, i didn't do any optimizing and have many logging action inserted... there is still a lot air
... too many ideas and too little time ... Related stuff written in Python:
Dragon 32 emulator / PyDC - Python Dragon 32 converter: https://github.com/jedie/DragonPy
DWLOAD server / Dragon-Lib and other stuff: https://github.com/6809
Dragon 32 emulator / PyDC - Python Dragon 32 converter: https://github.com/jedie/DragonPy
DWLOAD server / Dragon-Lib and other stuff: https://github.com/6809
Re: CPU perfomance...
OK... It has interest me... So i just comment the many logging output with a script:
https://github.com/jedie/DragonPy/commi ... a37f8fce6b
New values are:
* Running is a VirtualBox with PyPy: ~315.000 cycles/sec (before: 140.000 cycles/sec)
* in VM with CPython: ~165.000 cycles/sec (before 37.000 cycles/sec)
* CPython native under Windows: ~235.000 cycles/sec (before 55.000 cycles/sec)
It's still far from 894.000 on a real machine. But i only remove the logging activity and make no real optimizing.
And it's just enough to make some tests...
https://github.com/jedie/DragonPy/commi ... a37f8fce6b
New values are:
* Running is a VirtualBox with PyPy: ~315.000 cycles/sec (before: 140.000 cycles/sec)
* in VM with CPython: ~165.000 cycles/sec (before 37.000 cycles/sec)
* CPython native under Windows: ~235.000 cycles/sec (before 55.000 cycles/sec)
It's still far from 894.000 on a real machine. But i only remove the logging activity and make no real optimizing.
And it's just enough to make some tests...
... too many ideas and too little time ... Related stuff written in Python:
Dragon 32 emulator / PyDC - Python Dragon 32 converter: https://github.com/jedie/DragonPy
DWLOAD server / Dragon-Lib and other stuff: https://github.com/6809
Dragon 32 emulator / PyDC - Python Dragon 32 converter: https://github.com/jedie/DragonPy
DWLOAD server / Dragon-Lib and other stuff: https://github.com/6809
Re: CPU perfomance...
**update**
Currently TKinter doesn't work with PyPy under Windows. So i have implement a simple console with: https://github.com/jedie/DragonPy/commi ... debcf13f5d
So i can run the sbc09 ROM under Windows with PyPy... And so i go faster than realtime with ~1.370.000 CPU cycles/sec
And remember: I only just commend the few logging activity and not really optimizing code.
Currently TKinter doesn't work with PyPy under Windows. So i have implement a simple console with: https://github.com/jedie/DragonPy/commi ... debcf13f5d
So i can run the sbc09 ROM under Windows with PyPy... And so i go faster than realtime with ~1.370.000 CPU cycles/sec
And remember: I only just commend the few logging activity and not really optimizing code.
... too many ideas and too little time ... Related stuff written in Python:
Dragon 32 emulator / PyDC - Python Dragon 32 converter: https://github.com/jedie/DragonPy
DWLOAD server / Dragon-Lib and other stuff: https://github.com/6809
Dragon 32 emulator / PyDC - Python Dragon 32 converter: https://github.com/jedie/DragonPy
DWLOAD server / Dragon-Lib and other stuff: https://github.com/6809
Re: CPU perfomance...
You have to take into account that outputting debug data is quite a heavy process, so no wonder that without it, it goes faster...
Congratulations on having it running in real time!
Congratulations on having it running in real time!