exxos's Atari, Amiga & retro forum

We welcome retro users & hardware gurus alike! Come and join the party :)
https://www.exxosforum.co.uk/forum/

Granlunds FrankenSTein

https://www.exxosforum.co.uk/forum/viewtopic.php?t=2879

Page 15 of 19

Re: Granlunds FrankenSTein

Posted: 24 Nov 2020 12:48
by Atarian Computing
agranlund wrote: 24 Nov 2020 12:47
Atarian Computing wrote: 24 Nov 2020 12:34 This is me:
What benchmark tool is that and where can I get it? :)
Ah, this is the great Kronos: http://kronos.lutece.net/. Very useful.

Re: Granlunds FrankenSTein

Posted: 24 Nov 2020 13:37
by Atarian Computing
My Memspeed results:

Code: Select all

Type	L1	L2	Main
Read	47.1	31.2	31.6
Write	31.6	31.2	31.2

Re: Granlunds FrankenSTein

Posted: 24 Nov 2020 14:03
by Atarian Computing
Apparently I have an older firmware. Let’s see how @agranlund fares with the latest one.

Re: Granlunds FrankenSTein

Posted: 24 Nov 2020 14:27
by agranlund
At 50Mhz, you'll get somewhere close to 50mb/s on cache-hit 32bit reads. Cache-hit writes slower and about the same speed as cache-miss writes - makes sense seeing as the 030 cache is always operating in write-through mode.

The nembench numbers seem to be about the same between the PAK3 (with FRAK) and the TF534. For kicks I'll run those other benchmarking tools on the TF536 when I get a chance but I'd expect the numbers to be within a reasonable margin of error compared to these?

PAK3, 50Mhz, FRAK/2 TT-RAM:

Code: Select all

32bit read (100% hit)        -> 49.261 MByte/sec (~313%)
32bit write (100% hit)       -> 28.571 MByte/sec (~428%)

Linear 32bit read (ST-Ram)   -> 3.917 MByte/sec (~73%)
Linear 32bit write (ST-Ram)  -> 3.917 MByte/sec (~60%)
Linear 32bit copy (ST-Ram)   -> 1.682 MByte/sec (~52%)

Linear 32bit read (FastRAM)  -> 31.282 MByte/sec (~588%)
Linear 32bit write (FastRAM) -> 27.507 MByte/sec (~426%)
Linear 32bit copy (FastRAM)  -> 9.630 MByte/sec (~298%)
TF534:

Code: Select all

32bit read (100% hit)        -> 49.261 MByte/sec (~313%)
32bit write (100% hit)       -> 28.901 MByte/sec (~433%)

Linear 32bit read (ST-Ram)   -> 3.919 MByte/sec (~73%)
Linear 32bit write (ST-Ram)  -> 3.921 MByte/sec (~60%)
Linear 32bit copy (ST-Ram)   -> 1.964 MByte/sec (~60%)

Linear 32bit read (FastRAM)  -> 30.131 MByte/sec (~566%)
Linear 32bit write (FastRAM) -> 27.976 MByte/sec (~433%)
Linear 32bit copy (FastRAM)  -> 13.988 MByte/sec (~433%)

And some random TT machine:

Code: Select all

32bit read (100% hit) -> 31.1695 MByte/sec (~201%)
32bit write (100% hit) -> 13.1054 MByte/sec (~195%)

Linear 32bit read (ST-RAM) -> 7.867MByte/sec (~148%)
Linear 32bit write (ST-RAM) -> 7.867MByte/sec (~121%)
Linear 32bit copy (ST-RAM) -> 3.849 MByte/sec (~122%)

Linear 32bit read (FastRAM) -> 14.835 MByte/sec (~279%)
Linear 32bit write (FastRAM) -> 12.615 MByte/sec (~195%)
Linear 32bit copy (FastRAM) -> 4,849 MByte/sec (~150%)

Re: Granlunds FrankenSTein

Posted: 24 Nov 2020 14:30
by Atarian Computing
My Nembench for comparison's sake:

Code: Select all

NemBench v2.1 - precision CPU/FPU profiler.

Integer multiply (16bit)     -> 1.921 Mips (~313%)
Integer divide (16bit)       -> 1.135 Mips (~313%)
Linear (stalled) integer     -> 24.899 Mips (~312%)
Interleaved (piped) integer  -> 24.899 Mips (~312%)

16bit read (100% hit)        -> 24.600 MByte/sec (~313%)
16bit write (100% hit)       -> 14.409 MByte/sec (~239%)
32bit read (100% hit)        -> 49.140 MByte/sec (~313%)
32bit write (100% hit)       -> 28.735 MByte/sec (~431%)

Linear 32bit read (ST-Ram)   -> 3.156 MByte/sec (~59%)
Linear 32bit write (ST-Ram)  -> 3.917 MByte/sec (~60%)
Linear 32bit copy (ST-Ram)   -> 1.963 MByte/sec (~60%)

Linear 32bit read (FastRAM)  -> 32.524 MByte/sec (~611%)
Linear 32bit write (FastRAM) -> 27.798 MByte/sec (~430%)
Linear 32bit copy (FastRAM)  -> 13.415 MByte/sec (~415%)

Re: Granlunds FrankenSTein

Posted: 24 Nov 2020 15:53
by Atarian Computing
Some more benches:

Not having a Blitter takes a bit of a hit:
TF-STacy-GB.png
grab0085.png
grab0087.png
grab0088.png

Re: Granlunds FrankenSTein

Posted: 24 Nov 2020 17:06
by agranlund
memspeed shows this on my machine:

Code: Select all

	L1	L2	Main
Read:	48.0	31.6	32.0
Write:	31.6	31.6	31.6
xferrate isn't useful on my current setup I suppose. I'm not using the IDE interface on the TF536 due to height restrictions inside my setup. And to be honest I'm perfectly fine with my other, slower, IDE interface on the ST-bus so I'm not sure when/if I'll get around to making a relocator board for the TF536 IDE.

That Kronos application was really cool, thanks for the tip!

Re: Granlunds FrankenSTein

Posted: 30 Nov 2020 12:44
by agranlund
Maybe not super interesting for most people but STBlitter_RevC is now on Github:
https://github.com/agranlund/STBlitter_RevC

(along with the DIY rom+rtc and ram boards for the short ST model)

Re: Granlunds FrankenSTein

Posted: 01 Dec 2020 00:05
by Cyprian
great
thanks @agranlund

Re: Granlunds FrankenSTein

Posted: 02 Dec 2020 01:19
by agranlund
Now stable at 1280x1024, the native resolution of my BenQ monitor :yay2:
1280x1024.jpg
All times are UTC+01:00
Page 15 of 19

Powered by phpBB® Forum Software © phpBB Limited