Got a PPC card? I need you!

[rayt]

I have just downloaded it and tried it out but the "error 8" problem with p96 seems to be still there.

[Karlos]

Strange :-/

I'll have another look tonight. Probably I need to lower the cybergraphics.library version again. Or better still, make it prefer picasso96 directly if it finds it. I plan to add some more tests to it (scattered memory read/write for one), and change the test to allocate as large as possible memory buffers. This will allow the Peg users who tested it to get a fairer estimate of their memory speed (and not just their L2 caches ;-) ) as well as paving the way for an OS4 native version, which would have the same L2 cache issues on an A1.


Anyway, I'd like to say thanks to everybody so far for their time.

[PiR]

Don't shout, don't shout, I'm comming.

Cannot open intuition.library ver.40
My hardware: PPC603@210. ;-)

[Karlos]

Ah, b*llocks. I must have uploaded the original version twice instead of the recompiled library version fix :lol:

There is no other reason for it wanting to open v40 inuition, the last build *definately* requests v39.

I'll re upload it tonight (complete with some sort of large buffer allocation for those with plenty of RAM and/or Peg I/II).

[Karlos]

Hi again,

I've uploaded the (hopefully) P96 friendly version (uses CGX3 only). It might also work on OS3.0 (I lowered the graphics/intuition requirements to v39).

Didn't yet write a large memory version so anybody wanting to measure their L2 performance on a G4 peg can still do so :-D

I will fix that when I get a moment.

[poweramiga2002]

hows the OS4 version going ?

[Karlos]

poweramiga2002 wrote:
hows the OS4 version going ?

When it's done :-D

Seriously, the code needs updating a little before that. Benchmarking for A1 (and Peg) has to account for the large caches they have.

I was primarily interested only in the classic performance, but I will have to have to ask an A1 owner I know if I can write it on his since my own OS4 installation is a tad bit ropey at present :-(

[rayt]

I've uploaded the (hopefully) P96 friendly version (uses CGX3 only). It might also work on OS3.0 (I lowered the graphics/intuition requirements to v39).

Ok now it works on my system with P96, have just sent you the results.

[Karlos]

@all

Still no OS4 native version :-( but I did update the code in preperation for a port.

It now allocates up to 32MB buffers for testing (to mitigate the effects of L2 caches) and can be given width/height parameters on the command line to open a larger window (default is still 640x480) to get a larger VRAM surface (or indeed smaller for people using 4MB cards).

usage: test width height

Would those Pegasos users who ran the original consider running the updated version here ?

Thanks :-)

The previous version reported totally unrealistic memory speeds on systems with 512K L2 cache since the test buffer was only 512K ;-)

[itix]

Results sent

[Karlos]

itix wrote:
Results sent

Cheers :-)

[Acill]

I just emailed you my results as well from my Pegasos II G4 system.

EDIT: I guess your mail didnt like it, here is the results below:

Memory/Bus bandwidth estimation (WarpOS/PPC) (c) Karl Churchill 2004

System Info

CPU : 7447/7457 (G4) [PVR:0x80020101]
CPU : 1000.000 MHz
FSB : 133.333 MHz

Fast [cache normal] allocated : 33554432 bytes at 0x214AA960
VRAM [noncacheable] allocated :  1228800 bytes at 0xE8985C00

System running MorphOS WarpUP emulation

Estimating VRAM access bandwidth

Read  VRAM (64-bit) : 28914.40 KB/s
Read  VRAM (32-bit) : 16266.58 KB/s
Read  VRAM (16-bit) : 3615.66 KB/s
Read  VRAM  (8-bit) : 1414.92 KB/s
Write VRAM (64-bit) : 220626.04 KB/s
Write VRAM (32-bit) : 131799.60 KB/s
Write VRAM (16-bit) : 34544.66 KB/s
Write VRAM  (8-bit) : 16834.74 KB/s

Estimating fast RAM access bandwidth [normal]

Read  FAST  (64-bit) : 223107.84 KB/s
Read  FAST  (32-bit) : 223039.96 KB/s
Read  FAST  (16-bit) : 222790.42 KB/s
Read  FAST   (8-bit) : 200725.95 KB/s
Write FAST  (64-bit) : 446959.69 KB/s
Write FAST  (32-bit) : 446791.03 KB/s
Write FAST  (16-bit) : 442847.01 KB/s
Write FAST   (8-bit) : 225898.79 KB/s

Estimating fast RAM [normal] to fast RAM [normal] bandwidth

FAST -> FAST (64-bit) : 177491.57 KB/s
FAST -> FAST (32-bit) : 168491.36 KB/s
FAST -> FAST (16-bit) : 153469.40 KB/s
FAST -> FAST  (8-bit) : 121654.90 KB/s

Estimating fast RAM [normal] to VRAM copy bandwidth

FAST -> VRAM (64-bit) : 159743.28 KB/s
FAST -> VRAM (32-bit) : 107766.68 KB/s
FAST -> VRAM (16-bit) : 34014.99 KB/s
FAST -> VRAM  (8-bit) : 16397.64 KB/s

Estimating VRAM to RAM [normal] copy bandwidth

VRAM -> FAST (64-bit) : 28514.57 KB/s
VRAM -> FAST (32-bit) : 16066.22 KB/s
VRAM -> FAST (16-bit) : 3614.28 KB/s
VRAM -> FAST  (8-bit) : 1414.71 KB/s

[Karlos]

Maybe you spelled my name right when you emailed me. Alas I didnt spell it right when I registered it :lol:

The G4 figures look far more realistic now :-)

[JKD]

Same systems as Acill's except I have a Voodoo3 3500 AGP as opposed to
his Radeon 7000 (?) AGP:

Memory/Bus bandwidth estimation (WarpOS/PPC) (c) Karl Churchill 2004

System Info

CPU : 7447/7457 (G4) [PVR:0x80020101]
CPU : 1000.000 MHz
FSB : 133.333 MHz

Fast [cache normal] allocated : 33554432 bytes at 0x21437720
VRAM [noncacheable] allocated :   614400 bytes at 0xEC5AA970

System running MorphOS WarpUP emulation

Estimating VRAM access bandwidth

Read  VRAM (64-bit) : 17497.09 KB/s
Read  VRAM (32-bit) : 13054.57 KB/s
Read  VRAM (16-bit) : 7258.11 KB/s
Read  VRAM  (8-bit) : 3831.75 KB/s
Write VRAM (64-bit) : 224737.84 KB/s
Write VRAM (32-bit) : 140810.40 KB/s
Write VRAM (16-bit) : 39660.33 KB/s
Write VRAM  (8-bit) : 19243.93 KB/s

Estimating fast RAM access bandwidth [normal]

Read  FAST  (64-bit) : 222906.23 KB/s
Read  FAST  (32-bit) : 222879.27 KB/s
Read  FAST  (16-bit) : 222625.05 KB/s
Read  FAST   (8-bit) : 200591.57 KB/s
Write FAST  (64-bit) : 447184.60 KB/s
Write FAST  (32-bit) : 447251.85 KB/s
Write FAST  (16-bit) : 443442.91 KB/s
Write FAST   (8-bit) : 226074.08 KB/s

Estimating fast RAM [normal] to fast RAM [normal] bandwidth

FAST -> FAST (64-bit) : 177374.39 KB/s
FAST -> FAST (32-bit) : 168457.04 KB/s
FAST -> FAST (16-bit) : 153520.09 KB/s
FAST -> FAST  (8-bit) : 121715.30 KB/s

Estimating fast RAM [normal] to VRAM copy bandwidth

FAST -> VRAM (64-bit) : 172023.31 KB/s
FAST -> VRAM (32-bit) : 119516.86 KB/s
FAST -> VRAM (16-bit) : 39370.00 KB/s
FAST -> VRAM  (8-bit) : 19291.89 KB/s

Estimating VRAM to RAM [normal] copy bandwidth

VRAM -> FAST (64-bit) : 17421.77 KB/s
VRAM -> FAST (32-bit) : 13017.88 KB/s
VRAM -> FAST (16-bit) : 7222.08 KB/s
VRAM -> FAST  (8-bit) : 3826.98 KB/s

[Karlos]

Hi,

Does anybody know if the April fix has any effect on the memory write speed for Peg1 ?

The only peg1 results I have seen so far are for a non-april version (G3 600MHz / 100MHz FSB) and I was extremely surprised that the write times were much slower than read times. Normally you'd expect similar (or better) write performance.

In fact, the write performance, 80MB/s to normal copyback memory, was only fractionally higher than a CSPPC with (604e 266MHz / 66MHz FSB) at 72MB/s, which would actually put the CSPPC in front MHz for MHz (for FSB speeds) :-?