Page 4 of 8

Re: FrontBench: using Frontier as a benchmark (Discussion thread)

Posted: 16 Feb 2022 00:18
by exxos
Darklord wrote: 16 Feb 2022 00:10 Atari STacy (Pak 68/3 030 @ 40mhz, no FASTRAM) - 3163 frames
Well I'm confused.. That score shouldnt be possible without fastram..

Re: FrontBench: using Frontier as a benchmark (Discussion thread)

Posted: 16 Feb 2022 00:22
by Badwolf
Darklord wrote: 16 Feb 2022 00:10 I finally got around to setting this up here and ran it on 2 machines:

Atari Mega STe (stock) - 1227 frames

Atari STacy (Pak 68/3 030 @ 40mhz, no FASTRAM) - 3163 frames

BTW, watching the difference in Frontier between the Mega STe
and the Pak 68/3 equipped STacy is like night and day... :)
Thanks, Darklord. The Pak figures are great both here and Frank’s over on Atari-forum. Not sure how it’s doing that!

You do need to let me know your screen refresh rate, though. 50 or 60Hz?

BW

Re: FrontBench: using Frontier as a benchmark (Discussion thread)

Posted: 16 Feb 2022 01:18
by Darklord
exxos wrote: 16 Feb 2022 00:18
Well I'm confused.. That score shouldnt be possible without fastram..
I think we decided it's the Pak 68/3 board's level 1 and 2 caches making
the difference?

Love to have my STacy pumped up to 50mhz and FASTRAM! :)

Re: FrontBench: using Frontier as a benchmark (Discussion thread)

Posted: 16 Feb 2022 01:19
by Darklord
Badwolf wrote: 16 Feb 2022 00:22
You do need to let me know your screen refresh rate, though. 50 or 60Hz?

BW
Oh, sorry - should be 60Hz, in the US....

Re: FrontBench: using Frontier as a benchmark (Discussion thread)

Posted: 16 Feb 2022 01:22
by exxos
Darklord wrote: 16 Feb 2022 01:18 I think we decided it's the Pak 68/3 board's level 1 and 2 caches making
the difference?
Didn't even know the pak030 had L2 cache ?!

Do you have a image of it ?

Re: FrontBench: using Frontier as a benchmark (Discussion thread)

Posted: 16 Feb 2022 04:36
by Darklord
exxos wrote: 16 Feb 2022 01:22
Didn't even know the pak030 had L2 cache ?!

Do you have a image of it ?
I don't. This is what Frank Lucas said over in the FrontBench message thread at Atari Forum:

"The PAK68/3 have L1 Cache into the CPU and 32kB L2 Cache to speed up the transfer with the ST Ram ..."

HTH's...

Oh...it is listed on the Pak 68/3's jumpers - J4 SLC secondary level cache:


PAK Jumpers.jpeg

Re: FrontBench: using Frontier as a benchmark (Discussion thread)

Posted: 16 Feb 2022 05:51
by DoG
At least the PAK68/030 has 4 cache RAM chips under the ROM chips and 2 TAG ram chips. The PAK68/020 looks to be the same board but I don't know much about it.

Re: FrontBench: using Frontier as a benchmark (Discussion thread)

Posted: 16 Feb 2022 10:21
by exxos
DoG wrote: 16 Feb 2022 05:51 At least the PAK68/030 has 4 cache RAM chips under the ROM chips and 2 TAG ram chips. The PAK68/020 looks to be the same board but I don't know much about it.
I have a pak 20 but it doesn't like me :( be glad when we get the new MMU running with faster RAM. I had the MMU running along with @Icky a while back. But the GLUE was being sulky :( even so , ditch the DRAM and get some SRAM on there.. No need for caches then ;) They kinda obsolete now. The SEC booster can run at 64MHz. We just need to fix the RAM bottleneck and it will leave everything else in the dust ;)

It reminds me, I need to design a SRAM board for the H5 :lol: :roll:

Re: FrontBench: using Frontier as a benchmark (Discussion thread)

Posted: 16 Feb 2022 10:38
by rubber_jonnie
If I can find a suitable test machine, I'll see if I can test with my PAK 68/20.

Re: FrontBench: using Frontier as a benchmark (Discussion thread)

Posted: 16 Feb 2022 10:41
by exxos
rubber_jonnie wrote: 16 Feb 2022 10:38 If I can find a suitable test machine, I'll see if I can test with my PAK 68/20.
Good idea :)