exxos's Atari, Amiga & retro forum

exxos wrote: 30 Apr 2021 18:13 It does get to desktop at 8MHz. So must be something tripping up somewhere at the higher speed :shrug: I have had a 68000 running at 40MHz on the STE before now, but not 50Mhz.

Can the PLCC socket give you enough power, you think?

BW

I think something screwy is going on with ST-RAM timing... I did manage to run a few tests in GB6 just.. but it does not seem to matter where I latch DTACK something screws up. So the STE DTACK system must be different than what the STF is doing.. but not sure exactly what..

EDIT:

hmmm.. I think the problem could be my state machine for DTACK.. I copied the routines from my SEC BOOSTER code (64Mhz) , where I had to do some funky stuff to get things to behave... So I have simplified the STE version and its just run all tests in ST-RAM in GB6.. so will try fast-ram next...

TOS206 doesn't like running in alt-ram.. The floppy goes nuts..

EMUTOS runs fine, but scores are lower than the H4 :shrug: Though I think if I remember rightly, the H4 was running EMUTOS from a PRG.. that makes any difference, I need to find that version again and retry it..


STE


H4


EDIT:

Oddly when running EMUTOS from PRG (this one https://www.exxosforum.co.uk/forum/viewt ... 260#p61587) the floppy stops working afterwards :shrug: And yet EMUTOS seems fine running from ROM.

:dizzy:

Is pretty bonkers overall, I mean TOS206 works perfectly fine on a H4, but on a STE, the floppy malfunctions without the cache disabled ( I only tried the hardware jumper to disable). But AFAIK, caches are disabled for anything other than fast-ram anyway :shrug:

Something seems amiss with those benchmarks.

The STE ET one shows AES 1.4 and the H4 ET one show AES 3.2. Might be co-incidence but it's all the drawing routines that are slow.


As for the cache, I'm not sure what the 536 does, but in theory there's no reason cache can't work on the ST-RAM too.

Here's an example of my card working normally (a word or by read retrieves a longword every time and caches it all):-



And here with CIIN tied to GND -- presumably the same as your jumper (a word or byte read only uses one cycle):-




It could be if you've got some DTACK confusion, slowing down AS/DTACK by disabling cache (it'll only make one bus cycle if it only needs a word or byte rather than retrieve a longword every time and cache it) is enough to tip the balance?

BW

Badwolf wrote: 01 May 2021 11:40 Something seems amiss with those benchmarks.

The STE ET one shows AES 1.4 and the H4 ET one show AES 3.2. Might be co-incidence but it's all the drawing routines that are slow.

I might have been running EMUTOS on one benchmark and not the other. Theres that many OS versions on the go, i've lost track. I need to retest but not at home until Monday now.

TOS206 works fine on the H4, just not on the STE :shrug: Though TOS only goes nuts when running from fast-ram.

A interesting twist...

exxos wrote: 02 May 2021 21:39 A interesting twist...


IMG_6484.JPG

Cool. I didn't know an unpatched TOS 2.06 did auto-detecting. :-o

BW

Badwolf wrote: 03 May 2021 12:50 Cool. I didn't know an unpatched TOS 2.06 did auto-detecting. :-o

@Icky is busy patching it ;)

Its odd as I have been unable to get patched TOS206 to run in fast ram since the initial test. I've tried all sorts of variations as well . I've sent @Icky a built board to investigate also.

Meanwhile I have started a webpage on latest builds / firmware etc https://www.exxosforum.co.uk/atari/last/TF536/index.htm