exxos's Atari, Amiga & retro forum

DoG wrote: 13 Feb 2021 11:00

HigashiJun wrote: 13 Feb 2021 10:30...Just funny to see that Sysinfo reports the CPU speed as 10 Mhz...

Sysinfo don't display correct CPU speed. I don't know how it calculated it. It always seems off if you have an accelerator running.

Yes, and according to Sysinfo, we are in 2028...

Back to the future for free !

Cheers.

exxos wrote: 13 Feb 2021 10:55

HigashiJun wrote: 13 Feb 2021 10:30 As for now, I did not revert to originals SILs and still have the 2.2 K ones soldered in.

You should remove the MB ones as you will have 1.1K pullups in total , I found that can cause DMA issues on some machines.

HigashiJun wrote: 13 Feb 2021 10:30 No need either to use the 22uf capacitor in case the machine refuses to boot.

You still need it, or recap the reset circuit...

HigashiJun wrote: 13 Feb 2021 10:30 Just funny to see that Sysinfo reports the CPU speed as 10 Mhz...

Not surprised, effective speed, like 20% I guess overall. But ROM will work at much higher speeds, so desktop and apps will fly!


:bravo: :goodpost:

Sure, if I spot any glitches, I will apply these little tweaks, but as for now, no reason to do so.

I have the last DMA chip, so I might have no issues at all.


For this first test, everything works great.

I will keep you informed later on.

Thank you for this great piece of hardware Exxos !

:D

exxos wrote: 13 Feb 2021 10:55

HigashiJun wrote: 13 Feb 2021 10:30 As for now, I did not revert to originals SILs and still have the 2.2 K ones soldered in.

You should remove the MB ones as you will have 1.1K pullups in total , I found that can cause DMA issues on some machines.

Indeed, on my STFM on which I'm fitting the TF536, 2.2K pull-ups stopped the floppy from working at all and ACSI was flakey. With 4.7K it worked a lot better. I only changed them as I was getting corruption with the standard 10K on the floppy if the UltraSATAN was plugged in. Changing the DMA chip helped also. I think the UltraSATAN plays fast and loose and doesn't have open collectors on the data bus so the DMA constantly has to fight the unit trying to pull everything to 3.3V.

stephen_usher wrote: 13 Feb 2021 12:41

exxos wrote: 13 Feb 2021 10:55

You should remove the MB ones as you will have 1.1K pullups in total , I found that can cause DMA issues on some machines.

Indeed, on my STFM on which I'm fitting the TF536, 2.2K pull-ups stopped the floppy from working at all and ACSI was flakey. With 4.7K it worked a lot better. I only changed them as I was getting corruption with the standard 10K on the floppy if the UltraSATAN was plugged in. Changing the DMA chip helped also. I think the UltraSATAN plays fast and loose and doesn't have open collectors on the data bus so the DMA constantly has to fight the unit trying to pull everything to 3.3V.

You are totally right, guys...

I guess I will have some issues sooner or later and I have already ordered some 4.7K and 10K SILs, before testing further...

The "so far so good" does not really last with ATARIs. :D

The booster has 2.2K on it. So no need to replace the MB sils,just take them out.

BTW do you have other STEs to try it out on ?

Well got booster built but after removing my divide card my ste no longer boots as standard. So I've got another ste motherboard on its way so hopefully back end of next week will be able to provide some feedback

exxos wrote: 13 Feb 2021 14:26 The booster has 2.2K on it. So no need to replace the MB sils,just take them out.

BTW do you have other STEs to try it out on ?

Good to know, but as I have my PAK68/3 project pending, I will anyway need the original SILs.

Unfortunately, I only have one STE. :cry:

Right, as predicted by Exxos and stephen_usher, issues are starting to appear:

Random bombs while trying to access the Ultrasatan, freezes after disk formatting, corrupted file names on floppy disks, no disk change detection, etc.

I should receive the "original" value SILs I have ordered by after tomorrow, but I will first test again with all SILs removed (except P100).

I have de-soldered all the SILs (except P100) today and the Ultrasatan seems to work fine, but I have now issues with the floppy drive...

However, I think they are not related to the booster (because they are still present even with the booster removed) but to the IDE connector.

Indeed, if I test continuity with my multimeter, some pins are connected and some are not.

I have provisionally repaired the FDD connector pins before, and after having connected and disconnected the FDD flat cable several times, my repair is probably no longer alright.

I will then have to de-solder the FDD connector and do a full fledged repair using my PCB rivets.

My repair will take more time than expected...

What was my surprise when I de-soldered the FDD conncector: the previous owner had butchered all the vias when he installed it on the motherboard. :evil:

For the moment, all the vias are enlarged in order to insert the appropriate rivets, scars on the PCB are covered with epoxy. I also had to uncover some traces in order to make contact with the rivets.

To be continued...