TF536 on A2000

[GadgetUK164]

I have a rev 6.2 board (with factory mods applied to take it up to 6.4), another 6.0 board (with only mod done was to add the 3K3 pull downs on the 245's that buffer Zorro databus (in fact, all 3 boards have that mod applied). And my rev 4.x board with all the Commodore recommended mods done to that.

All 3 boards work with everything I have to test (A2091, A2630, GVP Combo (8MB + SCSI) ) - all 3 were also tested with other boards that I've since sold (ZRAM, Supra 8MB). And all 3 boards work with the TF534 too.

My rev 6.4 has the 8375 fitted. The 6.0 has an 8372A. The 4.x has an 8372A as well (but I've tested with 8371 on both the boards that have an 8372A).

[alenppc]

A new report came in from Amibay showing the board working on the A2000 in the CPU slot...

http://www.amibay.com/showthread.php?11 ... post989283

So it's seemingly random.

[8 Bit Dreams]

Seems 4.4 is the winner,
in both cases TF536 works on this revision

[GadgetUK164]

Seems 4.4 is the winner,
in both cases TF536 works on this revision

I believe my rev 4 is up to 4.4 now though - I went through various sources on the required mods and pretty sure I applied them all. I've ordered another CPU slot adapter, will try that to see if it behaves any different with another.

[8 Bit Dreams]

On the photo - there are no buffer PullUps for Zorro...
Not sure about other mods... would be nice to see the area around 68K CPU socket...

[matt020]

Yet Stephen did all his A2000 testing on a Rev 6.x board from what I see in the videos... explain that one?

[GadgetUK164]

On the photo - there are no buffer PullUps for Zorro...
Not sure about other mods... would be nice to see the area around 68K CPU socket...

I thought they were pull downs?

[alenppc]

Yet Stephen did all his A2000 testing on a Rev 6.x board from what I see in the videos... explain that one?

We had reports of the 536 working on some rev6s and NOT working on some rev 4s. So it’s really random it seems.

[terriblefire]

Put the rev2 on my A2000. What i observe.

1. Diagrom gives a purple screen.
2. no serial output except <>
3. On the LA DiagROM executes code from ROM.. but no output
4. On the LA DiagROM walks through the ram check sequence without issue. But no output at all.

Perplexed.

[terriblefire]

I've had a look and there is a minor change on the rev2 that could cause this but why it would only happen on the A2000 is not clear yet.

I suspect the Its the data bus buffers.

Basically on the rev1 these were auto sensing directional buffers... On the rev2 the direction is driven off the RW pin and they are 5v instead of 3.3v.

Its possible that writes arent managing to get that direction buffer all the way down if there is some kind of hard pull up or something else on the A2000 is driving that line. It would explain all the behaviours because code is reading from RAM and ROM fine but not managing to write even to the serial port which is highly unusual.

Unlikely but its all i can think of right now