DMA port "update" ideas..

[Icky]

I thought Icky got floppy and DMA working without a bodge?

I did but I also have a damaged DMA and WD1772.

The DMA port gets "delayed" by one clock cycle after FDCS goes high. This makes sense as the 1772 can hold the bus for that time. So if the DMA port was active, it could actually short out the 1772 bus to the DMA bus (same bus different DMA devices)

Based on chaos down the DMA / WD1772 path having the buffers should protect from frying any more chips.

[exxos]

I thought Icky got floppy and DMA working without a bodge?

Yes... Icky managed to get a blown DMA & 1772, this could happen due to the shorting of the DMA & 1772 databus.. so this fix will rule out and solve that issue. This will solve a issue which will likely bite someone in the ass at some point down the line anyway.. Atari used the buffering on each machine after the STFM.. So they clearly thought and knew of some problems there..

[troed]

Ah, now I get it .. then yes .. I would also like a patch board ..

[Icky]

Yep - I am not using HDD now until I get this buffer board in place

[exxos]

I had a small thought about the DMA HDD buffers.. I think its wise to add them, as its possible the HDD logic could be very slow to "let go" of the bus despite what the DMA is thinking is happening (similar problem to slow 1772 releasing the bus) so will definitely be added on next motherboard revision.

[PaulJ]

I think Icky told me he had his running fine without any changes. I we should ask him.

[PaulJ]

This board is as solid as the normal 1040 and I like the presentation better. Hopefully all the boards work as well!

I agree PaulJ. Since solving all my issues I have had the board running all the time. The only time it’s crashed is due to dodgy software I have tried to run on it.

Here's what Icky said when I got the DMA running. I have to say I've had no trouble with mine and I've been pounding on it a couple days. Exxos have you gotten to the point you can try yours?

[Icky]

I agree PaulJ. Since solving all my issues I have had the board running all the time. The only time it’s crashed is due to dodgy software I have tried to run on it.

Here's what Icky said when I got the DMA running. I have to say I've had no trouble with mine and I've been pounding on it a couple days. Exxos have you gotten to the point you can try yours?

I agree the board is very solid under use

However from a DMA perspective I feel safer with buffers :) in light of the chaos I had with the NASTY DMA.

[PaulJ]

Here's what Icky said when I got the DMA running. I have to say I've had no trouble with mine and I've been pounding on it a couple days. Exxos have you gotten to the point you can try yours?

I agree the board is very solid under use

However from a DMA perspective I feel safer with buffers :) in light of the chaos I had with the NASTY DMA.

Good morning Icky. If its questionable I totally agree however I haven't had any issues. Board real-estate is precious and we shouldn't add fixes that are just in case. When all the boards are built and running we will have a better sample of operation reliably. You said yours is running reliably. What did you have to do to achieve the reliable operation, I know you were using a various different chips?

[exxos]

. Exxos have you gotten to the point you can try yours?

I do need to try a HDD but not at home until Monday now.

The fixes are in part "just in case" but we know of issues now and they should be fixed. Let's do it right and not start down the path of the STE buggy DMA chaos ...

I have a imp chip set, that goes apesh*t on the ste, so be interesting if it works on my stf. The floppy does so far.