foft's dfb1 build

[Badwolf]

So far one FPU fully working, though they are all detected. I remember how fussy these are though putting them into the Falcon, perhaps some pins need bending out. So I need to sit down and try that when I get time. The working FPU sometimes fails, sometimes passed on fputest, so it seems marginal. I'm thinking I might install a slower crystal, since I don't really need a super-fast FPU.

If it's marginal, you could try putting some more capacitance on the FPU socket bottom -- there are four power pins board north and board south of the FPU socket that can have a capacitor soldered easily between them and ground.

If there's another board revision, it'll have more caps there, but this is an easy bodge.

As for the ROM, I just tried again with the 10s delay flash tool and AM29F800. It flashes ok and verifies successfully with emutos. Then it doesn't boot, I just get green stripes. Perhaps I used the wrong image: I used ETOS512U.IMG from the archive (md5 sum 9b441de13eb01d5f757eb4198888c566).

Try flashing TOS404 first?

BW

[foft]

Try flashing TOS404 first?
BW

Good thought :oops: That works :D

I'll try some extra caps in the coming days, good suggestion, thanks.

[Badwolf]

Good thought :oops: That works :D

Why not give my special sauce TOS4 a go next then?

viewtopic.php?p=94751#p94751

No MAPROM/FASTRAM needed -- if it works!

BW

[Badwolf]

I'll try some extra caps in the coming days, good suggestion, thanks.

This is what I did -- I didn't have a problem with my FPUs, but will show you the pins in question. Could just as easily be a through-hole cap given the positions.

NB. ignore the bodge wires, that was experimenting with something else.




BW

[foft]

Thanks. I popped on a couple of 0.1uF caps to the 'almost functional' board, that sometimes passes. Unfortunately it didn't help, but sure they aren't hurting!

I guess I'm best to try a slower crystal and see how I get on (edit: Although since the CPLD divides the FPU clock anyway I guess I can modify the firmware to divide it by 3 or 4)

In the meantime I'll try the other boards to see if they get any further with this.

edit: Tried the fpu clock at osc/4 and osc/8 ... still no dice on this one. So probably something else up.

[Badwolf]

I guess I'm best to try a slower crystal and see how I get on (edit: Although since the CPLD divides the FPU clock anyway I guess I can modify the firmware to divide it by 3 or 4)
...
edit: Tried the fpu clock at osc/4 and osc/8 ... still no dice on this one. So probably something else up.

I've found going below half CPU speed is fraught with trouble. In the latest firmwares I slow down the CPU when it accesses the FPU to let you run the FPU more slowly, therefore.

Should be possible to run the FPU at XCPUCLK (16MHz) with a small mod to the latest firmware.

BW

[exxos]

I've found going below half CPU speed is fraught with trouble. In the latest firmwares I slow down the CPU when it accesses the FPU to let you run the FPU more slowly, therefore.

If the CPU is running twice as fast as the FPU, could it trip up with DTACK ? I don't know how long the FPU keeps DTACK low for..

[Badwolf]

If the CPU is running twice as fast as the FPU, could it trip up with DTACK ? I don't know how long the FPU keeps DTACK low for..

That was my working assumption -- hence slowing the CPU during FPUCS these days.

The manual says there's no limit on the speed differentials, but it's hard to imagine how there couldn't be.

BW