so how does this compare with the real drive? it transfers...

so how does this compare with the real drive? it transfers a word in about 1.6us (compared to 2us for mine), so it is slightly faster for sustained data transfer with data that fits in the internal buffer. however, seek times (in the ms range) more or less cancel this out.

Like 20 Oct 2023 at 0:39 | Open on mastodon.social

26 comments

Tube🍂Time replied to Tube🍂Time

here's a DMA cycle for the real hard drive. it's using the bus more efficiently, requesting a new transfer as soon as the transfer completes. my design waits for the full DMA cycle to end before requesting a new one.

20 Oct 2023 at 0:43 | Open on mastodon.social

Tube🍂Time replied to Tube🍂Time

maybe theoretical maximum transfer rates would make the comparison easier:
My drive: ~1MB/s
stock HDD: ~1.25MB/s (neglecting seek)

if i put in a bunch of work and implemented burst mode DMA transfer, i might be able to hit 4.5MB/s.

20 Oct 2023 at 1:02 | Open on mastodon.social

Tube🍂Time replied to Tube🍂Time

no more debugging wires. the DBA-ESDI drive is now running stand-alone!

21 Oct 2023 at 2:31 | Open on mastodon.social

Darryl Ramm replied to Tube🍂Time

@tubetime Damn, that was impressive.

21 Oct 2023 at 2:32 | Open on hachyderm.io

Tube🍂Time replied to Tube🍂Time

I printed an angled carrier. the angle is because the socketed Teensy is really tall.

21 Oct 2023 at 2:33 | Open on mastodon.social

Tube🍂Time replied to Tube🍂Time

time to lay out a version of this board for other PS/2 models, like the 50Z. this is the somewhat less uncommon 72-pin DBA-ESDI form factor.

10 Nov 2023 at 23:35 | Open on mastodon.social

Tube🍂Time replied to Tube🍂Time

well that was a bit tricky, but it's more or less laid out now.

11 Nov 2023 at 1:54 | Open on mastodon.social

Mike Grusin replied to Tube🍂Time

@tubetime Nice! (And LOLing because I JUST asked a client whether they really need the big teensy or could they go with the smaller one.)

11 Nov 2023 at 3:39 | Open on mastodon.world

doragasu replied to Tube🍂Time

@tubetime Reading the silk screen of your boards is always so much fun! 😂

11 Nov 2023 at 8:19 | Open on mastodon.sdf.org

Tube🍂Time replied to Tube🍂Time

looks like there's a rare problem when reading sectors, around 1 in 100,000 reads. ugh these can be hard to track down

12 Nov 2023 at 2:17 | Open on mastodon.social

Tube🍂Time replied to Tube🍂Time

i thought it might be a timing issue caused by disabling USB debugging, but it still happens (with roughly the same statistics) with USB debugging turned on.

17 Nov 2023 at 15:20 | Open on mastodon.social

Tube🍂Time replied to Tube🍂Time

looks like some individual data bits are getting flipped. what's really odd is that it is repeatable and it seems to be the same bits at the same addresses.

17 Nov 2023 at 15:21 | Open on mastodon.social

Tube🍂Time replied to Tube🍂Time

also -- there seem to be two overlapping bugs. one causes a few bits to be corrupted for single sector reads, the other causes an entire sector's data to be shifted by several bytes.

17 Nov 2023 at 15:22 | Open on mastodon.social

/dev/rdsk/c5t1d0s2 replied to Tube🍂Time

@tubetime SD card’s cursed

17 Nov 2023 at 21:16 | Open on aus.social

Tube🍂Time replied to Tube🍂Time

huh, once I connect the logic analyzer, the problem takes much longer to reproduce; the probability of data corruption drops dramatically. interesting but it makes things much harder to debug.

17 Nov 2023 at 23:59 | Open on mastodon.social

/dev/rdsk/c5t1d0s2 replied to Tube🍂Time

@tubetime hey @azonenberg weren’t you looking for examples of probes changing behavior of the DUT?

18 Nov 2023 at 0:04 | Open on aus.social

Tube🍂Time replied to Tube🍂Time

looks like it is skipping a transfer here somehow. it transfers B66D, then it skips 6DDB and goes to DBB6. the preempt line seems to deassert right before arb/gnt goes high. that doesn't seem right.

18 Nov 2023 at 0:33 | Open on mastodon.social

Jon Dubovsky replied to Tube🍂Time

@tubetime A coincidence that DBB6 == 6DDB << 1 ? Could it be slipping a bit?

18 Nov 2023 at 0:37 | Open on mastodon.social

Tube🍂Time replied to Jon

@jond no that is just the pattern in the disk image. presumably a disk area that was never written to after being formatted and still contains this bit pattern.

18 Nov 2023 at 1:15 | Open on mastodon.social

Tube🍂Time replied to Tube🍂Time

the smoking gun: an invalid bus transfer. both rd and wr are asserted at the same time. before that we see the missing data on the bus: 6DDB. so the card thinks it is in a DMA cycle even though it is not, and it's putting data on the bus when it shouldn't be!