• Please review our updated Terms and Rules here

5150 Cannot format floppies

seaken

Experienced Member
Joined
Jun 20, 2016
Messages
293
Location
Shokan, New York
I just want to say that I learn a lot reading stuff like this. My guess was that is may have been the overclocking but it was just a guess. You guys have good skills and it's a great learning experience for me. Thank you for sharing your troubleshooting @maxtherabbit .

Seaken
 

maxtherabbit

Veteran Member
Joined
Apr 23, 2019
Messages
1,981
Location
VA, USA
I still can't think of a good reason why "verify" would fail when in turbo mode, but no other FDC command. Guess we'll just have to chalk it up to the overclocking game
 

maxtherabbit

Veteran Member
Joined
Apr 23, 2019
Messages
1,981
Location
VA, USA
I would test with one that is much newer and maybe 16-bit (the floppy port will still work when put into an 8-bit slot). Since the PC/XT only has one clock for both the CPU and the bus, you are running the bus with 7.x MHz as well, but PC/XT cards are made to run at 4.77 MHz. The floppy controller is probably acting up because of that. Note that DOS is VERY picky about track 0, which is most likely why IMD works but DOS format does not. Or maybe IMD is even able to run the floppy disk controller at the correct speed, as it accesses it on a lower hardware level.
Well I tried it - could scarcely believe my eyes but switching to a modern FDC fixed it. Can format disks with turbo on.

So all three of my discreet 765 based cards (2 NEC fab one SMC fab) can handle reads and writes just fine at the faster bus speed, but not verify. Bizarre
 
Last edited:

Chuck(G)

25k Member
Joined
Jan 11, 2007
Messages
40,851
Location
Pacific Northwest, USA
Depends. If it was from a foreign (i.e. not PC) that wrote the floppy without the Index Address Mark field, the first sector after the index couldn't be read. The issue was that the non-A 8272/765 goes "blind" for a whopping 1 millisecond after the index, which is enough to miss some ID address marks. It was reduced in the 8272A/765A. But that just moved the problem to not being able to read 10 sector alien formats.
 

Chuck(G)

25k Member
Joined
Jan 11, 2007
Messages
40,851
Location
Pacific Northwest, USA
Well, I'll have to dig up the schematic for that one; but only if you're willing to go at it with a soldering iron. Given that it's a bog-standard PC controller, I don't know if it's worth it to you.
 

Timo W.

Veteran Member
Joined
Nov 25, 2014
Messages
1,639
Location
Germany
Probably just an issue with the recovery time. It can't be much else that it not being able to switch fast enough from a write operation to a read operation when being driven with 7 MHz.
 

Chuck(G)

25k Member
Joined
Jan 11, 2007
Messages
40,851
Location
Pacific Northwest, USA
That's an interesting thought, but the 765 has 500 msec idle after index. The usual procedure is to write the track, then use a multi-sector verify on the next rev. Should not be an issue, but it's worth investigating.

Question that I'm not clear on is does the FDC format correctly at 4.77MHz?
 

SomeGuy

Veteran Member
Joined
Jan 2, 2013
Messages
4,254
Location
Marietta, GA
Out of curiosity, NFORMAT has the option to format without verify by specifying the "Q" parameter on the command line, will that create a usable disk?
 

maxtherabbit

Veteran Member
Joined
Apr 23, 2019
Messages
1,981
Location
VA, USA
That's an interesting thought, but the 765 has 500 msec idle after index. The usual procedure is to write the track, then use a multi-sector verify on the next rev. Should not be an issue, but it's worth investigating.

Question that I'm not clear on is does the FDC format correctly at 4.77MHz?
Yes
 
Top