• Please review our updated Terms and Rules here

Deluxe Option Board misbehaving

NeXT

Veteran Member
Joined
Oct 22, 2008
Messages
8,096
Location
Kamloops, BC, Canada
I thought this card was working but now I'm not really sure.
It's setup as per the manual for DMA 1, Port 0268 and the drive jumpers are set to AT/COMPAQ.

CGS_10697.JPG


I can verify that the floppy controller, the cabling and the floppy drive work. When TransCopy 5.4 is run it doesn't complain about not detecting the board either. The problems when you start doing board-level activities.

TC 5.4 will bump the heads but never step beyond track 0 when reading a floppy in. It will however step outwards correctly when writing, but you're going to create an unusable floppy.
TCM 5.4 will read in a floppy and step the heads correctly but then give a CHECK DISK when writing it out, even if the floppy is known good.
Macintosh disk functions all error out immediately. No idea if reading in a disk or formatting is being done correctly as it can't verify so it tries, then fails on the write.

TC 4.7 gives a CHECK DISK as soon as you try and read a floppy in. Again, known good floppy.
PK likewise fails with an 80h error

TC 3.3 immediately stops when you try to read in.

I can boot directly off a DOS floppy and try in case it's something loaded into ram but it's all the same. It's almost like the board is bad? I don't see any configuration options so I don't know if it's just misconfigured or something.
 
I'd have to check with my DOBs, but I thought (perhaps mistakenly) that the operation of track seek and motor control was left to the primary FDC and that STEP/ and DIR/ were pretty much straight through connections.
 
/STEP goes to pin 4 on the nearby SN7406N while /DIR goes to pin 6 on the same chip but yes both pins are passive to each cable header.
Any other function I try that does NOT require the DOB works fine.

egs_34_t3s4t_y34.png
 
Last edited:
For the heck of it I soldered wires on and investigated both pins and their respective inputs (from the DOB ASIC) and outputs (to the floppy interface) on the 7406.
Using the test function in IMD I can see /DIR simply shifts high or low depending on the stepper direction and /STEP is pulsing every time the motor steps.
Switching back to TC I can see both lines are being driven by the floppy controller as the inputs from the ASIC remain silent.
 
Interesting. I've had intermittent problems with mine where it decides it must step out instead of in, making a loud buzzing hitting a stop when it should be seeking track zero. Mashing the machine's reset button, however, takes care of that.
 
Youi've got to wonder what the ASIC does with the two step control lines. Clearly, it's passing them through when not active. Maybe someone with the older Option Board can shed some light on this one.
 
Another thing that's not at all ruled out is the machine I'm running it on is WAY too fast. So far everything else I've run that's DOS based and making direct hardware calls seems to be behaving well but I have no idea if that 1ghz CPU is throwing the software timing off for the DOB. Right now I do not have anything else available to test it in that is considerably slower so I'd love to hear what others have found when running in faster-than-AT machines.
 
It could also be that your ISA implementation isn't strictly on-standard. A 1GHz system sounds like a P3 minimum.

Recall that the DOB would work fine in a 4.77Mhz XT. I've used it on a 20 MHz 386 system, but nothing faster than that.
 
Well it's been a year and I went treasure hunting and dug up a 300mhz Socket 370 Celeron which is I believe damn near the worst of the early Celeron chips.
With this slug of a chip and turning the internal and external caches off the first two versions still don't want to behave but 5.4's tools have started to work. I can format and look at the contents of a mac floppy, I can duplicate an amiga floppy (albeit with length issues) and it seemed to also be able to copy a regular unprotected floppy. (again with length sizing issues but I was able to DIR the directory)

CGS_11330.JPG

CGS_11331.JPG

CGS_11332.JPG

CGS_11333.JPG


So you were correct about the CPU being way too fast, Chuck.
 
I suspect that it's also a DMA issue. Several later P3+ cards used an ISA-PCI bridge chip which generally didn't support DMA. The 440BX/GX, to the best of my experience can be depended upon to work. Anything later (e.g. 815), forget it.
 
Anything with a southbridge newer then a PIIX4E (440BX and 430TX) is going to be dicey with 8237 DMA usage.
 
My option board works really well on a 386SX. I couldn't get it to work on a Pentium 2 I tried it on at all. I just wish more tools used the TC (IMG) file format. I also wish that TC could write standard mac DSK image files to disk.
 
My option board works really well on a 386SX. I couldn't get it to work on a Pentium 2 I tried it on at all. I just wish more tools used the TC (IMG) file format. I also wish that TC could write standard mac DSK image files to disk.
I don't understand the fascination / selling prices for these dinosaurs.
 
I think it's mainly their ability to copy and handle the more oddball formats and some levels of copy protection, but otherwise I wouldn't really pay more than the $10 I got it for in RE-PC's card bin since most modern specialty floppy adapters can do all the same and more.
 
Back
Top