Media test density
When performing a media test, how does one specify the media density? I wouldn't think a one-size-fits all test would work reliably across 5.25" low density 300rpm, low density 360rpm, and high density 360rpm.
Whatever density is selected in the pull-down menu is what is used for the media test.
Interesting. For some reason the latest versions (at least 1.95 and 1.94) show completely random (not at all in the same place each time) track failures testing 5.25" disks that test fine with the earlier versions, and oddly switching density seems to make no difference at all.

I normally test 5.25" double density disks in a 1.2mb drive using low density mode and 80 tracks (AKA quad density and decent double density disks should pass that test).

At least versions 1.91 and earlier seem to still work properly, and disks that previously tested good still test good.

I did notice that 1.95/1.94 does not seem to give random errors when using a 360k floppy drive or a 1.44mb floppy drive. Only with my 1.2mb drives. And yes, I did try another 1.2mb drive and double checked the functionality of the current 1.2mb drive just to be sure.

Anyway for the moment I will just use 1.91 to get things done. If there was a 1.92 or 1.93 I didn't have those handy to test with.

This is why I don't like "upgrading" software. Tongue
The earlier versions used the exact same microseconds for every single track.  That resulted in a problem for me when hand testing each SuperCard Pro board because if the writing didn't work, the test would pass because of a previous write.  I changed the microseconds used for the test so it changes randomly for each track/side to prevent this problem from happening.  The window for failures was also narrowed a bit.

Take a look at a track that fails using the editor/analyzer.  Read the track that fails and look at the flux display.  It should be pretty obvious why it failed.  I have no issues here with the 1.2MB drives that I use - they all pass the media tests just fine with standard DSDD disks.  Most people use 1.2MB drives, and I have not had anyone else report this.
It's not quite so obvious to me. The tracks look good to me in the analyzer except for one invalid bit that occurs exactly 1/4th the way in to every track.

It took me a minute to figure out why this was, but I see that the media tester writes tracks that are longer than the actual track. Obviously to ensure there are no gaps on slower drives. At 300RPM it seems to overwrite about 1/16 of the first track, but on 360RPM (what I'm using) it overwrites about 1/4th. So I guess that part is normal.

Anyway, here is a dump of a media-tested disk.

I used a better quality (labeled 3M) disk just to be extra sure. Like the other disks it tests fine elsewhere.

I clicked "ignore" on each track so it would not re-write. It reported the following "bad" tracks and heads:
1 1
6 1
13 0
15 1
21 0
27 1
29 0
31 0
34 0
34 1
37 0
38 0
42 0
52 0
53 0
53 1
54 0
61 0
69 1
72 0
72 1
77 0
Viewing directly they look more or less normal to me.
But each time I test it will be completely random different tracks.

The 1.2mb drives in question are Teac FD-55GFR drives and the errors are reported at both 300RPM and 360RPM settings.
It could be that your power supply is flakey and when the write is turned off it smears the last bitcell written.  I am currently out of town, but I will be back tomorrow and check out your file.  Are ythey cleaning the heads frequently while using these old disks?  That is not an option.
It is powered by the PC's power supply, which is a fairly hefty supply. That's an interesting idea though, so I tried it with an external power brick. Same deal. Also tried it with a different computer. Same results. That also rules out some suspicions I had about the USB ports.

Of course dirty heads were the first thing I thought of. I've been at it multiple times with my cleaner disk, as well as gave it a q-tip rub down for good measure. I also verified that it worked normally when connected to a PC FDC - it formats/reads/writes perfectly, and tried another FD-55GFR from a different system, with the same results. The SCP seems to read disk images fine.

It does occur to me that whatever this issue is could be related to why I can't properly write write-spliced disks. Until just recently I have mainly just been reading disk images.

Would exact drive speed have anything to do with it? I did notice the drive actually reports around 361RPM, but there is no way to adjust that on this drive model.
Well, there is something definitely wrong with your drive when it comes to writing.  Take a look with the editor/analyzer at the tracks/heads that fail.  See attached - which is track 1, head 1.   The write splice is in fact being smeared.  I have never seen this before with any drive.

Very interesting. So that isn't supposed to be there. Since both Teac drives did this, it must be something in this specific revision. Both are the "149-U" revision.

I can't think of a normal use case from back in the day where drive makers would have taken this behavior in to consideration. So, logically this behavior could vary from drive to drive.
It's not a problem when a smear occurs between sectors or in the track gap. Maybe I should change the test program so that it is more laxed. I use 149-U drives though, and they work fine.

Users browsing this thread: 1 Guest(s)