At 05:28 PM 5/15/2007, you wrote: I found that I have to cover the hole or I get BDOS erros about missing sectors. I figured it wouldn't look for the hole... A freshly formatted disk gives me this: A>STAT B: BYTES REMAINING ON B: 69K A> 69k for a 1.4MB disk! I was ripped off! : ( It is running CP/M 1.4. I have the sources if anyone is interested. This is what it says when it boots up: TARBELL 24K CPM V1.4 OF 7-20-79 2SIO MINIFLOPPY VERSION. HOW MANY DISKS? 2 A> Were single sided single density 5.25" disks ~70k? The 3.5" drive does sound different than it usually does. : )

On Sunday 03 June 2007 10:47, Curtis H. Wilbar Jr. wrote: Doesn't this depend on the write current supplied by the drive electronics, rather than the format used? -- Member of the toughest, meanest, deadliest, most unrelenting -- and ablest -- form of life in this section of space, ?a critter that can be killed but can't be tamed. ?--Robert A. Heinlein, "The Puppet Masters" - Information is more dangerous than cannon to a society ruled by lies. --James M Dakin

On 3 Jun 2007 at 2:38, Grant Stockly wrote: Doesn't sound too far off from right. Consider that the Tarbell's probably using 35 cylinders on one side at the 250KHz clock single density. Let's run some numbers for 128 byte sectors, of which we can safely put 16 of them to the track: 35 * 128 * 16 = 71680 bytes. Now, if we take a 1K = 1024 bytes, that's 70K. Reserve a track or two for a boot track... You'd be better off calling your 3.5" drive an 8" unit to get 250K (SS) on a disk. Okay, you're going to say, but what about the deal with the 1.44MB drive being 300 RPM and not 360? Won't that cause a problem. Yes, but--you can fix the problem--several ways. About a year or two ago, I wrote to Herb Johnson on his retrocomputing site about how to modify some models of Teac FD-235HF drives to run at 360 RPM. You see, for many years, a lot of Japanese equipment requires a 360 RPM drive and many manufacturers have a way to accomplish this--even if they don't document it particularly well. I believe that FD-235HG's bring the speed select out to the edge connector, but they're harder to find on this side of the pond. Almost all NEC 3.5" HD drives can be modified for 360 RPM use. The other alternative is to swap the crystal on your Tarbell controller for something that's a bit slower to make the drive look as if it's running at 360 RPM. Say, for sake of argument that the crystal is 8 MHz. By substituting a crystal for (8*(300/360)) = 6.6666...7MHz, things will work as expected, albeit somewhat slower, but still faster than the 5.25" mode. You could also modify the formatting software to lay out a longer track (i.e. increase the number of end-of-track "run out" bytes) or even format the diskettes using something like 22Disk, which doesn't care if the physical track is too long. You could also modify the formatting software and CBIOS to handle 80 track 16-sector disks. Not as good as simulating 8" drives, but better than 60K. You could also modify it to use larger sector sizes. I don't know if the Tarbell controller handles double-sided drives, but that's another possibility. Early 5.25" drives were really pretty wimpy things compared to the 8" heavy iron. :) Cheers, Chuck

On Sun, 3 Jun 2007, Roy J. Tellason wrote: The MFM/Double Density. The single density was 40 track * 10 sectors per track * 256 bytes per sector. (100K) Allowing for the possibility of using 35, instead of 40 tracks (SA400), and 128 byte sectors (due to the "overhead"/gaps, could usually only fit 18 per track) would give 80640 bytes (78.75K), although that could also be obtained by 9 * 256. The single sided "360K" formats were, indeed 160K (DOS 1.xx) and 180K (DOS 2.xx) (40 * 512 * 8, 40 * 512 * 9) But, "half" is inaccurate. The name "Single Density" for FM did not exist until AFTER the name "Double Density" came about. Just as "The Great War" was not known as "World War I" until "World War II" began to be discussed. "Double Density" was a MARKETING name, produced by those who feared the dissemination of the actual information about it more than cannon. (Good sig quote!) FM provided 1 bit of info for every two pulses; hence, some recording engineers would prefer to see it called "Half Density". MFM permitted the omission of some, but not all of the clock pulses, thereby permitting squeezing twice the raw data transfer rate, and getting ALMOST, but definitely not quite, twice the info in the same space. Although you could get 10 * 256 on FM, you could only get about 18 256 byte sectors with MFM, and putting 10 512 byte sectors in MFM required seriously compromising the size of the intersector gaps to squeeze them in. -- Grumpy Ol' Fred cisin at xenosoft.com

The type hole on a 1.4M disk. No, but I did find a program: TARBELL MINI-FLOPPY DIAGNOSTIC OF 8-1-80 STANDARD VERSION 1.8 40 TRACKS 18 SECTORS What I would really like to do is find some way of connecting a 3.5" disk drive to the MITS hard sectored controller card set. This would give me the 300 or so K and allow me to run MITS basic off of the disk. Do you know anything about how to do that? Is it possible? Maybe with a PLD to deal with soft/hard sectored stuff? I have not investigated anything yet so I am not sure what it would take. If its possible I'm sure I could get it done. Grant

On Sun, 3 Jun 2007, Grant Stockly wrote: Can you modify the 3.5" drive to give pulses at various additional points in its rotation? I think that it would be easier to stick with the soft-sector controller. First, get it working with 8" (about 250K) Then put in a 5.25" 1.2M drive and lie to the machine and say it's 8" (1.2M was designed for doing that) Then put in a 3.5" drive and continue the lie. To minimize the lie, use a 3.5" that can be jumpered for 360 RPM OK, you've got over 200K. AND, all of the "standard" CP/M 8"SSSD stuff will match up. Will the controller support double sided? or can it be modified for it? If so, you can get over 400K. Now put a 300RPM drive in. Modify the BIOS to support 80 track 5.25" and tell the machine that that is what it has. The virtual 8" will have over 400K, and the virtual 5.25" 80 track will have between 300 and 400K depending on format choices. All without any significant hardware issues. -- Grumpy Ol' Fred cisin at xenosoft.com

At 03:19 PM 6/3/2007, you wrote: Getting a drive working with MITS software? I have a lot of MITS software on 8" hard sectored disks and I would like to move that onto other disks without having to modify boot code. One example is MITS Music Basic, or the Music Operating System. I found someone who has it and it is in the "FedEx" right now. : ) MITS was working on a 4 voice music card and had a special version of basic written for it. There is even a version of Star Trek with sound effects! I want to be able to preserve this stuff as historically accurate as possible. It would be MUCH easier to use a flash memory chip and emulate the entire disk subsystem at the I/O port level. I can't seem to find any MMC, SD, or compact flash cards in the 1-4MB range dirt cheap. I thought that with flash memory prices being low I could find some. I want to retain the physical piece per disk feel and not have all the disks in one memory image. So this means I have to have a cheap media source.