Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Path: utzoo!utgpu!water!watnot!watmath!clyde!rutgers!seismo!nbires!hao!gaia!zhahai From: zhahai@gaia.UUCP Newsgroups: comp.sys.ibm.pc Subject: Re: Can 360k be written on 1.2M drive? Message-ID: <287@gaia.UUCP> Date: Tue, 24-Feb-87 02:34:26 EST Article-I.D.: gaia.287 Posted: Tue Feb 24 02:34:26 1987 Date-Received: Fri, 27-Feb-87 06:05:52 EST References: <409@thumper.UUCP> Reply-To: zhahai@gaia.UUCP (Zhahai Stewart) Organization: Gaia Corp., Boulder, CO Lines: 66 In article <409@thumper.UUCP> steve@thumper.UUCP writes: >A friend claims to have heard of software that will enable the AT 1.2 Mb >drive to write on floppy formatted for 360k. I would like to know if >anything like this exists. Even a hardware patch would be welcome. > -Steve Miller ihnp4!bellcore!thumper!steve First off, standard PC-DOS for the AT will both read and write 360K floppy diskettes on the 1.2M drives; this is automatic and needs no special setup. However.... the 1.2M drives have thinner head gaps (96 tracks per inch) than the 360K drives (48 tpi). This means that when a track is written on the 96 tpi drive and then read on the 48tpi drive, the latter will also pick up some "noise" on either side of the recorded data, thus reducing the noise margin. The result is that this is not always reliable; some people have written hundreds of disks on a 96 tpi drive (like the AT's 1.2M) and read them on 360 drives with no problems, other people have had many problems. The former can turn into the latter if either of the drives drifts out of alignment a little bit (or some other factor changes and reduces the noise margin just that little bit more to push it over into unreliablity). I have never heard of problems with writing on a 360K drive and reading on a 1.2 M drive. Theoretically, the thinner head would pick up no extra noise, but somewhat less signal, but this seems to always stay within the margins. (By the way, all of this applies to 720/800K 96 tpi drives as well as 1.2M 96 tpi drives). Many people believe that you reduce the problem by formatting the 360K disks on a 360K drive before writing via a 1.2M drive. This way at least the address marks (which are written only once, at format time) will be reliable. This is what I do, though I have no empirical evidence that it works better (I happen to have good results between my 360/1.2 M drives most of the time, but I don't count on it continuing). After all, it is guaranting that there will be wide track remnants around the edges of the new narrow track laid down by the 1.2 M drive. The answer: if you do not have handy 360K drives for your AT, go ahead and write 360K disks on the 1.2 M drives - but don't delete anything until you successfully read the disks on your 360K only drives; copy the files to a hard disk or to 360K disks if you want to be sure it will still be readable next week (ie: never user 360K disks written on a 1.2 M drive for archives). Oh, yes, 360K disks written on a 1.2 M drive seem to read well on another 1.2M drive (both have thin head gaps). Or for that matter, I have a 96 tpi drive on my PC (gets 720/800 K on a standard diskette, got this before the 1.2 M "HD" drives came out in this country with the AT) - this drive reads 360K disks written on the AT fine, as you would expect. With the proper software (which I have heard rumors of but have not found, sigh), I could read and write my 720/800K diskettes on the AT's 1.2M drives. (The 800K figure comes from Jformat's ability to cram 10 rather than 9 sectors per track by reducing the inter-sector gaps. Jformat and Qdrive are a format program and installable device driver pair from Tall Tree which I use on my PC to handle the 96tpi drive there. The 96 tpi drive on my PC (TEAC 55F) writes data in the same linear density as the normal 48 tpi 360K drives, and uses the same controller; the 1.2M drive in the AT packs bits more tightly to get the higher capacity, using a different controller as well as different diskette coatings to get 1.2M - but includes the ability to read/write the old linear density for compatiblity). ~z~ -- Zhahai Stewart {hao | nbires}!gaia!zhahai