tim@cit-vax.Caltech.Edu (Timothy L. Kay) (10/25/87)
It seems to me that when I format a double-density floppy on a high- density system, the first half of the tracks always format OK. This is because the actual cylinders are longer on the outside of the disk compared to the inside of the disk, so the media density is greater. After about 60-70 tracks, lots of bad sectors start showing up. Because of this, I am reluctant to use low-density floppies formatted as high- density floppies. Here is a new idea which is mine. My idea, which now follows is my idea. :-) How about defining a new floppy format where the outside of the disk is formatted high-density, and the inside is formatted low-density. This way, you should be able to comfortably and reliably pack about 800k-1000k of data on the dirt-cheap floppies. All you would need to do is rewrite the DOS block device driver to recognize this special format automatically. Any comments? Is there anybody out there sufficiently familiar with all the necessary stuff to do what I recommend? Tim
dhesi@bsu-cs.UUCP (Rahul Dhesi) (10/27/87)
In article <4289@cit-vax.Caltech.Edu> tim@cit-vax.UUCP (Timothy L. Kay) writes: >Here is a new idea which is mine. My idea, which now follows is my idea. :-) > >How about defining a new floppy format where the outside of the disk is >formatted high-density, and the inside is formatted low-density. The Macintosh already does this on 3.5-inch disks. Except that the disk head doesn't see the change in density, because the drive rotates at different speeds depending on which cylinder is being accessed. -- Rahul Dhesi UUCP: <backbones>!{iuvax,pur-ee,ter.lcohowe'rp!sun!d
Dion_L_Johnson@cup.portal.com.UUCP (10/29/87)
I think the Victor PC used a type of floppy drive that had a variable rotation speed, in order to accomplish more uniform (and therefore greater) density recording on diskettes. In response to Timothy Kay's query: Here is a new idea which is mine. My idea, which now follows is my idea. :-) How about defining a new floppy format where the outside of the disk is formatted high-density, and the inside is formatted low-density. This way, you should be able to comfortably and reliably pack about 800k-1000k of data on the dirt-cheap floppies. All you would need to do is rewrite the DOS block device driver to recognize this special format automatically.
davidsen@steinmetz.steinmetz.UUCP (William E. Davidsen Jr) (10/31/87)
In article <1155@cup.portal.com> Dion_L_Johnson@cup.portal.com writes: |I think the Victor PC used a type of floppy drive that had a variable |rotation speed, in order to accomplish more uniform (and therefore |greater) density recording on diskettes. True. The Victor kept the linear speed of the media constant relative to the head. This allowed the data on the outer tracks to be packed as densely as the data on the inner tracks. As I recall the number of sectors per track was variable. [ ... ] |Here is a new idea which is mine. My idea, which now follows is my idea. :-) | |How about defining a new floppy format where the outside of the disk is |formatted high-density, and the inside is formatted low-density. This |way, you should be able to comfortably and reliably pack about 800k-1000k |of data on the dirt-cheap floppies. All you would need to do is rewrite |the DOS block device driver to recognize this special format automatically. High density diskettes have different coatings. The material used has a higher hysterysis (it is harder to magnetize and demagnetize). I doubt that a diskette which used one oxide on the inner tracks and one on the outer would be dirt cheap. Victor got the only free lunch there was, they pushed the outer tracks as hard as the inner. -- bill davidsen (wedu@ge-crd.arpa) {uunet | philabs | seismo}!steinmetz!crdos1!davidsen "Stupidity, like virtue, is its own reward" -me
tim@cit-vax.Caltech.Edu (Timothy L. Kay) (11/02/87)
In article davidsen@crdos1.UUCP (bill davidsen) writes: >|How about defining a new floppy format where the outside of the disk is >|formatted high-density, and the inside is formatted low-density. This >|way, you should be able to comfortably and reliably pack about 800k-1000k >|of data on the dirt-cheap floppies. All you would need to do is rewrite >|the DOS block device driver to recognize this special format automatically. > >High density diskettes have different coatings. The material used has a >higher hysterysis (it is harder to magnetize and demagnetize). I doubt >that a diskette which used one oxide on the inner tracks and one on the >outer would be dirt cheap. Victor got the only free lunch there was, >they pushed the outer tracks as hard as the inner. Bill, I think you've missed the point. Your comment about different coatings on the outside and inside of the disks doesn't make any sense. Consider an existing high-density floppy. The medium has to have the higher hysterysis so that it can provide a higher *linear density* on the inner tracks. However, because the outer tracks are moving more quickly, their linear density is much higher than necessary. Now consider a double-density floppy on a high-density machine. The effective linear density of the outer tracks is higher than that of a high density floppy's inner tracks. I was proposing that we do just like they are doing on the Victor, but rather than increase the speed of the drive on the outer tracks, we slow down the rate of the bits being recorded when we get to the inner tracks. (We do this by switching the drive to and from high-density mode, which is something the Victor machine couldn't do.) This way, we could get more than a megabyte reliably on floppies that cost $.25 each. Have I made myself clearer? Is anybody interested in having such a device driver? Tim
davidsen@steinmetz.steinmetz.UUCP (William E. Davidsen Jr) (11/05/87)
In article <4386@cit-vax.Caltech.Edu> tim@cit-vax.UUCP (Timothy L. Kay) writes: |In article davidsen@crdos1.UUCP (bill davidsen) writes: |>|How about defining a new floppy format where the outside of the disk is |>| [ many things ] |>High density diskettes have different coatings. The material used has a |>higher hysterysis (it is harder to magnetize and demagnetize). I doubt |>that a diskette which used one oxide on the inner tracks and one on the |>outer would be dirt cheap. Victor got the only free lunch there was, |>they pushed the outer tracks as hard as the inner. | |Bill, I think you've missed the point. | |Your comment about different coatings on the outside and inside of |the disks doesn't make any sense. Consider an existing high-density [ I didn't use ;-) on that one because I assumed that everyone would understand that I was kidding. Obviously you didn't ] |floppy. The medium has to have the higher hysterysis so that it can |provide a higher *linear density* on the inner tracks. However, because |the outer tracks are moving more quickly, their linear density is |much higher than necessary. [ this is exactly what Victor did, they put more data on the outer tracks than the inner by slowing the disk rotational speed and thereby putting the bit (flux changes) closer together, as they would be on the inner tracks. I'm glad you agree with me, but I wish you wouldn't phrase it as though you were correcting what I said. You can't use high density writes to a regular disk. You need a high density disk for high density writes because the coating is different. You *can* slow the speed so that the flux changes are closer together at normal density. ] -- bill davidsen (wedu@ge-crd.arpa) {uunet | philabs | seismo}!steinmetz!crdos1!davidsen "Stupidity, like virtue, is its own reward" -me
burton@parcvax.Xerox.COM (Philip M. Burton) (11/06/87)
At the risk of sounding like an old f---, I think the last thing the world needs is Yet Another Floppy Disk Format (YAFDF). Even in today's IBM/clone world, we already have 4 currently viable formats: 5 1/4" "standard" DS/DD 5 1/4" high-density 3 1/2" 720 KB 3 1/2" 1.44 MB Guys, that's enough!! When I worked at Shugart, the now gone disk drive company, we did a calculation that demonstrated that the maximum increase in capacity from using constant-bit density recording was 100%. In practice, the increase would be much less because of finite sector sizes, complexity of motor speed controls, etc. The 4X increase in capacity of the 5 1/4" high-density disk, at a much lower increase in cost, definitely is the way to go. Much higher capacities are available in 5 1/4" in the laboratory, which are read-compatible with today's formats. However, even four formats is too many, if you have a big population of systems or you are a software publisher. Now, for those in the audience who never knew the joys of CP/M systems, the lack of disk interchange was one of the notorious problems of that environment. Every different company, and there were so many in the pre-IBM PC days, had a slightly different format. I even have a file of about 100 different 5 1/4" and 8" formats, specifying all the different parameters that uniquely defined YAFDF. The practical effect was to kill interchangeability for all but 8" systems that could support the old SS/SD. There were, and still are, commercial programs that can read and write "foreign" formats on CP/M systems. All in the name of "progress", the customer was screwed. That's not progress, that's NIH-ism at its worst. (Maybe not, RS232C may be worse.) Flames to /dev/dysan and /dev/verbatim. -- Philip Burton burton.osbunorth@parcvax.COM Xerox Corp. .... !hplabs!parcvax!burton 408 737 4635 ... the usual disclaimers apply ...
ugfailau@sunybcs.uucp (Fai Lau) (11/06/87)
In article <7779@steinmetz.steinmetz.UUCP> davidsen@crdos1.UUCP (bill davidsen) writes: > >[ this is exactly what Victor did, they put more data on the outer >tracks than the inner by slowing the disk rotational speed and thereby >putting the bit (flux changes) closer together, as they would be on the >inner tracks. I'm glad you agree with me, but I wish you wouldn't >phrase it as though you were correcting what I said. > Does anyone know what the exact format of the 3.5 diskettes is? When you slow the disk down, you'd have more cluster per track. It seems that there are still disk space wasted because you cannot put an additional cluster (or sector, depending on what the formatting method is) on the next track out until the track is "out" enough and the speed is slowed down enough to allow enough space for it. What you'd get is a "staircase" effect. Am I right? >You can't use high density writes to a regular disk. You need a high >density disk for high density writes because the coating is different. >You *can* slow the speed so that the flux changes are closer together at >normal density. ] > The difference in the coating is the size of the magnetic particles and their density. Otherwise the coatings are pretty much the same for normal disk and high-dens disk. They might also change the coating for the particles (not the disk coating which *contains* the particle) for the difference in roatation speed, if there is any. And off cause, you can always write a high density disk with a normal drive.. Fai Lau SUNY at Buffalo (The Arctic Wonderland) UUCP: ..{mit-ems|watmath|rocksanne}!sunybcs!ugfailau BI: ugfailau@sunybcs
farren@gethen.UUCP (Michael J. Farren) (11/06/87)
In article <6343@sunybcs.UUCP> ugfailau@joey.UUCP (Fai Lau) writes: > The difference in the coating is the size of the magnetic >particles and their density. Otherwise the coatings are pretty >much the same for normal disk and high-dens disk. They might >also change the coating for the particles (not the disk coating >which *contains* the particle) for the difference in >roatation speed, if there is any. And off cause, you can >always write a high density disk with a normal drive.. No you can't. Not reliably. Regardless of whether or not the coating on a HD diskette is physically similar to the coating on a normal drive, the media has a higher coercivity, thus requiring a different head assembly, one which can produce the levels of flux required to change the state of the media. Normal drives don't have it. My intuition is that the coatings of a normal diskette and a HD diskette are, in fact, quite different. I don't have proof of this, but they look different to my eye. -- ---------------- Michael J. Farren "... if the church put in half the time on covetousness unisoft!gethen!farren that it does on lust, this would be a better world ..." gethen!farren@lll-winken.arpa Garrison Keillor, "Lake Wobegon Day!po
ugfailau@sunybcs.uucp (Fai Lau) (11/08/87)
In article <300@gethen.UUCP> farren@gethen.UUCP (Michael J. Farren) writes: ... >>roatation speed, if there is any. And off cause, you can >>always write a high density disk with a normal drive.. > >No you can't. Not reliably. Regardless of whether or not the coating >on a HD diskette is physically similar to the coating on a normal >drive, the media has a higher coercivity, thus requiring a different >head assembly, one which can produce the levels of flux required to >change the state of the media. Normal drives don't have it. > Thanx for the comment. You're quite right. However, it only applies in the situation where the coercivity of the media is indeed higher than that of its regular density counterpart. In order to make a magnetic media HD suitable, there are two things that should be done. One is to reduce the size of the magnetic particles. Second is to increase its coercivity. When the size of the magnetic particles are reduced, a narrower track of a higher frequency signal can be recorded on the media, while the maximum output level of the media increases. However, since the increase of this output level due to the smaller particle size and more densly packed particles generally does not adequately compensate the decrease in the output level resulted from the increase of signal frequency and narrowing of the recording track (the higher frequency the signal you write on a magnetic media, the lower it's output would be. Except where the frequency is *very* low), soemthing is usually done to the coercivity of the magnetic material to increase its output level (and of course, to make it more stable). For a 5.25 HD disk instead of the mini 3.5 disk, I don't know if the increase of the coercivity is warranted to generate enough output signal strength. I am not sure. If no increase is made, then the disk is indeed usable in a regular density drive. But for something like 3.5 minidisk, such increase is a must. BTW, most of the time when the density of the magnetic particles is increased on a media, their "shape" is changed, too. They're more needle like and more uniform. >My intuition is that the coatings of a normal diskette and a HD diskette >are, in fact, quite different. I don't have proof of this, but they >look different to my eye. > When a different type or chemistry of the magnetic particles is used on a HD disk, it'll look different. But a mere change in the size and shape of the particles should yield no visual difference. Fai Lau SUNY at Buffalo (The Arctic Wonderland) UUCP: ..{mit-ems|watmath|rocksanne}!sunybcs!ugfailau BI: ugfailau@sunybcs
davidsen@steinmetz.steinmetz.UUCP (William E. Davidsen Jr) (11/09/87)
In article <6343@sunybcs.UUCP> ugfailau@joey.UUCP (Fai Lau) writes: [ ... ] >which *contains* the particle) for the difference in >roatation speed, if there is any. And off cause, you can >always write a high density disk with a normal drive.. I have not been able to write high density media in my 360k drives, have others had other results? I tried Maxell and Verbatum, in a real AT, PC, a Sperry IT and a Compaq 386. Could you mention which diskette brand and disk or computer brand you used? -- bill davidsen (wedu@ge-crd.arpa) {uunet | philabs | seismo}!steinmetz!crdos1!davidsen "Stupidity, like virtue, is its own reward" -me