consp21@bingsuns.cc.binghamton.edu (Ken Hoover) (10/21/89)
I have a Seagate ST225 in my XT (no flames on my obsolete machine); and
was wondering if getting an RLL controller for it would allow me to format
it out to 32 megs instead of the 20 I get now. My roommate and I were
puzzling over this one (he has an ST238, and we couldn't find any physical
differences), so I thought I'd submit it to the net.
- Ken
------------------------------------------------------------------------------
Ken Hoover [ consp21@bingsuns.pod.binghamton.edu | consp21@bingvaxa.BITNET ]
Resident computer jock and Mac hacker, SUNY-Binghamton Bio dept.
Senior undergraduate consultant, SUNY-Binghamton Computer Center
------------------------------------------------------------------------------unkydave@shumv1.uucp (David Bank) (10/22/89)
In article <2546@bingvaxu.cc.binghamton.edu> consp21@bingsuns.cc.binghamton.edu (Ken Hoover) writes: > > I have a Seagate ST225 in my XT (no flames on my obsolete machine); and >was wondering if getting an RLL controller for it would allow me to format >it out to 32 megs instead of the 20 I get now. My roommate and I were >puzzling over this one (he has an ST238, and we couldn't find any physical >differences), so I thought I'd submit it to the net. > > - Ken > >------------------------------------------------------------------------------ >Ken Hoover [ consp21@bingsuns.pod.binghamton.edu | consp21@bingvaxa.BITNET ] > Resident computer jock and Mac hacker, SUNY-Binghamton Bio dept. > Senior undergraduate consultant, SUNY-Binghamton Computer Center >------------------------------------------------------------------------------ DISCLAIMER: This is ** NOT ** a flame! Sure, Ken. Go for it. Hook that MFM (i.e. NOT RLL-Rated) drive up to that RLL controller, format it, and instant 50% more drive room. You CAN do it. It'll even format. If you're lucky, it might even hold data for a month before it starts developing amnesia. RLL is a MUCH more demanding format and less tolerant of errors. MFM drives are not RLL rated for a very good reason -- they can't deliver the performance RLL-controllers demand. They are very simply not engineered for the rigors of RLL usage. So...you CAN do what you want. But you had better hope and pray and sacrafice virgins to Mammon or that drive WILL crash. Not might. WILL. It is only a matter of time before you get "Disk Boot Error" or "Invalid Drive Specification" or some similar, sinister, error message. KIDDIES! Do NOT TRY THIS AT HOME!!! If a drive is not SPECIFICALLY RLL-Rated by its manufacturer, do NOT try hooking it to an RLL controller to boost storage. You are playing Russian Roulette with your data and there is a VERY high probability that you will lose everything. If you want the space THAT badly, go out and buy yourself an RLL-rated drive! Unky Dave unkydave@shumv1.ncsu.edu
rogers@SRC.Honeywell.COM (Brynn Rogers) (10/22/89)
In article <4265@ncsuvx.ncsu.edu> unkydave@shumv1.ncsu.edu (David Bank) writes: > DISCLAIMER: This is ** NOT ** a flame! > Sure, Ken. Go for it. Hook that MFM (i.e. NOT RLL-Rated) drive up to >that RLL controller, format it, and instant 50% more drive room. You >CAN do it. It'll even format. > If you're lucky, it might even hold data for a month before it starts >developing amnesia. > RLL is a MUCH more demanding format and less tolerant of errors. MFM >drives are not RLL rated for a very good reason -- they can't deliver >the performance RLL-controllers demand. They are very simply not >engineered for the rigors of RLL usage. > So...you CAN do what you want. But you had better hope and pray >and sacrafice virgins to Mammon or that drive WILL crash. Not might. >WILL. It is only a matter of time before you get "Disk Boot Error" >or "Invalid Drive Specification" or some similar, sinister, error >message. > KIDDIES! Do NOT TRY THIS AT HOME!!! > If a drive is not SPECIFICALLY RLL-Rated by its manufacturer, >do NOT try hooking it to an RLL controller to boost storage. You >are playing Russian Roulette with your data and there is a VERY high >probability that you will lose everything. ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Dave's blanket statement > If you want the space THAT badly, go out and buy yourself an >RLL-rated drive! >Unky Dave >unkydave@shumv1.ncsu.edu I think that this is a little overboard. (or Dave sells drives) One Manufacturer in particular (Maxtor) does not rate many of it's drives RLL and yet I have had one running for over a month at almost double its rated capacity. (A XT-1140 formats to 120 Meg MFM, I formatted it as a XT-1240 RLL drive and got 204 Meg) I talked to no less than six people who have been running MFM-rated Maxtors as RLL drives and not one of them had a single complaint (and they had a aggregate total time of more than ten years) As I understand it, Maxtor uses plated media, and I have heard that the main requirement for RLL is plated media. I talked with the tech support people at maxtor and they said 'the XT-1140 is not RLL rated', but in another sentence said 'Many of our customers run our drives RLL with no problem' (But I had to drag it out of them.) I don't doubt that taking a low end MFM drive and running it RLL can give problems, But I strongly object to the blanket statement that you CANNOT use a MFM drive with a RLL controller. Some MFM drives will work RLL, some won't. About the only way to be sure is to try it (But keep everthing backed up, and keep your MFM controller) After getting an informal poll of people that use your drive RLL. Brynn Brynn Rogers Honeywell S&RC rogers@src.honeywell.com home 612 874-7737
gjh@galen.acc.virginia.edu (Galen Hekhuis) (10/23/89)
In article <4265@ncsuvx.ncsu.edu} unkydave@shumv1.ncsu.edu (David Bank) writes: }In article <2546@bingvaxu.cc.binghamton.edu> consp21@bingsuns.cc.binghamton.edu (Ken Hoover) writes: }> }> I have a Seagate ST225 in my XT (no flames on my obsolete machine); and }>was wondering if getting an RLL controller for it would allow me to format A fairly common question, followed by } DISCLAIMER: This is ** NOT ** a flame! } KIDDIES! Do NOT TRY THIS AT HOME!!! Including advice like sacrificing virgins, etc. Oh poo. While I agree with the advice that it is a bit risky, and one is better off getting an RLL certified drive for the extra bytes, it works a fair percentage of the time, in my (humble) experience. As a matter of fact, I currently have more (not by many) difficulties with genuine 238R drives on RLL controllers than the 225s that I have stuck on them. It isn't like you are going to blow up if you try this, just be aware that it isn't the most reliable setup. -- Galen Hekhuis UVa Health Sciences Center Box 449, Jordan Hall Charlottesville, VA 22908 (804)982-1646 gjh@virginia.edu ...there I was, standing in front of the fan...
unkydave@shumv1.uucp (David Bank) (10/23/89)
,. In response to Brynn Roger (rogers@src.honeywell.com):
Nope. I don't sell hard drives. I merely install and maintain
them and diagnose them when they go bad.
Contrary to your assertion, plated media is NOT the sole
basic difference in RLL-rated drives. The biggest single difference
is in the use of VOICE COIL technology in the armature control
mechanism. This is what allows the mush more closely controlled
head movement demanded by RLL. The plated media simply makes
for a more reliable rusty pie plate.
Sure, I know people running MFM drives on RLL too. Its
like the car you drive off the lot and it never needs to see a
mechanic. Some RLL drives are just able to do it. You may be running
them at the edge of tolerance, but you are just on this side of
their operational capacity. I never meant to say it was IMPOSSIBLE
to run an MFM drive on RLL. Merely dangerous with your data.
No, not a blanket statement at all. As I recall, I
qualified it with "probably".
Oops. Two paragraphs back, 3rd line, "RLL" should be "MFM"
These remote connections are hell on editing.
Unky Dave
unkydave@shumv1.ncsu.edukeithe@tekgvs.LABS.TEK.COM (Keith Ericson) (10/23/89)
In article unkydave@shumv1.ncsu.edu (David Bank) writes: <In article <consp21@bingsuns.cc.binghamton.edu (Ken Hoover) writes: <> <> I have a Seagate ST225 in my XT (no flames on my obsolete machine); and <>was wondering if getting an RLL controller for it would allow me to format <> < Sure, Ken. Go for it... < < If you're lucky, it might even hold data for a month before it starts <developing amnesia. < < RLL is a MUCH more demanding format and less tolerant of errors. MFM <drives are not RLL rated for a very good reason -- they can't deliver <the performance RLL-controllers demand. They are very simply not <engineered for the rigors of RLL usage. Well, it all depends on the quality of the drive (manufacturer?). I have had absolutely no problems (zero, zip, nada) connecting an Adaptec 2372B RLL controller to either Micropolis or Maxtor drives. Miniscribe: yes, problems galore. Seagate: I wouldn't even _think_ of trying it... kEITHe
unkydave@shumv1.uucp (David Bank) (10/23/89)
"Oh, poo" indeed!
As a professional, I would much rather overexaggerate the danger to
a slight degree than to underrate it.
There is a reason MFM drives are not RLL-rated: they were not
designed to meet RLL standards or performance. Period.
Yes, you CAN get away with it. But your hard drive is living on
borrowed time. Some drives can hack it, most can't.
I think I'd rather see 20 people go short drive room than I would
see 1 person lose his papers/research project/thesis/whatever to a
hard drive pushed beyond its limits.
(Yes, I realize I do not control other people or their computers.
The previous paragraph is merely my rationale for action.)
Unky Dave
unkydave@shumv1.ncsu.edublitter@ele.tue.nl (Paul Derks) (10/23/89)
In article <4273@ncsuvx.ncsu.edu> unkydave@shumv1.ncsu.edu (David Bank) writes: > Contrary to your assertion, plated media is NOT the sole >basic difference in RLL-rated drives. The biggest single difference >is in the use of VOICE COIL technology in the armature control >mechanism. This is what allows the mush more closely controlled >head movement demanded by RLL. The plated media simply makes >for a more reliable rusty pie plate. Voice Coil technology has NOTHING to do whether a drive can do RLL or only MFM. Most Seagate RLL drives use stepper motors and these drives are fully RLL certified. Mechanically there is no difference between an RLL or MFM drive. The main difference is that RLL uses a 7.5 Mbit/s datastream and MFM 5 Mbit/s. This means that the electronics, the heads and (maybe) the magnetic surface should be different (better). The reason why better drives (Maxtor, Priam, Rodime) do RLL without problem is because they have better specs. However, I do agree that it is dangerous to use an MFM drive with an RLL controller. In my experience ST 225's with RLL go bad after some time. Even ST 238's go bad after some time. (I had one) I have friends who use PRIAM V185 drives at RLL and this works perfectly. (But they are RLL certified, I think). The drive I use is a Rodime 203E and I have heard that it is RLL capable but not from a reliable source. (i.e. the manufacturer self). Fact is that the drive works perfectly with a WD 1003 RLL controller. Paul Derks
cth_co@tekno.chalmers.se (CHRISTER OLSSON) (10/23/89)
> KIDDIES! Do NOT TRY THIS AT HOME!!! > > If a drive is not SPECIFICALLY RLL-Rated by its manufacturer, > do NOT try hooking it to an RLL controller to boost storage. You > are playing Russian Roulette with your data and there is a VERY high > probability that you will lose everything. > > If you want the space THAT badly, go out and buy yourself an > RLL-rated drive! > Check with SPINRITE (from CORETEST package). If the tolerance is lower than 1%, you can use the drive with RLL. If the tolerance is near 1% or over, you shouldn't use drive with RLL. Most drives is between 0.1 and 0.5% and can use RLL. For example, the seagates typically has lower than 0.5% even if they are only MFM-rated. (ST225, ST251 ..)
Ralf.Brown@B.GP.CS.CMU.EDU (10/23/89)
In article <4273@ncsuvx.ncsu.edu>, unkydave@shumv1.uucp (David Bank) wrote: > Contrary to your assertion, plated media is NOT the sole >basic difference in RLL-rated drives. The biggest single difference >is in the use of VOICE COIL technology in the armature control >mechanism. This is what allows the mush more closely controlled >head movement demanded by RLL. The plated media simply makes >for a more reliable rusty pie plate. RLL *does*not* need more closely controlled head movement, since it does not change the number of cylinders on the drive. What RLL *does* need is more accurate and cleaner amplification/processing of the data stream coming off the drive. RLL and MFM both use the same number of flux changes per second, but RLL requires more accurate timing on the flux changes, since the timing carries information. This is analogous to 1200 and 2400 bps modems. 1200 bps modems are actually 600 baud, since each change in the carrier transmits two bits of information, not just one. 2400 bps are also 600 baud, but transfer four bits of information per signal change, and thus require more accurate signal processing to recover the information from the signal changes. Unless your RLL controller magically increases the number of cylinders on the drive, it will not demand any more accurate head movement than an MFM controller for the same drive. -- UUCP: {ucbvax,harvard}!cs.cmu.edu!ralf -=-=-=-=- Voice: (412) 268-3053 (school) ARPA: ralf@cs.cmu.edu BIT: ralf%cs.cmu.edu@CMUCCVMA FIDO: Ralf Brown 1:129/46 FAX: available on request Disclaimer? I claimed something? "How to Prove It" by Dana Angluin 6. proof by omission: "The reader may easily supply the details." "The other 253 cases are analogous."
davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) (10/23/89)
If your 225 is the late model version it will probably work with the RLL controller. See if the LED is the same color as your friend's 238. I suggest the WD1006VSR2 controller if you have an AT bus. -- bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) "The world is filled with fools. They blindly follow their so-called 'reason' in the face of the church and common sense. Any fool can see that the world is flat!" - anon
aqm@mace.cc.purdue.edu (Steve Weinrich) (10/23/89)
In article <2546@bingvaxu.cc.binghamton.edu> consp21@bingsuns.cc.binghamton.edu (Ken Hoover) writes: > I have a Seagate ST225 in my XT (no flames on my obsolete machine); and >was wondering if getting an RLL controller for it would allow me to format >it out to 32 megs instead of the 20 I get now. My roommate and I were >puzzling over this one (he has an ST238, and we couldn't find any physical >differences), so I thought I'd submit it to the net. I have read most of the replys to this poor man's question, and I think that you have missed a point that I believe he was asking... Is there a physical differnce? NO. The ST225 physical makeup is the same as the ST238, the only difference is the quality (haha) of it's plates. The only reason I know this is that when I sent in a ST225 for repair, the person I spoke with on the phone said that they (the re-work shop) could simply put in RLL rated plates and it would be fine. [This is when I was working for a computer firm as thier repair engineer] It would only increase the cost by a little bit, so I said.. fine, do it. Sure enough, we hooked that puppy up to an RLL and it worked just fine. Of course it did! A perfect analogy for this whole situation is that of the difference between a 360K disk and a 1.2 Meg disk. Sure, you can throw a 360K rated disk (Double density) into a 1.2Meg and make it work, but you are simply playing with fire... think about it, it's like recording your favorite albums on the cheapest cassette that you could find. It *WILL* store it, but then again, how good of a recording were you looking for? Ah well, so much for replys... I think I will go back to bed... -Steve
larry@jc3b21.UUCP (Lawrence F. Strickland) (10/23/89)
From article <35883@srcsip.UUCP>, by rogers@SRC.Honeywell.COM (Brynn Rogers): > In article <4265@ncsuvx.ncsu.edu> unkydave@shumv1.ncsu.edu (David Bank) writes: >> Sure, Ken. Go for it. Hook that MFM (i.e. NOT RLL-Rated) drive up to >>that RLL controller, format it, and instant 50% more drive room. You >>CAN do it. It'll even format. > >> If you're lucky, it might even hold data for a month before it starts >>developing amnesia. > > I think that this is a little overboard. (or Dave sells drives) > > One Manufacturer in particular (Maxtor) does not rate many of it's drives > RLL and yet I have had one running for over a month at almost double its ... > Some MFM drives will work RLL, some won't. About the only way to be sure > is to try it (But keep everthing backed up, and keep your MFM controller) This particular question seems to pop up over and over again, but there are rarely any real solutions to it. Much of the problem seems to be related to the drives used, not whether MFM drives will work as RLL in general. Yes, the RLL encoding scheme is more intensive on the disk. Yes, formatting an MFM drive as RLL may or may not work. About that question of it working for a month or so and forgetting, well... I thought for a considerable period of time before attempting to reformat my 72 Meg Miniscribe drive as RLL. However, the 70 Megs was too small for my application and I really didn't have the money to upgrade to a bigger drive and several people said they didn't _think_ there would be any problem, so I went ahead and reformatted. It has been running for most of a year now with no data loss. The controller card I used was the Adaptec which has been beaten up on the net but which has worked well and, because of its on-board ROM, has allowed me to use both XENIX and DOS together on the same drive. My experience has been very positive and I recommend the Miniscribe line for other reasons as well. So, if you have a Miniscribe, it seems likely that it will work (at least the full-height 72 Meg drive :-) -l -- +--------------------------------------+-- St. Petersburg Junior College --+ | Lawrence F. Strickland | P.O. Box 13489 | | ...!uunet.uu.net!pdn!jc3b21!larry | St. Petersburg, FL 33733 | +--------------------------------------+-- Phone: +1 813 341 3321 ---------+
terry@eecea.eece.ksu.edu (Terry Hull) (10/23/89)
In article <4273@ncsuvx.ncsu.edu> unkydave@shumv1.ncsu.edu (David Bank) writes: > >[bunch of stuff about RLL vs MFM] I have been running a Priam ID-130 (MFM) drive formatted RLL for over 2 years under Novel with NO PROBLEMS. I have a 71 MB full height Miniscribe that I used for 1 year MFM before I changed it to RLL, and it has been running happily for 1.5 years now as a /usr/spool filesystem on a computer that runs news. I also have a Maxtor 2190 formatted to 244 MB that has been running for 6 months with no problems. Yes, there are problems with some drives, but just because the drive is not rated RLL, does not mean that it will loose data after a month or so. The bottom line is try it. It may work it may not. BTW: I did have a Maxtor 2190 go flakey after 3 weeks when formatted RLL, but I discovered it was flakey when formatted MFM also. The only thing I do differently when using a MFM drive RLL, is I tend to run extra passes through the diagnostic software to make sure I get all the bad blocks before I put data on the drive. -- Terry Hull Department of Electrical and Computer Engineering, Kansas State University Work: terry@eecea.eece.ksu.edu, rutgers!ksuvax1!eecea!terry Play: terry@tah386.manhattan.ks.us, rutgers!ksuvax1!eecea!tah386!terry
pete@Octopus.COM (Pete Holzmann) (10/23/89)
In article <4275@ncsuvx.ncsu.edu> unkydave@shumv1.ncsu.edu (David Bank) writes: > As a professional, I would much rather overexaggerate the danger to >a slight degree than to underrate it. This I can agree with. But how much should we 'professionals' exaggerate the danger? In my rather extensive experience with using MFM "rated" drives under RLL, I can give you the following facts: (1) Old Seagate ST-225's (and the identical ST-238's) are not good under RLL. Don't try it! Supposedly, Rev 3 and up are better. I'm still not ready to waste my time on one. (2) Most, if not all (I've yet to see any hard facts that contradict my 'all' here...) MFM/RLL pairs are absolutely identical hardware. The only difference is that the ones sold with an RLL part number have been specifically tested for RLL and are guaranteed to work with an RLL format. ***What you get when you pay the extra bucks is a guarantee that RLL will work. You do not get different hardware!*** (3) The **only** drive that I have ever personally seen fail to work reliably with an RLL format is a Toshiba 56MB drive. And it *was* rated for RLL!!! (Note that I avoid ST-225 upgrades...) I've done a zillion Maxtors, ST-251's, and Micropolises. A few Miniscribes, a few Priams. Maxtor 1140's are especially nice, because a person can usually get a full 1224 cylinders out of them (i.e. treat it as a 2190) instead of the 'rated' 918. Yes, the extra cylinders are not under warranty, so you need to do a good job of pre-testing before using them; most customers are willing to take the time, in order to turn what is guaranteed as a 120MB drive, into a very nice 230MB drive! (4) The only drives I've heard of having any significant trouble are Seagate and Miniscribe. Does this mean they are lower quality? You judge for yourself! (5) This is not a 'short term' problem or solution. I've been doing this for a few years now, ever since RLL controllers became available. The format does not deteriorate any more than MFM formatting does. (6) Those who tell you that RLL requires more careful head positioning or media capable of higher storage densities are blowing smoke. They do not know what they are talking about. I don't have a copy any more, but I wrote a few extensive articles a year ago explaining how RLL works. Maybe somebody can send me a copy; I'll edit a few of the typos and repost. > There is a reason MFM drives are not RLL-rated: they were not >designed to meet RLL standards or performance. Period. Not true, except for earlier ST-225 drives. And the ST-238 labeled versions of those drives, even though 'rated' for RLL, were very unreliable in an RLL format. Those drives are what gave RLL a bad name. > Yes, you CAN get away with it. But your hard drive is living on >borrowed time. Some drives can hack it, most can't. Unky Dave, you've got to be careful with your wording. The *worst* that can happen is that your 'format is living on borrowed time'. RLL formatting *never* will do physical damage to a drive! > I think I'd rather see 20 people go short drive room than I would >see 1 person lose his papers/research project/thesis/whatever to a >hard drive pushed beyond its limits. I'd rather let all 20 decide for themselves, with full knowledge of the minimal risk involved. >unkydave@shumv1.ncsu.edu Pete -- Peter Holzmann, Octopus Enterprises |(if you're a techie Christian & are 19611 La Mar Ct., Cupertino, CA 95014 |interested in helping w/ the Great UUCP: {hpda,pyramid}!octopus!pete |Commission, email dsa-contact@octopus) DSA office ans mach=408/996-7746;Work (SLP) voice=408/985-7400,FAX=408/985-0859
gordon@eecea.eece.ksu.edu (Dwight Gordon) (10/24/89)
In article <mace.cc.purdue.edu> aqm@mace.cc.purdue.edu (Steve Weinrich) writes: >In article <2546@bingvaxu.cc.binghamton.edu> consp21@bingsuns.cc.binghamton.edu (Ken Hoover) writes: > > >> I have a Seagate ST225 in my XT (no flames on my obsolete machine); and >>was wondering if getting an RLL controller for it would allow me to format >>it out to 32 megs instead of the 20 I get now. [ etc. ] > > I have read most of the replys to this poor man's question, and I >think that you have missed a point that I believe he was asking... > > Is there a physical differnce? NO. The ST225 physical makeup >is the same as the ST238, the only difference is the quality (haha) >of it's plates. [ etc. ] It didn't use to be so. Back when the ST238 was first introduced the technicians (at the repair company for which I was consulting at that time) did a comparison between the ST225 and the ST238. (Remember, the RLL coding scheme originally was supposed to yield a 50% improvement on _existing_ MFM drives.) The ST238's data circuit board was different than the ST225s. (We did not have the equipment to check the media at that time.) This may only have been a "standard" upgrade. But, depending on the age of the ST225, it may have that "other" board. Seagate advised us not to use the ST225 with a RLL controller, as they had had some problems with some of them on the RLL controllers. (Note: this whole experience was about 4 years ago. Things may have changed since then.) - Dwight - -- Dwight W. Gordon, Ph.D. | 913-532-5600 | gordon@eecea.eece.ksu.edu Electrical & Computer Engineering Department | dwgordon@ksuvm.bitnet Kansas State University - Durland Hall | rutgers!ksuvax1!eecea!gordon Manhattan, KS 66506 | {pyramid,ucsd}!ncr-sd!ncrwic!ksuvax1!eecea!gordon
mrichey@orion.oac.uci.edu (Mike Richey) (10/24/89)
In article <4273@ncsuvx.ncsu.edu> unkydave@shumv1.ncsu.edu (David Bank) writes: > > ,. In response to Brynn Roger (rogers@src.honeywell.com): > > Nope. I don't sell hard drives. I merely install and maintain >them and diagnose them when they go bad. > Contrary to your assertion, plated media is NOT the sole >basic difference in RLL-rated drives. The biggest single difference >is in the use of VOICE COIL technology in the armature control >mechanism. This is what allows the mush more closely controlled >head movement demanded by RLL. The plated media simply makes >for a more reliable rusty pie plate. This is not true. The difference between the ST225 and the ST238, the miniscribe 8425 and 8438 is certainly not plated media any more, but whether or not in the factory it passed the RLL tests. Neither of these drives has a voice coil. RLL PUTS NO MORE BITS ON A SURFACE than MFM. the DIFFERENCE is the 2,7 ENCODING -period-. > Sure, I know people running MFM drives on RLL too. Its >like the car you drive off the lot and it never needs to see a >mechanic. Some RLL drives are just able to do it. You may be running >them at the edge of tolerance, but you are just on this side of >their operational capacity. I never meant to say it was IMPOSSIBLE >to run an MFM drive on RLL. Merely dangerous with your data. > RLL has been aound for a very long time. It has been used with mainframe drive systems since the early seventies. It's only recently that RLL came to the PC market (like 1984). Rodime made a number of drives (some that had plated media, some that didn't) that were NOT rated for RLL, that simply worked with RLL. Novell used to sell the Maxtor XT1140 hard disk drive in their NDS-2 and NDS-4 hard disk subsystem. This drive enclosure included a SCSI (Adaptec 4070a) that had an ST412 interface that used RLL encoding. The Maxtor XT1140 is not or ever has been RLL spec'd. There were a series of tests that Novell would run and these tests were distributed with Netware to make sure the drive that was installed was capable of handling RLL encoding. The installer was encouraged to run these legthly tests on a drive that was to be installed. Know what, I've got dozens of these drives on sites that have been going for >3 years, without a bit of trouble. In fact, Maxtor began marketing a drive called the XT1240, that was an RLL certified XT1140. Same drive, Same logic, same everything. Well different serial number. One of the differences from drive to drive is a device known as the pulse detector. This device sits on the read/write head. It detects the flux pulses that pass under it from the surface of the disk. This device is a very important one. Silicon Systems is the company that supplies them to Seagate and Miniscribe. The one that's used in the ST225 and the ST238 is not a current design. Silicon Systems still manufactures it specifically for Seagate. Silicon Systems doesn't support the part anymore and has advised Seagate to change the design because of faults that S.S. has fixed in newer designs. But, Seagate uses the old one. Why? because the newer one probably costs .12 more (my opinion). This little device is suspect of causing intermittant problems with the ST225s and ST238s that drives sometimes experience. I've had problems with ST2XX series where low level formatting fixes the intermittant problems (or spinrite). Well, without rambling on anymore...... Seagate tests all of their ST225 and ST238 drives at RLL first. If they pass RLL, they're ST238s, if they fail, they test tham at MFM and ship them as ST225s if they pass. This information I was told by seagate. This only make economic sense if not logical. Hey, you have a lot to offer, but no one is an expert on anything. No matter how much you know, there's always some smart@ss out there that thinks they know more 8-). So lighten up. And have a good today. Michael S. Richey Internet: mrichey@orion.oac.uci.edu Bitnet: MRichey@UCI CompuServe: 71650,3132 Voice: (714) 856-8374 University of California, Irvine Network and Telecommunications Services 342 Computer Science Irvine, CA 92717
davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) (10/24/89)
In article <3544@orion.cf.uci.edu>, mrichey@orion.oac.uci.edu (Mike Richey) writes: | Seagate tests all of their ST225 and ST238 drives at RLL first. If they | pass RLL, they're ST238s, if they fail, they test tham at MFM and ship | them as ST225s if they pass. This information I was told by seagate. This | only make economic sense if not logical. Are you *very* sure about that? What I thought they told me was that they tested drives until they had enough 238's. If a drive failed as a 238 it was retested as a 225. Therefore a 225 which you buy might be either a drive which failed as RLL or one which was never tested. This would explain why so many pass as RLL in actual service. -- bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) "The world is filled with fools. They blindly follow their so-called 'reason' in the face of the church and common sense. Any fool can see that the world is flat!" - anon
mrichey@orion.oac.uci.edu (Mike Richey) (10/24/89)
In article <1342@crdos1.crd.ge.COM> davidsen@crdos1.UUCP (bill davidsen) writes: >In article <3544@orion.cf.uci.edu>, mrichey@orion.oac.uci.edu (Mike Richey) writes: > >| Seagate tests all of their ST225 and ST238 drives at RLL first. If they >| pass RLL, they're ST238s, if they fail, they test tham at MFM and ship >| them as ST225s if they pass. This information I was told by seagate. This >| only make economic sense if not logical. > > Are you *very* sure about that? What I thought they told me was that >they tested drives until they had enough 238's. If a drive failed as a >238 it was retested as a 225. Therefore a 225 which you buy might be >either a drive which failed as RLL or one which was never tested. This >would explain why so many pass as RLL in actual service. >-- >bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) >"The world is filled with fools. They blindly follow their so-called >'reason' in the face of the church and common sense. Any fool can see >that the world is flat!" - anon We're saying the same thing here, except, I'm saying they do all of them right away. To save money, test them all at RLL. If they need 225s they label them as such. Another thing is that the controllers that go into PCs are not so quite as tight on the specs as the devices that are used to test the drives in the final test. The winchester testing units I'm speaking of. Regardless, the point was that there isn't any mechanical difference between RLL and MFM drives. IMHO, i wouldn't buy a seagate of any king. Drat CDC-Imprimis was just bought out by who??? yes, Seagate.... Well I'll take a positive outlook. The seagate product line will improve drastically, rather than the CDC quality falling!! Michael Richey mrichey@orion.oac.uci.edu
phil@diablo.amd.com (Phil Ngai) (10/24/89)
In article <4273@ncsuvx.ncsu.edu> unkydave@shumv1.ncsu.edu (David Bank) writes: | Contrary to your assertion, plated media is NOT the sole |basic difference in RLL-rated drives. The biggest single difference |is in the use of VOICE COIL technology in the armature control |mechanism. This is what allows the mush more closely controlled |head movement demanded by RLL. The plated media simply makes Here we have a self-proclaimed professional who hasn't the faintest idea what he's talking about. I think you're just a parts-swapper, at best. Tell me, smart guy, what is RLL and how does it work? Then you can tell me what voice coils have to do with it. -- Phil Ngai, phil@diablo.amd.com {uunet,decwrl,ucbvax}!amdcad!phil Just say NO to the "War on Drugs".
unkydave@shumv1.uucp (David Bank) (10/24/89)
Reply to Pete Holzmann:
As I have said a number of times, you CAN get away with it and I
in fact know people who have done so. I also know people who have had
their drive disappear from the list of ones DOS recognizes.
When I said "your drive is living on borrowed time" I ** NEVER **
meant to intimate that RLL would do physical damage to an MFM drive.
That is a ridiculous notion. You may be right in that a better choice
of words would have been "format" instead of "drive".
Unky Dave
unkydave@shumv1.ncsu.eduunkydave@shumv1.uucp (David Bank) (10/24/89)
In article <27860@amdcad.AMD.COM> phil@diablo.AMD.COM (Phil Ngai) writes: >In article <4273@ncsuvx.ncsu.edu> unkydave@shumv1.ncsu.edu (David Bank) writes: >| Contrary to your assertion, plated media is NOT the sole >|basic difference in RLL-rated drives. The biggest single difference >|is in the use of VOICE COIL technology in the armature control >|mechanism. This is what allows the mush more closely controlled >|head movement demanded by RLL. The plated media simply makes > >Here we have a self-proclaimed professional who hasn't the faintest >idea what he's talking about. I think you're just a parts-swapper, at >best. Tell me, smart guy, what is RLL and how does it work? Then you >can tell me what voice coils have to do with it. > >-- >Phil Ngai, phil@diablo.amd.com {uunet,decwrl,ucbvax}!amdcad!phil >Just say NO to the "War on Drugs". Well, Mr. Ngai, I could respond in the same immature vein you did and say: "I think you're just a <INSERT RACIAL\ETHNIC\WHATEVER SLUR>" but I think I'll exercise more self-control and intelligence than you did. No point in sinking down to your level. I generally have the respect for the net to keep flames in E-Mail, but hey, if you want to start something, I'll happily finish it. I feel absolutely no need to explain one damn thing to such a rude and boorish person such as yourself. Next time try engaging a CPU before your I/O. Unky Dave unkydave@shumv1.ncsu.edu
davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) (10/24/89)
In article <4265@ncsuvx.ncsu.edu>, unkydave@shumv1.uucp (David Bank) writes: | | RLL is a MUCH more demanding format and less tolerant of errors. MFM | drives are not RLL rated for a very good reason -- they can't deliver | the performance RLL-controllers demand. They are very simply not | engineered for the rigors of RLL usage. Fact: RLL rated drives are made on the same lines as MFM. The RLL drives are tested for rotational speed regularity and accuracy of the PLL. An MFM drive may have failed those tests, or may not have been tested. There are no engineering diferences. | | So...you CAN do what you want. But you had better hope and pray | and sacrafice virgins to Mammon or that drive WILL crash. Not might. | WILL. It is only a matter of time before you get "Disk Boot Error" | or "Invalid Drive Specification" or some similar, sinister, error | message. This is true, if misleading. Every drive will eventually fail. People who tell you that RLL is "harder on the drive" are kidding themselves. The only failure which hits RLL before MFM on any given drive would be bearing drag. If you go MFM you might run another 100 hours or so before they fail. I have never seen this happen, usually either the electronics go or a head comes in for a landing. Backups are improtant with ANY disk. -- bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) "The world is filled with fools. They blindly follow their so-called 'reason' in the face of the church and common sense. Any fool can see that the world is flat!" - anon
phil@diablo.amd.com (Phil Ngai) (10/25/89)
In article <4286@ncsuvx.ncsu.edu> unkydave@shumv1.ncsu.edu (David Bank) writes: | Well, Mr. Ngai, I could respond in the same immature vein you did |and say: | | "I think you're just a <INSERT RACIAL\ETHNIC\WHATEVER SLUR>" This is yet another lie. I used no racial\ethnic\whatever slurs. Although I can believe that you often do. | but I think I'll exercise more self-control and intelligence than |you did. No point in sinking down to your level. | | I feel absolutely no need to explain one damn thing to such a |rude and boorish person such as yourself. Next time try engaging |a CPU before your I/O. This is the kind of response you would expect from someone with little knowledge and a big ego. I know your type. You're in a low-paying, demeaning job swapping black boxes that you know nothing about. Maybe you even graduated from high school. You're the lowest Indian on the totem pole and your only sense of self-importance comes from snowing any customers unlucky enough to talk to you. Unfortunately, Mr. Bank, there are too many people on this net who do know what they're talking about for you to get away with the kind of stuff you probably pull on your customers. -- Phil Ngai, phil@diablo.amd.com {uunet,decwrl,ucbvax}!amdcad!phil Just say NO to the "War on Drugs".
eb2e+@andrew.cmu.edu (Eric James Bales) (10/25/89)
All of this talk about the ST225's and 238's not being very
good at RLL format has made me wonder about my drive. I
swapped the 20 Meg in my PS/2 Model 30 for a ST138R when I
bought the computer used about eight months ago. Does the
ST138 (3.5 inch model of 238) have as much trouble as the
ST238? I am formatting(sp?) the drive RLL.
-Eric Kirkbride-
Carnegie Mellon
Disclaimer: Who, in their right mind, would believe me?!?
"I have a firm belief that all computers are the children of
Satan...", Shannon Clineblitter@ele.tue.nl (Paul Derks) (10/25/89)
In article <27883@amdcad.AMD.COM> phil@diablo.AMD.COM (Phil Ngai) writes: >In article <4286@ncsuvx.ncsu.edu> unkydave@shumv1.ncsu.edu (David Bank) writes: >| Well, Mr. Ngai, I could respond in the same immature vein you did >|and say: >| >| "I think you're just a <INSERT RACIAL\ETHNIC\WHATEVER SLUR>" >any customers unlucky enough to talk to you. Unfortunately, Mr. Bank, >there are too many people on this net who do know what they're talking >about for you to get away with the kind of stuff you probably pull on >your customers. It already has been pointed out by several people on the net that Mr. Bank absolutely doesn't know what he is talking about. It would have been wise of Mr. Bank to say nothing more about the subject. I sympathize with Mr. Ngai but I do think his response should have been less rude. Perhaps it would be best if both Mr. Bank and Mr. Ngai reread the rules about flames on the net. Paul Derks.
jeh@simpact.com (10/26/89)
Has anyone had notable success or notable problems using a Micropolis
1325 in "RLL mode"? (I called the manufacturer and got the same answer
that someone else reported from Maxtor: "We don't support it, but a
lot of people are doing it.")
--- Jamie Hanrahan, Simpact Associates, San Diego CA
Internet: jeh@simpact.com, or if that fails, jeh@crash.cts.com
Uucp: ...{crash,scubed,decwrl}!simpact!jehJLI@kuhub.cc.ukans.edu (10/26/89)
In article <4273@ncsuvx.ncsu.edu>, unkydave@shumv1.uucp (David Bank) writes: > > Contrary to your assertion, plated media is NOT the sole > basic difference in RLL-rated drives. The biggest single difference > is in the use of VOICE COIL technology in the armature control > mechanism. This is what allows the mush more closely controlled > head movement demanded by RLL. The plated media simply makes > for a more reliable rusty pie plate. > > Sure, I know people running MFM drives on RLL too. Its > like the car you drive off the lot and it never needs to see a > mechanic. Some RLL drives are just able to do it. You may be running > them at the edge of tolerance, but you are just on this side of > their operational capacity. I never meant to say it was IMPOSSIBLE > to run an MFM drive on RLL. Merely dangerous with your data. > > No, not a blanket statement at all. As I recall, I > qualified it with "probably". > > Oops. Two paragraphs back, 3rd line, "RLL" should be "MFM" > These remote connections are hell on editing. > > Unky Dave > unkydave@shumv1.ncsu.edu This is NOT a flame, but I don't think the RLL-rated drives differ from the MFM-rated ones based on the use of voice coil technology. I am not sure how they related, but as far as I know, there are many RLL-rated drives still using stepper motor to move the heads (such as Miniscribe 8225, 8450, 8438, etc.) and MFM-rated drives using the voice coil (such as Seagate 251-1, 4096, etc.). I haven't done much research (at least not enough) on hard drives and their specifications, but I failed to see how one can differetiate RLL and MFM rated drives based on the use of voice coil. By the way, I only see one person mentioned the different encoding schemes that RLL and MFM systems (the encoding is done on the controller, I think.) using, which should be the main difference between these two types of systems. In order to meet the RLL specification, a drive has to have better physical characteristics and media quality. M.C. Baldwin
aland@infmx.UUCP (Dr. Scump) (10/26/89)
In article <4273@ncsuvx.ncsu.edu| unkydave@shumv1.ncsu.edu (David Bank) writes: | | ,. In response to Brynn Roger (rogers@src.honeywell.com): | | Nope. I don't sell hard drives. I merely install and maintain |them and diagnose them when they go bad. | | Contrary to your assertion, plated media is NOT the sole |basic difference in RLL-rated drives. The biggest single difference |is in the use of VOICE COIL technology in the armature control |mechanism. This is what allows the mush more closely controlled |head movement demanded by RLL. The plated media simply makes |for a more reliable rusty pie plate. |... |Unky Dave |unkydave@shumv1.ncsu.edu What does voice coil have to do with it? There are MFM voice-coil drives (e.g. ST40XX) and RLL stepper-motor drives (e.g. ST2XXR, I think). These are two separate issues. -- Alan S. Denney @ Informix Software, Inc. {pyramid|uunet}!infmx!aland
unkydave@shumv1.uucp (David Bank) (10/26/89)
Mr. Ngai:
Oh, I just LOVE this. You "know my type:
Pity you don't know yourself.
I'd've been more than happy to Post the information you said I
didn't have. All it would have taken from you was a response devoid of
name calling, hysterics, and pathetic attempts at insults.
Instead, you CHOSE to attack me personally on the net. And then
you seem not to understand why I shouldn't have a charitable response
back.
Let's go over some facts, shall we, Mr. Ngai???
1) You don't know a damn thing about me. A few hundred bytes of
text does not a face-to-face meeting make.
2) You instigated the present state of affairs, complete with
the bandwidth waste as you Post ever more inaccurate
summations
of what I do and don't do for a living and how well I am
employed. Do you really think anyone out there in netland
gives a damn??? You're a bigger fool than you have already
made of yourself if you do.
3) My response to you has been the same as for anyone else
who would attempt to use the net as their own personal
soapbox in childish attempts to denigrate others. The
subject matter at hand is irrlevant.
Perhaps one day you'll learn how to be a human being. Let us know
when that happens.johne@hpvcfs1.HP.COM (John Eaton) (10/27/89)
<<<<< < RLL and MFM both use the same number of flux < changes per second, but RLL requires more accurate timing on the flux < changes, since the timing carries information. ---------- This is analogous to what happened when the floppy world went from Single density to Double density. With single density a controller would write a 8 us pulse for a 0 and two 4 us pulses for a 1. Double density added a 6 us pulse and was able to fit one bit in every 4 us by suitable encoding. Note that the high and lows were the same and that they simply added a pulse in the middle. A drive could smear a pulse by 2 us and work with single density but could only smear by 1 us if it had to work with Double density. John Eaton !hpvcfs1!johne
plim@hpsgpa.HP.COM (Peter Lim) (10/27/89)
Here's my two-cent's worth on RLL.
From what I read from a book titled "????'s IBM-PC Hardware Bible",
can't quite remember the author's name, RLL differs from MFM on two
counts:
[1] The encoding method. RLL re-arrange the data using the 2,7
encoding method so that you always get a certain number of
consecutive 1's (I think 5 ??) in a 16 bit word. This arrangement
does two things: increase the number of raw bits to be written on
disk for the same amount of data, and reduce the flux density
for the same amount of raw bits to be written on the disk.
[2] Increase the number of sectors per track from 17 (MFM) to 26;
which gives about 50% more capacity.
You increase the amount of data to be written to disk and put more
sectors on one track, but reduce the inter-bit flux density.
According to the author, it works out that for an increase of flux
stress of about 14%, you get 50% more capacity. Reasoning goes that
this 14% extra stress is well within the tolerance of most recent
MFM drives, so they can be used as RLL drives.
My brother has been using Seagate MFM drives for RLL for more than
a year and had no problem at all. I haven't got my own system yet,
but I think I will just go ESDI instead :-). I hope the above explains
something. Sorry I didn't memorize the numbers and probably phrase my
sentenses badly........ but that's what you get of two-cent's worth
:-).........
Regards,
Peter Lim.
HP Singapore IC Design Center.
E-mail address: plim@hpsgwg
Snail Mail address: Peter Lim
Hewlett Packard Singapore,
(ICDS, ICS)
1150, Depot Road,
Singapore 0410.
Telephone: (065)-279-2289wayne.ho@f526.n250.z1.fidonet.org (wayne ho) (10/27/89)
> So...you CAN do what you want. But you had better hope > and pray > and sacrafice virgins to Mammon or that drive WILL crash. > Not might. > WILL. It is only a matter of time before you get "Disk > Boot Error" > or "Invalid Drive Specification" or some similar, sinister, > error > message. > > > KIDDIES! Do NOT TRY THIS AT HOME!!! > Hello there, I do agree with you that putting RLL controllers on MFM drives is quite a risk, but I must say, it isn't impossible. I have RLLed an ST-225 and a Miniscribe 3650 (which is exactly the same drive as the RLL rated 3650.. 'cept its not certified) for a year or so now without a hitch. Yes, I may be slightly crazy, but most of my data is not EXTREMELY important. I have backups.. From what I know Seagate no longer allows their non-RLL certified drives to use RLL. They have done something with their drives to disallow this. In their older ST-225's RLLing is possible, but it isn't very safe. I suppose I was lucky. From what I know you need a ST-225 above revision 3. I'm not sure when they started changing their drives to disallow RLLing. Oh, BTW, I will tell you if my drives crash.. but by that time, it may not be due to the RLL (these drives are pretty old). Wayne --- ConfMail V3.31 * Origin: MeTaStAsIo'S -`Not ready error reading drive A' (416)487-9093 (1:250/526)
hacker@isadora.ikp.liu.se (Goran Larsson [Hacker of Hackefors]) (10/29/89)
In article <713.2545af86@simpact.com> jeh@simpact.com writes: >Has anyone had notable success or notable problems using a Micropolis >1325 in "RLL mode"? (I called the manufacturer and got the same answer >that someone else reported from Maxtor: "We don't support it, but a >lot of people are doing it.") One of the disks in my UNIX machine (not pc :-) is a 1325 and it has been running continuosly with a RLL controller for the last 6 months without any problem. The root filesystem is located on this disk so the disk is used heavily. I'm using the Adaptec ACB4070 RLL controller. I'm also using RLL on a Quantum 640 without problems. ! _ ! ! Goran Larsson [The Hacker of Hackefors] --+-! Hackefors, Linkoping, SWEDEN (See) +46 13-155535 (Hear) !-+-- ...!uunet!sunic!liuida!prodix!isadora!hacker (UUCP) ! ! hacker@isadora.ikp.liu.se (Internet) ! Mmh, Yes
davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) (10/30/89)
In article <89102911043898@masnet.uucp>, wayne.ho@f526.n250.z1.fidonet.org (wayne ho) writes: | From what I know Seagate no longer allows | their non-RLL certified drives to use RLL. They have done something | with their drives to disallow this. Where did you get this info? 225's up to date code May89 seem to work okay. Seagate has *always* told you not to do it, but as far as I can tell that's the only thing they've done. Let's verify or squash this story. -- bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) "The world is filled with fools. They blindly follow their so-called 'reason' in the face of the church and common sense. Any fool can see that the world is flat!" - anon
jca@pnet01.cts.com (John C. Archambeau) (10/31/89)
wayne.ho@f526.n250.z1.fidonet.org (wayne ho) writes: >Hello there, > I do agree with you that putting RLL controllers on MFM drives > >is quite a risk, but I must say, it isn't impossible. I have RLLed >an ST-225 and a Miniscribe 3650 (which is exactly the same drive as >the RLL rated 3650.. 'cept its not certified) for a year or so now >without a hitch. > Yes, I may be slightly crazy, but most of my data is not EXTREMELY > >important. I have backups.. From what I know Seagate no longer allows >their non-RLL certified drives to use RLL. They have done something >with their drives to disallow this. In their older ST-225's RLLing >is possible, but it isn't very safe. I suppose I was lucky. From >what I know you need a ST-225 above revision 3. I'm not sure when >they started changing their drives to disallow RLLing. You mean the drive won't format RLL or what? I can tell you since I work for an authorized Seagate dealer that Seagate NEVER condoned RLL formatting a non-RLL certified drive. Same goes in reverse, if you MFM format an RLL drive, the warranty is invalidated. But I do know one thing, Seagates if formatted the way they're supposed to be never gave me any trouble. I have a two year old ST225 to prove it. The only time it crashed was when *I* had the pleasure of trashing the partition table myself. Of course, that ST225 is now in the hands of a friend of mine whom I sold it to and replaced with an ST151. Oh, by the way. Seagate is not going to develop anymore MFM or RLL drives. The ST151 was Seagate's last MFM drive developed. Seagate is now focusing on SCSI drives so now this problem of MFM and RLL will be done with as soon as the older MFM and RLL drives die out from normal usage. Seagate will still do servicing of these drives of course, but gradually, they will cease manufacturing of MFM and RLL drives. There will be no 80 Mb 3.5" MFM or RLL drive, only a SCSI drive. A shame since the newer 3.5" drives have an average access time of 24ms (ST151, ST177N, and ST1096N). Guess I'd better buy my 2nd ST151 while I can. Got another hard drive bay to fill up. :) /*--------------------------------------------------------------------------* * Flames: /dev/null (on my Minix partition) *--------------------------------------------------------------------------* * ARPA : crash!pnet01!jca@nosc.mil * INET : jca@pnet01.cts.com * UUCP : {nosc ucsd hplabs!hd-sdd}!crash!pnet01!jca *--------------------------------------------------------------------------* * Note : My opinions are that...mine. My boss doesn't pay me enough to * speak in the best interests of the company (yet). *--------------------------------------------------------------------------*/
edhall@rand.org (Ed Hall) (11/01/89)
In article <1514@crdos1.crd.ge.COM> davidsen@crdos1.UUCP (bill davidsen) writes: >In article <89102911043898@masnet.uucp>, wayne.ho@f526.n250.z1.fidonet.org (wayne ho) writes: >| From what I know Seagate no longer allows >| their non-RLL certified drives to use RLL. They have done something >| with their drives to disallow this. > > Where did you get this info? 225's up to date code May89 seem to work >okay. Seagate has *always* told you not to do it, but as far as I can >tell that's the only thing they've done. Well, when I tried to reformat my year-old ST-4096 with a WD-1006V-SR2 controller (a 1:1 RLL controller), just about *half* (50%) of the tracks came up bad, according to WD's ROM-based formatter. Seagate claims this is a plated-media, voice-coil-positioned drive--so by what I've read (here and elsewhere) it *should* run RLL despite any lies Seagate has been telling about MFM/RLL incompatability. But nothing I did would make it take an RLL format. So, I popped back in the old WD MFM controller and reformated. ZERO bad tracks! (I had to mark the bad tracks listed on the drive manually.) One more go with the RLL controller yielded the same sad results as the first time. So the disk and MFM controller are now running happily in my father's AT, and I'm running an old, noisy, slow (and too small!) MiniScribe 3650 formatted RLL until I can replace it with an RLL-rated (or at least RLL-quality) drive. (The 3650 RLL'd perfectly!) Now, I understand the difference between 1,3 RLL (aka MFM) and 2,7 RLL, and can see no reason why the latter would perform so poorly on the ST-4096 unless Seagate either used exceptionally poor head electronics (much worse than a MiniScribe drive a year older and three times cheaper), or they intentionally engineered that electronics to prohibit 2,7 RLL. Perhaps I should post to alt.conspiracy. I've certainly read of ST-4096's running RLL. I might just be unlucky. (The drive was manufactured in October of 1988, for those who care.) -Ed Hall edhall@rand.org
wayne.ho@f526.n250.z1.fidonet.org (wayne ho) (11/02/89)
Hello,
I have read information for a hard drive conference about this.
I should have said, I HEARD this, rather than said I KNOW this. I
can't say what they have or have not done, for sure. If I find anymore
information regarding this topic, I'll let you know.
Wayne
--- ConfMail V3.31
* Origin: MeTaStAsIo'S -`Not ready error reading drive A' (416)487-9093 (1:250/526)perry@ccssrv.UUCP (Perry Hutchison) (11/02/89)
In article <624@crash.cts.com> jca@pnet01.cts.com (John C. Archambeau) writes: > I work for an authorized Seagate dealer ... Seagate NEVER condoned RLL > formatting a non-RLL certified drive. Same goes in reverse, if you MFM > format an RLL drive, the warranty is invalidated. I asked this once before, but now that we have someone with direct Seagate knowledge maybe we can get a semi-authoritative answer. It's pretty clear from previous postings that using RLL on a drive not designed for it is not always going to work, and one can therefore not expect a manufacturer to warrant the results. What I don't see is why Seagate objects to MFM formatting of an RLL-rated drive. Obviously it reduces the capacity by 1/3, but if someone wanted to do it, would it really impair the reliability so as to justify voiding the warranty?
pete@Octopus.COM (Pete Holzmann) (11/03/89)
In article <1989Oct31.224410.272@rand.org> edhall@rand.org (Ed Hall) writes: >Well, when I tried to reformat my year-old ST-4096 with a WD-1006V-SR2 >controller (a 1:1 RLL controller), just about *half* (50%) of the >tracks came up bad, according to WD's ROM-based formatter.... >So, I popped back in the old WD MFM controller and reformated. ZERO >bad tracks!... >Perhaps I should post to alt.conspiracy. I've certainly read of >ST-4096's running RLL. I might just be unlucky. These symptoms are very similar to what I saw the one time I had a drive that simply refused to 'take' RLL. In my case, it was certified for RLL, so I just yelled until I got a replacement. (It was actually pretty tough- the distributor had an MFM tester, not an RLL tester. The drive worked perfectly under MFM; I had a time convincing them it was bad!) I never got a truly satisfactory answer out of the manufacturer's field engineer. Supposedly it had something to do with a portion of the mechanical assembly not being tightened down quite enough; the drive would eventually have failed under MFM also. By the way, it had a huge pile of defects listed on its RLL defect map also. I suppose I should have been suspicious from the start! Pete -- Peter Holzmann, Octopus Enterprises |(if you're a techie Christian & are 19611 La Mar Ct., Cupertino, CA 95014 |interested in helping w/ the Great UUCP: {hpda,pyramid}!octopus!pete |Commission, email dsa-contact@octopus) DSA office ans mach=408/996-7746;Work (SLP) voice=408/985-7400,FAX=408/985-0859
davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) (11/03/89)
In article <1989Oct31.224410.272@rand.org>, edhall@rand.org (Ed Hall) writes: | | Now, I understand the difference between 1,3 RLL (aka MFM) and 2,7 | RLL, and can see no reason why the latter would perform so poorly on | the ST-4096 unless Seagate either used exceptionally poor head | electronics (much worse than a MiniScribe drive a year older and three | times cheaper), or they intentionally engineered that electronics to | prohibit 2,7 RLL. I think the complete text of my posting included the word "most." Since the drives sold for MFM are not tested for RLL timing accuracy, you could get a unit which doesn't hack it. Remember that the TOTAL timing jitter in drive, drive electronics, and controller is the determining factor. You could have a controller which is in spec but not better than spec. | | Perhaps I should post to alt.conspiracy. I've certainly read of | ST-4096's running RLL. I might just be unlucky. A lot of us have run 4096s and 251s RLL happily. I have done about six of these installations, and I haven't had any trouble. The sixhub archive server is running on one, and that has been very reliable. -- bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) "The world is filled with fools. They blindly follow their so-called 'reason' in the face of the church and common sense. Any fool can see that the world is flat!" - anon
wayne.ho@f526.n250.z1.fidonet.org (wayne ho) (11/04/89)
> You mean the drive won't format RLL or what? Hello, I may have been mistaken here. I heard read it somewhere in another echomail conference and thought it came from a reliable source. I have no proof with this, so until anyone else has information on SEAGATE and RLL stuff, I won't worry about it. > Oh, by the way. Seagate is not going to develop anymore > MFM or RLL drives. > The ST151 was Seagate's last MFM drive developed. Seagate > is now focusing on > SCSI drives so now this problem of MFM and RLL will be > done with as soon as > the older MFM and RLL drives die out from normal usage. So SCSI is the way to go?? I was thinking about purchacing a SCSI controller and drive for my 386sx, but I have been warned against purchacing a 16-bit SCSI controller. Again, I don't know if this is true or not, but someone told me that the 16-bit SCSI doesn't not work that well. I was considering an 8-bit SCSI controller, what do you think? Also, what types of SCSI drives does Seagate have out at the moment? If you have any specs such as speed and size that would be nice to know. Thanks in advance. Wayne --- ConfMail V3.31 * Origin: MeTaStAsIo'S -`Not ready error reading drive A' (416)487-9093 (1:250/526)
jca@pnet01.cts.com (John C. Archambeau) (11/05/89)
perry@ccssrv.UUCP (Perry Hutchison) writes: >I asked this once before, but now that we have someone with direct Seagate >knowledge maybe we can get a semi-authoritative answer. It's pretty clear >from previous postings that using RLL on a drive not designed for it is >not always going to work, and one can therefore not expect a manufacturer >to warrant the results. > >What I don't see is why Seagate objects to MFM formatting of an RLL-rated >drive. Obviously it reduces the capacity by 1/3, but if someone wanted to >do it, would it really impair the reliability so as to justify voiding the >warranty? I don't know the exact reason why. I do know that somebody posted here that he formatted an RLL drive MFM and it didn't quite work the way it was supposed to. I do know Seagate's warranty policy well and I follow it to the letter. It might deal with the fact that an RLL drive is designed to do a 7.5 Mbit/sec transfer and an MFM drive is only 5.0 Mbit/sec. But I do know that it is documented on page 6 of the Seagate Universal Installation Guide. The copyright date on it is '87, but we have been getting this installation booklet with every drive we have ordered. /*--------------------------------------------------------------------------* * Flames: /dev/null (on my Minix partition) *--------------------------------------------------------------------------* * ARPA : crash!pnet01!jca@nosc.mil * INET : jca@pnet01.cts.com * UUCP : {nosc ucsd hplabs!hd-sdd}!crash!pnet01!jca *--------------------------------------------------------------------------* * Note : My opinions are that...mine. My boss doesn't pay me enough to * speak in the best interests of the company (yet). *--------------------------------------------------------------------------*/
keithe@tekgvs.LABS.TEK.COM (Keith Ericson) (11/06/89)
In article <1538@crdos1.crd.ge.COM> davidsen@crdos1.UUCP (bill davidsen) writes: >Since the drives sold for MFM are not tested for RLL timing accuracy, My _guess_ would be that "the drives sold for MFM" really _are_ tested for RLL and fail... kEITHe
keithe@tekgvs.LABS.TEK.COM (Keith Ericson) (11/06/89)
In article <89110410221851@masnet.uucp> wayne.ho@f526.n250.z1.fidonet.org (wayne ho) writes: > > Oh, by the way. Seagate is not going to develop anymore > > MFM or RLL drives. > > Seagate is now focusing on > > SCSI drives so now this problem of MFM and RLL will be Seagate has purchased/absorbed/taken over Imprimis, formerly known as CDC, one of the foremost manufacturers of SCSI drives. I only hope the Seagate-ness doesn't rub off on Imprimis... >So SCSI is the way to go?? I was thinking about purchacing a SCSI >controller and drive for my 386sx, but I have been warned against purchacing >a 16-bit SCSI controller. Again, I don't know if this is true or not, >but someone told me that the 16-bit SCSI doesn't not work that well. >I was considering an 8-bit SCSI controller, what do you think? Also, >what types of SCSI drives does Seagate have out at the moment? If >you have any specs such as speed and size that would be nice to know. A 16-bit controller has a better chance of achieving maximum transfer rate than does an 8-bit. I measure (using Core27) a transfer rate of about 1.4Megabytes (yes, 1,400 kilobytes) per second with an Adaptec 1542A/Imprimis94181-702 (617.4 Megabytes formatted) in an Everex STEP/25. With the Seagate ST-)2 I got (as I recall) certainly no better than 450 Kbytes/sec (if even that good; as I recall it acted like a 2:1 MFM drive as far as transfer rate is concerned). My suggestion is to only purchase the ST-02 (or -01) if you simply cannot afford one of the better (16 bit) controllers. It's like hooking up a 50-amp power supply with a 1/4 amp fuse. kEITHe
jca@pnet01.cts.com (John C. Archambeau) (11/07/89)
wayne.ho@f526.n250.z1.fidonet.org (wayne ho) writes: >Hello, > I may have been mistaken here. I heard read it somewhere in another >echomail conference and thought it came from a reliable source. I >have no proof with this, so until anyone else has information on SEAGATE >and RLL stuff, I won't worry about it. > > > Oh, by the way. Seagate is not going to develop anymore > > MFM or RLL drives. > > The ST151 was Seagate's last MFM drive developed. Seagate > > is now focusing on > > SCSI drives so now this problem of MFM and RLL will be > > > done with as soon as > > the older MFM and RLL drives die out from normal usage. > >So SCSI is the way to go?? I was thinking about purchacing a SCSI >controller and drive for my 386sx, but I have been warned against purchacing >a 16-bit SCSI controller. Again, I don't know if this is true or not, >but someone told me that the 16-bit SCSI doesn't not work that well. >I was considering an 8-bit SCSI controller, what do you think? Also, >what types of SCSI drives does Seagate have out at the moment? If >you have any specs such as speed and size that would be nice to know. >Thanks in advance. Not quite yet. I wouldn't buy a SCSI hard drive unless your motherboard has a built-in SCSI port (I have seen some) or until companies such as Western Digital design a good 16-bit or 32-bit SCSI host adaptor. There might be such a beast out now. I think I remember seeing an ad for a 32-bit SCSI host adaptor in a magazine (MIPS?), but I can't for the life of me remember who made it. As for which SCSI drives Seagate has out now, they have the following: 5.25" (these are basically SCSI implementations of their MFM/RLL counterparts): ST225N (20 Mb), ST251N (40 Mb), ST277N (60 Mb), ST296N (80 Mb) 3.5" Drives: ST125N (20 Mb), ST138N (30 Mb), ST157N (46 Mb), ST177N (60 Mb) ST1096N (80 Mb) These are the more common units out there that I can think of from the top of my head. I had the latest data booklet from Seagate a week ago but I think my one of our assinine sales reps ate it or gave it to a customer. The newer 3.5" drives by Seagate have an average seek of about 24ms which is better than the old STblah-1 units with only 28ms. I do recall seeing some drives with a 20ms seek in that booklet, but that's just a vague memory since my primary concern was with the newer 3.5" drives. I just bought an ST151 which is basically a 3.5" version of the ST251-1 with a seek of about 24ms. I like it and until the SCSI host adaptor card is resolved with the IBM compatables entirely, I will stick with MFM and RLL drives for the time being. I'm not convinced of the reliability of SCSI ports for the IBM compatable domain as of yet. I do know that Seagate does make pretty sloppy drive controllers and host adaptors. I base this on the fact that they recommend an interleave of 3:1 on an AT and 4:1 on an XT while I can easily get a 2:1 with my WA3-16 (a WD1003-WA2 clone more or less) or a WD1006V-MM2 that I just put in an Arche Rival 286-12 today. I do admit that Seagate makes good drives, but I don't like their controller and host adaptor boards. In that department I'm partial to Western Digital. When WD makes a 16-bit SCSI host adaptor that's been around for awhile, then I might contemplate a SCSI hard drive, but not till then. When Seagate Dealers and Distributors can't order the MFM and RLL drives anymore, then I would start to break my piggy bank for a SCSI host adaptor and SCSI hard drive, but not until I see that bridge on the horizon which probably won't be for a few years to maybe a decade depending on where the demand is. /*--------------------------------------------------------------------------* * Flames: /dev/null (on my Minix partition) *--------------------------------------------------------------------------* * ARPA : crash!pnet01!jca@nosc.mil * INET : jca@pnet01.cts.com * UUCP : {nosc ucsd hplabs!hd-sdd}!crash!pnet01!jca *--------------------------------------------------------------------------* * Note : My opinions are that...mine. My boss doesn't pay me enough to * speak in the best interests of the company (yet). *--------------------------------------------------------------------------*/
wayne.ho@f526.n250.z1.fidonet.org (wayne ho) (11/11/89)
Thanks for the information on the SCSI drives. Have you heard of any
future plans from IBM or other companies to make a true SCSI port (bus?)
onto an IBM type computer?
Wayne
--- ConfMail V3.31
* Origin: MeTaStAsIo'S -`Not ready error reading drive A' (416)487-9093 (1:250/526)jca@pnet01.cts.com (John C. Archambeau) (11/14/89)
wayne.ho@f526.n250.z1.fidonet.org (wayne ho) writes: >Thanks for the information on the SCSI drives. Have you heard of any >future plans from IBM or other companies to make a true SCSI port (bus?) >onto an IBM type computer? There are already a couple of boards out there with a built in SCSI port. The Tiny Turbo Motherboards use them. But as for IBM, Compaq, AST, et. al. Nah, they'll leave it to Western Digital, Adaptec, and the other drive controller manufacturers to produce them. There's no guarrantee that the port built into your board would be the ideal port. I'd wait for WD to do it, they make good controller boards. /*--------------------------------------------------------------------------* * Flames: /dev/null (on my Minix partition) *--------------------------------------------------------------------------* * ARPA : crash!pnet01!jca@nosc.mil * INET : jca@pnet01.cts.com * UUCP : {nosc ucsd hplabs!hd-sdd}!crash!pnet01!jca *--------------------------------------------------------------------------* * Apple Computer, Inc. is really the Anti-Christ! *--------------------------------------------------------------------------* * Note : My opinions are that...mine. My boss doesn't pay me enough to * speak in the best interests of the company (yet). *--------------------------------------------------------------------------*/
ruiu@dragos.uucp (dragos) (11/21/89)
In article <703@crash.cts.com> jca@pnet01.cts.com (John C. Archambeau) writes: >wayne.ho@f526.n250.z1.fidonet.org (wayne ho) writes: >manufacturers to produce them. There's no guarrantee that the port built into >your board would be the ideal port. I'd wait for WD to do it, they make good >controller boards. > * INET : jca@pnet01.cts.com I dunno... my roommate arrived at work one day to find the fire deparment there. His WD controller board spontaneously combusted ovenight, melting a chunk of the motherboard with it and setting off every smoke detector in the place. When contacted about the possibility of getting a replacement controller, WD politely told him to go away - they have recently been taken over and are not too concerned about old products, according to the PR droid. I'll think twice before buying a WD product. -- Dragos Ruiu (ruiu@dragos.uucp) All system administrators should hand out alberta!dragos!ruiu a bottle of valium with every news-reader uunet!myrias!dr man page.
emmo@moncam.co.uk (Dave Emmerson) (11/25/89)
In article <1989Nov21.062846.6164@dragos.uucp>, ruiu@dragos.uucp (dragos) writes: > In article <703@crash.cts.com> jca@pnet01.cts.com (John C. Archambeau) writes: > > I dunno... my roommate arrived at work one day to find the fire deparment > there. His WD controller board spontaneously combusted ovenight, melting a > chunk of the motherboard with it and setting off every smoke detector in > the place. > [omission] > I'll think twice before buying a WD product. > To be fair, I have yet to see a PC add-on board which has UL approval, and some may even be potentially flammable. Isn't it about time individual boards had fuses on their power rails, rather than relying on the half kilowatt or so limiter in the power supply? -Yes I know why they don't, I'm 'in the trade' myself, but we'd find a happy solution if we HAD to, instead we rely on parts which are considered 'safe' under those conditions, and make no attempt at *prevention*. Just MHO, my employer obviously disagrees... Dave E.
jca@pnet01.cts.com (John C. Archambeau) (11/26/89)
ruiu@dragos.uucp (dragos) writes: > I dunno... my roommate arrived at work one day to find the fire deparment > there. His WD controller board spontaneously combusted ovenight, melting a > chunk of the motherboard with it and setting off every smoke detector in > the place. > > When contacted about the possibility of getting a replacement controller, > WD politely told him to go away - they have recently been taken over and > are not too concerned about old products, according to the PR droid. > > I'll think twice before buying a WD product. Then what you do is get around the PR droid. I do it all the time as a when doing my duties that involve tech support. Of course, you have to yell louder than Morton Downey, Jr., but often times you can get to a system engineer that will help you out. The marketing and PR people never know what the hell is going on, you need to get to a system engineer and if she or he says we don't support it, then you are screwed. Never deal with a PR or sales droid when dealing with an RMA or something involving tech support because they don't know or really care. They just want to make their commission and/or sales quota. But a good system engineer will want to get to the bottom of the problem and solve it because chances are she or he will encounter it again (Murphy's law of error propogation). Now if the controller is out of warranty and just up and died, there's nothing you can do even if the company still supports the product. They'll either want a swap out fee or just tell you to buy another one. /*--------------------------------------------------------------------------* * Flames: /dev/null (on my Minix partition) *--------------------------------------------------------------------------* * ARPA : crash!pnet01!jca@nosc.mil * INET : jca@pnet01.cts.com * UUCP : {nosc ucsd hplabs!hd-sdd}!crash!pnet01!jca *--------------------------------------------------------------------------* * Apple Computer, Inc. is really the Anti-Christ! *--------------------------------------------------------------------------* * Note : My opinions are that...mine. My boss doesn't pay me enough to * speak in the best interests of the company (yet). *--------------------------------------------------------------------------*/