• Please review our updated Terms and Rules here

need some mfm / rll help!

J

jc179

Experienced Member
Joined
Apr 21, 2007
Messages
99
Hi Guys

I have been tinkering with computers for some time and started doing so on an XT computer. I recently have aquired this hardware:

St 251-1 and st 251-0 drives 41 meg each
A wd1004 and WD1002 8 bit isa card
A wd1006 isa card

I really want to use them in my computer for logging, and low disk utilizition applications.

So far I have configured the Wd1004 and Wd1002 on address CA000, or 0x324, the secondary address, and have them running happily in my computer (abit very slowly).

This is where I need some of your help. I've used a program called diskman for format them, and also G=C800:5 in the bios. MFM i am assuming. (it doesn't say or give me the option to change to rll (I know its bad to format rll, but It is faster, but more unreliable if the drive will do it)). Anyways same result.

I have a lot of questions that ive been trying to figure out but im not getting anywhere. Im just going to be simple for now as I get carried away lol.

First one would be are the wd1004 and wd1002 rll compatabile?

In the pdf for the 1004 for example it says this:

IDENTIFY DRIVE The WD1004-27X and WD1004A-27X boards work with any hard disk drive that has the ST506/412 drive interface and no more than 1024 cylinders and 16 heads. The Seagate 238 hard disk drive unit is such a drive.
Look in your drive manual to verify that you have an ST506/412 RLL drive or
check with your dealer. Then check the drive manual for the drive parameters, such as those listed in Table 1-1. You'll need this information to select the proper low level format. Table 1-1. WD1004 and WD1004A Default Drive Parameters Formatted Capacity Number of Heads Number of Cylinders Sectors per Track
32MB 4 615 26
This board does not work with RLL drives. If you have an MFM drive you
need either the WDXT–GEN2 or WD1004A–WX1 boar
Click to expand...


SO

The WD1004-27X and WD1004A-27X boards work with any hard disk drive that has the ST506/412 drive interface and no more than 1024 cylinders and 16 heads.The Seagate 238 hard disk drive unit is such a drive.
Look in your drive manual to verify that you have an ST506/412 RLL drive or
check with your dealer.


and

This board does not work with RLL drives. If you have an MFM drive you
need either the WDXT–GEN2 or WD1004A–WX1 boar

... have me just confused. So verify I have an St506 std drive (example of this is a 238.. ok). Its an RLL drive.... and now your telling me it doesn't work with RLL in the next line ..lol. Im confused. AAH!.

So question 1, Anyways im assuming this is just an MFM and the RLL (7.5 mhz bus are 16 bit isa)

Which brings me to question number 2 .. getting there atleast

I realize that MFM is 5 mega-bits-per-second (5,000,000 bits/sec). Which is approximately 625,000 bytes a second, or well basically 625,000 *60 seconds * 1024 bytes * 1024 bytes = 1 minute transfer rate of 35.7 megs in a minute. TO me that just seems way too fast.

Anyways My problem #1 is, its is very slow, as expected, but seems too slow im told be people with experience with these drives. I've bench'd the drives at approx 1.3 megs in a minute (just large file transfer). It takes 13 minutes to transfer 17 megs !

So my question is .. is this normal or way too slow ? Did I mis configure something like the interleave ? Where ever you set that. !


And Question 3... the wd1006. Has no G=C800:5 in debug, *however* apprently you can access it by G=CC00:5 . I fail and have no luck doing so.

So what am I doing wrong. ?

When I use the XT 8 bit ctrl... I don't setup anything in the bios, msdos sees the drive fine. When I use the AT I set the pri IDe interfaces to disabled, (as well as FDD disabled on the card with an FDD, my wd1006 doesn't not have the fdd ctrl) and set the correct user type / drive paramters... and get no fixed disks found, nothing to preformat in disk man, and nothing to run in debug.. I've hit a wall ! doh.

I have tried all the different bios ranges and dumped the 0000 -> CCCC range looking for bios string with no luck. I take it that this is the next step to go if I want the RLL size and speed?

If anyone can help me out ID really appreciate it !

I am not using a legacy system for this, I am using a P2 350 mhz with an ISA, and also a P2B asus mobo /w a 450 mhz cpu. SAme results in both.

P.s. this page has really good helpful advice for jumper settings of this should google not help! ,


http://artofhacking.com/th99/

Cheers,

Thanks for any help guys!

Jonathan
 
jc179 said:
Anyways My problem #1 is, its is very slow, as expected, but seems too slow im told be people with experience with these drives. I've bench'd the drives at approx 1.3 megs in a minute (just large file transfer). It takes 13 minutes to transfer 17 megs !

So my question is .. is this normal or way too slow ? Did I mis configure something like the interleave ? Where ever you set that. !
Click to expand...

You set interleave in the low-level BIOS routine for formatting. If you don't have one in your BIOS (or can't find it) then you can use Spinrite or Norton Utilities' CALIBRAT to test for the best interleave setting and then reformat the entire drive with the setting (non-destructive; it reads data, formats, puts data back).

For a fast controller, 1:1 interleave works. But most XTs hit about 3:1 or 4:1.

If you low-level format a drive, keep it with the controller you did it with -- MFM/RLL controllers do NOT low-level format all the same way, and you risk not being able to read the disk with a different controller.
 
Hey

Sorry I didn't realize I had put it in the wrong thread!

Anyhow the data on the drive isn't important, but that doesn't seem to make a difference... Since im using a more modern system, could I approach interleave 1:1, or would I need to goto the 16 bit card? I take it that the spinrite is a very different version than whats available now for drives? Is there an abandon ware page with them ? If not im sure I can find them, was just asking :)

Thanks for your info ! I Still have lots to figure out it would seem :|

JC
 
I have formatted ST-251 drives to 1:1 interleave, and works just fine, if your controller can keep up. Running the wrong interleave can slow the transfer rate down to a crawl, but no ST-251 MFM drive is going to transfer 40Mb in just over a minute. I think you need to re-check your math if you're expecting that kind of speed, you've made an error somewhere.

--T
 
yea thats what I thought.. .Can I ask you what kind of controller your using? 8 bi or 16 bit isa bus, and did you too set the interleave with spinrite ?

And yes I did make an error, I knew I had :) Happens when you post at 5 am after you've spent the night tickering with stuff lol

Thanks for the help, let me know if you can!

P.s. how come the interleave isn't an avilable option with the ontrack disk manager that pre-formats the hard drive? I was hoping to find it in there... It seems that without spinrite your stuck with a perhaps a poor interleave setting!

Jonathan
 
jc179 said:
yea thats what I thought.. .Can I ask you what kind of controller your using? 8 bi or 16 bit isa bus, and did you too set the interleave with spinrite ?

And yes I did make an error, I knew I had :) Happens when you post at 5 am after you've spent the night tickering with stuff lol

Thanks for the help, let me know if you can!

P.s. how come the interleave isn't an avilable option with the ontrack disk manager that pre-formats the hard drive? I was hoping to find it in there... It seems that without spinrite your stuck with a perhaps a poor interleave setting!

Jonathan
Click to expand...


The controller would have been a 16-bit ISA. I don't recall exact models, but I was partial to Seagate & WD back then (especially the WD Gen-2). And yes, Spinrite would be the method used for changing the interleave, but as for why Spinright does things the way it does, you'd have to ask Steve Gibson, I ain't got a klew, mate.

--T
 
hey guys


thanks for the info. I tried the interleave setting in spinrite (or what thought was it), but I had to create a fat16 dos partition for spinrite to even work...ugh so that wasn't going so hot, I should have mentioned that im using linux and not msdos. Anywho i managed to find an interleave setting in the ontrack diskmanager software, I tried that but xfer speeds still didn't increase. I will check in the G=C800 , perhaps I missed something, but for now it looks like its going to be pokey... !

Ill let you know if I do get the 4 drives going, and how I managed to do it, and if I get it going with ide's inside as well, thanks for that info!

Give me a little time :) Many projects on the go, just blew out valve #5 last night so I have to get that fixed first ;)

Good day mates,

JC
 
I'm looking for information on the following controller

It has a Compaq Sticker on an eeprom chip.

I'm told the card is a WD1002-HX4

Wondering if I can use a SEagate ST-238R with this card.

Thanks for any insight.. :>
 
Wd1002-hx4

Wd1002-hx4

gerrydoire said:
I'm looking for information on the following controller
It has a Compaq Sticker on an eeprom chip.
I'm told the card is a WD1002-HX4
Wondering if I can use a SEagate ST-238R with this card.
Thanks for any insight.. :>
Click to expand...
Why not a new thread ?

Your Compaq labeled WD1002-HX4 could be from a Compaq Deskpro, because the second page of a Compaq Deskpro document at ftp://ftp.compaq.com/pub/supportinformation/techpubs/qrg/deskpro.pdf has info on the WD1002-HX4.

It has switch settings and from my reading of the entire page, informs us that there are various releases of the controller:
1. Early version - no switches - supports type 1 drives only
2. Later version with revision E ROM - has switches - supports types 1 and 2 drives only
3. Later version with revision F ROM - has switches - supports types 1, 2 and 6 drives

A bit of a puzzle. The Compaq Deskpro is an XT clone but the use of type numbers for hard drives is an AT thing.
But since the document contains, "Support Discontinued", the document must postdate the Compaq Deskpro, and because of
the use of type numbers for hard drives, written after the release of the AT.

So how do we determine what drives types 1/2/6 are in the ROM of your Compaq WD1002-HX4 ?

Appendix A of various editions of Mueller's Upgrading & Repairing PCs book reveal the drive types in the motherboard BIOS of the various Compaq Deskpro XXX models that came after the original Compaq Deskpro. Through the various models, Compaq are consistent with their use of types 1/2/6. So I think we can confidently say that the types 1/2/6 in your Compaq WD1002-HX4 are:

Type 1: cyl=306, heads=4, WPC=128, LZ=305, S/T=17 (10MB) (eg. ST-412)
Type 2: cyl=615, heads=4, WPC=128, LZ=638, S/T=17 (10MB) (eg. ST-225)
Type 6: cyl=697, heads=5, WPC=128, LZ=696, S/T=17 (30MB) (IMPORTANT: Different to the type 6 used in the IBM 5170)
 
First one would be are the wd1004 and wd1002 rll compatabile?
Click to expand...

No--RLL is a different encoding method and data rate. RLL controllers don't understand MFM and MFM controllers don't understand RLL.

IF a drive can handle RLL (and many can), I've found no difference in reliability from MFM encoding. The boost in performance and storage is well worth the change. RLL works most reliably with high end drives. By "high end', I mean CDC, Priam, Maxtor, Miniscribe, Atasi, etc., not Seagate--although I've got a 30MB Quantum running RLL in an XT that's never given a problem--just dumb luck, I guess.

And Question 3... the wd1006. Has no G=C800:5 in debug, *however* apprently you can access it by G=CC00:5 . I fail and have no luck doing so.
Click to expand...

Check your jumpers. On my WD1006V-SR2s, it's jumper W2 that must not be present for the BIOS to appear. AFAIK, the BIOS position is fixed at CC00:0. SInce the BIOS is pretty much only used for formatting, I couild send you a snapshot of mine, which you could load with DEBUG and execute if worse came to worst.

Hope this helps.
 
Last edited:
modem7 said:
Why not a new thread ?

Your Compaq labeled WD1002-HX4 could be from a Compaq Deskpro, because the second page of a Compaq Deskpro document at ftp://ftp.compaq.com/pub/supportinformation/techpubs/qrg/deskpro.pdf has info on the WD1002-HX4.

It has switch settings and from my reading of the entire page, informs us that there are various releases of the controller:
1. Early version - no switches - supports type 1 drives only
2. Later version with revision E ROM - has switches - supports types 1 and 2 drives only
3. Later version with revision F ROM - has switches - supports types 1, 2 and 6 drives

A bit of a puzzle. The Compaq Deskpro is an XT clone but the use of type numbers for hard drives is an AT thing.
But since the document contains, "Support Discontinued", the document must postdate the Compaq Deskpro, and because of
the use of type numbers for hard drives, written after the release of the AT.

So how do we determine what drives types 1/2/6 are in the ROM of your Compaq WD1002-HX4 ?

Appendix A of various editions of Mueller's Upgrading & Repairing PCs book reveal the drive types in the motherboard BIOS of the various Compaq Deskpro XXX models that came after the original Compaq Deskpro. Through the various models, Compaq are consistent with their use of types 1/2/6. So I think we can confidently say that the types 1/2/6 in your Compaq WD1002-HX4 are:

Type 1: cyl=306, heads=4, WPC=128, LZ=305, S/T=17 (10MB) (eg. ST-412)
Type 2: cyl=615, heads=4, WPC=128, LZ=638, S/T=17 (10MB) (eg. ST-225)
Type 6: cyl=697, heads=5, WPC=128, LZ=696, S/T=17 (30MB) (IMPORTANT: Different to the type 6 used in the IBM 5170)
Click to expand...


Now all I need is the address to access the low level format..
 
Last edited:
modem7 said:
I don't know whether or not you've picked it up yet, but you're answering posts that are almost 2 years old.
Click to expand...

Good grief! :eek: I picked up on Gerry's post and didn't realize that the one before was almost 2 years earlier.

Maybe there needs to be some automatic "archiving" done of threads. This is just plain silly.
 
Chuck(G) said:
Good grief! :eek: I picked up on Gerry's post and didn't realize that the one before was almost 2 years earlier.

Maybe there needs to be some automatic "archiving" done of threads. This is just plain silly.
Click to expand...

Stamping a date on each message would solve that problem!
 
You must log in or register to reply here.
Back
Top