Questions about RLL controllers

[Karl Denninger]

In article <3700022 at eecs.nwu.edu> skrenta at eecs.nwu.edu (Richard Skrenta) writes:
>I've got a Miniscribe 6053 (40 meg) and a Seagate ST-4096 (80 meg) on a
>standard Western Digital MFM controller.  I've heard that switching to
>an RLL controller could boost the available space on each drive by 50%
>as well as increasing the data transfer rate.

Yep.  If your board can handle the increased transfer rate (WD1006-VSR2), or
has a true track buffer, you'll get much higher throughput.  The basic data
rate is 750kbytes/second rather than 500kbytes/second as with MFM.

HOWEVER:  A cheap RLL board will only give you increased capacity, 
	  not performance.  Ask the magic question: 
		"Is it truly track buffered"?

>Are there any dangers associated with using an RLL controller on drives
>only certified for MFM?  Could the controller damage the drive in any
>way, or is it just a matter of mapping a few more bad tracks during the
>format?  I noticed there are some nasty warnings in the Seagate manual
>about voiding your warranty if you use the wrong controller with a drive.

I've never seen a drive that was physically damaged by RLL formatting.  You
may find some, particularly older drives with non-plated media, that won't
work with RLL formatting.   There are a few that will appear to work, but
won't hold up over time; that is best checked out with a thorough diagnostic
or some analog equipment.  Some controllers have on-board surface analysis
software that does a _real_ good job, the WD1006VSR2s come to mind again.
The Adaptec's surface scan is just a read/write test from what I can tell.
If you run a "dos level" surface analysis utility, let it run 48 hours at a
_minimum_ with worst-case patterns before you say everything is ok.  I've 
never had a problem with RLL as long as I do the surface analysis and make 
sure that I've got all the bad areas locked out.

Seagate doesn't like people knowing that many of their drives are the same
in both RLL & MFM configurations (they sell both; RLL is more expensive), 
so they put those nasty warnings in the manual.  In reality, if you were to 
reformat with MFM they'd never be able to tell you ran RLL format......  
The only time it would be an issue is if the drive were to fail 
catastrophically but in such a manner that Seagate could determine the 
format later (but I hear rumors that they don't inspect or check returns 
anyways......)

>Also, where does the extra space come from?  Do RLL controllers increase
>the sectors per track or the number of tracks?  How would I get Xenix to
>recognize the new space?

They run 26 sectors per track instead of 17; this is why some drives don't
work.  The density of information on the tracks is raised by 50% to
accomplish the increased storage; this means the drives must be able to
resolve a "smaller" magnetic domain (and write a smaller one too). 

Most current drives _are_ up to the requirement.  We've been using two of
'em in our Xenix machine now for about a year and a half with no trouble at
all (ST4096 and ST251).

If you're paranoid you can pay Seagate to guarantee that the ST4096 you get
will work with RLL, the only difference is that the model number then is
"ST4144" and you get to pay more money; the same trick can be pulled with a
ST251 (you buy an ST277 instead)  Other than the certification level, there
is no difference between these drives.  :-)

Disclaimer:  We sell track-buffered RLL controllers, as well as MFM boards.
	     Of course I'm biased, but we do put our own data on our
	     internal systems where our mouth is. :-)

My understanding and personal experience is that the following tend to work:

	Seagate:
	ST4096 (late models)
	ST251-x (for any value "x", as long as it's a suffixed drive with
		an "MLC" sticker it should be ok)
	ST225 (some; I wouldn't trust these!)

	Maxstor:  
	XT1140 (most units)
	XT2190 (Nearly 220MB formatted!  Yikes!  Once again, most units)
	RLL certified drives are also available from Maxstor.

	CDC WREN:
	Some Wren IIIs (not all) will work. 

	Priam:
	Most models, the ID-xx series are usually ok, some are certified.

	Miniscribe:
	Most 60xx series work fine (plated media and whitney
	actuators are probably responsible here).  Check older units in
	particular!


--
Karl Denninger (karl at ddsw1.MCS.COM, ddsw1!karl)
Data: [+1 312 566-8912], Voice: [+1 312 566-8910]
Macro Computer Solutions, Inc.    	"Quality solutions at a fair price"