31 Mar
2015
31 Mar
'15
3:41 a.m.
On 2015-03-31 01:57, Paul Koning wrote:
Right, right and right. Except I don't know if the servo mechanism
necessarily is marginal. The RL drives are "funny" in that there aren't
even any absolute seeks. When you seek on those drives you simply tell
it "seek n tracks in or out". The value of "n" is something you need
to
calculate based on what your current track is, and what track you want
to get to. And when the seek is done, you repeat if you detect that you
are not on the right track. The drive itself have absolutely no checking
if it got to the right track or not, since it don't even know what the
"right" track is.
And a switch of head select does an implicit seek to whatever track is
closest to the other heads, which means a head switch can cause the
drive to move to a different track. So you always need to do a seek
cycle after a head switch.
While disks would be formatted with the tracks on both side of the
platter aligned, there is no guarantee that the heads in the drive are
actually perfectly aligned, which is why you get the head movement on
head switch.
Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
On Mar 30, 2015, at 7:50 PM, Christopher Parish
<christopher.parish at parishcomputers.com> wrote:
...
Second, I've noticed that the drive seems to mis-seek on occasion. I command the
drive to walk forward or back a single track, and the heads move but sometimes land on the
same starting track. This results in duplicate data for the next 10kB or so, and is
heavily pack dependent. Some packs don't exhibit it, others do. I imagine this is
related to the drive/pack runout condition described in the manual. Until I figure out
more, my plan is to add additional verification to the incoming data, checking the track
it landed on and re-commanding the difference if necessary.
As I recall, RL01/02 have embedded servo (one of the first DEC drives to do so). That
would suggest you have a marginal servo mechanism. As for wrong data, isn?t there a track
(cylinder) number in the headers? I would expect there to be one, so a wrong seek should
be detectable.