PE> I may have confused the issue by running HPFS386 on and off though.

RS> You sure like living dangerously dont you ?

PE> It's not the same version of HPFS386 that screwed my data many moons ago,

Thats a separate argument to the risk of just using it with what its not meant
to be used with tho. Let alone with what its been adequately tested with.

PE> and I sort of have a complete backup at the moment.

With a lot of emphasis on the sort of. You may well have the
data that matters in more than one place, but its likely that
both of those are exposed to potential risks with the HPFS386 too.

RS> And you STILL leave out half the vital info,
RS> you dont say which version of OS/2 you are using.

PE> 3.0.

And are you saying that you have the appropriate latest HPFS386 for that
too or was it one intended for use with one of the earlier versions etc ?

PE> I remembered that I got those sorts of results before,
PE> and didn't complete the investigation, but did remember
PE> the ballpark I had got to.  That was sqpackp.

From memory tho you did always see a
significant improvement with HPFS386 in use.

PE> I do not yet have a theory to explain the observed behaviour.

RS> Until you spell out the full OS2 version and HPFS386 story,
RS> the most likely explanation is just that its showing that
RS> the HPFS386 is producing some weird results with that OS/2.

PE> Like I said above, I had the same problem, ages ago.

You didnt say that above at all on the performance relativitys.
Or maybe you think that para says that. Classic Paul Edwards cryptic.

PE> As for the HPFS386 story, what do you want to know?  I installed
PE> HPFS386.ZIP followed by LS30*.*, both available from here.

The most obvious question is what they
are supposed to be used with OS/2 wise.

RS> Most likely the problem is just that that HPFS386 is fucked in that config.

PE> Fucked doing what?

Caching, to give a better result than the plain HPFS.

PE> If I knew what was screwing it up, I
PE> could probably adjust some parameters.

Welp, my recollection was that you previously got a dramatic improvement
with it and that you binned it for reliability reasons. I dont recall
you ever said it gave the same lousy performance that this config does.

So if you are only getting lousy performance now, thats most likely due to
some interaction quirk with an OS/2 version it isnt meant to be used with.

Have you tried other HPFS/HPFS386 tests with the same
cache size, and never seen HPFS386 better ?  If so its
unlikely to be a simple parameter adjustment problem.
@EOT:

---