Re: DMA or not DMA...

[pubpc1@library.ucla.edu]

nathan@visi.com (Nathan Mates) wrote,
in response to my reply:
>>True.  The point is, however, that in random read/write, a major
>>part of the time is taken to position the read/write head over
>>the platters.  Thus, in such operations the speed gained by DMA
>>is only a fractional gain (since the majority of the time is still
>>spent on head positioning).  In sequential access, head positioning
>>time is mostly taken out of the equation, so the DMA speed gains
>>are more apparent.
>
>   Looks like you haven't read a thing I've said, so blinded are you
>with your "sour grapes" covering up for your choices. I provided hard
>numbers on DMA On/Off booting, with facts as to the nonlinearity of
>that. You shoot back with irrelevant hot air.
>
>   Scott, get your head out of wherever it is, and listen to the
>facts: DMA HELPS A LOT.

Nate, I was replying to your statement that I "got it all wrong as
to what DMA helps" in regards to random versus sequential access.
It helps more in sequential than random is my contention.


>   I'll say it again: 25% slowdown on booting IS significant, without
>DMA. Adding 5 seconds to the boot (18->23 seconds) is based on one
>thing and one thing only: no DMA. I don't care if you think the HD is
>seeking a lot, it's STILL got to transfer that data around,
>eventually, and that's where DMA helps a damn lot.

I read you.  What I am saying is that it is going to help _more_ in
a sequential access situation than a random access situation.


>   Judging by the amount of used memory, probably somewhere between
>3/4 and 1.5MB was transferred from my HD to the GS's memory. It
>matters not what speed that data reached my HD card; the transfer took
>longer. Significantly and noticeably longer.
>
>   Scott Gamon, if and when you have some hard numbers, not vaporous
>bull**** to try and defend an stupid position you've staken out for
>yourself, then you might be able to speak on this subject. Until then,
>stop living in your fantasyland that you can lose DMA without any
>noticeable performance hit.

Since you don't believe me when I said that I noticed no difference,
I opened up the GS and ran some tests, with DMA off and flipping the
switch of the Apple HS SCSI card to turn DMA on (powering down each
time first).  I did this with 13.75MHz/16K cache acceleration and
without.  Here are the results of the shift-boot times measured from
the moment the "Welcome to the IIgs" screen popped up to the time
the Finder appeared.  My copy of System 6.0.1 was installed after the
volume had been used for a while, so there was fragmentation.


Time to boot (sec)      with DMA on     with DMA off
-------------------------------------------------------------------
@13.75MHz/16K cache             11              12
@non-Zip speed                  19              21


I ran the tests, booting three times in each case to make sure I was
measuring a real and reproducable difference.  There, I teased out
that 1 second difference, something that's _not_ noticeable in
everyday use.

-Scott G.