Re: DMA or not DMA...

[spec@vax2.concordia.ca (Mitchell Spector)]

In article pubpc1@library.ucla.edu writes...

>spec@vax2.concordia.ca (Mitchell Spector) wrote,
>in response to my reply:
>>
>>>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
>>>------------------------------------  [side snipping]
>>>@13.75MHz/16K cache             11
>>>@non-Zip speed                  19
>>
>>    That's odd, on my system, which is very close to yours (my Zip is
>>1.25 MHz faster with all cache active, but I doubt there is a major
>>significant difference). I get the following:
>>
>>Time to boot (sec)      with DMA on
>>------------------------------------
>>@15MHz/64K cache                 5
>>@non-Zip speed                  11
>>
>>    Either the RamFAST makes a huge difference or you have a really old
>>and slow SCSI hardrive (my past two drives have been 4500 RPMs and have
>>had an access time of around 11-12ms). I think my new Seagate drive is
>>slightly slower than the Quantum, but I'm getting roughly double the
>>performance over your setup.
>
>Probably a combination of a slow drive and the speed advantage of
>the RamFAST over the Apple HS SCSI card.

    Probably the hardrive itself more than anything else (the number
of rotations per minute, seek time, cache). I seem to recall there
is a minimal difference between the Apple High-Speed SCSI and RamFAST
SCSI under GS/OS (often the same, but at most you would see a 2x
increase). The real advantage comes when using the RamFAST under
ProDOS 8, I heard you could see as much as a 8x increase in speed.

    Of course I've never used a High-Speed card, just the original
Apple Rev C and several RamFAST (rev D) boards, so I can't do any
personal comparison.

>My drive is a Quantum ELS 170MB drive, which was rated among the
>slowest, but most cost effective in a MacWorld or MacUser review
>a few years back.

    That drive is at least 5 years old, I think it has an average
seek/access time of 14-18ms and spins at 3600 RPMs. You can examine
its technical specifications at http://www.quantum.com (look in the
file archives for discontinued drives). I'm sure that accounts for
some of the loss of speed performance compared with my system.

>I wonder what speed you would get with DMA off and Zip on?
>The Apple HS's DMA advantage seems to be paltry compared to the
>benefits of Zip speed.

    Hmmm, I just tried it now (going into the RamFAST's 'options'
menu and toggling DMA to 'no') and I don't see any difference in
performance from the numbers I originally posted. Does that option
really disable DMA?

    DMA is going to show different performance increases depending
on what your doing, in my case booting my hardrive doesn't seem to
make much (if any) difference. Howevering, copying or writing files
will probably show some noticable speed hits with DMA disabled.

    The DMA issue aside for a moment (but for the record, I do _not_
think DMA is an overrated feature)  I think an accelerator makes a
world of difference to hardrive performance. My hardrive virtually
crawls if I switch the GS back to 2.6 MHz, but flies when I put it
back up to 15 MHz. I have a spare RamFAST card kicking around which
I could install in my ROM 01, but I'm not sure it's worth finding a
hardrive for it unless I can repair my TWGS (currently sitting in a
box) or find a cheap Zip GS. :)

Mitchell Spector
spec@vax2.concordia.ca