| From: | Craig_Daines |
| Date: | 18 Mar 2001 at 21:17:08 |
| Subject: | Re: PowerFlyer & MakeCD (Buffer Under-Run Problems) |
Hi Mike 8-D
Thanks for your comments (below). I list some other tests I have
performed...
--- In amigactive@y..., "Mike O'Hara" <mike@k...> wrote:
> Hi Craig_Daines, on 04-Mar-01 21:59:26 you wrote:
>
> >One reader recommended I placed the CDRW on the same channel on the
> >same channel as my HD (which I have also tried):
>
> Placing the writer on the same channel as the source drive will be
much slower
> than if the source and writer are on different channels. At least,
this was the
> case on my Power Flyer (A1200) and my zip and hardrive.
I tried switching Master / Slave pairs on different channels and that
doesn't cure the problem and restarting 'FastATAPrefs' to inform the
'Fastata.driver' the drives have been switched to different channels,
but that doesn't cure anything :-(
> BTW, try the latest beta (3.2d) of makecd, works fine here. Also try
toggling
> parralel hardware compatibility section of the settings window.
>
> >Burning Speed:
> >This has been set automatically by MakeCD. I am using the default
> >settings (slowest).
>
> Try dropping this down to 1x, i guess.
I've tried that too....
> >Buffer Size:
> >I am using the default buffer size which MakeCD has set up. I've
also
> >ticked 'Ignore Overflow' in the 'Hardware Compatability Settings'
> >since my CDRW recognizes buffer overflow intelligently (the CDRW
has
> >it's own internal 1 megabyte buffer as well).
>
> I set a 32meg buffer simply so I can do other stuff while burning
CDs.
> Try increasing the buffer size to 16meg or so, and seeing what
happens.
Hmmm, I haven't tried that, so it might be the next thing to try.
As it stands write now, I tried fiddling with Mask settings, adding
additional buffers to my partition (where my ISO images are created),
booting with 'No-Startup-Sequence' *then* running MakeCD (to free up
additional CPU time), one reader also recommended adjusting MakeCD's
tooltypes to:-
MAKECD_PRI_READ=7
MAKECD_PRI_WRITE=2
Still I receive 'buffer-underrun' 8-(
> Even if it doesn't solve the problem, it might lead to a solution...
650meg of
> ram is *not* a solution btw. :-)
> >I am at a complete loss to find a possible solution, and have
> >exhausted all other possibilities. Any suggestions would be most
> >welcome. HELP !!!!
>
> SCSI. :)
It should be noted that Hisoft's Squirrel and the majority of other
third party SCSI controllers for the A1200 *do not* offer transparent
DMA (processor time is used during disk transfers). As far as I know,
the *only* card that offers 'true' transparent DMA is a Z3 FastLane,
and I don't own an A4000 :-( SCSI can't be without it's problems
either ;-) From my own observations, the PowerFlyer is *very*
processor intensive compared to C= IDE... try running PCx / Fusion
using 'hardfiles' to simulate virtual PC / Mac HD's! I believe this
problem with regard to excessive CPU usage that the PowerFlyer relies
upon is the root of my problems which are causing 'buffer-underun'
with MakeCD 3.2b...
Many thanks for the suggestions. I'll let you know how I get on...
Kind regards,
Craig Daines
Technical Editor For 'The Crypt' Disk Magazine
Visit 'The Crypt' Homepages at: www.thecrypt.uklinux.net
------------------------ Yahoo! Groups Sponsor ---------------------~-~>
Make good on the promise you made at graduation to keep
in touch. Classmates.com has over 14 million registered
high school alumni--chances are you'll find your friends!
http://us.click.yahoo.com/l3joGB/DMUCAA/4ihDAA/d8AVlB/TM
---------------------------------------------------------------------_->
Quote carefully and read all ADMIN:README mails
Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/