CD-RW's Drive Problems.

[Ronald Maggio]

----- Original Message -----
From: "Gregory J. Neumann" <gjn at certainlywood.com>
To: <nflug at nflug.org>
Sent: Tuesday, March 12, 2002 6:27 PM
Subject: Re: CD-RW's Drive Problems.


> Hi, Ron!
> Just to confuse things, here's my $.02: ;-)
> I'd try the Linux route myself, but be sure to read the HOWTO and follow
the kernel
> / module recommendation carefully!  Mandrake may come w/ the kernel
already pre-
> compiled properly.  mkisofs maks an image on the CD and cdrecord uses RAM
(you can
> specify how much, too!) as the buffer, so buffer underruns are very rare,
unless
> you want to do Quake or something equally CPU intensive while burning.  On
my ol'
> 486 w/ 32 Meg of RAM I used to burn CD's while browsing w/ Lynx and never
had a
> problem.  This is where Linux's superior memory usage really shines! :-D
The only
> problem I've had w/ cdrecord is a bug in the early 2.4.x kernel ide-scsi
emulation.
> IIRC, it was fixed in 2.4.13 or thereabouts.
>
> I'd also try moving the HD's to the primary bus and set up the CD-RW and
CD-ROM to
> the secondary bus.
<---------------------------------snip--------------------------------------
--->
Thanks for the info and to all that helped so far. I thought that you could
not take data from a cdrom and send the data to the writer on the same IDE
channel. I thought that having both reader and writer on the same channel
would cause a signal collision. Even when one is set to slave (reader) and
the other master (writer) I tryed it that way but never had any luck in the
hardware working correctly.
I've run into this from time to time and have always put them as slaves on
different channels to avoid what I assumed was data signal collision, but I
could be total wrong?
So I've got used to the idea of dual masters HDD and dual slave optical disk
drives.

Well I'm going to read what you suggested and try to burn in Linux also as
cybersource pointed out. Sorry the person did not leave/use Sir name. I will
try these out next as suggested:

1. slow the write speed down to the slowest setting, if that works, you can
gradually move up in speed for trials
check the media.

(2. If you are using 800mb, try the 650's and check the write speeds on the
media as well!)

I think this is most likely the problem. Its a 24x10x40 cdrw and I might be
using 2x to 12x media. I just baught some 16x cdr's and will try them. I'll
have to use up the slower speed cdr's on the Acer 2x2x6x The old girl is
sitting on my P1 box with Windows NT Workstation and Mandrake 8.0 with a
Hitchi 16x reader. Have had better luck with this Acer & Hitchi (spelling?)
set up than using the Acer w/win98 & 48x reader. (reader was to fast for the
writer) Talk about bad burns. I wasted more cdr's taying to get it to burn,
but I have no problems with this box that they are placed in now. Although
it takes about an hour and a half to format and burn an average diskcopy. If
this does not work I'll just move it to the AMD box as master on IDE 2
channel and try it there. If I still have any problem then the hell with it
then its a defective product.

Thanks:)

Ron

> Linux cares not at all whether the drive it's on is a master or
> if the partition it's on is set "bootable".  That sort of thing is from
the
> arrogant Microsoft mindset that the only O/S on the system is Microsoft.
I've also
> run into at least one CDRW that refused to work until I moved the CD-ROM
it was
> "bussed"(?) with to a different bus, though.  Seems to me that there's a
potential
> bottle neck, as the CDRW would interfere w/ the ATA-100 (or 33 or 66?)
because they
> can't run at that bus rate.  Things may have changed, but WAY back in the
old days
> (486's were VERY fast then!), the ide bus ran at the speed of the slowest
device on
> the bus, and CD-R(OM)(W)'s are VERY slow compared to any HD.  But back in
the
> 386/486 days, who could tell? ;-)
>