cd burning speeds and audio

[Julian Knowles <j.knowles@xxxxxxxxxx>]

At 12:16 PM +0100 10/2/01, bjeannin wrote:
> Hello,
> I'm a bit late on this one, but I thought it could be
> interesting to add a few things I recently heard
> during my audio engineering course :
>
> - If you plan to send a CD-r to a mastering studio,
> you should choose metalazo-cyanine CD-rs (blue ones I think) which
> apparently produce less drops than others. A few good
> names : Verbatim, Taiyo Yudens (hard to find), and Mitsui as
> other people already said.

Its not clear from the post, and correct me if i am wrong, but did 
the audio engineering school really tell you to send a red book audio 
cd???? This is not a good format to submit masters -  the error 
correction is not good... partly due to compromises made to fit the 
length of program that was desired at the time..  Ever wondered why 
you only get 650MB on a CD blank that can fit 74 minutes of stereo 
audio (= 740MB)? The answer is redundant error correction. Plenty of 
it in the CD ROM format, very little in red book audio.. This means 
that media defects etc.. are much more serious on red book audio cd 
than cd rom, so cd rom is a much more robust storage format.

The recommended practice for delivering masters to a mastering studio is

ISO9660 format CD ROM (can be read cross platform). I would make all 
files 'wav' because many PCs seem to behave unpredictably with other 
formats... Macs are fine with everything. Further to this... if you 
have worked at 24 bit resolution, then stay at 24 bit through the mix 
stage and burn the 24 bit master mixes to CDROM. Then let the 
mastering house dither down to 16 bits at the end of the processing 
stage. That way all mastering is done at 24 bit resolution - much 
better. The same principle applies to sample rate conversion, if for 
some reason you have worked @ 48k (for broadcast/DVD whatever...). 
Let the mastering house do it... and finally, never ever normalise 
you files...

Now if you are bypassing a mastering house and are producing a red 
book audio cd as a 'duplication master', then you need to find out 
which speed and media gives you the lowest error count. As i have 
said before, this is not easy to determine without specialist 
equipment. Many people do not know there is  a problem until they get 
a disk rejected by the plant due to unacceptably high block error 
counts.. (even thought the cd plays fine on their own system... Like 
i said in my previous post, it is not a simple question of *media* - 
more of the matching of media to specific drive mechanisms. Verbatims 
may produce great results on one mechanism, terrible on another. 
Conclusions can't be drawn from media alone.

If you are doing a lot of this work, and it is critical..it is 
probably an idea to try and get your hands on a burner which has been 
measured with specific media to produce a very low error count... 
thats why we are using a Yamaha CDE100 - the older plextors also give 
extremely good results for *certain media*.

Better still, if you are a cdr/small volume label doing in house 
mastering on a frequent basis... consider buying an exabyte drive and 
write out your disks in DDP (disk description protocol) format. 
Masterlist CD 1.4 (not current) and some other packages allow this. 
This is the safest and most reliable method of all and the format 
used by most professional mastering facilities.

Hope that covers it and is of some use.. don't want to be didactic 
but the information is really important


--
**************************
J u l i a n  K n o w l e s
Lecturer/Co-ordinator Music Technology
Course Co-ordinator Electronic Arts
School of Contemporary Arts (Music), University of Western Sydney
Social Interiors web: http://www.geocities.com/socialinterior