PDA

View Full Version : GoldWave audio editor - performance? Comparably priced alternatives?


Todd H.
July 31st 03, 07:38 PM
Greetings,

My question is partly a Goldwave question, and partly a "do other
editors work faster?" question.

I'm demoing Goldwave 5.0 on a 1.7GHz 512MB Intel machine, and was
disappointed that it's not faster than it is in editing. My last
experience was with Goldwave 4.x on an Athlon 800/256Mb machine, and
frankly while I do see a speed boost, I don't see as much of a speed
boost from my new hardware as one might think. It feels maybe 40%
faster at most, and some operations still take quite a while.

My application is editing stereo 44.1khz WAV files around 700Mb in
total size from live recordings, applying some compression, maximizing
volume, and splitting them using cue points. Gone with my new
machine/v5 is the "deflashing to disk" thing that took so long in v4
and the old machine.

However, compression, maximizing volume, and saving still take a
pretty long time (same ata100 ide drives in both machines, both on
ata100 controllers)...and what I found interesting is that the
Goldwave program never seems to use more than a small fraction of my
available physical RAM according to task manager. CPU is about 75-80%
during the compression and maximization operation, so it doesn't
appear to be limited by CPU. I'm assuming I'm waiting on disk i/o.

Is anyone aware of a way to have Goldwave use more RAM (assuming that
this will speed things up considerably!), or recommend an audio editor
that is comparably priced, but leverages RAM better to give a faster
editing experience. :-) If CoolEdit Pro is hte answer, and performs
quickly, then maybe that's what I need to do...

Best Regards,
--
/"\ ASCII Ribbon Campaign | Todd H
\ / | http://www.toddh.net/
X Promoting good netiquette | http://triplethreatband.com/
/ \ http://www.toddh.net/netiquette/ | "4 lines suffice."

Todd H.
July 31st 03, 10:15 PM
(Todd H.) writes:

> Greetings,
>
> My question is partly a Goldwave question, and partly a "do other
> editors work faster?" question.
>
> Is anyone aware of a way to have Goldwave use more RAM (assuming that
> this will speed things up considerably!), or recommend an audio editor
> that is comparably priced, but leverages RAM better to give a faster
> editing experience. :-) If CoolEdit Pro is hte answer, and performs
> quickly, then maybe that's what I need to do...

I blind copied the author on this question...and lo and behold he
responded within an hour and indicated these very helpful tidbits:

gw> You can change that by using the RAM storage option under the
gw> "Options | Storage" command. However, you'll need a lot more RAM
gw> installed if you'll be working with 700MB files. You'll probably
gw> have to set the Undo levels to zero as well. In general, you'll
gw> need at least twice as much RAM as the largest file size before
gw> things become faster (1.4GB), otherwise the bottleneck is the hard
gw> drive and it does not matter how fast the CPU is.
gw>
gw> Are you using v5.04? If so, try v5.05, which speeds up saving
gw> slightly.
gw>
gw> Chris



--
/"\ ASCII Ribbon Campaign | Todd H
\ / | http://www.toddh.net/
X Promoting good netiquette | http://triplethreatband.com/
/ \ http://www.toddh.net/netiquette/ | "4 lines suffice."

Roseb441702
August 6th 03, 03:38 AM
It's hard to say. I have both programs and Cool-Edit seems to be the faster of
the two but not by much.


Rose
***************************************
"Give A ShoutOut To The World!"
Put Your Voice Message Online at:
The ShoutOut Page
http://members.aol.com/Roseb441702/shoutout.htm

Clive Backham
August 6th 03, 08:54 AM
On 06 Aug 2003 02:38:51 GMT, (Roseb441702)
wrote:

>It's hard to say. I have both programs and Cool-Edit seems to be the faster of
>the two but not by much.

That's interesting. I have both, and I'd say that on balance, if a
process is CPU-bound (eg. parametric eq), then Goldwave tends to be
faster that CoolEdit. (I don't know, but would guess, that CoolEdit
might use more accurate algorithms that require more calculations).

Disk-bound processes (eg. normalisation) tend to run about the same
speed on both (which you would expect).