Skip to main content

Hi,
I experienced a major problem working with Sonar, so I wanted to see if any of you had similar kind of problems at some point...
The problem is, that when for example, making a drum track by multiplying (copying) the snare drum sample (clip) in one track and Hi-hat in another (any sound with fast attack / transients will do) the program sometimes destroys (compresses or fades-in) the transients only on the first clip of the audio (the first snare drum sample), and the others play well... The problem is obviously somewhere in the mixing engine (buffering), since it was not present in Cakewalk Pro Audio 8, but CWPA9 & Sonar (even 2.0) have that problem... I've tried everything to solve it (from tweaking all the parameters in the audio section to completely reinstalling the system and putting a different soundcard, MME or WDM drivers), but the only thing that seems to “improve” the situation is lowering the buffer to 20ms in the mixing latency section of Sonar's audio options, and it still doesn't solve the problem completely...

The technical support suggested that I should :

- Download and install the 1.3.1 update from the Cakewalk site – Which I already have done
- Check if there were linked clips in my projects – I did, and there were none...
- Copy the whole project and paste it into a new document – I did, it didn’t solve it
- Select the whole project and then Apply Trimming – I did, it didn’t solve the problem

I'm worried since sometimes the problem is not so obvious & audible, for example when using sounds with slow attacks (like pads & stringy sounds), but it STILL IS present and it drastically spoils the sound... The other thing is that the problem appears somewhere in the process of making or mixing the song – it is not always there when I begin the project. I am in contact with many Cakewalk users working on different platforms and many of them experienced the problem at one level or another... Maybe we are doing something wrong, but I really don’t know what, since I’ve read every darn book about it and I’ve been in the computer-audio world for years now... I’ve also tried to turn off all the software synthesizers, pluginsand any other software running or memory resident – but it did nothing. I still get the “Boom-Css-Chh-Css”-rhythm sound more like “Woom-Css-Chh-Css” =). It doesn't matter what location I start the song from - (start, middle, near end) - the first clip played looses transients - others play well...

So, If any of you have experienced this problem or by some miracle know how to solve it, PLEASE let me know... The current Sonar configuration I'm using is : Asus A7V KT266, AMD Athlon 1.33GHz, 2xMaxtor UDMA133/60GB 7200 RPM, M-Audio Delta 1010 with WDM drivers, G-Force 2 MX400 64MB, 256MB DDR Ram on Win98SE.

Thanx, and all the best
Hypo

Comments

anonymous Wed, 09/04/2002 - 13:26

Many of my friends working on totaly different systems experience it every day... I'm currently waiting for the people from the technical support to answer my e-mail. They said that they had sent the program to QA Lab for testing, and promised that I would hear from them in a couple of days but it's been over two weeks from that now, so it would be a great idea if other users wrote to the tehnical support too, showing them that this is not an isolated case. I think that this is a serious problem, since it is not only present in the realtime monitoring engine, but in mixing (export-mixdown) engine too. If you have any additional info on this (or remembered how you solved it), please let us know ! All the best,

Hypo

anonymous Wed, 09/25/2002 - 17:04

Just a suggestion that might not work: leave some quiet space at the start of your composition - one measure or so. This will insure that your audio engine is up and running in time for the first transient when the composition starts playing. Another possibility is that initial few milliseconds of the first clip is cut off in the pasting or dragging process, perhaps because it is right up against the start of the first measure. Of course, my theory falls apart if the first clip that is degraded is not at the very start of the track. A potential work-around is to use a junk measure at the start of the composition to place clips that you know will degrade, so that the rest of the composition (the useful part) has only undegraded clips. You can then eliminate the first measure upon mixdown, and the first clips will be undegraded. It means you have to start compositions at measure 2, and 12 bar blues will not repeat bar 13, but that is easy to get used to.