Problem replicated with
- write caching on the hard drive on and off
- 64  bit processing engine on and off
- FX global bypass checked and unchecked...
 
To replicate, I start playing a clip and while playing the clip I normalize the track.
I have replicated with projects at 24 bit 44.1 and 24 bit 88.2.
I have not replicated the problem doing a straight bounce to tracks, so I am beginning to doubt myself on that. It seems normalizing during playback is the only way I can consistently replicate, which means that it's very easy to avoid the problem.

First of all, I LOVE your re-interpretation of that tune.  Very kewl concept.

I only have had one time where s certain part was out of whack - but it was inadvertently my own step sequencing that caused it for me.

Sorry I cannot give you any help in figuring this out.

On a side note - if you ever need or want any keyboard tracks for any remakes just give me the word, and I would cheerfully donate my time to help you.  Just send a private message and we can go from there.

Question - do you experience the same problem if you just play back one track at a time?  Also, try exporting each track separately, then playing them using Windows Media Player, one at a time, to see if the problems show up that way.  And lastly, try exporting each track again separately, then firing up a new project and importing each track back in - sort of a fresh reload of the same previously recorded material.  Then play within Sonar and see if that did anything different.

I am just tossing ideas out to see if any behavioral changes can be observed, as that may lead to a better understanding as to why this is occurring.  

Bob Bone

It seems normalizing during playback is the only way I can consistently replicate

FWIW, I would never commit to doing something like this during playback - you're simply asking for the sort of trouble you're seeing.

And, just as an aside, why are you normalising anyway?

Bristol_Jonesey >
I'm normalizing to improve the signal to noise ratio when burned to CD (from CD to stereo). That's all. I know you're against it. I have the concrete limiter before it, but if it doesn't use every available bit, I go ahead and normalize before exporting to 16 bit. Maybe just a foolish thing to do.

So, I recognize I was asking for trouble. I'm haven't normally done it, but if it's an option, it should work as expected, I think. But it's no big deal. The first time it happened, I thought it happened on a bounce to tracks but I haven't been able to replicate that over some time now. Also, I don't know if I might have been accidentally playing the project when executing bounce to tracks (not even sure if this is possible).

At this point, I was only replicating because I was on to a bug. Thought it a hole that could be plugged if they get available bandwidth. Perhaps not necessary. More just exploring it. I found it happened to me > I looked into why > I created this thread as I studied it.

I know you think it's never necessary to normalize, but normalizing can be part of gain staging, can't it? Especially if I record at the super low levels people mention, gain staging helps get the feed into your compressors and fx to be reasonable so your dials for setting compression makes sense, right? When I record with reasonable levels, which is most of the time, I don't normalize the tracks themselves before bouncing, but I might normalize the final mix before exporting, assuming the concrete limiter never had to limit anything.

Bob Bone>
The original recordings are fine. I only encountered this on a bounced track. At first, it appeared to have happened on a 'bounce to tracks' but since I have only been able to repeat it reliably using normalize while concurrently playing the clip. I'm totally into trying to collaborate... I'll PM.

Everyone > Thanks for the help. I totally don't want to burn your time unnecessarily and appreciate the time you've given. This is just an oddity and I'm always interested in running down oddities like this just to be helpful to Cakewalk if I can be. This is not a problem for me since I've had an easy time avoiding the problem.

gswitz


Bristol_Jonesey >
I'm normalizing to improve the signal to noise ratio when burned to CD (from CD to stereo). That's all. I know you're against it. I have the concrete limiter before it, but if it doesn't use every available bit, 

How can improve the signal-to-noise ratio using normalize? Normalizing inevitably raises the noise level exactly the same amount as it raises the signal.

I was talking about the signal to noise ratio on the equipment you use to  play back the CD.

What plugins are you using. I recall older versions of Kontakt for instance had to be bounced down without fast bounce enabled as it wasn't able to stream the audio that quickly. Do you have fast bounce enabled if so might be worth trying to bounce with it turned off?

gswitz


I was talking about the signal to noise ratio on the equipment you use to  play back the CD.

My bad. I did not read carefully enough.
I do "mastering" inside the project, and exporting the final mix I use Ozone to get the desired levels.
I hardly ever need normalizing, but I think it's quite an "acceptable" method at places, as it really doesn't do any harm, if you know what/why you're using it for.

@Kalle Ozone is clearly very popular. I haven't purchased it as of yet. I think it is on my long list, but that means it may be years yet. . @Dorism There are no FX in play at the time. Fast bounce is definitely enabled, but I'm Applying Normalization. It doesn't usually ask if you want to do a fast bounce when you are normalizing. . BTW, I am totally happy to let this thread drop. . Thanks to everyone for your help!

Cakewalk refused my bug report on the grounds that normalizing while listening is a bone stupid thing to do and I should know better.

Haha.

:-)

Not exactly their words to me, but probably what they said to each other.