Cumulative Clipping?

[mark4man]

In my sequencing software (Cakewalk Pro Audio 9), I mix several audio tracks into a .wav file (Cakewalk calls the process "Mixdown Audio to File".) The levels are all just right. When I open the .wav file in Sound Forge (v4.5), the waveform is incrediblly fat, with the upper & lower extremes clipped off. Thing is...it sounds fine!, just a (very) little bit muddy, but it certainly doesn't sound nearly as distorted as the waveform looks. It sounds almost identical to playing the tracks simultaneously in Cakewalk. Very little difference. Is this a cumulative effect in graphic representation only? I don't know how this could be, since the waveform is a representation of the audio, but I would hear clipping, right? Is it somehow a graphics adjustment? I would appreciate some input. Thanks.

Sincerely,

mark4man

 

[gnarled]

Well, I'll attempt to answer your question since nobody else has tried yet. So, I'm not sure about this but here goes. Generally when I hear clipping its when I'm recording a at a level that is too high and its overloading the inputs. I never hear clipping if I crank the gain on a track after its recorded. I bet whats going on in your situation is that cakewalk is simply hard-limiting your track before it clips. There is no way the software could mix down a clipping file anyway and still clip - it would have to stop at 0dB since there is no room above that.

When you say the levels are all just right, I assume you mean that the individual track levels never clip. Well, waveforms do add and you could have two tracks that don't clip on their own but when mixed together go over 0dB.

Hope that helped a little.

 

[mark4man]

gnarled,

I opened the problem .wav file in Cakewalk; & it looked the same. This eliminates Sound Forge. I then listened to it at lower volumes; & I can definitely now hear the clipping. Has to be the Buss in Cakewalk (which had been set at 0db.) I'm going to pull back the output gain 1db at a a time to mixdown new .wav files, unit I get a file with little or no clipping. Thanks.

mark4man

 

[SPINSTERWUN]

Or..... instead of "mixing to new track".

You could play the un-mixed down song directly out of your soundcard, then record it 'realtime' into Sound Forge. That would probably eliminate your distortion problems (because you can adjust the incoming signal into Sound Forge.

 

[mark4man]

Spinsterwun,

You know...that's not a bad idea! And it's so obvious (I feel dopey that I didn't think of it.) I'm not even sure of approach (since this would represent a data transfer between apps.) I suppose I could just set my sound card (Audigy) output ports correctly; & record directly (varying input signal strength, as you suggested.) I still intend to pull down the buss in cakewalk a bit, though...cause I can now hear the clipping in the cumulative .wav file. Thanks much; & have a great holiday weekend, man.

mark4man

 

[AlChuck]

I don't really understand how sending your Cakewalk tracks out through the digital-to-analog converters, then back in to the analog-to-digital converters to record the signal with Sound Forge is a reasonable thing to do.

First, would it even work at all unless your soundcard has multi-client drivers?

Second, if you do have a multi-client soundcard, you can just set Sound Forge to record from the soundcard's internal mixer. No routing out and back in required.

Finally, ultimately doing this is more-or-less identical to doing it by mixing to another track within Cakewalk. You can adjust the bus level in Cakewalk just as well as you can adjust the input leel in Sound Forge.

Sound like a process my mother-in-law might have cooked up (she's notorius for introducing all kinds of twists and redirects and stuff into the simplest of errands). Unless maybe I am just missing something here, which I am notorius for...

 

[SPINSTERWUN]

.....Sound like a process my mother-in-law might have cooked up (she's notorius for introducing all kinds of twists and redirects and stuff into the simplest of errands).....

I was just giving mark4man a different solution/outlook to his problem.

I myself would not do it that way, unless I was trying to "fatten" the sound.

Hopefully, it will give m4m a different angle to view the situation.

 

[mark4man]

Fellas,

I utimately fixed it by pulling back the output buss in Cakewalk -1db at a time for each mixdown to .wav, until I had a .wavfile with no clipping (backing up a little bit, I found that I actually could hear the clipping in the original fat .wavfile...the vocal being affected the most.) Thanks for the help, everyone.

mark4man