waveform basics

  • Thread starter Thread starter dobro
  • Start date Start date
dobro

dobro

Well-known member
I've got two newbie questions about waveforms when you record digitally.

* What's the significance of a fat waveform? When I record the AKG C-1000 and the Rode NT-1, the Rode has a fatter waveform, although both of them are tracking just short of clipping. Ever since I started using a compressor, the AKG waveform is much fatter and closer in fatness to the Rode waveform.

* What's the significance of differences in the bottom and top halves of a waveform? Sometimes there's a spike in the top half that isn't matched in the bottom half of a waveform. What does it mean when the waveform isn't symmetrical?

I've been tracking since January and haven't paid any attention to this until now...
 
Good question. I thought the positive and negative peaks represented stereo centering... But when you record with one mic, that doesnt make sense... I dont know...
A fat waveform will be more mid-heavy than a thin one... but I dont know why...
A "scooped out" dry sound will be represented by a thin, spikey waveform... again, I dont know why...
Waveforms seem to be misleading to me... I try to ignore them when mixing... Turn your back to the monitor and remix things... Then try it facing the monitor with your eyes closed...
I dont understand 100% of what is going on with visual waveforms... but I do know that they dont seem to correlate with what my ears pick up...
Maybe I'm iggnant.
 
Since the waveform is a composite of the entire signal and most sounds consist of more than one frequency playing at the same time, don't be surprised when you see weird assymetric shapes instead of a pure sine wave.
The top half of the wave is the "push" and the bottom half of the wave is the "pull"
of the motion that generates sound waves.
Or the other way around if your speakers are wired that way.
The "fatness" of the waveform simply indicates a larger proportion of samples that are close to the maximum sample value.
The "loudness" is a function of this and the actual frequencies involved because human hearing has a big midrange hump.
 
Laughing out loud... You know, there are two kinds of responses I like - the ones that answer my question in a way I understand and the ones that answer my question in a way that surprises me and makes me laugh. You both get a prize for the latter. This bbs is fun, but I wish we were actually talking in the same room with real beers in real hands right now.

S8-N: there's all kinds of shit that happens on my screen that I worry about while my ears are saying 'that's okay, let it go'. It's got worse since I got a compressor.

Drstawl - "a larger proportion of samples that are close to the
maximum sample value."

Well, okay, is this larger proportion of samples that are close to the maximum sample value good news then? It *seems* to be good news, and when I look at the waveform it *looks* like good news. Is it better, or just different? If you say, it's 'louder', I'll go home happy. Or is it qualitative?
 
S8-N is right that a low-mid heavy sound will tend to have a fuller look.

doc is correct that a fuller look means that the wave form has more samples closer to maximum. And is correct that the upper part is "push" and the lower is "pull, although, these are not technically correct terms, as doc well knows. He was just keeping it simple.

The upper and lower parts of the wave form are the + and - polarity of the sound. All sound waves have a + and - value that make one cycle. If you start at the center (no value) the wave form starts by going up (possitive value, how high means amplitude, or volume) then crosses back over 0, then goes down (negative value, how low means amplitude). So, the process of going up, then down once is a cycle, or hertz, or frequency. You need to see several in a row to get a feel for the actual frequency, because it means nothing unless you have several in a row. 1 cycle happens to fast for the ear to detect it. So, just one hi hat hit would have several of these up and down movements. It is the accumilation of all of them that makes it sound like a hi hat hit to our ears. Anyway, frequency is how many of these up and downs happen per second. A hi hat hit would be somewhere around 2-12KHz, or 2000-12000 or these up and down cycles per second. Even though a hi hat hit doesn't last a second, it is still a frequency of 2-12KHz per second because if the hit WAS a second long, that is how many up and down movements it would make.

Wave forms and how the correlate to sound? Well, like all the others are suggesting, don't worry too much about it.

The wave form being symetrical? Possibly something to worry about. This could suggest a DC offset somewhere in your recording chain. The most likely culprit is your soundcard. Cheaper ones tend to have pretty bad DC Offset problems. My old yucky Yamaha soundcard suffered from it pretty bad. DC Offset WILL cause distotion to a certain degree. The more of it there is, the more distortion you have. If you soundcard has a option to recalibrate the DC Offset of the converters, use it often. I know that higher quality digital recorders reset DC Offset when the tape is not moving. So the offset is constantly adjusted and updated. Soundcards to not enjoy this feature. But my Lynx One sound card has a converter calibration function that I seldomly need to use. But once a week I use it just in case... :)

But the offset can also be from other things going into your soundcard. A preamp, mixer board output, mixer line in OP amp, compressors, etc.... Quality components means less offset. I have mp3's that have some very severe DC Offset problems, as well as .wav files from others with the same. Cheap units will suffer from it. Cheap tube devices can be particualily bad too, as tubes deal a lot of DC Offset. Usually means a tube going bad, or is not biased correctly.

You will notice that big boy recordings tend to have a fuller wave form when viewed in an editor. You will also notice that it sounds a heck of a lot better too... :) So, there is a correlation between fat looking wave forms and your audio. But, few, if any of us are dealing with the quality of equipment that the big boys have, so expectations need to be lowered.

The reason that the wave form looks different between your two mics suggest the qualitive difference between the two. The Rodes is picking up a lot more of the infomation accurately, thus, fuller wave form.

The day you get your first really killer sounding mix (I mean the one that REALLY sound close to a big boy recording) you will see a very full looking wave form in your editor too.

Ed

[This message has been edited by sonusman (edited 05-23-2000).]
 
The day I get my first really killer sounding mix, you will see a very full looking smile all over the front of this face. Thanks you guys. Thanks, Ed - now I understand what doc and S8-N were talking about.
 
Back
Top