How can I remedy crackling / static / popping in rendered audio?

  • Thread starter Thread starter alaskatroll
  • Start date Start date
A

alaskatroll

New member
Hey guys, I just discovered this forum! I use Live, and I haven't had much luck on the forums on the Ableton website. I've been struggling with a particular issue.:

Almost every time, when I try to export a project and render to disk, although it sounds perfect when playing in Ableton 8.2.7, the exported audio will have crackles and bits of staticy noise around the louder parts. I've already got limiters and compressors set up right, and I've tried recording all the midi tracks to one audio track and just rendering that track, still no luck. Anyone know how I can fix this?
My in/out sample rate is set to 44100, buffer size 512 samples.
My Export settings are normalize on, wav, 48000, 24bit, no dither.
I've tried tweaking the dither, bitrate, and sample rate, but haven't gotten anything to work. There are still crackles and static around the peaks in the rendered WAV. I tried routing the output into an input on my usb audio interface, and recording through audacity, but that made the problem worse.
I'm puzzled that the exported audio is no good when the audio coming straight from Live is just fine. Is there something I'm missing, or something I can try?
I'm running it on a Macbook pro, 2.26 GHz intel dual core, 2 GB ram
 
I think I know what crackles you're talking about. It's only happened when I exported audio with normalize on so I don't export with it on anymore. So far I haven't had any issues exporting with normalize off.
 
Normalize is probably not doing you any practical favors. Especially if it's peak normalizing. That's the same as pulling your fader up or down until your audio peaks are where you want them. RMS normalization is somewhat like compression but a dumber type of compression, and regular compressors are dumb enough as they are. Compressors at least have user input for different parameters, whereas RMS normalization does the same thing without any regards to what you want.

With all that said, I don't know that that's necessarily your problem in this case. But it very well may be. It sounds as if you are clipping your digital to analog converter. It's possible that Ableton keeps our audio below -0dbfs due to a limiter, or by design. Likely due to a limiter. But when you use a separate audio player, that audio player is taking your -0dbfs signal and playing it louder, and anything that gets above that will clip.
 
"I've already got limiters and compressors set up right..."

I'm willing to bet not.

"recording all the midi tracks to one audio track and just rendering that track..."

This is NOT a good practice. Keep in Mind that a MIDI signal is digital information and nothing else. If you are playing back multiple MIDI tracks at the same time, your CPU is allocating a certain amount of resources per track instance. More so if you are recording all of them. I would suspect that this is where a major part of your problem lies. I would recommend converting each MIDI track to audio first, then rendering those. Why are you consolidating them, anyway? Remember, once you've combined more than one track, you can't make any more adjustments to those sounds as individual sounds.

"My in/out sample rate is set to 44100, buffer size 512 samples..."

I run my daw on a desktop with similar specs as yours. You can set your sample rate to 48kHz and you should be pretty good. I don't Think the buffer size is an issue unless you're noticing latency.

"My Export settings are normalize on, wav, 48000, 24bit, no dither..."

Don't normalize. It's a specialty tool that, to my Understanding up to this point, is not necessary as a standard practice. You'll find that most mixers GENERALLY export at 41.1Hz and 16-bits because that is the standard format that CDs and .mp3 are sold at and played back at at the consumer level.

The way you hear audio in your DAW is NOT an exact representation of what your final product is going to sound like elsewhere. Learning to mix and finding the universal sound is a process that one Learns over time.

Try the above first and see what happens.

Peace.
 
Last edited:
^^ see I had a feeling normalize was a bad idea, thanks for clearing that up for me! Very well put btw
 
Back
Top