This is a screenshot of two 32-bit audio files added to Vegas Pro 18.
The audio on the left sounds good after reducing the gain to -30dB. But the peaks look incorrect, they are too small and display clipped.
On the right is the same audio but normalized in another app.
Is there a way Vegas can reproduce correct visual peaks after reducing gain on 32-bit files? (like on the right)
Audio on the left looks clipped.
How did you bring the level down in Vegas?
How did you bring the level down in Vegas?
@RogerS Looks like he did with the event volume, see close to his mouse pointer in the screenshot.
@Ustik Audio is not clipped, only the visual peaks are squashed. Select the audio event and type shift "up arrow". That will increase the visual peaks. However, it will do it for all events.
AFIAK, the max bit depth in VP (File > Properties).is 24 fixed point. For audio that originated from a 32 floating point bit recorder (Sound Devices MixPre II series or Zoom F6 for instance), I peak normalize the files in Sound Forge first if the waveform 'appears' abnormal . Normalizing 32 bit* audio in VP does not work as it does in Sound Forge.
Audio on the left looks clipped.
It is not clipped, just visual misrepresentation.
But it shouldn't look this way, and that's why I created this topic.
How did you bring the level down in Vegas?
@RogerS Looks like he did with the event volume, see close to his mouse pointer in the screenshot.
@Ustik Audio is not clipped, only the visual peaks are squashed. Select the audio event and type shift "up arrow". That will increase the visual peaks. However, it will do it for all events.
It will remain clipped, just in other scale factor of peaks...
AFIAK, the max bit depth in VP (File > Properties).is 24 fixed point. For audio that originated from a 32 floating point bit recorder (Sound Devices MixPre II series or Zoom F6 for instance), I peak normalize the files in Sound Forge first if the waveform 'appears' abnormal . Normalizing 32 bit* audio in VP does not work as it does in Sound Forge.
That's why I asked. I want to buy a Zoom F2 recorder, but converting audio in other software doesn't look adorable to me.
Have you tried setting your volume using the simple Volume FX, it doesn't change the visual peaks.
The Volume Fx worked perfectly to bring down audio that was recorded 30db too hot. However, visually there is no change to the representation of the wav, so it appears clipped. I couldn't find a way to fix this. I added a YouLean loudness meter which lets you see the actual levels and waveform.
I change the volume in Sound Forge mostly because the flattened waveform display in VP is visualy unnerving.
I think this would be a good time for the OP to post his file properties.
The ambiguity for me arises with the description of "32 bit."
Except for W64, all PCM is 32-bit architecture.
On the other hand, 32 bit depth (float or integer) audio is less common than 16, 24, etc. Also important to distinguish from sample rate, which is another independent variable.
Also, loading an original file sample to Drive or Dropbox will assist others trying to help, rather than guessing around the problem.
I recall vaguely from discussions fully two decades ago that the waveform display resolution is very low, so maybe this inquiry is better framed as a feature request, rather than a support issue.
Based on the file names on the OP's timeline it's clearly 32-bit float and the user is using these sample files from SoundDevices.
We're seeing Zoom, Tascam, and SoundDevices come out with reasonably-priced 32-float recorders and mixers now.
I added improving waveform display as a feature request yesterday : )
General
Complete name : C:\24bit_32bit_files\32_bit_float.WAV
Format : Wave
File size : 2.48 MiB
Duration : 6 s 754 ms
Overall bit rate mode : Constant
Overall bit rate : 3 079 kb/s
Producer : SoundDev: MixPre-3 II QB05172031
Description : sSPEED=030.000-ND / sTAKE=032 / sUBITS=$00000000 / sSWVER=4.00.6664 / sSCENE=BBB / sFILENAME=BBB-032.WAV / sTAPE=190826 / sCIRCLED=FALSE / sTRK1=MixL / sTRK2=MixR / sNOTE=
Encoded date : 2019-08-26 15:01:56
Encoding settings : A=PCM,F=48000,W=32,M=stereo,R=48000,T=2 Ch;Ambisonics=off
Producer_Reference : USSDVQB0517203123190826F1S000201
Audio
Format : PCM
Format profile : Float
Codec ID : 3
Codec ID/Hint : IEEE
Duration : 6 s 754 ms
Bit rate mode : Constant
Bit rate : 3 072 kb/s
Channel(s) : 2 channels
Sampling rate : 48.0 kHz
Bit depth : 32 bits
Stream size : 2.47 MiB (100%)
Based on the file names on the OP's timeline it's clearly 32-bit float and the user is using these sample files from SoundDevices.
We're seeing Zoom, Tascam, and SoundDevices come out with reasonably-priced 32-float recorders and mixers now.
I added improving waveform display as a feature request yesterday : )
General
Complete name : C:\24bit_32bit_files\32_bit_float.WAV
Format : Wave
File size : 2.48 MiB
Duration : 6 s 754 ms
Overall bit rate mode : Constant
Overall bit rate : 3 079 kb/s
Producer : SoundDev: MixPre-3 II QB05172031
Description : sSPEED=030.000-ND / sTAKE=032 / sUBITS=$00000000 / sSWVER=4.00.6664 / sSCENE=BBB / sFILENAME=BBB-032.WAV / sTAPE=190826 / sCIRCLED=FALSE / sTRK1=MixL / sTRK2=MixR / sNOTE=
Encoded date : 2019-08-26 15:01:56
Encoding settings : A=PCM,F=48000,W=32,M=stereo,R=48000,T=2 Ch;Ambisonics=off
Producer_Reference : USSDVQB0517203123190826F1S000201Audio
Format : PCM
Format profile : Float
Codec ID : 3
Codec ID/Hint : IEEE
Duration : 6 s 754 ms
Bit rate mode : Constant
Bit rate : 3 072 kb/s
Channel(s) : 2 channels
Sampling rate : 48.0 kHz
Bit depth : 32 bits
Stream size : 2.47 MiB (100%)
Correct, these are the same sample files. Wayform looks clipped in all circumstances...
Thanks for clarifying. I think the problem with providing higher resolution graphics in the waveform preview is already evident in the time it takes to build the previews in their current state. People complain about that a lot.
Maybe the feature request should include a high/low graphic resolution switch in Preferences, which would serve the purpose of either speed or accuracy depending on the user's needs.
I generally don't pay too much attention to the audio graphic preview, since most of my audio ends up in Sound Forge anyway at some point.