BlixYZ
Max Output Level: -74 dBFS
- Total Posts : 805
- Joined: 2010/12/31 16:45:54
- Location: Barrington, NJ
- Status: offline
SOLVED*** Perplexing question regarding audible vs. inaudible clipping?
When I changed the audio settings from 48khz to 44.1khz the problem went away. Now there is the headroom I am accustomed to. What's weird is that in my profire 2626 settings, the sample rate is set to 48khz. In any event, I'm happy :) My new setup is kicking ass. I am in the process of switching to a new computer. For many years I have noticed how forgiving sonar is when transients "clip"; there is no audible distortion for an occasional peak at or even above 0.0. When using my new machine yesterday, I very quickly found out that a peak at 0.0 caused VERY AUDIBLE distortion ) like I haven't heard in YEARS of using sonar). My level settings, hardware, mic, outboard gear, these things were all set in the same way I have countless time before. Within Sonar, some (audio) settings may not be identical- I'm not entirely sure. I was using 32bit Sonar (I have only just started to use 64bit and only on big projects that need full RAM usage). This particular project was started on my old machine and we were re-cutting vocals on the new computer. THE OLD VOCALS RECORDED ON THE OLD MACHINE HAD THE SAME 0.0 CLIPPING WITHOUT ANY AUDIBLE DISTORTION!!!! How? What? The new and old vocals were side by side looking completely identical but the new ones had loud digital distortion at or above 0.0. I tried disabling all plugins and it still happened. Anyone know what is going on here? Seriously- I have wondered for years how sonar lets you get away with slight transient clipping with no consequences.
post edited by BlixYZ - 2012/08/01 23:21:10
James W BlixYZ Recording Studio BlixYZ Records Audient ASP800 thru Focusrite Saffire Pro 40 Mackie Control Universal + C4 Yamaha HS50's plus Matching Sub, Tannoy 501a Blue Baby Bottle, AT 4050, Neumann TLM 103, etc. UA 610, Focusrite/ART/Neve 2CH. Windows 10
|
Bristol_Jonesey
Max Output Level: 0 dBFS
- Total Posts : 16775
- Joined: 2007/10/08 15:41:17
- Location: Bristol, UK
- Status: offline
Re:Perplexing question regarding audible vs. inaudible clipping?
2012/07/17 09:26:15
(permalink)
What are you looking at, your tracks or your main outs? If it's your tracks, it doesn't matter how far into the red you go - you won't clip them individually, but if you clip your mains, the result will sound ugly & harsh AT BEST! The bigger question is, why are you recording so hot? You can safely afford to leave 12dB of headroom in a 24 bit system (you're recording at 24 bit right?). There's no need to get anywhere near clipping on your tracks
CbB, Platinum, 64 bit throughoutCustom built i7 3930, 32Gb RAM, 2 x 1Tb Internal HDD, 1 x 1TB system SSD (Win 7), 1 x 500Gb system SSD (Win 10), 2 x 1Tb External HDD's, Dual boot Win 7 & Win 10 64 Bit, Saffire Pro 26, ISA One, Adam P11A,
|
g_randybrown
Max Output Level: -40 dBFS
- Total Posts : 3522
- Joined: 2003/12/24 11:30:04
- Location: Las Cruces, NM, USA
- Status: offline
Re:Perplexing question regarding audible vs. inaudible clipping?
2012/07/17 09:33:38
(permalink)
Can you see squared off tops on the wav graphic?
G. Randy Brown Windows 10, 64 bit, PlatinumIntel Core i7-3770S Asus P8Z77-V LK mobo 4X8GB Corsair XMS3 memory 500 GB Crucial BX100 SSD (OS)two WD Black 1 TB HDDSAPPHIRE DUAL-X 100314-4L Radeon HD 6970 2GB 256-Bit GDDR5 Presonus AudioBox 22VSLyoutube.com/crystalclearnm
|
BlixYZ
Max Output Level: -74 dBFS
- Total Posts : 805
- Joined: 2010/12/31 16:45:54
- Location: Barrington, NJ
- Status: offline
Re:Perplexing question regarding audible vs. inaudible clipping?
2012/07/17 12:56:22
(permalink)
I am looking at the waveform and at the track meter The waveforms are squared off if you zoom in on the transients in BOTH THE OLD CLIPS AND THE NEW but only the new clips distort audibly (loudly and unpleasantly as a matter of fact). While the meters show red as the signal reaches 0.0, it does't "hold" because it hasn't exceeded the threshhold. the master bus shows no clipping
|
bitflipper
01100010 01101001 01110100 01100110 01101100 01101
- Total Posts : 26036
- Joined: 2006/09/17 11:23:23
- Location: Everett, WA USA
- Status: offline
Re:Perplexing question regarding audible vs. inaudible clipping?
2012/07/17 12:59:41
(permalink)
What Jonesey is alluding to is the difference between floating-point versus integer data. Within SONAR, data is represented as floating-point, which means that there is practically no upper limit for levels. 0db is just a reference, and exceeding it does not cause distortion. Your audio interface, however, can only handle integer data. For it, 0db isn't just some arbitrary reference, but rather an absolute limit that literally cannot be exceeded. Attempting to do so will result in clipping and very nasty-sounding distortion. (You can actually tolerate a few sequential clipped samples before it becomes audible, but since it's a fine line between what you can and cannot hear, it's best to simply avoid overs altogether.) This is why exceeding 0db sometimes results in distortion and sometimes it doesn't. Right up until you send the signal to the main outs, it's safe to exceed 0db. But if you're clipping at that point, you risk audible distortion. If you're just a db or two over the limit, you can simply lower the level going into your master bus using the trim/gain slider or the input fader on your bus compressor/limiter. However, it's a much better practice to keep your track levels low enough that when the mix hits the master bus you've got at least 6db headroom, and preferably 12db. This is true whether you're doing your own mastering or not. None of this applies if the track was distorted as it was recorded, though. If it was your audio interface or mic preamp that was distorting on the way in, then you're screwed. I think that's why g_randybrown asked if the waveform looks clipped, which would indicate distortion that happened during the initial recording.
 All else is in doubt, so this is the truth I cling to. My Stuff
|
g_randybrown
Max Output Level: -40 dBFS
- Total Posts : 3522
- Joined: 2003/12/24 11:30:04
- Location: Las Cruces, NM, USA
- Status: offline
Re:Perplexing question regarding audible vs. inaudible clipping?
2012/07/17 13:03:27
(permalink)
BlixYZ I am looking at the waveform and at the track meter The waveforms are squared off if you zoom in on the transients in BOTH THE OLD CLIPS AND THE NEW but only the new clips distort audibly (loudly and unpleasantly as a matter of fact). While the meters show red as the signal reaches 0.0, it does't "hold" because it hasn't exceeded the threshhold. the master bus shows no clipping There are programs (ie Sound Forge) that will help somewhat but Jonesy has the best advice so far...don't record anywhere near that hot (as much as you probably don't want to hear it).
G. Randy Brown Windows 10, 64 bit, PlatinumIntel Core i7-3770S Asus P8Z77-V LK mobo 4X8GB Corsair XMS3 memory 500 GB Crucial BX100 SSD (OS)two WD Black 1 TB HDDSAPPHIRE DUAL-X 100314-4L Radeon HD 6970 2GB 256-Bit GDDR5 Presonus AudioBox 22VSLyoutube.com/crystalclearnm
|
pwal
Max Output Level: -46 dBFS
- Total Posts : 2909
- Joined: 2004/08/24 07:15:57
- Status: offline
Re:Perplexing question regarding audible vs. inaudible clipping?
2012/07/17 13:22:23
(permalink)
is it an old 32 bit project in x64 sonar, or are you loading it in 32? could it be a driver difference? (between win32 & win64)
|
BlixYZ
Max Output Level: -74 dBFS
- Total Posts : 805
- Joined: 2010/12/31 16:45:54
- Location: Barrington, NJ
- Status: offline
Re:Perplexing question regarding audible vs. inaudible clipping?
2012/07/17 13:25:05
(permalink)
I don't mind recording less hot, and I'm aware of the difference between audio that was recorded too hot and has digitalclipping and audio that is turned up too high and is clipping the meters in sonar (but is not actually compromised) What I'm perplexed about is the obvious differnce when using one computer vs. another when all other variables are the same (so far as i know). These two waveforms are BOTH squared off on some of the transients but one is magically fine sounding (as it's been in my experience for many years) and the other has the typical "digital distrotion" sound that cannot be confused with analog distortion. It's seriously bugging me out. It's not a problem, per se, since i can just chill on my levels- it's just that I've been getting away with it for so long and now it's different and no one can explain why or even really understand what i mean! Am i the only one who has been getting away with occasional transient clips that do not distort? It's been like a benevolent, magical limiter in my studio and I miss it!
|
g_randybrown
Max Output Level: -40 dBFS
- Total Posts : 3522
- Joined: 2003/12/24 11:30:04
- Location: Las Cruces, NM, USA
- Status: offline
Re:Perplexing question regarding audible vs. inaudible clipping?
2012/07/17 13:25:12
(permalink)
pwal is it an old 32 bit project in x64 sonar, or are you loading it in 32? could it be a driver difference? (between win32 & win64) Good call...but what about the squared off tops?
G. Randy Brown Windows 10, 64 bit, PlatinumIntel Core i7-3770S Asus P8Z77-V LK mobo 4X8GB Corsair XMS3 memory 500 GB Crucial BX100 SSD (OS)two WD Black 1 TB HDDSAPPHIRE DUAL-X 100314-4L Radeon HD 6970 2GB 256-Bit GDDR5 Presonus AudioBox 22VSLyoutube.com/crystalclearnm
|
BlixYZ
Max Output Level: -74 dBFS
- Total Posts : 805
- Joined: 2010/12/31 16:45:54
- Location: Barrington, NJ
- Status: offline
Re:Perplexing question regarding audible vs. inaudible clipping?
2012/07/17 13:26:20
(permalink)
it was recorded in sonar e 32 on a vista machine and then updated in sonar e 32 on a w7 machine. weird, right?
|
pwal
Max Output Level: -46 dBFS
- Total Posts : 2909
- Joined: 2004/08/24 07:15:57
- Status: offline
Re:Perplexing question regarding audible vs. inaudible clipping?
2012/07/17 14:52:54
(permalink)
g_randybrown pwal is it an old 32 bit project in x64 sonar, or are you loading it in 32? could it be a driver difference? (between win32 & win64) Good call...but what about the squared off tops? the squared-off tops could be played back differently in different environments(drivers)? weird for sure... but when the going gets weird, the weird turn pro  maybe you can check the squared-off wavs on another machine?
|
Bristol_Jonesey
Max Output Level: 0 dBFS
- Total Posts : 16775
- Joined: 2007/10/08 15:41:17
- Location: Bristol, UK
- Status: offline
Re:Perplexing question regarding audible vs. inaudible clipping?
2012/07/17 15:14:00
(permalink)
The waveforms are squared off if you zoom in on the transients in BOTH THE OLD CLIPS AND THE NEW All I can really think is that it's your old system that's the anomaly here - not your new one. I would expect squared off clips to sound distorted
CbB, Platinum, 64 bit throughoutCustom built i7 3930, 32Gb RAM, 2 x 1Tb Internal HDD, 1 x 1TB system SSD (Win 7), 1 x 500Gb system SSD (Win 10), 2 x 1Tb External HDD's, Dual boot Win 7 & Win 10 64 Bit, Saffire Pro 26, ISA One, Adam P11A,
|
bitflipper
01100010 01101001 01110100 01100110 01101100 01101
- Total Posts : 26036
- Joined: 2006/09/17 11:23:23
- Location: Everett, WA USA
- Status: offline
Re:Perplexing question regarding audible vs. inaudible clipping?
2012/07/17 18:06:47
(permalink)
The waveform could have a squared-off appearance and still sound OK. It depends on how the clipping occurred. If it was the result of overdriving a mic preamp, it might sound fine because the artifacts generated by analog clipping are harmonically related and therefore less audible and less annoying. Artifacts generated by digital clipping, OTOH, are not correlated to the original signal, making the distortion more audible and more objectionable, and may also cause aliasing, which is even more obvious and objectionable.
 All else is in doubt, so this is the truth I cling to. My Stuff
|
BlixYZ
Max Output Level: -74 dBFS
- Total Posts : 805
- Joined: 2010/12/31 16:45:54
- Location: Barrington, NJ
- Status: offline
Re:Perplexing question regarding audible vs. inaudible clipping?
2012/07/17 23:25:32
(permalink)
yes bitf;ipper, that is what i'm thinking too- but the mic, pre, levels, vocalist, compressor- everything was the same and the waveforms look the same but they were done on different computers......... I promise to post the answer if i even discover it!
|
BlixYZ
Max Output Level: -74 dBFS
- Total Posts : 805
- Joined: 2010/12/31 16:45:54
- Location: Barrington, NJ
- Status: offline
Re:Perplexing question regarding audible vs. inaudible clipping?
2012/08/02 00:04:16
(permalink)
SOLVED! When I changed the audio settings from 48khz to 44.1khz the problem went away. Now there is the headroom I am accustomed to. What's weird is that in my profire 2626 settings, the sample rate is set to 48khz. In any event, I'm happy :) My new setup is kicking ass.
|