Subject:Clipping confusion

Posted by: cyberbeat

Date:1/15/2008 11:56:46 AM

The other day I had a standard 16-bit stereo wave file loaded in Sound Forge 8. According to the statistics, the maximum sample value (DB) was -0.010 for each channel. However, when I used the find feature under Tools/Find/clipping, it found a small section at the beginning of the file which was clipped. So, if there is any clipping, wouldn't the max DB be 0 or higher? I'm wondering how the statistics could show the max sample as being under 0 DB, but still have clipping.

Subject:RE: Clipping confusion

Reply by: cyberbeat

Date:1/16/2008 8:24:28 PM

Doesn't anyone know the answer to this. Maybe I didn't explain it well enough?

Subject:RE: Clipping confusion

Reply by: Grazie

Date:1/17/2008 12:32:42 AM

Yes, I understand WHAT you are saying, but I don;t know WHY it is happening beyond guessing:

it found a small section at the beginning of the file which was clipped.

Is it possible that the this small section is just TOO small to figure on the SF stats RADAR? And maybe there is a setting tat would make it more sensitive? Dunno.

Like I said, I understood WHAT you said, but I don't have a knowledgeable, definitive, user explanation. Maybe when people don't have an answer they don't wish to come here and say so, and potentially clog-up the bandwidth? Sorry guy . . .

Oh yeah, what do you mean by: "I'm trying not to get rid of SF completely, . . " - Maybe SONY would like you to ask them this question before you make any "further" decisions. Yeah?

Grazie

Subject:RE: Clipping confusion

Reply by: Steven Myers

Date:1/17/2008 2:41:29 AM

I think the Find Clipping preset is actually just looking for levels of 0dB, not clipping, although the two may coincide.
Your supposed max level is .01. A hundredth of a dB difference is pretty dang small. Negligible, IMO. Perhaps the SF software is not sensitive enough to be accurate in that range.

Take a look at the peaks that supposedly are clipped. Do they look flattened for more than two samples in a row? If not, don't worry, be happy.

Subject:RE: Clipping confusion

Reply by: ForumAdmin

Date:1/17/2008 7:27:54 AM

1) Was there a time selection when you ran Statistics? Statistics only analyzes the selected portion.

2) Was the threshold less than 0 dB in Detect Clipping? If so, it will detect any clip-like plateau above the threshold.

3) Have you tried zooming in and looking at the sample values themselves? Try opening the Levels toolbar and see what the actual sample values are.

J.

Subject:RE: Clipping confusion

Reply by: cyberbeat

Date:1/17/2008 10:38:52 AM

Thanks for the responses. I apologize for coming across so rude in my last message. I just got really frustrated.

>1) Was there a time selection when you ran Statistics?

Actually, I had everything selected.

>2) Was the threshold less than 0 dB in Detect Clipping? If so, it will >detect any clip-like plateau above the threshold.

I used the standard 'Clipping' system preset under Tools/Find.

>3) Have you tried zooming in and looking at the sample values >themselves?

I did this and didn't notice anything 0 or above. There wasn't any audibly noticeable clipping either, so I'm not going to worry about it. I ended up lowering the volume in that one spot and now all is good.