Topic: SBE correction on 24/48

[pool]

Is there a programm (windows) which performs Sector Boundery errors fixing (forward fix or pad) directly on 24/48 wave file?

Regards

[pool]

PS
I can do it via compressing to flac and vs..
I'm ytrying to avoid this by working on raw 24/48 directly

[bgalizio]

You don't need to worry about SBEs unless you're doing 16/44 CD burning. In that case, you don't do the SBE correction until you've already resampled and dithered.

[JasonSobel]

if you track out a 24/48 (or a 24/96) file with CD Wave, and then do a batch resample/dither to 16/44.1, your files should be clean and not have any SBE's.  I like to double check with SHNtool after the resample.  works every time for me.

[scb]

there is no such thing as a sector boundary on a 24 bit file

[pool]

there is no such thing as a sector boundary on a 24 bit file

1) Thanks for the replies. however there IS such thing as sbe on a 24 bit file. If you stop a recording and restart lets say on the MT, all files will have SBE at the end.

2) checking SBE is simple- copy this string  and go
foobar>prefs>display>title formatting >playlist> make space at end of existing strings and paste. Tracks with SBE will have red numbers at end of track lenght.

3) I need to work on 24/48 so that i remove the ebe before I join the 2 wavs as in No 1.



My aim to work on

[scb]

there is no such thing as a sector boundary on a 24 bit file

1) Thanks for the replies. however there IS such thing as sbe on a 24 bit file. If you stop a recording and restart lets say on the MT, all files will have SBE at the end.

A sector boundary refers to the boundary of sectors on a CD.  a cd sector (also called a frame) is 1/75th of a second, or 588 samples.  A "sector boundary error" refers to a track that has a total number of samples that is not a multiple of 588.  It has an incomplete final sector. 

So there is no such thing as a sector boundary error on a 24 bit file, because the definition of a sector boundary specifically deals with 16 bit, 44.1khz files

[pool]

scott brown wrote:
"A sector boundary refers to the boundary of sectors on a CD."

Fair enough. That refers to the actual physical bounderies "inherent" by the cd medium.

But a wave file (using foobar and shntool proves this) can be cut in a way that does/will  not  match the CD predifined physical bounderies. I have various 24 bit files that have a sector boundery error. Wheather they will be d'sampled for cd or not the SBE is still reported.

[pool]

PS Sector boundary refers to actual physical pre-defined boundery on the CD.

Sector boundary error is a "virtual" error on the raw wave file itself, resulting from random cutting, that will translate iteself in clicks/gap when burnt on CD.

I think SB and SBE are different terms.

[JasonSobel]

Tim & Scott - you guys are of course, correct.  a sector boundary error, by definition, only relates to 16/44.1 kHz files.

however, I think I understand what pool is getting at.  it is possible to split 24 bit files in exact increments of 1/75th of a second.  if you then resample/dither those 24 bit files, you will get 16/44.1 files with no SBE's.  I do this all the time with CD Wave and then batch-resampling in WaveLab, verified with SHNtool.

what pool is wondering is this:  is there a tool, similar to SHNtool, that will take split 24 bit files, and make sure that each track is an exact increment of 1/75th of a second.  so that when the files are resampled to 16/44.1, there are no SBE's.

I don't know of a program that does that for 24 bit files.  it could be done with a little programming skills, but I don't know if that program is out there already.

here are a couple options for you:
1) track out your 24 bit files with CD Wave, so everything gets cut on multiples of 1/75th of a second.  then you won't have to "fix" your 24 bit files, because they will already be good to go when they are resampled/dithered to 16/44.1.

or

2) don't worry about the random cutting of files for 24 bit files, because, as stated in this thread, SBE's are not applicable to 24 bit files.  then, after you convert everything to 16/44.1, use SHNtool or FLAC Frontend to align everything on sector boundaries.

[pool]

I agree with JasonSobel. My aim not not to track a single wave file. My aim to fix the SBE's (yes i sbes are reported on 24/48 using foobar) and then JOIN two 24/48 waves after. It was just a question, so I can learn more. Nothing meant to overblow. There still are no audible glitches when joing 2 waves 24/48 as above.

[JasonSobel]

I agree with JasonSobel. My aim not not to track a single wave file. My aim to fix the SBE's (yes i sbes are reported on 24/48 using foobar) and then JOIN two 24/48 waves after. It was just a question, so I can learn more. Nothing meant to overblow. There still are no audible glitches when joing 2 waves 24/48 as above.

wait, huh?  if you're no trying to track out a file, but rather, you need to join two files together, why even worry about where they are split?  as long as they are seamless, it won't matter whether it's split on a multiple of 1/75th of a second or not.  just join the two files together.

[wbrisette]

Maybe this will be helpful.

As I understand it, Pool, you're trying to join two files that the MT device split. The MT being a consumer device doesn't include any timecode info for you align tracks like the 744T, Deva, or Cantar does. This means you have to manual align the tracks. <b>IF</b> the MT is accurate you should be able to simply join these by butting them up together.

The big mistake you're making is you're using tools designed for 16-bit files on 24-bit files. Ignore what they are saying for the moment. Wait until you track out the files and convert them to 16-bit files.

Wayne

[pool]

Yes i have joined the 2 wave 24/48 without problem many times. I was just asking if it was possible to align before joining. thanks all for repling. appreciated.