Help Encode 24 Bit to FLAC

[Toadinohio]

So my friend started taping in 24 Bit.
I've been doing the transfers-
I'm on Window XP Pro
I use Audacity, then spit out 2 whole wav files, one in the original 24 bit and then one dithered down to 16 Bit.
I then put each one in CD Wave Editor and split the tracks.
I use Traders Little Helper to convert to FLAC.
The 16 bit converts fine.
The 24 bit gives me an error message-
found non-standard 'fmt' sub-chunck which has length = 40

The files are 2 Channel Stereo 24 bit/48 kHz- the same format they were recorded in.

Is CD Wave doing something to them?
Does TLH not do 24 Bit?
Is the 48 khz the problem?
Should I use another FLAC program?

Thanks in advance!

[xjsb125]

Weird, I've never had an issue with 24 bit in TLH. Maybe try the actual FLAC Frontend program and see if you get a similar error.

[Toadinohio]

I thought about doing that but figured I would ask for advice here before installing another app that may or may not work.
After reading your post about using TLH I started thinking maybe it's because I have an older version-
I am still using version 1.1.1 build 90 and I see the newest is 2.6.0 Build 168
I will try updating, then FLAC frontend if that doesn't work

[rspencer]

When it gives you that message, does it still convert the file? FLACFrontend will give that sometimes as well. It's just due to nonstandard info left in the header, usally by editing software (in this case, Audacity).

[optiplex2]

i have a questions. I have a show taped in 96k/16bit. Would the recording become lossy if i converted it to 96k/24bit?

what exactly would it do, and would it be lossy? or would it effect the tempo? just wondering.

[Toadinohio]

When it gives you that message, does it still convert the file? FLACFrontend will give that sometimes as well. It's just due to nonstandard info left in the header, usally by editing software (in this case, Audacity).

Does not encode :(

If new TLH or flac frontend doesn't work I'll try using something other than audacity!

[weedwacker]

You could import and then export the audio out audacity. Doing that should clean the header of any junk in wav file. You just need to set the output preferences to 24 bit audio having the header set as wav microsoft and encoding as signed 24 bit.

Though normally the flac encoder will automatically remove any substandard junk in the header when encoding. My only guess is TLH is treating warnings as errors which will abort the encoding process. The other thing you can try is the strip audio header tool in TLH using both options which should clean the junk out of the header.

[roann]

I thought about doing that but figured I would ask for advice here before installing another app that may or may not work.
After reading your post about using TLH I started thinking maybe it's because I have an older version-
I am still using version 1.1.1 build 90 and I see the newest is 2.6.0 Build 168
I will try updating, then FLAC frontend if that doesn't workDo _not_ use TLH v1.1.1.90 (anymore) - this is a very outdated version. The "fmt sub-chunck error" was fixed in a later release. So please upgrade...

[Toadinohio]

Do _not_ use TLH v1.1.1.90 (anymore) - this is a very outdated version. The "fmt sub-chunck error" was fixed in a later release. So please upgrade...

Thanks!
This is the 1st problem I've had with TLH so I've never updated.

[Toadinohio]

Problem solved!
Updating TLH was the fix.
Will be upping shows in 16 and 24 soon :)
Thanks for the replies-
and thanks once again to roann for TLH!
It is my favorite encoder/decoder and the thought of installing flac frontend or anything else just seemed wrong :)

[rspencer]

:thumbsup

[GRC]

i have a questions. I have a show taped in 96k/16bit. Would the recording become lossy if i converted it to 96k/24bit?

what exactly would it do, and would it be lossy? or would it effect the tempo? just wondering.

I take 'lossy' to mean that the audio has, from its original form, been through a file-size reduction process that throws some data away to leave a reduced file.

In your example, you won't actually be losing anything, or throwing anything away, but you'll have to 'up-sample' to fill from 16-bit to 24-bit, I reckon......

So in short - no, I wouldn't call the up-sampled version 'lossy'

[direwolf-pgh]

to my understanding, if you took a digital 16bit file and saved it directly as 24bit.. it would pad the data with 8bit of zeros (identical sample rate / bloated file size)

..there may be some pro audio software that could upsample/alter the original - but I'm unfamiliar.
I do know that many new amps/receivers have an audio upsample option for playback (not unlike video upscaling).. but they deal with the sample rate (kHz) not the bit/data unit.