Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Best way to convert 24/48 to 16/44 (Read 5303 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Best way to convert 24/48 to 16/44

I want to convert masters from 24/48 to 16/44.1. The sources are in 24/48. The better way would be to feed a very good ADC at 16/44 with a very good DAC at 24/48 but I have to do it in the digital domain.

My first idea was to make everything during the downmix in Nuendo (16/dithering with Waves L2 and resample 44.1 after).

I noticed that the ssrc in foobar was better. So I thought that I have to reconsider the question...

A better way might be to resample to 44 first and then downsample with dithering to 16.

Any idea?

Best way to convert 24/48 to 16/44

Reply #1 –
 No anwser on HA!?

Best way to convert 24/48 to 16/44

Reply #2 –
Did you try SSRC ?
You can find it on RareWares
Look for 'Shibatch sampling rate converter' there.

Best way to convert 24/48 to 16/44

Reply #3 –
Quote
Did you try SSRC ?
You can find it on RareWares
Look for 'Shibatch sampling rate converter' there.
[a href="index.php?act=findpost&pid=224710"][{POST_SNAPBACK}][/a]


I'd prefer foobar2000's SSRC combined with its ATH shaped dithering for this task.
"To understand me, you'll have to swallow a world." Or maybe your words.

Best way to convert 24/48 to 16/44

Reply #4 –
Quote
Did you try SSRC ?
You can find it on RareWares
Look for 'Shibatch sampling rate converter' there.
[a href="index.php?act=findpost&pid=224710"][{POST_SNAPBACK}][/a]

Sounds very good. I'm abxing it. In my case (>5000 conversions), the command line is more than a bonus.

Best way to convert 24/48 to 16/44

Reply #5 –
Quote
Quote
Did you try SSRC ?
You can find it on RareWares
Look for 'Shibatch sampling rate converter' there.
[a href="index.php?act=findpost&pid=224710"][{POST_SNAPBACK}][/a]


I'd prefer foobar2000's SSRC combined with its ATH shaped dithering for this task.
[a href="index.php?act=findpost&pid=224714"][{POST_SNAPBACK}][/a]

I'm trying to abx both.

Any reason to prefer foobar?

Best way to convert 24/48 to 16/44

Reply #6 –
I tried foobar (resampling ssrc slow mode, stong ath ns) and ssrc (--dither 3).

I can't really abx them...

Any theoretic difference between them?

I can abx both 16/44 with the original 24/48 though. Everything was converted and played at 96/24 during this abxing.

Best way to convert 24/48 to 16/44

Reply #7 –
Quote
Any theoretic difference between them?
[a href="index.php?act=findpost&pid=224728"][{POST_SNAPBACK}][/a]

foobar2000's dithering code was based on SSRC's and further tuned by Garf, so it should be one of the best implementations around.

Quote
I can abx both 16/44 with the original 24/48 though. Everything was converted and played at 96/24 during this abxing.

Could you supply ABX results for your 24/48 vs. 16/44 tests? There only have been few reported cases of reliable ABXing different bitdepth/samplerate combinations above 16/44.
Please avoid further resampling in your tests. Dithering/Resampling twice will surely cause detectable differences.
"To understand me, you'll have to swallow a world." Or maybe your words.

Best way to convert 24/48 to 16/44

Reply #8 –
Quote
Please avoid further resampling in your tests. Dithering/Resampling twice will surely cause detectable differences.


Do you think it's better to let the hardware upsample for ABXing 44 kHz, 16 Bit versus 48 kHz, 24 Bit ?
I don't. The error you get by converting 44,1 kHz 16 Bit back to 48 or 96 kHz, 24 Bit is far far less compared to the error you get by converting 48 kHz, 24 Bit to 44,1 kHz, 16 Bit and won't be noticable. It's IMHO important to feed the hardware with files in the same format when ABXing. You never know - it may handle different formats differently well.

BTW: Did anyone try to adapt the dithering/noise shaping to the signal ? Pure ATH-shaping (without accounting for the masking effects) seems to be suboptimal. One could implement a simple psychoacoustic model which calculates a signal adaptive noise shaping filter. (kind of a time domain psychoacoustic coder)

just my two cents.

bye,
Sebi

Best way to convert 24/48 to 16/44

Reply #9 –
Here is the result for the 16/44 vs 24/48.
I don't know how to avoid resampling because I can't change the clock of the DAC during abx test... I thought that the best way was to feed the DAC with 24/96.

foo_abx v1.2 report
foobar2000 v0.8.2
2004/07/11 02:12:45

File A: file://C:\Documents and Settings\marcan\Desktop\airlock tst\test 24.wav
File B: file://C:\Documents and Settings\marcan\Desktop\airlock tst\001 test 24.wav

02:13:17 : Test started.
02:15:19 : 01/01  50.0%
02:15:25 : 02/02  25.0%
02:15:33 : 03/03  12.5%
02:16:15 : 04/04  6.3%
02:16:22 : 05/05  3.1%
02:16:25 : Test finished.

----------
Total: 5/5 (3.1%)