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: 24-bit Packed vs. 32-bit Unpacked (Read 4990 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

24-bit Packed vs. 32-bit Unpacked

I'm using foobar v0.62 on my computer (WinXP, Duron 950, 384MB PC133), and I have the output set for 32-bit Unpacked (24-bit data + 8-bit padding).  My soundcard is a Terratec EWX 24/96, and the drivers support both 24-bit packed and 32-bit unpacked sample formats.

The reason I've been doing it this way is that the manual for the card states that padding to 32-bit is easier on the CPU (and slightly higher on utilization of the PCI bus) than 24-bit packed. (They have a nice graphic depicting DWORDs of the different modes...)

The foobar v0.666 changelog states that the state of the experimental output modes checkbox isn't saved between instances... I take that to mean that it will start in 24-bit packed mode (as opposed to 32-bit unpacked) when foobar loads up next time.

So my question is, does it really matter which 24-bit mode I use or is it just splitting hairs? I don't notice a difference when watching CPU usage in Task Manager, but I do like to tweak performance...

TIA!
godzilla525

24-bit Packed vs. 32-bit Unpacked

Reply #1 –
Was wondering about those settings as well for a long time.  Really have no idea what they mean.  Would be really greatful if someone could explain them roughly.  Do they only have to do with performance or with output quality as well.  Am really slightly lost here.

Am using a Terratec SixPack Soundcard(no the most high end, but neat for the price), in case that matters.
 

24-bit Packed vs. 32-bit Unpacked

Reply #2 –
It has no effect on sound quality at all.

As for the differences, hopefully this picture will explain things better:



With the packed 24-bit format, 4 samples can be transferred in 3 cycles, which is better in terms of PCI bandwidth utilization—no bits are wasted.  However, the CPU has to chop up the samples and reassemble them first.

With unpacked, the samples are padded with 8 bits worth of 0s (or those 8 bits are simply ignored).  This wastes 8 bits per cycle, but does not require the CPU to split up the samples.

Hope this helps...
godzilla525

24-bit Packed vs. 32-bit Unpacked

Reply #3 –
Thanks, that explains it very well.

24-bit Packed vs. 32-bit Unpacked

Reply #4 –
Quote
The foobar v0.666 changelog states that the state of the experimental output modes checkbox isn't saved between instances... I take that to mean that it will start in 24-bit packed mode (as opposed to 32-bit unpacked) when foobar loads up next time.

It saves the ouput format.  It just doesn't save he state of the "Show all options" checkbox.