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: Lossless codec comparison - part 3: CDDA (May '22) (Read 16904 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Re: Lossless codec comparison - part 3: CDDA (May '22)

Reply #50
Yes, this is all very complicated. However, the main conclusion I can draw is that adding buffering isn't necessarily an improvement for all common usage scenario's.

The things I tested (WAV->FLAC->WAV) on a USB-stick that is almost full and on an external HDD that is almost full are both very normal things to do. For the first case a performance improvement up to 10x was seen, which is *huge*. For the second case, a performance deterioration of 40% was seen, which is quite large.

It isn't particularly clear what should be chosen.
Music: sounds arranged such that they construct feelings.


Re: Lossless codec comparison - part 3: CDDA (May '22)

Reply #52
NTFS for the harddisk, exFAT for the USB-stick
Music: sounds arranged such that they construct feelings.

Re: Lossless codec comparison - part 3: CDDA (May '22)

Reply #53
Performance might be file system specific - and cluster size specific?
Windows defaults to 128 kibibytes clusters when formatting 32 GB and more to exFAT. And, I remember I recently checked a big (4 or 5 or 6 or 8 TB it must have been) USB hard drive out of the box - it came formatted as exFAT with even larger cluster size than default. That is quite some difference to a 4096 byte NTFS cluster, if that size matters at all.

Sure there is a need for a reasonable default, whatever that may be. But if you provided for a --buffer-size option, and a way to set default buffer size upon compile, then you would get some test results around here I guess, at least there were a few at https://hydrogenaud.io/index.php/topic,123025.0.html


(This is turning very FLAC-specific, but I guess that mods make better judgement as to whether to split the thread, and if so, where.)

Re: Lossless codec comparison - part 3: CDDA (May '22)

Reply #54

I've rerun the tests with two different CPUs, for exactly the same codecs (same versions, same compile, same executable etc.) so we can compare. Took me a few weeks
I'm now running the encodes/decodes on yet another CPU, on Windows 7 and on Linux through wine, to establish whether wine adds any noticeable overhead. CPU is an Haswell i7 by the way. Results should be very close to the i3 tested previously.

I have those numbers, took a while. Bottom line: not much difference except for Shorten. There is a small difference in file size, but that is probably because one of the tests is measuring actual file size and the other file size on disk or something like that. For the wine results WMA Lossless is missing (because it is an OS component) and for the other TTA is missing, because I couldn't get it to work properly.

Music: sounds arranged such that they construct feelings.