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.
Recent Posts
1
Scientific Discussion / Re: One updated and one new audio processing DSP library
Last post by bryant -
Quote from: bennetng on
Convolution version 4 without using any double. Files within the same group are identical (also true for version 1).
Quality wise, version 4 is obviously cleaner and more consistent among different compiler settings.
Interesting that the canonical version with the "safest" compiler settings produces the worst result (If I'm reading your graphs correctly)!

I suspected that the 4th version might (jn theory) produce the best results (because it orders the adds from the smallest to the largest values), but I was never able to measure it. What's more surprising is that (again, if I'm reading correctly) the more advanced SIMD options are producing progressively better results compared to the "correct" ones. Have no guess what that would be.

It seems though that it might be reasonable to switch to version 4 as the default.
2
Scientific Discussion / Re: One updated and one new audio processing DSP library
Last post by bryant -
Quote from: bennetng on
Quote from: bryant on
BTW, is this with your "double" change for precision? And which of the provided convolutions implementations is this with?
I specifically tried mingw64 because I hate the MSVC limitations. There is no change in the original source code (i.e. 32-bit with canonical convolution). Because the comparison among different versions would be unfair for example in the 4x unroll version:
Code: [Select]
sum += (A[0] * B[0]) + (A[1] * B[1]) + (A[2] * B[2]) + (A[3] * B[3]);
Everything is still in float before adding to sum, and if I cast A and B to double within such a loop things will slow down a lot which nullify any unrolling effort.
My thinking was that the unrolled versions will let the compiler know that there are (in the case you show) always a multiple of 4 samples available, which should make the SIMD easier. But maybe it just throws in extra code to check the counts and in the case of 1024 terms it makes no difference. I'm not sure what you mean by "unfair". The fastest version would be the best unless it generated bad results (which I think in this case would just be slightly more noise).
3
Polls / Re: 2024 Lossy format poll
Last post by a.ok.in -
Quote from: Squeller on
they don't display embedded art from aac files, whereas mp3 works.
Same, I have an old TV (from 2010) with USB support; artwork from mp3 is fine, but doesn't show anything for aac. Also it seems to detect .ogg Vorbis files, but cannot play them.
Anyway, just confirms that MP3 is the damn standard everywhere :)...
4
3rd Party Plugins - (fb2k) / Re: foo_discogs
Last post by heliosphaner -
Quote from: BiOMeX on
Quote from: heliosphaner on
Any idea how to get rid of it?
Same thing here:
"(FATAL) Error: Page Deleted or Missing (404)(url: https://api.discogs.com/artists/15037826)

[ESCAPE to close]"

foo_discogger 1.0.21.1

Yes, I used this one before I wrote. Eventually it worked after installing fb2k-component, but the one installed from zip caused errors.

Also, today I'm having error 500:

(FATAL) Error: Error loading release 31743548: HTTP Error (500)(url: https://api.discogs.com/releases/31743548)

[ESCAPE to close]
5
3rd Party Plugins - (fb2k) / Re: ReplayGain DSP - Alternative ReplayGain implementation by Case
Last post by Case -
Quote from: Squeller on
a) If it is set to "Rescan everything", it will be enforced to know gain/peak levels?
In theory. It will try to scan the track gain and peak levels. But remember that the scanner will abort the job if it takes too long. The scanner will only be allowed to run while the output buffer still contains audio data so that playback remains gapless. If that time isn't enough to scan the entire track, then the new checkbox will essentially treat the track as if its RG gain value is 0.0 and peak info is missing, if the RG scanner returned above 0.0 gain.
Note that easy way to give more time for the scanner is to increase foobar2000 output buffer size.

Quote from: Squeller on
b) If set to "Prevent clipping only" the checkbox has no effect?
That's correct.

Quote from: Squeller on
c) With positive gain you mean "volume increase", correct? Because first I read it like "Allow end result to be +0.5db"/i.e. "Allow clipping".
Actually at the moment the new checkbox can still allow volume increase in certain situations. Namely if the target loudness is higher than the ReplayGain reference level and the gain value is zero or less negative than the volume boost. For example if estimated gain is -1.0 dB and you have selected Spotify playback level of -14 LUFS. That means the playback level is reduced by 1 dB but boosted by 4 dB, so +3 dB increase.
I will need to change that to prevent volume increase instead of just positive RG result. But that will have to wait until tomorrow, too many quick fixes already released today. Let's see if there are more things that need adjusting.
7
3rd Party Plugins - (fb2k) / Re: ReplayGain DSP - Alternative ReplayGain implementation by Case
Last post by Squeller -
Quote from: Case on
New version out with an option to prevent loudness increase for tracks with missing or guessed peak.
But do I understand correctly:

a) If it is set to "Rescan everything", it will be enforced to know gain/peak levels?
b) If set to "Prevent clipping only" the checkbox has no effect?
c) With positive gain you mean "volume increase", correct? Because first I read it like "Allow end result to be +0.5db"/i.e. "Allow clipping".

I may be confused about terminology.
9
Polls / Re: 2024 Lossy format poll
Last post by Squeller -
Mostly LC AAC (always created by qaac/Apple), sometimes mp3 - because of cars. It's not even because they don't understand AAC (one car of mother-in-law is just stupid and needs m4a renamed to mp4), but e.g. they don't display embedded art from aac files, whereas mp3 works. Car audio - a sad story, often just an afterthought and a mystery. E.g. embedded album art exceeding 800px doesn't display in my car. It takes time to find that out. OK, that's offtopic.