71
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.
72
General Audio / Re: Need some advice on resampling 192Khz/32bit float.
Last post by Case -73
FLAC / Re: Multithreading
Last post by Porcus -Also, from a bit more fiddling around with speed:Looks like the explanation was as easy as: "takes more time to do more!" Especially with the --tweak, that will re-tweak for as long as it improves above a given threshold.
It looks like flaccid --gasc becomes slower for longer files, and thus, the one I aligned to roughly match flac -p is now beaten at speed.
@cid42 :
* how to set it to do "nothing but --tweak"? A setting that I can put to 4096 (to compare to reference flac) and then it just tweaks the block boundaries? I tried --mode-gasc --blocksize-limit-lower 4480 --queue 16 --tweak 64 ... cannot be much "gasc" in that, given subset? (It errs out on 4608, and I can understand why.)
* Also: one that does "nothing at all" but compression - this to get any idea of overhead from using a proof-of-concept application. If there is nothing such: one that does "nothing at all but compress once for analysis and once for writing the file"?
* I even wonder if console output slows it down. Yes, Windows does that sort of stuff. A verbosity switch to turn off the tweak(n) output could test that.
* The effort numbers: are those relative to output(1.000)?
As for --mode gasc, it seems to me that you don't need a heavy analysis setting. Sure you save bytes running analysis at -8 (or more) over at -3m, but not much per second extra run, compared to invoking flaccid at all. Heck, when I had run it as low as -3m, I tried
-- mode gasc --blocksize-limit-lower 2560 --queue 16 --tweak 0 --merge 0 --analysis-comp 0 --output-comp 8
and it did beat -8p relatively convincingly (I mean relatively, the numbers aren't big):
11 minutes -8
18 minutes that flaccid-gasc I just quoted, improves 0.12 percent (yeah I mean not much) over -8
35 minutes -8p, is 0.05 worse (meaning 0.07 better than -8)
The explanation then cannot be anything else than "signal changes and even a change among fixed-only predictors is enough to detect it"?
The flac-1.5.0-x64-gcc14.2.0-O3+someflags.7z in @Wombat 's top post has a directory called "AVX-512", and honestly that is all that I know.Build with AVX 512: https://hydrogenaudio.org/index.php/topic,127408.0.htmlAre those flac builds actually using avx512? I only see source code for axv2, sse2, sse3 and sse4
(... you are capable of compiling for several platforms, so ... what for flaccid with both this and that and the "Exact Rice" calculations and whatnot?)
74
General Audio / Re: Need some advice on resampling 192Khz/32bit float.
Last post by mycroft -You should only have once instance of a limiter and only at the very end. Normally as the last step but if you do dithering, dither should be the last.
Advanced Limiter will not touch the signal unless it would clip. Zero-crossing or non-zero crossing will go through as-is, unless it's so loud that it would clip. SRC, SoX and RetroArch resampler components should handle non-zero crossing track boundaries nicely. They extrapolate the existing signals, do the resampling, and then cut away the extrapolated bits which should produce virtually perfect transitions.
This is nonsense. resampling single sample takes samples before and after, and at track boundary there are not such samples, so it uses silence instead probably. This still can be heard if not processed with keeping state between tracks.
75
3rd Party Plugins - (fb2k) / Re: foo_skip: skip tracks that match a specified search query
Last post by Seren -Like I said, I can't reproduce any behavior like that. I asked for your help in replicating the issue. And I can't comment about skip behavior when I don't even know your skip string. Nor do I see console to know if anything relevant is printed there.
https://www.youtube.com/watch?v=Mr4XYJ0ed-o
Skip track NEEDS to be triggered for the bug to occur.
My skip track settings: (%last_played% DURING LAST 5 HOURS AND "$mod($rand(),100)" GREATER 18) OR (%last_played% DURING LAST 10 HOURS AND NOT %last_played% DURING LAST 5 HOURS AND "$mod($rand(),100)" GREATER 60) OR (NOT %last_played% DURING LAST 10 HOURS AND ((%MYRATING% IS "1" AND "$mod($rand(),100)" GREATER 16) OR (%MYRATING% IS "2" AND "$mod($rand(),100)" GREATER 25) OR (%MYRATING% IS "3" AND "$mod($rand(),100)" GREATER 35) OR (%MYRATING% IS "4" AND "$mod($rand(),100)" GREATER 46) OR (%MYRATING% IS "5" AND "$mod($rand(),100)" GREATER 58) OR (%MYRATING% IS "6" AND "$mod($rand(),100)" GREATER 70) OR (%MYRATING% IS "7" AND "$mod($rand(),100)" GREATER 80) OR (%MYRATING% IS "8" AND "$mod($rand(),100)" GREATER 86) OR (%MYRATING% IS "9" AND "$mod($rand(),100)" GREATER 93)))
Obviously that skip settings only works with custom "MYRATING" but you can probably just set it to randomly skip 50% and to try and reproduce.
I'll try message you my config.sqlite if that helps?
EDIT: Also tried the setting "Slow but accurate seeking" and it doesn't occur with that, but instead of seeking back it just goes onto the next song. I don't know if it makes a difference but I have equalizer dsp enabled and only other component is foo_abx and foo_playcount.
76
Support - (fb2k) / Re: foobar2000 2.25 Preview 2025-05-09 -> UPnP
Last post by wojak -If it does not are there any plans to implement it? The best option would be Android phone being a remote control for PC.
77
3rd Party Plugins - (fb2k) / Re: foo_truepeak True Peak Scanner
Last post by regor -Also just "LRA" for track doesn't really work for album, it will have to have a much longer tag field. I'd highly encourage you to consider switching your tag fields to these too.I don't need album LRA, so that's a tag I would probably nuke. But yes, I could choose another name or just adopt that standard in that case.
There were no objections to removing the LRA configuration a year ago. But if you must have it configurable, I can restore it.
Anyway I still have the same opinion, you can apply your desired standard as default tags without compromising configuration.
I have experienced so many cases where designers thought the chosen standard was good enough and then that was wrong as soon as you start using third party software, outside their software/hardware ecosystem or after some years. I would never say that forcing an standard is preferred over configuration + standards by default, but for sure it helps development.
And this discussion just reminded me I should also use that standard tag name on my tool on future releases, even if I allow to change it.
78
General - (fb2k) / Re: DarkOne 4.0 - modified skin
Last post by Defender -Lots of new users over the last two months, and based on their feedback I did some updates, created some vids and also provided documentation about ELP Field Definitions being used.
More changes to JS3_UTD (basically Marc2k3 sample scripts, thx again), to be a little less in your face (my bad, so toned them down a bit). Latest version will be posted separately, which has a more friendly configuration section for changing fonts for the displayed lines in the UTD (NFO/NFS) panel.
I also implemented Br3tt's Smooth Playlist, modified for transparency, alpha and dropshadows, in order to provide an easy way to do bulk rating of tracks, since ELP cannot provide buttons at all. So I now do my bulk ratings in SPL then switch back to ELP to have all visual bells and whistles available again.
All in all also a lot of extra effort was put in creating a visually appealing presentation (imo) for loved/ratings. All optional and configurable, check the attached spreadsheet for options for presentation within ELP.
Also a lot of work in streams again, which I really love since I get to know a lot of really good new stuff because of that.
In case you are wondering, this is fooBar 32bit only since PSS and ELP do not support 64bit and there is currently no wrapper available to run those UI plugins in x64 native (one of my dreams ;-) ).
Some other components are 32bit only too, but those could be easily changed to 64bit alternatives.
79
General Audio / Re: Need some advice on resampling 192Khz/32bit float.
Last post by Wombat -Is there the chance to add the option in foobar changing the gap behaviour for the CUE from nearest to CD frame to always backwards?
80