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: Crossmix DSP (foo_dsp_crossmix) (Read 5391 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Crossmix DSP (foo_dsp_crossmix)

Quote from: Case on
You had something like this in mind? Attached is a component for desktop foobar2000 but if people want it it could finds its way to mobile.

Crossmix DSP (foo_dsp_crossmix) on foobar2000.org
or
https://foobar.hyv.fi/?view=foo_dsp_crossmix

Yeah, very nice! After a couple of months of using CrossMix component I can confirm that it does the trick. Thank you very much.

I would love to have it available on fb2k components Web page, for the others to discover this little gem and to make sure it will not fall into incompatibility abyss in some future release.  ;)

Re: Crossmix DSP (foo_dsp_crossmix)

Reply #1 –
I have installed this plugin.

https://foobar.hyv.fi/?view=foo_dsp_crossmix

I have noticed that at the end of the last sequential track of a playlist, there are a number of anomalous noises at the end.

Re: Crossmix DSP (foo_dsp_crossmix)

Reply #2 –
I don't see anything wrong in the sources, At the end of playback all buffered samples should just be played out.

I tried creating a few playlists from different albums and played the last two or three tracks of these with the Default playback mode. I failed to replicate the noise bug, playback just ended smoothly.

Would you kindly confirm if the bug requires some special settings to replicate? Or special input.

Re: Crossmix DSP (foo_dsp_crossmix)

Reply #3 –
I did some tests.

The problem seems to be related to the [exclusive] mode, both with the default sound card and with the dac.

By removing the [exclusive] mode, the problem does not arise.

Re: Crossmix DSP (foo_dsp_crossmix)

Reply #4 –
Thank you. This turned out to be a problem in foobar's output code and should be fixed for the next build.
For now you can circumvent it by enabling the smooth fader in the output config.

Re: Crossmix DSP (foo_dsp_crossmix)

Reply #5 –
Fixed in the freshly released v2.1 preview.

Re: Crossmix DSP (foo_dsp_crossmix)

Reply #6 –
THX  :D

Re: Crossmix DSP (foo_dsp_crossmix)

Reply #7 –
I want to express my thanks to the developer of the Crossmix DSP component. It's easy to configure and works very well.

As I couldn't find a reference to the developer on the forum, I'll post it here.

I ask you to keep developing the component.

Thank you and see you later.


wagner reatto

Re: Crossmix DSP (foo_dsp_crossmix)

Reply #8 –
I have been encountering problems with this component, while using along with ReplayGain alternative component. RG component works but not this. Please fix the compatibility issue.

Re: Crossmix DSP (foo_dsp_crossmix)

Reply #9 –
Do you refer to the pausing if RG DSP scans the track? Or do you suffer from something else?

The pausing is certainly something the DSP was never meant to cause, though its auto-scanning was intented to be kind of a last resort ear protection method when accidentally playing something not previously scanned.
I started testing different scanning strategy that would solve this issue and hopefully offers better experience with streams and other slow sources too.

Re: Crossmix DSP (foo_dsp_crossmix)

Reply #10 –
No, I am using Alternative ReplayGain 3rd party component. (https://www.foobar2000.org/components/view/foo_arg)

This component has no problem with any others I am using alongside, like skip silence or crossfader. It interferes only with your crossmix DSP, making it not work. Please fix it.

Also, in your next update, can you please increase your threshold to crossmix from -20db to 0db? So that users will have more control when to perform crossmix. -20db is still low for me.

Re: Crossmix DSP (foo_dsp_crossmix)

Reply #11 –
DSP components have absolutely no effect on the component you linked. It is an alternative ReplayGain scanner. That component can not be broken by a DSP or by any other rule following component.
However that component breaks the built-in ReplayGain scanner causing it to produce wrong results and also prevents anyone from using the SDK-provided ReplayGain scanner service.

Re: Crossmix DSP (foo_dsp_crossmix)

Reply #12 –
That's what I am saying too. It doesn't affect any other component except yours, that is crossmix. When I remove Alternative ReplayGain and use crossmix with other components I have, it works fine. But I need Alternative ReplayGain too. I am not using built-in ReplayGain scanner; I have disabled it.

Re: Crossmix DSP (foo_dsp_crossmix)

Reply #13 –
I have no idea why you think you need foo_arg. It claims to make RG scanning faster by doing the math incorrectly, but at least on my computer the official RG scanner is much faster. And obviously with foo_arg installed none of the options of RG scanner have any effect since its functionality is replaced by a broken replica.
There is nothing for me to fix in relation to this.

Re: Crossmix DSP (foo_dsp_crossmix)

Reply #14 –
i use foo_arg because it eliminates the need to scan the tracks first for rg gain, for any newly added tracks. for some reason, official rg doesn;t work for me, i guess because i dont scan the tracks in my library. foo_arg doesnt ask me to scan anything; i enable it and it works depending on lufs i set it to.
 

Re: Crossmix DSP (foo_dsp_crossmix)

Reply #16 –
I am so sorry about the mixup. you are right. its the same component authored by you. This and crossmix doesn't function together.

Re: Crossmix DSP (foo_dsp_crossmix)

Reply #17 –
Hi Case, just want to get acknowledgement from you if there is a problem you see here. If so, can you please work on this fix?
Another want is to increase the threshold from -20dB to -10 or -5dB. It's closed source, so I am requesting you. Thanks

Re: Crossmix DSP (foo_dsp_crossmix)

Reply #18 –
Like I said on the 12th day, I am aware of a problem with foo_dsp_replaygain. And the problem is in foo_dsp_replaygain, though in a backup feature that isn't meant to replace proper ReplayGain scanning. I was working on a fix, but it got buried as unimportant. I can resume it now.

I uploaded a new Crossmix version that allows increasing the threshold up to 0 dBFS.

Re: Crossmix DSP (foo_dsp_crossmix)

Reply #19 –
CASE, hello.
I always appreciate using the Crossmix DSP.
I may have submitted a request before, but it might have been deleted due to posting etiquette or other reasons.
Let me explain again.

As I understand it, this DSP triggers the next track when the entire waveform from a specific point to the end of the current track falls below the threshold.
My request is to start playback of the next track earlier-specifically, from its beginning up to the point where it exceeds the threshold.
For example, when transitioning between a track without a fade-out and a subsequent track with a fade-in, the current mechanism results in an abrupt switch to silence. Implementing this adjustment would allow for a mixed transition, which I believe would improve the listening experience.

I’m unsure if this is algorithmically feasible, but if possible, I’d greatly appreciate it if this could be implemented.

Re: Crossmix DSP (foo_dsp_crossmix)

Reply #20 –
Thanks Case. It works fine for "most" audio files. But I could still find some audio when transitioning, it completes play fully and then goes to next; crossmix doesn't work only for certain tracks, though the amplitude reduces towards the end for those particular songs. I cannot identify it it's a bug. But much better than previous version where upper limit was -20db. I will wait for the foo_dsp_replaygain fix. Appreciate it.


tasamagos, your request may or may not fix my above problem, but we will see if Case can implement it.

Re: Crossmix DSP (foo_dsp_crossmix)

Reply #21 –
Sorry if you are already aware of this,
but it appears that two consecutive tracks with different sample rates cannot be mixed.
Therefore, I place a resampler at the front of the DSP chain to unify the sample rates.
With this approach, I find that this component operates extremely stably for me.
If mixing were possible without a resampler, that would be welcome, but I imagine it would be technically difficult.

Re: Crossmix DSP (foo_dsp_crossmix)

Reply #22 –
Thanks, I was not aware of this. Now it works for me with a resampler. But I fear if it would result in audio losing data or quality. I also appreciate if crossmix can do this without resampler, because it's usual and definitely expected to have multiple tracks with different sample rates and I thought crossmix, like crossfade was built to handle this already.

Re: Crossmix DSP (foo_dsp_crossmix)

Reply #23 –
New version released that hopefully does what @tasamagos wants. Currently only released at https://foobar.hyv.fi/?view=foo_dsp_crossmix as I'm on the road.

You need common sample rate and number of channel with all mixing, including crossfader. You could get somewhat decent results by doing resampling inside the mixers, but it would never be perfect.

Re: Crossmix DSP (foo_dsp_crossmix)

Reply #24 –
Hello,

I'd like to tell you that I've found Crossmix DSP to be the best mixing component for linking my music tracks on a PC.

I get exactly the mixes I used to make in the 70s and 80s.

However, I also sometimes listen to music in the car, and I don't know if this plugin would work with the Android version of foobar2000; I don't think so.

I'm currently using Neutron Player on my smartphone, but I'm not satisfied with the mixes produced by this application (based on the start/end duration of the track).

So my question is: Do you know of an Android application that allows you to create links as perfect as those made with the Crossmix DSP plugin?

So, to achieve what I want, I use my PC and foobar + Crossmix to create a file of several tracks into a single one, with the mixes that suit me. This is a backup solution that still has drawbacks, such as not knowing which song is playing and difficulty moving forward or backward within the file being played.
My largest file created this way has nearly 110 hours of continuous listening time in FLAC (it takes a long time to build, but it works.)

Thanks for reading this far.