IPB

Welcome Guest ( Log In | Register )

3 Pages V  < 1 2 3  
Reply to this topicStart new topic
foo_musical_spectrum, Musical Spectrum
Sunrays
post Jan 14 2014, 16:54
Post #51





Group: Members
Posts: 10
Joined: 31-December 13
Member No.: 113678



Pitty its using lots of CPU, even up to 50% sad.gif (goes away when foobar minimized)

This post has been edited by Sunrays: Jan 14 2014, 16:57
Go to the top of the page
+Quote Post
EpicForever
post Feb 8 2014, 09:44
Post #52





Group: Members
Posts: 706
Joined: 14-September 11
From: Szczecin, PL
Member No.: 93712



Here is uses simply 0%. Do you use something like intel's GMA450 video card prosthesis on laptop?
Go to the top of the page
+Quote Post
Philip Goddard
post Feb 15 2014, 12:10
Post #53





Group: Members
Posts: 6
Joined: 16-March 12
From: Exeter, UK
Member No.: 97850



I've recently discovered this add-on, and am finding it extremely useful - and my great thanks to the developer. However, I have certain particular requirements that I would like to make known with regard to possible future developments of this program.

I have a little issue in that really I want this program or an equivalent of it as a VST plugin rather than a FB2000 component, because:

(a) I use a graphic EQ to counter a nasty speaker resonance that I get (quite possibly a room resonance rather than the speakers' fault), and that EQ shows on the Musical Spectrum display; if I had the latter as a VST, I could then set the order of the VSTs in FB2000's DSP manager, presumably enabling Musical Spectrum to get the signal prior to its going to the EQ, which is really what I want, for generally I want to see the spectrum of the recording, not of the output from DSPs. At the moment there seems to be no means to arrange things so that Musical Spectrum gets the signal before the VSTs.

(b) If Musical Spectrum were VST, with any luck I could then also use it with WavePad, my preferred editor for most tasks.

I've searched online and have not found an equivalent VST plugin, which is why I mention this here.

Anyway, keep up the excellent work!


--------------------
-- Philip.
Go to the top of the page
+Quote Post
godrick
post Mar 18 2014, 19:09
Post #54





Group: Members
Posts: 305
Joined: 31-December 10
Member No.: 86948



I generally use this component for its visual effect, but occasionally I try to understand the actual peak values of songs - in such cases, it seems that the peaks shown by this component are about 10 to 15 dB lower than peaks shown by the DUI peakmeter and foo_uie_vis_peakmeter_spectrum (0.2.0.0 beta). The peaks from the last two elements always agree to the extent I can tell, and disagree with foo_musical_spectrum. I believe I've set the FFT size/samples per analysis, channels, refresh interval, etc to be the same as foo_uie_vis_peakmeter_spectrum to the extent possible.

Since the DUI peakmeter element and foo_uie_vis_peakmeter_spectrum seem to report the same peak info, it seems there is an issue with my musical spectrum settings or the component itself, and I've done everything I can to eliminate a settings issue. I don't care about any matching between any specific bar or frequency - I am just trying to compare and understand the peak dB in the audible spectrum across an entire song. I've inserted an example snapshot of the foo_uie_vis_peakmeter_spectrum and musical spectrum outputs for a sample song showing the issue.

Has anyone else seen this, or are you seeing the same peaks across these visualisations?

fismineur, thanks again for a great component!
Go to the top of the page
+Quote Post
romor
post Mar 18 2014, 20:38
Post #55





Group: Members
Posts: 668
Joined: 16-January 09
Member No.: 65630



QUOTE (godrick @ Mar 18 2014, 19:09) *
I don't care about any matching between any specific bar or frequency - I am just trying to compare and understand the peak dB in the audible spectrum across an entire song. I've inserted an example snapshot of the foo_uie_vis_peakmeter_spectrum and musical spectrum outputs for a sample song showing the issue.


Godrick, you can't compare apples and oranges to deduce something you don't even seem to care about. Was it even worth posting?


--------------------
scripts: http://goo.gl/M1qVLQ
Go to the top of the page
+Quote Post
godrick
post Mar 18 2014, 21:22
Post #56





Group: Members
Posts: 305
Joined: 31-December 10
Member No.: 86948



I don't understand why the highest overall spectrum peaks don't match between the images I showed, and I think I made it clear it's only these peaks I care about. I do not care about the peak of any individual bar because I already realize the individual bars will display slightly differently by element. FWIW, the peaks in Replaygain tags also match what I see in DUI peakmeter and foo_uie_vis_peakmeter_spectrum but not in musical spectrum. If I failed to make any of that clear, I apologize.

Do you not see the roughly 10 db difference in the max peak across the two elements in the image I provided? Better yet, if you see the same peaks across all of the three elements I mentioned in your setup, please let me know.

This post has been edited by godrick: Mar 18 2014, 21:26
Go to the top of the page
+Quote Post
romor
post Mar 18 2014, 21:36
Post #57





Group: Members
Posts: 668
Joined: 16-January 09
Member No.: 65630



Godrick, not that just those components use different range scales, but they use different slicing methods to show to us the data that's rendered later as colorbars (and/or peaks).
I'm only questioning your quest, to where?

This post has been edited by romor: Mar 18 2014, 21:38


--------------------
scripts: http://goo.gl/M1qVLQ
Go to the top of the page
+Quote Post
godrick
post Mar 18 2014, 23:03
Post #58





Group: Members
Posts: 305
Joined: 31-December 10
Member No.: 86948



So you're stating that each of the 4 elements (Replaygain track peak, DUI peakmeter, foo_uie_vis_peakmeter_spectrum, musical spectrum) show significantly different overall peak values in a song (regardless of the exact displayed frequency/bar/time in song) in your setup? That's not what I see in my setup: I see the first three match as far as I can tell. Please share how all 4 significantly differ in your setup.

This post has been edited by godrick: Mar 18 2014, 23:09
Go to the top of the page
+Quote Post
romor
post Mar 18 2014, 23:12
Post #59





Group: Members
Posts: 668
Joined: 16-January 09
Member No.: 65630



Godrick, I can't see how you compare ReplayGain track peak with momentary value of peakmeter, nor how you compare peakmeter with spectrum, or spectrums from differently set programs (components), and I already stated that I assume that's not going to get you anywhere. But by any means and your firm state, do continue. Who know...

This post has been edited by romor: Mar 18 2014, 23:13


--------------------
scripts: http://goo.gl/M1qVLQ
Go to the top of the page
+Quote Post
godrick
post Mar 18 2014, 23:38
Post #60





Group: Members
Posts: 305
Joined: 31-December 10
Member No.: 86948



You replied with yet another strawman argument. I could care less about the momentary peak, or the peak for any given bar or frequency at any moment. I only care about the overall peak across all frequencies and the entire duration for a song. I've stated that several times several different ways.

I will try one final time for you: I can see the Replaygain track peak occur at some point during a song using the DUI peakmeter and the CUI peakmeter spectrum. I never see the Replaygain track peak ever occur in the musical spectrum, and the musical spectrum seems to report something typically about 10 dB lower than the other 3 elements. I believe the held peaks in the image I provided provided a data point consistent with that.

I hope we just have a language barrier issue here. I don't think you've compared all 4 elements in detail as I have. I encourage you to do so.
Go to the top of the page
+Quote Post
romor
post Mar 18 2014, 23:56
Post #61





Group: Members
Posts: 668
Joined: 16-January 09
Member No.: 65630



Thanks you man, I was just trying to assist, but english conversation is indeed my barrier, so maybe I sound to you a bit harsh...

I'll let others tell you what's involved in computing those numbers that you see as bars and how fragile those are to initial parameters, let alone trying to hammer peakmeter that tries to show you the notes and another one ranging the hertz...

Cheers


--------------------
scripts: http://goo.gl/M1qVLQ
Go to the top of the page
+Quote Post
Bob Brines
post Mar 19 2014, 12:57
Post #62





Group: Members
Posts: 65
Joined: 11-May 07
Member No.: 43395



Peak meter integrates the power in all frequencies into a single value. Music spectrum shows the power at each individual frequency. The music spectrum plug-in is working correctly. To demonstrate this, locate a test file with a single frequency, preferably around 1kHz or so. When you play this file, you will see that music spectrum and peak meter display the same value.

Bob

This post has been edited by Bob Brines: Mar 19 2014, 12:58
Go to the top of the page
+Quote Post
kamild1996
post Apr 23 2014, 19:36
Post #63





Group: Members
Posts: 1
Joined: 23-April 14
Member No.: 115709



For me, the best Foobar2000 component ever made, I can actually watch music smile.gif
But you know what would be awesome? An option to make the spectrum work as wallpaper!
Is that something possible or rather something to dream of?
Go to the top of the page
+Quote Post
ashwolters
post May 13 2014, 09:30
Post #64





Group: Members
Posts: 1
Joined: 13-May 14
Member No.: 116051



QUOTE (geckolizard @ Aug 9 2013, 10:55) *
QUOTE (8823 @ Mar 31 2013, 10:48) *
Feature request: show notes (A2,A3 etc) as Hz

Thank you!!


Please provide this option. Thanks!



Yes please....it would be even better with an option for displaying frequency smile.gif
Go to the top of the page
+Quote Post
EpicForever
post May 13 2014, 16:20
Post #65





Group: Members
Posts: 706
Joined: 14-September 11
From: Szczecin, PL
Member No.: 93712



When you stop mouse pointer at certain bar it will show you both note and frequency... It is even shown ion first post in this thread.
Go to the top of the page
+Quote Post
staps
post Jul 6 2014, 12:37
Post #66





Group: Members
Posts: 2
Joined: 6-July 14
Member No.: 116604



Sorry for my English.
This plugin is very useful but has one error.
FFT size must be in seconds rather than in the samples, because changing the sampling frequency of files, analyzer behaves differently.
Useful times are 100, 150, 200, 250, 300 miliseconds.
Go to the top of the page
+Quote Post
Wander
post Jul 11 2014, 09:27
Post #67





Group: Members
Posts: 46
Joined: 25-May 11
From: Germany
Member No.: 90967



No, I don't think so - directly affecting the frequency resolution is (imho) more important than directly affecting the time resolution.
Go to the top of the page
+Quote Post
xnor
post Jul 11 2014, 10:33
Post #68





Group: Developer
Posts: 496
Joined: 29-April 11
From: Austria
Member No.: 90198



???

It makes sense.

Resolution = Sampling Rate / FFT size

44100/2048 = ~22 Hz
96000/2048 = ~47 Hz

50ms with 44100 Hz would be closer to 2048 (~22 Hz) and with 96000 Hz closer to 4096 (~23 Hz).
The result is more consistent frequency resolution.

This post has been edited by xnor: Jul 11 2014, 12:02
Go to the top of the page
+Quote Post

3 Pages V  < 1 2 3
Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 



RSS Lo-Fi Version Time is now: 30th July 2014 - 06:44