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: foo_loudness_peakmeter 2.x.x.x feature requests (Read 6670 times) previous topic - next topic - Topic derived from Re: foo_loudness_peak...
0 Members and 1 Guest are viewing this topic.

foo_loudness_peakmeter 2.x.x.x feature requests

Quote from: TF3RDL on
BTW, to not clutter this thread like I did on a forum thread for foo_enhanced_spectrum_analyzer component, I had to create a separate thread if I continue posting almost endless FRs
Here we go, another FR thread (same as the last FR thread) but this time, for @Crossover's other component Loudness Peakmeter (foo_loudness_peakmeter):
  • An option to hide time/amplitude graphs
  • Alternate ways for graph to move, like true static scroll mode like in foobar2000's built-in Spectrogram visualization as well as right-to-left scrolling, again like the built-in spectrogram visualization
  • Upgrade "Loudness Peakmeter" area into actual loudness meter with momentary, short-term, integrated, and LRA bars appearing at once
  • I know it is definitely possible to run infinite average level metering at 60fps as in my AudioWorklet-based peakmeter, so I would for consistency with others, increase the update rate for both LRA and integrated loudness to same as the rest (something like 60fps, 120fps, or above)
  • A feature to show Peak/RMS meter bars like foo_uie_peakmeter but with an optional feature to also display Mid/Side representations for each channel pairs and can be easily combined with LUFS meter bars
    • A true peak bar showing on top of sample peak bar would be cool with separate clipping indicators (orange for ISP clipping and red for sample clipping)
    • Also an option to display crest factor for each channel and optionally its M/S representations
    • A phase correlation and stereo balance meter like this one
    • Bonus: Optionally includes infinite average level meter (not the same as integrated LUFS metering) that determine the RMS level over the entire song since playback start or last seek

I hope that Loudness Peakmeter (foo_loudness_peakmeter) gets a similar significant update and/or redesign as Enhanced Spectrum analyzer (foo_enhanced_spectrum_analyzer) component

Re: foo_loudness_peakmeter 2.x.x.x feature requests

Reply #1 –
A new audio impact plotter graph thing in Vizzy made me think about drawing modes for the graph part of the Loudness Peakmeter (foo_loudness_peakmeter) component, which is stroke drawing mode, which I already have and used on my own audio spectrum analyzer and spectrogram project

X

Also, since I've wanted to see true peaks in a graph (while still showing sample peaks), I'd prefer the "shaded" drawing (similar to a spectrum analyzer thing found in iZotope Ozone 5) where the main one above shaded one is a true (4x or 8x oversampled) peak, and the latter one is a sample (no oversampling) peak, and I wanted the graph to optionally show both sample and intersample clipping at the same time where orange and red area in the graph are intersample and sample peaks exceeding 0dBFS line respectively

Re: foo_loudness_peakmeter 2.x.x.x feature requests

Reply #2 –
How about Leq(a), Leq(m), or something like that? Could be either a new bar visualization showing current Leq levels in an upcoming Loudness Peakmeter 2.0.0.0 component, or something that replaces K-weighting with other weighting types in a loudness meter?

Basically, the LUFS meter is just Leq(k), which is K-weighted RMS meter and I'm sure there are other frequency weighting types like A, B, C, D, and M that can be useful (e.g. using A-weighting instead of K-weighting on a loudness peakmeter to simulate what would the level of mic input close to a "standard" but cheap loudspeaker on PC would look like, albeit inaccurately and excludes background noise not coming from speakers entirely)

BTW, if you don't know about Leq, it is equivalent continuous sound level, which is used to measure noise levels on the environment as well as how loud movie trailers it really is (so long as it is properly calibrated)

Re: foo_loudness_peakmeter 2.x.x.x feature requests

Reply #3 –
A Jones & Torick's CBS loudness meter algorithm (used in Orban loudness meter program) would be nice addition for loudness peakmeter component as an alternative to ITU-R BS.1770 loudness measurement and can be shown alongside existing LUFS measurement

BTW, I've made a mockup image of foo_loudness_peakmeter 2.x.x.x with phase correlation and stereo balance meter as well as Peak/RMS meter and three LUFS bars (which I've liked it more than text readout of loudness values):
X
And without the graph (this image is an unedited screenshot of multichannel peakmeter visualization, unlike the first case, which is photoshopped to merge multichannel peakmeter screenshot with foo_loudness_peakmeter for the mockup purposes):
X

Re: foo_loudness_peakmeter 2.x.x.x feature requests

Reply #4 –
It would be cool if we can have different coloring options (besides the static gradient) for the graph part of foo_loudness_peakmeter:
  • Spectral peaks: This uses the same waveform coloring by frequency algorithms found in REAPER DAW (as spectral peaks feature) and some DJ software programs like Serato and Rekordbox to color the graph by frequency at that point
  • Current loudness: Can be either momentary, short-term or even integrated loudness, used as an input for colorization of the graph by current LUFS value
  • Intensity: This works similarly to Horizontal B gradient mode in old foo_musical_spectrum, where the color is determined by current scaled amplitude at that point
  • Crest factor: Shows the difference between peak and RMS values

Also, how about an option to calculate LUFS values independently for all channels (which means multiple LUFS values and channel weighting becomes irrelevant when enabled) and also its Mid/Side representations? And also, what about custom Mid/Side and balance/correlation pair (where first and two channels to use for balance/correlation and M/S metering are user-selectable) as in a recently-updated filter bank-based spectrum analyzer project I've made?

BTW, I think the @Crossover is probably active at certain month range; starting from February and ending in June this year hopefully

[APRIL FOOLS] Loudness Peakmeter 2.0.0.0 (foo_loudness_peakmeter 2.0.0.0)

Reply #5 –
Version 2.0.0.0 of foo_loudness_peakmeter:
  • NEW: Peak/RMS meter bars with optional display of true peaks
  • NEW: LUFS bars
  • NEW: Phase correlation and stereo balance/pan meter. Can also be shown on the graph
  • NEW: Graphs can be hidden (e.g. to only show peak/RMS and LUFS bars)
  • NEW: Two new scrolling modes for the graph; the scrolling and alternate static scroll mode similar to foobar2000's built-in spectrogram
  • NEW: Short-term, integrated (as a line graph) and even loudness range (LRA) can now be shown and also RMS are optionally shown on this graph
  • NEW: Clipping indicators for the graph, displaying both sample peaks and true peaks exceeding the threshold, shown as red and orange areas for sample and intersample clipping respectively
  • NEW: True peaks can now been shown on the graph either alone or in conjunction with sample peaks
  • NEW: Mid/Side mode and can be shown in-addition to left and right channels, and it also works with multichannel audio up to 11.1.6 (18 channels/17.1 surround) Dolby Atmos format
  • NEW: Expanded color scheme customization; infinite color gradient stops with adjustable colorstop position and multiple graph coloring modes, and background color can be made dynamic
  • NEW: Simultaneous channel display on single graph; each channel have its own color and in case of L/R + M/S, has two graphs (one for stereo and other for sum/difference representation of the same audio)
  • IMPROVEMENT: Ditched GDI+ rendering in favor of OpenGL hardware-accelerated 2D rendering
  • IMPROVEMENT: Multiple graphs mode now show more than two graphs when playing multichannel audio
  • IMPROVEMENT: Some of measurements like integrated LUFS and LRA now update at higher framerates (or more accurately, exactly the same as the rest). Optional but enabled by default
  • IMPROVEMENT: Channel weighting now properly account for different channel configuration for the same channel count (e.g. 4 = FL FR BL BR, FL FR FC LFE or FL FR SL SR)
  • FIX: A slow memory leak, resulting in potential stability issues in 32-bit versions of foobar2000

Video showcase by @Crossover:
https://youtu.be/q86MBMphQ88?feature=shared

Spoiler (click to show/hide)

Re: foo_loudness_peakmeter 2.x.x.x feature requests

Reply #7 –
Quote from: darkflame23 on
April Fools?
Yeah alongside this April Fools post about foo_enhanced_spectrum_analyzer over YouTube

Happy April Fools day everyone! and I hope the real @Crossover actually returns and develops a new update for two existing components and maybe a brand new one