LA School Distance Schematic CCBY

[wlangfor@uoguelph.ca]

Hi everyone,
I'd made another post:
http://flowstone.co.uk/support/viewtopic.php?f=2&t=13660&sid=ea...

where I had described the mathematical phenomenon in regards to the decline of bass up close and treble far away. I created a "roughed" prototype which achieves this effect and I wanted to share it with You. You can use the code for commercial purposes I don't care really and I don't intend to have any legal worries about this prototype.

BUT, I mean wow; This is a new technology that might re-define the audio world. Where Pan is not by amplitude but instead by frequency and delay.

A Bass wave You see can be heard fairly close but not too close. I have not added a High Pass to this example yet; I thought it should be an option. Special thanks to Loopeytunes, Dozius for a stereo width code example which I butchered and to BobF for an example of stereo pan using delay. You inspired Me BobF and I made this, thanks for sharing the open sourced idea on SM forums.

I feel that this idea if mathematically proportioned and made into a unique filter can be a brand new audio science in the scientific world regarding Pan Law. I wanted to give it the name Ellis Law, It's the name of My Mother's father who was a WW2 Bomber Mechanic/Gunner/Operator in the RAF.

Here's the file:

LA School Mini Distance v16 - Changed Unity Levels.fsm

Distance Schematic

(1.25 MiB) Downloaded 1466 times

I'd like to replace the legendre filter with a mathematical grade that follows exactly the mathematics of wave lengths and their rarefaction but this is a very good start and a working example of the phenomenon.

[RJHollins]

Will check it out

[wlangfor@uoguelph.ca]

I was going to use this example in mathematics to decide the new filter's configuration:
http://www.sengpielaudio.com/calculator-distance.htm

[wlangfor@uoguelph.ca]

After a lot of research - I can probably perform these calculations. But some of that would have to do with k-weighting.
K-weighting You see relies upon decibels by db(C) rating with which I can perfect perform the calculations.

The question is will I need to continue to use K-weighting so it is dynamic or statically for merely gathering the results.
I suppose it comes down to results that I would test against the EBU. I'm using Youlean's testing schematic to derive the results:
http://flowstone.co.uk/support/viewtopic.php?f=3&t=3018&sid=ad57e47cbaf8eaca389ee4a558c719b6

I got the test set and can make sure that the results are interpretable via world standard.
https://tech.ebu.ch/publications/ebu_loudness_test_set

My goal is to document a simple way to achieve the pan with test results and then I can submit the idea to the ATSC and EBU. Ultimately; What a system like this would infer is that there was a base distance, for it is only with that distance that the pan's distance can be decided by percentile.


It is like a raster image that has no defining measurement and with the distance related to 0.4 meters I believe the math is achievable. As yet, However, dB(C) is a somewhat rough outline (it's suggested that there is no exact configuration or math surrounding it, only that the math is achievable "Per Frequency only"), I'll first have to create plots of distance and theoretically desired effect by a lowpass summating to a unit of measurement.
http://www.sengpielaudio.com/calculator-dba-spl.htm
https://www.noisemeters.com/help/faq/frequency-weighting.asp

With that scheme it should be possible to consider the amount of error using the EBU test set by then comparing the K-Weighted peak dB by the then reduced dB by frequency.

It's a lot of work and yet We play with these technologies everyday; It's really quite annoying when there is no set standard. And I will also make sure to test the same theory with various makes of lowpasses and will test for integrity so as to give some play and flex to this invention.

Thanks a lot to Youlean, who provided a schemetic which includes K-Weighting, I'll be using it to test the decibels.
Please don't feel that I am rushing to do this; I believe it might take Me two or three weeks to plot the curve, but in the meantime; I'll finish My other project; Universal Slider Text/Touch Screen/Help Popup/Tooltip in first the old style of primitives and then summated to Ruby. More than likely the approach will combine the two technologies.

[wlangfor@uoguelph.ca]

So, this math is soon becoming a priority; And that's after I finish the colour-less knobs and the touch screen input.

But, I wanted to point out something: If the decibels of "very loud" are matched with DBFS's "very loud", I mean as long as we agree that -13 youtube LU, -8 CD Master LU, -23,-24 ATSC/EBU is the LOUDEST it goes then; As long as the realm of decibels matches with dbfs (assuming how LOUD something is) then We can create a grid. This grid can cover distance, proximity, movement, equalization of treble and could be used for multiple channels.

In fact there are huge amounts of dBSPL calculations that can be used with K-Weighted stuff. As long as we agree that the volume on a tv can be turned down and up, and loudest is loudest then We have this possibility.

A linear grid of measurement can leave dolby in the dust. In fact, it could make video games so amazing; Ducks flying by in the sky You could hear just the right amount of treble and bass by how much closer they fly towards You (or away). You could hear the sky wooshing by, You could virtualize for each channel; And with this math - Any amount of fluctuation is possible.

This math can re-write audio and If You're into this idea and You can find any examples of LU's and LKFS/ LUFS's idea of loudness (whisper, construction machinery) in the formulative stages then We will have documented findings which can be used to form a means of placing the dBSPL scale upon the LU scale. Though arguably, the interpreted loud will always vary - As long as the scale of decibels matches dBFS by measurement, it won't matter. It's like the metric system: if 10 played at 20 then thirty, it's still on the same scale (just like volume).