Draft for Audio Support

[moritzstaffel]

No description provided.

[moritzstaffel]

I have added the information so fro that we researched. What is missing is how we handle attachments, hanging material and other stuff.
Also we need to get more input on the things that we researched.

[petrvanekrobe]

additional items needed for audio support:

need to support a range of angles for the sound angle. for their speakers they have a set of angles within the range that they support. for others they can support any angle in that range

for sound angle, need to support asymetric dispersion with a different angle possible in each direction

also need to support rotation of the sound angle

max SPL is not always in dB. There are different dB formats like a weighted dB.

for mvr, need different material types to support reflectiveness for the audio analysis

[petrvanekrobe]

The Frequency Min and Max should be at the -6dB mark (instead of the proposed 5dB)
I can confirm, that the MaxSPL taken from 1 meter from the speaker is common practice.

For Listening Planes, I would find it more useful to all polygon with any number of points that make up an infinitesimally thin layer. I believe this is how our software treats listening planes (with no range of height, just a single height measurement hovering over a surface).

[ILane-VW]

• we should treat the sound angle almost like a beam of light with a range of values in both the X and Y direction since the sound dispersion is not always symmetrical. using sound angle min X, min Y, max X, max Y. not all speaker manufacturers have this capability to change the angle dynamically on a speaker but we should support it
• they also need to be able to rotate the sound, but don't need support for a range of rotations, they just need to be able to set the rotation on the sound "beam" since each speaker is different. an asymmetric dispersion at a 45 deg angle for example
• we will get the list of SPL formats that we want to support for the gdtf, but we do not need to worry about converting the values from one format to another, this is not something that is done. speakers are specified with a certain format. so the user just needs the option to choose the format and enter the value, if they change the format later it should remove the value and force them to input a new one

[ILane-VW]

Received the following information about different formats to support for sound level:

Here is a table with different excitation signals and weightings.
Excitation Signals and weightings are independent from each other and can be used in any combination.

Excitation Signal	Frequency Weighting	Unit	Time Weighting	Unit
Pink noise	None	DBZ, dB or DBSPL	Fast	F
IEC60268 "Program Simulation Noise"	A-Weighting	dBA	Slow	S
EIA426B "Loudspeaker test signal"	C-Weighting	dBC	Leq	Leq, Laeq, LCeq
AES75 "M-Noise"			Peak

[petrvanekrobe]

GDTF: For sound angles, we will not do a range - we will define the horizontal Left, horizontal Right, vertical Up, vertical Down, and Rotation.
MVR: Define materials by names (string).

[petrvanekrobe]

I have updated the proposal.md file (not yet the gdtf-spec) with the latest information.

[klinzey]

From the MVR standpoint I think we can use overrides and adjust min X, min Y, max X, max Y to the appropriate values when the speaker has an adjustable beam? I don't think we need an additional field in the MVR to store that data. Do we need a flag in the GDTF to mark it as adjustable or can we just use the fact that Min<>Max signifies that it is adjustable?

[klinzey]

We also need to define the orientation of the speaker geometry like we did with a light. There are a few things in the spec that depend on the orientation.

[klinzey]

On the todo list, we should also look at supports. We need to classify the speaker bumper and other connections as supports. If it doesn't exist already we may need to add optional subtypes to supports so we can separate Truss, Bumpers, Connections, Structure, etc. so we can specifically identify the type of structure.

[JGaudin]

Hello,

Here a few notes from review of the proposal at Meyer Sound :

Meyer Sound exclusively builds self-powered loudspeakers among now many others manufacturers, with some also including DSP (but that seems to be taken care of in the amp part). We should think in how to reconcile this, either by having the amp part of the description of the amplifier (prefered, as it will be 1 "device") or link them.

There is also a use case where the amplifier is in the loudspeaker, but the power supply is outside of the enclosure, generally for 2 reasons : space, or using longer low-voltage cabling that can be plenum rated for fixed installations.

[JGaudin]

About the current description :

Sound rotation angle :
This is confusing, and is generally driven by the horn rotation of the loudspeaker. Some loudspeakers have rotatable horns, the horn provides the directivity in the high frequencies. Should be more explicit to a sound person.

M-Noise in AES-75 context is now called Music Noise suggesting AES75MusicNoise

Max SPL weighting :
Generally accepted that "none" = Z

Max SPL :
Max SPL = dB SPL there is no other unit available, but it is described using the above weighting.

For example if an A or C filter is applied to the signal, the value will be written as 200 dB SPL A

Averaging time (time weighting):
Fast (F), Slow(S) and Peak(Pk) are defined. L is the "Length of the average and needs to be specified in minutes.

For example 200 dB SPL LAeq, 30 is A weighted signal averaged over 30 minutes.
Which I believe we could write as 200 dB SPL L A, 30
For peak values we'd have 200 dB SPL Pk Z or sometimes (Z)

Impedance is not used for self-powered loudspeakers.
Assuming that non-relevant fields can stay empty ?

About Sound Angle :
Loudspeakers are not producing energy forward with left and right boundaries, they would produce a bubble of energy in from the loudspeaker connected to it, that would behave different a various frequencies. The angle coverage is an average of all these and we should be careful of the future use of these values : it does not describes the power distribution of the energy in this case.

Line-arrays have proportional vertical angles, meaning they vary with frequency. Low frequencies are very wide, higher frequencies very narrow, up to a factor 10 sometimes.
We should have a field for "Proportional" or "Variable" that could also covers loudspeakers that have variable horns, like the L-Acoustics K2, for example.

[JGaudin]

Proposed new fields :

Internal amplifier (or Self-Powered) - Yes / No
External Power Supply - Yes / No
External Power Supply Voltage - value in VDC
External Power Supply compatible models - Text field / link (?)
Power, Maximum Long Term Continuous - Watts
Power, Idle - Watts
Operating voltage range Minimum - Voltage
Operating voltage range Maximum - Voltage
Operating voltage frequency - Hz
AC Power connector -
DC Power connector -
Power loop output - Yes / No
Audio Input - Analog, AES3, Dante, MILAN, AES67, other
Audio Connectors - XLR3, XLR5, Phoenix, Ethernet
Audio loop output - Analog, AES3, Dante, MILAN, AES67, other
User DSP - Yes / No
Latency - ms
IP rating
Rigging accessories - U-Braket, Single arm, Rotoclamp, Yoke, Eyebolt
Max SPL Reference RMS input level - dBV
Splay angle capability

[JGaudin]

The case for line-arrays needs to be considered here as they only work as an assembly of loudspeakers (geometries), but there are also other cases like sub arrays, and point-source arrays. All of these will have to flow through the MVR.

A line array, is an assembly of these properties :

• Amount of loudspeakers
• Models
• Rigging equipement (Top Grid, Transition frame (to other models), Pull back frames, etc)
• Inter-element splay angles
• Model splay angle capacity
• Total Weight
• Weight at forward and rear pickup points

[petrvanekrobe]

@JGaudin thank you. Can you please post an image/schematic view of a speaker and label for us the individual elements? What we want to get clear on is what is the terminology for individual parts of the speaker. Thank you!