You are not logged in. Please login or register.
To all,
I seem to be getting requests for using Nion's as a speaker controller, along with DSP needs.
Given that many systems are stereo or LCR, it might be helpful to have filter/EQ blocks that can have multiple I/O, with the filters locked together across channels?
Anyone else think this would be helpful?
I guess this could also be expanded to any DSP block typically used in speaker processor...
Also, would this lower the DSP resources needed?
Thanks,
Greg
Hi Greg-
This is already something that is incredibly simple to build. The magic control functions are handled by the "Auto Group" device. This device will manipulate, in a peer fashion, all controls that contain the same RUID. So, if you are working on a stereo device (or muti-band device), you can install an 8-band parametric, a compressor, high pass, low pass, limiter, etc on each channel, then add the "Auto Group" device and all similar devices (for example the two 8 band parametric EQ's) will be able to manipulate each other. So, you only need to copy out the controls for one of them to control both of them.
The one place this kind of thing could get sticky, and thus should be implemented by nesting DSP Blocks, is where you have high pass and low pass filters that you do NOT want to be set the same. When you use the Auto Group object in a DSP Block, if there are two high pass filters, they will be set the same. If you are trying to do different bandpasses with the high pass filters, you need to move them outside the block that contains the Auto Group device.
This would not reduce the DSP resources required.
Oh yes... I should also mention a special case where simply using the Auto Group does NOT work as you would expect...
Graphic Equalizers.
If you dig into the schematic of the graphic eq's you will note that each graphic eq is made up of two processing objects, one for the even bands and one for the odd bands.
The catch here is that the first control in both the "Even" and "Odd" has the RUID "band_gain_1" since they are both the first control in that device.
So, since the Auto Group device essentially creates a "Control Group" for all devices with the same RUID, you will end up with both filters moving at the same time even though they are at different frequencies because they both have the same RUID.
The simplest way to get around this is to just copy all the controls to a tab, set Peer wiring nodes on them and wire them accordingly. Perhaps a better fix would be to assign even numbered RUIDs to the "Even Bands" device and odd numbered RUIDs to the "Odd Bands" device, then it would just work without issue. However, there are some much bigger fish to fry to continue to bring newer and better features to NWare.
(I'm just going to leave out my commentary about why I think using Graphic Equalizers in the DSP domain is a complete waste of resources and how anyone who cares to learn about parametric EQs can easily get a MUCH better result with a little bit of critical listening and knob twiddling.)
Josh,
Real men use FIR filters
Parametrics are soooo old school...
Greg
Greg-
The difference is that I prefer to use parametric EQ's when I am mixing monitors. No one that I know of has an FIR Filter that could be used in the same manner.
hmm... product idea.
NOTE: You read about it here first.
Josh, why would a Peavey employee discourage the use of DSP-hungry algorithms? Think how many n6s you could sell if everyone used 1/3 octave EQs on every signal path? I should expect v1.6 to include 31 band graphics as a default on all mixer inputs and outputs at the very least.
Well, graphic EQ's are extremely limited in their functionality. Why not use a tool that actually accomplishes what you want to do instead of being forced into a pigeon hole.
Someone who uses 1/3 (or worse, 1/6) octave Graphic EQ's on every signal path, without good reason, needs to be taken out back and shot, repeatedly. Then dismembered and fed to the local stray dog population. 
When I first learned about parametric EQ's (wow... it's already been over 15 years???) I got all kinds of excited and got my hands on a Symetrix 551 (or whatever the Symetrix mono 5 band parametric was in the 500 series) parametric EQ. I still have it today, and you're not going to pry it out of my hands without a fight (along with the 565 Comp/Limiter). I learned all about how useful it is to be able to adjust the frequency and the Q of the filter in addition to just running the level of the filter up and down. Of course, I'm also the kid who had a dual 10 band graphic EQ on my home stereo when I was growing up... complete with LED's on the faders and the LED spectrum display (I guess I learned a lot playing around with that thing!)
Anyhow, sorry I got a bit off topic there... While the business is all about selling hardware, there is a lot to be said for intellegent design. I've seen far too many examples of system designs made FAR too complicated for no good reason. When you buy NIONs, we give effectively give you pallets of rope. If you want to use it to hang yourself and your customers, you can... or, you can knit it together into a sturdy and robust net that covers everything the customer needs and then some. The same can be said for most of the DSP products out there that allow you to build your own signal flow.
I suppose you are correct however, Jason. I should encourage the use of as many 1/6 octave EQ's as possible. After all, the more 61 band parametrics that are being used with fixed frequency and fixed Q, the more DSP everyone will need. Why do you suppose 1/6 octave EQ's never really caught on with the touring crowd? Even today, it's tough to get them interested in a parametric.
Powered by PunBB, supported by Informer Technologies, Inc.
Generated in 0.033 seconds, 12 queries executed