ProSoundWeb Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: 1 2 3 [All]   Go Down

Author Topic: Lake linear phase crossover  (Read 4272 times)

Helge A Bentsen

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1766
  • Oslo, Norway.
Lake linear phase crossover
« on: September 21, 2019, 10:35:54 AM »

When using a linear phase crossover in a Lake product, is there any difference in behavior between a 48dB/oct filter and a 49dB/oct brick wall filter, apart from the 1dB difference in slope?

I can get a steeper brick wall filter but then I have to double the alignment delay.
Logged

Frank Koenig

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1318
  • Palo Alto, CA USA
Re: Lake linear phase crossover
« Reply #1 on: September 21, 2019, 11:06:58 AM »

When using a linear phase crossover in a Lake product, is there any difference in behavior between a 48dB/oct filter and a 49dB/oct brick wall filter, apart from the 1dB difference in slope?

I can get a steeper brick wall filter but then I have to double the alignment delay.

I would guess there is not much difference except maybe deep in the stop-band, which is of little relevance for speaker crossovers. I would use Smaart, etc. to look at the filters' frequency responses. From this one could even back out the filter coefficients that Lab/Lake do not make available to mortals.

If you measure, post what you find. I'm curious.

--Frank
Logged
"Nature abhors a vacuum tube." -- John Pierce, Bell Labs

Helge A Bentsen

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1766
  • Oslo, Norway.
Re: Lake linear phase crossover
« Reply #2 on: September 22, 2019, 01:48:52 PM »

Logged

Peter Morris

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1467
Re: Lake linear phase crossover
« Reply #3 on: September 23, 2019, 07:17:12 AM »

When using a linear phase crossover in a Lake product, is there any difference in behavior between a 48dB/oct filter and a 49dB/oct brick wall filter, apart from the 1dB difference in slope?

I can get a steeper brick wall filter but then I have to double the alignment delay.

They will be similar, a bit different in shape, but it depends on how Lake have implemented the 48dB LR flat phase type.  Here is a comparison, note the brick-wall stop band in this example is 50 dB down.  Both HF and LP filters 1K3 ... My Guess at about 48dB for the Brick-wall and the other is a 48dB IIR
« Last Edit: September 23, 2019, 07:57:52 AM by Peter Morris »
Logged

Helge A Bentsen

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1766
  • Oslo, Norway.
Re: Lake linear phase crossover
« Reply #4 on: September 23, 2019, 01:12:38 PM »

Have Powersoft implemented FIR-filters you can load directly in Armonia now or do you still have to create a FIR-filter file in a different program and load the file?
Logged

Peter Morris

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1467
Re: Lake linear phase crossover
« Reply #5 on: September 23, 2019, 07:40:19 PM »

Have Powersoft implemented FIR-filters you can load directly in Armonia now or do you still have to create a FIR-filter file in a different program and load the file?

It now almost the same as Lake ... raised cosine, all-pass & FIR crossovers except you can't create linear phase 24 dB or 48 dB LR slope types.
Logged

Jeff Bankston

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2568
Re: Lake linear phase crossover
« Reply #6 on: September 24, 2019, 02:21:49 AM »

Does this apply to electronic crossovers in biamped and triamped systems ?
Logged

Helge A Bentsen

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1766
  • Oslo, Norway.
Re: Lake linear phase crossover
« Reply #7 on: September 24, 2019, 02:26:18 AM »

It now almost the same as Lake ... raised cosine, all-pass & FIR crossovers except you can't create linear phase 24 dB or 48 dB LR slope types.

Ok, I didn’t know Powersoft is this close now.
Still, probably faster to do dsp settings on Lake and transfer that to Powersoft when it’s done.
Logged

Helge A Bentsen

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1766
  • Oslo, Norway.
Re: Lake linear phase crossover
« Reply #8 on: September 24, 2019, 02:27:06 AM »

Does this apply to electronic crossovers in biamped and triamped systems ?


Yes.
Logged

David Sturzenbecher

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1966
  • So. Dak.
    • Sturz Audio
Re: Lake linear phase crossover
« Reply #9 on: September 24, 2019, 09:15:02 AM »

Ok, I didn’t know Powersoft is this close now.
Still, probably faster to do dsp settings on Lake and transfer that to Powersoft when it’s done.

....Now?  This have been a thing for powersoft for a few years.   The Armonia software was far more advanced the lake until their latest major rev.   I am not a fan of the new workflow.
Logged
Audio Systems Design Engineer
Daktronics, Inc.
CTS-D, CTS-I
AES Full Member

Helge A Bentsen

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1766
  • Oslo, Norway.
Re: Lake linear phase crossover
« Reply #10 on: September 24, 2019, 10:52:36 AM »

....Now?  This have been a thing for powersoft for a few years.   The Armonia software was far more advanced the lake until their latest major rev.   I am not a fan of the new workflow.

Don't get me wrong, Powersoft is great. I run my system on X8/K20.

It's just that for on the fly-dsp-development I find Lake to be faster. Some things you can do in Lake Controller directly needs a 3rd party program to generate a FIR-filter file and then import it into Armonia.
Logged

David Sturzenbecher

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1966
  • So. Dak.
    • Sturz Audio
Re: Lake linear phase crossover
« Reply #11 on: September 24, 2019, 11:25:13 AM »

Don't get me wrong, Powersoft is great. I run my system on X8/K20.

It's just that for on the fly-dsp-development I find Lake to be faster. Some things you can do in Lake Controller directly needs a 3rd party program to generate a FIR-filter file and then import it into Armonia.

The same can be said going the other way as well. In order to import raw FIR coefficients, you need a third party software for Lake, while they import directly into Armonia.
Logged
Audio Systems Design Engineer
Daktronics, Inc.
CTS-D, CTS-I
AES Full Member

Mark Wilkinson

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1104
Re: Lake linear phase crossover
« Reply #12 on: September 24, 2019, 11:34:56 AM »


It's just that for on the fly-dsp-development I find Lake to be faster. Some things you can do in Lake Controller directly needs a 3rd party program to generate a FIR-filter file and then import it into Armonia.

Don't all processors require an external FIR file to do anything beyond a few linear phase crossovers.... brick wall, pseudo LR24 or LR48 ?
I guess you're saying the on-the-fly linear phase crossover implementation is faster with Lake?

I got to spend a brief time with Armonia for the first last week....felt pretty clunky really.
Linea has the slickest linear phase crossover implementation I've encountered.  It auto manages necessary delays as you move crossover frequencies.  But alas, it only has a psuedo LR24.
Logged

Helge A Bentsen

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1766
  • Oslo, Norway.
Re: Lake linear phase crossover
« Reply #13 on: September 24, 2019, 11:39:54 AM »

Don't all processors require an external FIR file to do anything beyond a few linear phase crossovers.... brick wall, pseudo LR24 or LR48 ?
I guess you're saying the on-the-fly linear phase crossover implementation is faster with Lake?

I got to spend a brief time with Armonia for the first last week....felt pretty clunky really.
Linea has the slickest linear phase crossover implementation I've encountered.  It auto manages necessary delays as you move crossover frequencies.  But alas, it only has a psuedo LR24.

Yes, Lake have Linear Phase 24dB, 48dB/oct, Brick Wall and All-pass as "drag'n drop".

Powersoft and Lake both performs on the highest level, for me it's a difference in work flow. I'm happy running a system on either platform. 
Logged

Mark Wilkinson

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1104
Re: Lake linear phase crossover
« Reply #14 on: September 24, 2019, 11:44:26 AM »

Hey David, I know you have experience on a lot of platforms, including q-sys.
I have a number of speakers working well with q-sys processing, and I'm thinking of selling my processors and using a Core110f for all speaker management.
Any concerns you see for live ? (other than  needing a racked switch)

Helge, sorry for the swerve....pls pardon..
Logged

David Sturzenbecher

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1966
  • So. Dak.
    • Sturz Audio
Re: Lake linear phase crossover
« Reply #15 on: September 24, 2019, 12:31:33 PM »

Hey David, I know you have experience on a lot of platforms, including q-sys.
I have a number of speakers working well with q-sys processing, and I'm thinking of selling my processors and using a Core110f for all speaker management.
Any concerns you see for live ? (other than  needing a racked switch)

Helge, sorry for the swerve....pls pardon..

A post on the Facebook Qsys user group pages a few months back reviled that a transfer function through a 110f wasn't perfectly flat.   I couldn't immediately dig up the post...and can measure mine 110f when i get some time, but I don't recall the deviations being horrible.   It was clearly nothing you would worry about in a conference room situation.   You would have to decide for yourself given your exact application.  IIRC other users indicated this transfer function anomaly wasn't present in a 510i.

That aside, the 110f, with the forthcoming soft dante integration would be a nice speaker management unit.   You could even tie it into full system management, adding multiple inputs for console switching, MC and BGM inputs (over dante...or not).   

Qsys 8.1 added the "Flat Top Parametric EQs" aka Lake Mesa Filters which is a nice touch. It already had the "drag and drop" style FIR crossovers.   They don't list them as db/oct rolloffs, but give a stop band attenuation parameter (40dB, 60dB, 80dB, or 100dB) and a Transition Bandwidth of anywhere in between 1 octave and 1/4 octave.

I currently use a BSS BLU-806 for the duty of switching desks and mixing in background music, that is automatically being ducked by an MC mic for festival applications.   It works great and does the job, but if i were to do it all over again, i would be looking at a 110f.

The major downside of the 110f, that many people can't seem to get over....a fixed 48kHz clock...
Logged
Audio Systems Design Engineer
Daktronics, Inc.
CTS-D, CTS-I
AES Full Member

Frank Koenig

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1318
  • Palo Alto, CA USA
Re: Lake linear phase crossover
« Reply #16 on: September 24, 2019, 01:33:40 PM »

Don't all processors require an external FIR file to do anything beyond a few linear phase crossovers.... brick wall, pseudo LR24 or LR48 ?

Yes. So far as I'm concerned FIR filters only get interesting when you can use them to implement custom transfer functions. Having a few stock constant-delay filters to play with never hurt, I suppose, but I also assume that tables of coefficients for them are widely available for those who don't have the tools set up to generate them.

While the Powersoft interface (Armonia) is still more arbitrary and complicated than it needs to be (in my opinion), and lacks a little in UI dressing (colors, etc.), at least I can use it. I tore my hair out trying to use the Lake interface, and not being able to load custom FIR filters makes it useless for my little experiments. Lab Gruppen builds great amplifiers. They just need to write a new interface for the Lake stuff and open it up for developers.

--Frank
Logged
"Nature abhors a vacuum tube." -- John Pierce, Bell Labs

David Sturzenbecher

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1966
  • So. Dak.
    • Sturz Audio
Re: Lake linear phase crossover
« Reply #17 on: September 24, 2019, 02:11:59 PM »

I tore my hair out trying to use the Lake interface, and not being able to load custom FIR filters makes it useless for my little experiments. Lab Gruppen builds great amplifiers. They just need to write a new interface for the Lake stuff and open it up for developers.

--Frank

You can now do this with FIR designer.  It's still not easy, but it can be done.
Logged
Audio Systems Design Engineer
Daktronics, Inc.
CTS-D, CTS-I
AES Full Member

Frank Koenig

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1318
  • Palo Alto, CA USA
Re: Lake linear phase crossover
« Reply #18 on: September 24, 2019, 02:27:22 PM »

You can now do this with FIR designer.  It's still not easy, but it can be done.

Good to know. Thanks. -F
Logged
"Nature abhors a vacuum tube." -- John Pierce, Bell Labs

Mark Wilkinson

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1104
Re: Lake linear phase crossover
« Reply #19 on: September 24, 2019, 04:44:35 PM »

A post on the Facebook Qsys user group pages a few months back reviled that a transfer function through a 110f wasn't perfectly flat.   I couldn't immediately dig up the post...and can measure mine 110f when i get some time, but I don't recall the deviations being horrible.   It was clearly nothing you would worry about in a conference room situation.   You would have to decide for yourself given your exact application.  IIRC other users indicated this transfer function anomaly wasn't present in a 510i.

That aside, the 110f, with the forthcoming soft dante integration would be a nice speaker management unit.   You could even tie it into full system management, adding multiple inputs for console switching, MC and BGM inputs (over dante...or not).   

Qsys 8.1 added the "Flat Top Parametric EQs" aka Lake Mesa Filters which is a nice touch. It already had the "drag and drop" style FIR crossovers.   They don't list them as db/oct rolloffs, but give a stop band attenuation parameter (40dB, 60dB, 80dB, or 100dB) and a Transition Bandwidth of anywhere in between 1 octave and 1/4 octave.

I currently use a BSS BLU-806 for the duty of switching desks and mixing in background music, that is automatically being ducked by an MC mic for festival applications.   It works great and does the job, but if i were to do it all over again, i would be looking at a 110f.

The major downside of the 110f, that many people can't seem to get over....a fixed 48kHz clock...

Thanks David,
Yeah, I got wind of the 110f transfer issue, but quickly decided it's no big deal. Attached is line-in straight to line-out, to compare with yours.
I've heard that the phase tails disappear when using an I/O frame input card, or the inputs of the CXD-Q amps. I guess I should check that, but phase looks good enough to me, so i have a hard time caring.

Really glad to hear you think it has speaker / system management potential.
The coming Dante integration is what's pushing me to jump whole hog into it.
The Mesa EQ is nice.  Did you see FIR tap count allows up to a whopping 16,384 taps ?(I'm using 8ch with up to 4096 taps per channel before maxing out a 110f)
The limiters now work like PLD amps with regular voltage settings....(about time)
Been using snapshots which allow instantaneous switch over from different speaker processing, including full linear phase to full min phase, for comparisons. Can swap designs and FIR files on the fly, with no clips, pops, interruptions.
I applaud the QSC folks for steadily improving the System Designer software.... I'm sold  :)

Oh, and a quick plug for FirDesigner.   I have no idea how its direct-to-processor capability works for Lake and Powersoft, but the files it builds for standard csv and wav FIR have worked super, whether building FIR files for linear phase processing or low latency min-phase processing.

Logged

Peter Morris

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1467
Re: Lake linear phase crossover
« Reply #20 on: September 24, 2019, 05:13:14 PM »

You can now do this with FIR designer.  It's still not easy, but it can be done.

Yes for the PLM's and no for LM26 at the moment as I understand :-(
Logged

ProSoundWeb Community

Re: Lake linear phase crossover
« Reply #20 on: September 24, 2019, 05:13:14 PM »


Pages: 1 2 3 [All]   Go Up
 



Site Hosted By Ashdown Technologies, Inc.

Page created in 0.043 seconds with 23 queries.