Welcome, Guest
Username: Password: Remember me
This is the optional Forum header for the Suggestion Box.
  • Page:
  • 1

TOPIC:

44/48k plugin for 2x4HD 5 years 2 months ago #27830

  • denver8me4dinner
  • denver8me4dinner's Avatar Topic Author
  • Offline
  • Elite Member
  • Elite Member
  • 3038564518
  • Posts: 285
  • Thank you received: 58
For us car audio folks the 1024 taps isn't always enough when we can make better filters at a lower sample rate.

The 2x4HD can do some pretty nifty stuff would be nice to be able to use some 3rd party convolution software to fix more than just a crossover alone.

Is there any way to make a 2x4HD plug-in for 44k or 48k?

Thanks in advance
Andrew :)

Please Log in or Create an account to join the conversation.

44/48k plugin for 2x4HD 5 years 1 month ago #28008

  • devteam
  • devteam's Avatar
  • Offline
  • Administrator
  • Administrator
  • Posts: 8984
  • Thank you received: 1306
Hi Andrew,

It's a good suggestion indeed.. We'll consider it for the future as it would allow a lot more taps indeed! :-)

Devteam
miniDSP, building a DSP community one board at a time.

Please Log in or Create an account to join the conversation.

44/48k plugin for 2x4HD 4 years 8 months ago #30945

  • astroguy1
  • astroguy1's Avatar
  • Offline
  • New Member
  • New Member
  • Posts: 13
  • Thank you received: 1
I have my 2x4HD working well. I'm new to FIR filtering and was surprised at how many Taps are needed to do any useful Hi-Pass filtering at low frequencies, eg. 80Hz (crossover from subs to mains). I've assigned the number of taps asymmetrically : 1500/channel for 80Hz Hi-Pass and 500/channel for 80Hz Lo-Pass. This only allows me to define 12dB/oct slopes.

My request is to be able to select a 44.1 or 48 kHz sample rate in the plug-in to effectively double the number of Taps I have available. Since my source material is all 44.1 kHz, I get no benefit from running the 2x4HD at 96kHz.

thanks,

Please Log in or Create an account to join the conversation.

44/48k plugin for 2x4HD 4 years 8 months ago #30994

  • john.reekie
  • john.reekie's Avatar
  • Offline
  • Platinum Member
  • Platinum Member
  • Posts: 3775
  • Thank you received: 1588
At 80 Hz, just use IIR (i.e. the regular crossover).
I am not miniDSP support.

"You must ask the right questions." - Dr. Alfred Lanning's hologram.
-> Have you read the User Manual??
-> Have you drawn and posted a diagram?
-> Have you posted a screenshot?
-> Have you posted your config file?

Please Log in or Create an account to join the conversation.

44/48k plugin for 2x4HD 4 years 8 months ago #31034

  • astroguy1
  • astroguy1's Avatar
  • Offline
  • New Member
  • New Member
  • Posts: 13
  • Thank you received: 1
That is indeed what I have been using for many years. My reason for buying the 2x4HD is to experiment with FIR filtering. The motivation is to have linear phase cross-overs to simplify the summing of the drivers through the crossover region- in this case my mains and subwoofers. Where the system is minimum phase, I still use IIR filters.

Please Log in or Create an account to join the conversation.

44/48k plugin for 2x4HD 4 years 8 months ago #31153

  • denver8me4dinner
  • denver8me4dinner's Avatar Topic Author
  • Offline
  • Elite Member
  • Elite Member
  • 3038564518
  • Posts: 285
  • Thank you received: 58
An IIR crossover at80hz would still have significant spacial time smear .

Try using rephase and making a linearized IIR crossover. It uses less taps and should be able to achieve a linear phase crossover at80hz LR4

i would use energy centering and either a rectangle or hann window (or one of the terms)

Keep in mind one your past about 40db stop band attenuation any small ripples in the responce are not heard. But you should still be able to get a clean filter just linearized.

80hz LR4 you should be able to get away with the 1024 taps on each side with a less aggressive window applied, whatever window you use on sub use the same type window on midbass as its centering and FFt length will be slightly different if it's not.
The following user(s) said Thank You: astroguy1

Please Log in or Create an account to join the conversation.

Last edit: by denver8me4dinner.

44/48k plugin for 2x4HD 4 years 8 months ago #31240

  • astroguy1
  • astroguy1's Avatar
  • Offline
  • New Member
  • New Member
  • Posts: 13
  • Thank you received: 1
Thanks for the info! I will try that next.

Please Log in or Create an account to join the conversation.

44/48k plugin for 2x4HD 4 years 6 months ago #32038

  • Leif
  • Leif's Avatar
  • Offline
  • Junior Member
  • Junior Member
  • Posts: 27
  • Thank you received: 1
Hello!

I am using this way. I create the crossovers with the internal IIR filters and then apply FIR filters to get rid of the phase problems of the IIRs and to save taps. This works quite well for frequencies above 150 Hz.
For the LR 4th order crossover at 90 hz I have 1693 taps left, which creates this FIR filter in rePhase.



It looks good regarding the amplitude, but the phase is showing some ripple. If you import that filter into REW and have a look at the group delay that does not look very nice.



A 48 kHz plugin would be ideal for lower frequencies.
Attachments:

Please Log in or Create an account to join the conversation.

Last edit: by Leif.

44/48k plugin for 2x4HD 4 years 6 months ago #32057

  • pos
  • pos's Avatar
  • Offline
  • Platinum Member
  • Platinum Member
  • Posts: 354
  • Thank you received: 200
Hello Leif,

That group delay curve might not look nice, but how this relates to perceived sound quality is another matter altogether, and not an easy one (as is phase linearity).
I suppose you are using iterative optimization here? This process only focuses on magnitude, thus potentially letting the phase trace drift away from target...
In your situation you could try linearizing only part of the phase shift (eg LR24 linearization for a LR48, etc.) without optimization.
The following user(s) said Thank You: devteam

Please Log in or Create an account to join the conversation.

44/48k plugin for 2x4HD 4 years 6 months ago #32060

  • Leif
  • Leif's Avatar
  • Offline
  • Junior Member
  • Junior Member
  • Posts: 27
  • Thank you received: 1
Hello!

Thank you for the tip with no optimization. That looks better in the group delay, but the response is +/-0.45 dB.
Since the crossover is only 24 dB it does not make sense to only compensate the 12 dB phase shift, as this would make the summing more complicated / worse.
With a sample rate of 48 kHz no compromises would be necessary and that is the real cause of the problem.

Maybe it is possible to control the DSP with Sigma Studio directly. With that you could use all the functions, which are possible with this DSP including use of 48 kHz. Has anyone tried Sigma Studio with any MiniDSP product yet?

Please Log in or Create an account to join the conversation.

44/48k plugin for 2x4HD 4 years 5 months ago #32085

  • Leif
  • Leif's Avatar
  • Offline
  • Junior Member
  • Junior Member
  • Posts: 27
  • Thank you received: 1
Hello!

Today is a good day, I found a way how to reduce the phase / group delay ripple.
After I have done the first run, I had the result with the ripple in the phase. Then I used the phase EQ to put the same ripple in the phase target curve. After they looked the same I inverted the phase EQ.
With this new target curve the phase response has less ripple.

Please Log in or Create an account to join the conversation.

Last edit: by Leif.
  • Page:
  • 1
Moderators: devteam