Welcome, Guest
Username: Password: Remember me
  • Page:
  • 1

TOPIC:

Rephase linear phase filter /2x4HD 3 years 4 hours ago #52224

  • gmanoovaloo
  • gmanoovaloo's Avatar Topic Author
  • Offline
  • New Member
  • New Member
  • Posts: 4
  • Thank you received: 0
HELLO from Mauritius
i am using rephase 1.4.3 to generate FIR to minidsp 2x4hd. The purpose is to replace the 2 xovers that I initially set up in minidsp:
1: low pass at 100 hz LR 48 db and 2:high pass at 100 hz LR 48 db
2000 and 1000 taps allocated respectively. Once generated using default parameters ,the graph looks quite acceptable. After upload and 'send to dsp' the curve starts sloping down at 300hz for the high pass and at around 150hz for the low pass. The resulting sound is worse (thicker midbass) than with the bypassed native xovers.
I tried 48khz and 96khz but no change .
What am I doing wrong ? Already researched the forum but could not find any tip. Could 'optimization' help?

please see the screen shots (oops getting request entity too large ..will send pics separately )
Awaiting any help. THANKS
Gilbert

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

Rephase linear phase filter /2x4HD 3 years 4 hours ago #52225

  • gmanoovaloo
  • gmanoovaloo's Avatar Topic Author
  • Offline
  • New Member
  • New Member
  • Posts: 4
  • Thank you received: 0
Attachment file size exceed limit allowed by configuration :(

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

Rephase linear phase filter /2x4HD 2 years 11 months ago #52350

  • pos
  • pos's Avatar
  • Offline
  • Platinum Member
  • Platinum Member
  • Posts: 366
  • Thank you received: 206
Hello,

I might be wrong, but I think the 2x4HD will only run at 96kHz, so you need to select this sample rate.
As for the slopes being different between the rephase preview (in red) and the one shown in the 2x4HD interface, this might be due to a difference in scale, but also to a truncation of the FIR if the number of taps exceed what the 2x4HD expects.

Also note that if the shape of the resulting LR filter (in red) is different that the target one (in blue) then it will not be an LR anymore, and the low-pass and high-pass will most probably not sum properly.

And of course, what matters in the end is the final acoustical curves (ie "natural" rolloff + electrical filter), and this is where you want to get an LR (or whatever filter you are aiming for) and a complementary crossover.

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

Rephase linear phase filter /2x4HD 2 years 11 months ago #52351

  • gmanoovaloo
  • gmanoovaloo's Avatar Topic Author
  • Offline
  • New Member
  • New Member
  • Posts: 4
  • Thank you received: 0
Hello pos

Your input is much appreciated.
As i mentionned before, i did try the 96khz sample, but the slopes/graphs did not match. But there were in rephase déviations between the blue and red graphs. So maybe not sufficient taps allocated..
2x4hd has 4096 taps available in total.
I used 2000 for the low pass (left and right bass are summed in the minidsp) and 1000 Taps per channel for High pass. So 4000.
Could it be that the 2x4hd has limitations to replicate exactly the rephase xover ?

I am not technically savvy at this level but am passionate about music reproduction.

Mauritius being so remote does not help
Regards

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

Rephase linear phase filter /2x4HD 2 years 11 months ago #52352

  • gmanoovaloo
  • gmanoovaloo's Avatar Topic Author
  • Offline
  • New Member
  • New Member
  • Posts: 4
  • Thank you received: 0
I Just tried in rephase. The blue and red graphs become practically identical when using 4000 taps for LR 24db low pass at 100hz.
But then i Will have not enough taps for the High pass crossover when sending to minidsp...
Additionally minidsp 2x4hd does not allow more than 2048 taps for a FIR. So 4000 taps exceed the limit even though 4096 taps available.

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

  • Page:
  • 1
Moderators: devteam