Go Back   Home > Forums > >

PC Based Computer music servers, crossovers, and equalization

CamillaDSP - Cross-platform IIR and FIR engine for crossovers, room correction etc.
CamillaDSP - Cross-platform IIR and FIR engine for crossovers, room correction etc.
Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools
Old 26th February 2021, 05:46 PM   #1641
sarcastic1 is offline sarcastic1
diyAudio Member
 
Join Date: Oct 2020
Location: mayday
NO WAY!
There are known unknowns and unknown unknowns...



I know which one I prefer,

so, so long as I know what I don't know then that's fine by me.
  Reply With Quote
Old 26th February 2021, 06:55 PM   #1642
HenrikEnquist is offline HenrikEnquist  Sweden
diyAudio Member
 
Join Date: Apr 2016
Location: Lund
So the issue is simply this: the measured effect of the IIR filters isn't what theory says it should be.

TNT has a point, there are too many unknowns in this to figure out at which stage things go wrong.
I would start by applying a simple config with a single Peaking filter. Then I would try to measure that with REW directly on the analog output, without involving amplifiers or speakers. If that works and produces the expected result, then it's time to add something, like the amplifier. The once that works I would add the speaker and put the mic very close to it (few cm). Finally after that works, it's time to do the real measurement.

You wrote that you had measured on the analog out. Could you show something of those measurements?
__________________
CamillaDSP - cross platform dsp engine
Reporting an issue with CamillaDSP? Please attach the config file and the log from a run with "debug" log level.
  Reply With Quote
Old 26th February 2021, 09:24 PM   #1643
emailtim is offline emailtim  United States
diyAudio Member
 
Join Date: May 2005
Location: USA
Quote:
Originally Posted by HenrikEnquist View Post
... I have been working on trying to optimize the RustFFT library, and have managed to get some quite nice improvements.
This shows the speed of my new version compared to the current release, for all lengths 1-1024:
Attachment 900518
The really large improvements are for prime lengths, which aren't really used by CamillaDSP. The power-of-two lengths were already very good so I have not managed to improve them. The benefit for CamillaDSP is when using the synchronous resampler, where conversions between the 44.1 and 48kHz groups should become about 40% faster.


....
Would this be of any value or is it too green ?

Released RustFFT 5.0.0-experimental.1: Now faster than FFTW!
  Reply With Quote
Old 27th February 2021, 08:09 AM   #1644
sarcastic1 is offline sarcastic1
diyAudio Member
 
Join Date: Oct 2020
Location: mayday
Quote:
Originally Posted by HenrikEnquist View Post
So the issue is simply this: the measured effect of the IIR filters isn't what theory says it should be.
YEP.
Quote:
Originally Posted by HenrikEnquist View Post
I would start by applying a simple config with a single Peaking filter. Then I would try to measure that with REW directly on the analog output, without involving amplifiers or speakers.
I did that with the loopback test.

Output XLR is connected to Input XLR. Simple.

Did you understand, that's what connecting the output to the input of the sound card does??

That gives you ONE reference channel (say the LH).
In the event I detected a slight 1dB-1.5dB roll off at 19-20khz...


At the same time the amp is fed with the output of the SPDIF, which gives it the stereo output via an (external) DAC.
(so you don't lose your LH channel output, you duplicate it).

That LH channel output is fed to the AMP, then back from the output of the amplifier via a voltage divider network (say 10:1), and now hey presto you can visualise any part of the audio spectra in real time via "scope" or record it as a 24bit audio .wav file LH (original) versus RH (amp playback) for posterity.

This gives you an exact comparison of input versus output of any stage of the process you care to mention,
no unknowns,

what you see is what you get.

Then do a SCAN 20hz-20khz.


The camilla DSP if engaged shows you the exact result on both source and destination, which can then be fed into REW as a .wav audio file for analysis.
You can compare it with "no filter" or "biquad with all values set to zero".

Quote:
Originally Posted by HenrikEnquist View Post
If that works and produces the expected result, then it's time to add something, like the amplifier.
Don't need to. see above it DOES.

Quote:
Originally Posted by HenrikEnquist View Post
Then once that works I would add the speaker and put the mic very close to it (few cm). Finally after that works, it's time to do the real measurement.
Exactly, which is why you can compare the AMP output, the source and the mic output.

Quote:
Originally Posted by HenrikEnquist View Post
You wrote that you had measured on the analog out. Could you show something of those measurements?
Maybe my explanations were not so clear.
Another problem you CANNOT place a mic a few cms from a multi driver speaker, it won't work.
But that is a pro audio engineer/acoustician issue, which I don't need to describe
  Reply With Quote
Old 27th February 2021, 08:24 AM   #1645
fluid is online now fluid  Australia
diyAudio Member
 
Join Date: Jan 2009
Quote:
Originally Posted by sarcastic1 View Post
I did that with the loopback test.

Output XLR is connected to Input XLR. Simple.

Did you understand, that's what connecting the output to the input of the sound card does??

That gives you ONE reference channel (say the LH).
In the event I detected a slight 1dB-1.5dB roll off at 19-20khz...
I don't think you are understanding what Henrik is suggesting.

A loopback is to measure the card against itself to allow that to be calibrated.

Henrik is suggesting that you add a known filter to the output so that when you record it with a loopback you can see if the electrical output of the soundcard gives the response of that filter.

Quote:
Originally Posted by sarcastic1 View Post
But that is a pro audio engineer/acoustician issue, which I don't need to describe
Please dial this back you are not a master lecturing an apprentice.
  Reply With Quote
Old 27th February 2021, 09:27 AM   #1646
sarcastic1 is offline sarcastic1
diyAudio Member
 
Join Date: Oct 2020
Location: mayday
Quote:
I don't think you are understanding what Henrik is suggesting.
A loopback is to measure the card against itself to allow that to be calibrated.

Henrik is suggesting that you add a known filter to the output so that when you record it with a loopback you can see if the electrical output of the soundcard gives the response of that filter.


I most certainly DO understand.
maybe this little photo "simple comme bonjour" will make it clear?


CamillaDSP - Cross-platform IIR and FIR engine for crossovers, room correction etc.-harness_loop_d-jpg

calibrating a sound card + v the amplifier output simply cannot be easier using this arrangement.



Once you have got the Amplifier response sorted and recorded on the sound file (detected a -5dB roll off from 17-20khz), you add the filter via Camilla, perform another scan test, and hey presto you edit the output of that (copy paste) in your favourite stereo audio editor putting that in the RH channel compared with the flat output from the LH channel (which you just got back from the harness above.


You can't get it wrong.


You can do any comparison you want.
Output of amp before and after camilla filter, output of DAC before and after filter, sine wave, square wave anything you like.


You can then compare it with the mic returns.
In practice my positioning of the mics in that "dead" room were validated because 2 of the (JBL) speakers came out as nigh on identical, while the other test (Vandersteen) showed them as wildly different.


Validation is everything.

That test system works.


Listening to them (audio scan test) after this aural re-education immediately made us recognise the difference, which was not obvious before.

What don't you understand?
Attached Images
File Type: jpg harness_loop_D.jpg (97.2 KB, 219 views)

Last edited by sarcastic1; 27th February 2021 at 09:31 AM.
  Reply With Quote
Old 27th February 2021, 09:54 AM   #1647
fluid is online now fluid  Australia
diyAudio Member
 
Join Date: Jan 2009
Quote:
Originally Posted by sarcastic1 View Post
What don't you understand?
Your attitude towards people who are trying to help you with a problem.
  Reply With Quote
Old 27th February 2021, 11:20 AM   #1648
TNT is offline TNT  Sweden
diyAudio Member
 
Join Date: Apr 2003
Location: Sweden
CamillaDSP - Cross-platform IIR and FIR engine for crossovers, room correction etc.
Quote:
Originally Posted by sarcastic1 View Post
What don't you understand?
Honestly - you are babbling - get to the point. No need to be sarcastic.

//
__________________
More distortion to the people! Timing . . . . is paramount.
  Reply With Quote
Old 27th February 2021, 11:50 AM   #1649
sarcastic1 is offline sarcastic1
diyAudio Member
 
Join Date: Oct 2020
Location: mayday
Quote:
babbling
Quote:
Your attitude ...trying to help you with a problem
I don't "babble" and I don't see a "problem".
I have been trying to describe in detail the validation strategy.
I thought this was useful, but OK, if you don't like it, I won't bother.

1/ The validation strategy of the speakers clearly matches.
(it's not the first time, and it's by far the most complex to manage). This is good.

2/ The validation of the DSP, also clearly is throwing up interesting results, but which appear to be different from 1/

3/ The validation of the amps & i/o calibration, clearly functions, but needs a little more work to be sure.
Astonishingly, it appears all amps tested were absolutely the same from 30hz>around 8-9khz, which is the limit of human male hearing over anything exceeding 45yrs old.


I will do another test again to see what happens at the "business end" into a resistive load.
We deployed camilla for my last resonance "room resonance removal" test (which was attended by a number of people at a "hi end" audio shop).
This was at LF.

Some liked it, some didn't.

I suspect REW is buggy, and doesn't display stuff properly.
(which I found on their previous release versions).
  Reply With Quote
Old 27th February 2021, 02:19 PM   #1650
TNT is offline TNT  Sweden
diyAudio Member
 
Join Date: Apr 2003
Location: Sweden
CamillaDSP - Cross-platform IIR and FIR engine for crossovers, room correction etc.
If you think there is apriblem with one part of a system, you isolate that part an verify it in that particular parts external interfaces - not in the systems external interfaces. 101 trouble shooting 1st lesson.

//
__________________
More distortion to the people! Timing . . . . is paramount.
  Reply With Quote

Reply


CamillaDSP - Cross-platform IIR and FIR engine for crossovers, room correction etc.Hide this!Advertise here!
Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
The room correction or speaker correction? What can we do with dsp power now availabl Raimonds Full Range 233 28th January 2017 07:51 AM
Introducing OpenDRC, Open Digital Room Correction engine minidsp miniDSP 20 20th January 2016 05:37 PM
What the difference between dsp room correction eq and software correction erez1012 PC Based 0 10th March 2014 07:07 PM
Writing a Cross-Platform, Free Software Modeling Tool and TS-Parameter DB justinzane Software Tools 6 31st December 2013 06:55 AM
FS: DAC, room-correction, active crossovers, amp, speakers! taloyd Swap Meet 4 14th April 2009 03:16 PM


New To Site? Need Help?

All times are GMT. The time now is 11:32 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2021 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2021 DragonByte Technologies Ltd.
Copyright ©1999-2021 diyAudio
Wiki