Best Way to Measure Speaker Response (Delay) When Using DSP w/Convolver

Sliderule

Member
Thread Starter
Joined
Apr 1, 2024
Posts
31
Location
South Carolina
More  
Front Speakers
Soundlab Majestic 745's
I am currently using an Acourate convolver and DSP filters with a laptop. The signal processing delay is abour 2.7 seconds I am currently using acoustic timing with 2.7ms delay (3ft.) for swept frequency measurements.
I have looked at loopback but I don't understand how they would work with a convolver. Any assistance to make better measurements would be appreciated.

Slidrule
 
Hello @Sliderule, let me see if I understand you correctly... Your current kit consists of a laptop, the Acourate convolver some DSP filters and some unknown pre/amp and SoundLab speakers...

It would be helpful if you could say what laptop, player app, DAC? and measurement microphone you are using...

Off hand, if you are trying to measure your entire kit including your current convolution solution I would suggest you create sweep tone files that include a timing reference using the Generator function of REW... Then you can play the sweep files through your player and entire signal chain... Separately REW can listen for the Timing Reference and record an accurate measurement...

See Measuring with file playback onward at https://www.roomeqwizard.com/help/help_en-GB/html/makingmeasurements.html
Signal Generator Help... https://www.roomeqwizard.com/help/help_en-GB/html/siggen.html
More in this thread https://www.avnirvana.com/threads/t...ustic-treatment-time.15075/page-2#post-115416
This thread and the methods presented my be of some interest... https://www.avnirvana.com/threads/convolution-room-correction-and-room-curve.14005/
Last, but not least... A master class in Digital Room Correction...
 
Last edited:
I did a complete reply but for some reason it did not get posted. I am using Sony ES5400 Cd player, custom built phono preamp and Benchmark Preamp feeding a Lynx Hilo 2 DAC/ADC which is connected to a laptop unning an Acourate concolver to modify the signal which goes to the power amps thru the Hilo. The ABH2 amps are bridged and feed Soundlab electrostats. I do use acoustic timing reference when measuring. I have attached a screenshot of the frequency response ((DSP modified) at the listening position in my room. I have watched Mitch Barnett's video several times. I have also read his book focusing on the Acourate software and ignoring DSP crossovers. The biggest problem with his book is the 4-5 point type is too small. It really needs to get into detail about target curve creation and tuning which is lacking and I haven't found much in the way of detail on modifying the target curves on the Acourate forum. I will try creating a swept frequency file to test if it resolves the lconvolver latency issue.
 

Attachments

  • Screenshot (95).png
    Screenshot (95).png
    456.1 KB · Views: 3
What software are you using to create your FIR filters? You can create a target curve in REW... If you are creating you FIR filters with another tool maybe you can create in REW and export/import into whatever tool you are using...

The EQ function window in REW use the Target Settings to create your target curve...
BTY in REW a "house curve" is just a txt file...

* DD's Target House Curve - based on Bruel & Kjaer's
* Optimum Curve for HIFI equipment from B&K's Application Notes figure 5 page 4
*
* C-weighting compensation: Off
* Target level: 75.0 dB
* Smoothing: None
* Frequency Step: 10.0 Hz
* Start Frequency: 20.0 Hz
*
*
* Freq(Hz) SPL(dB)
*
20.000 3.000 0
30.000 3.250 0
etc...
 
Last edited:
I am using Acourate software to create the filters. The target curves I believe can be edited in text format but I haven't seen any detailed instructions on that and the additional subfilters on tweaking. I have attached a screenshot of one of my filters. The filters are very difficult to graphically edit as they move as the mouse is released. I am continuing to experiment to try to fine tune the target curves. one of the tests I did earlier was tol compare the REW frequency response to that of Acourate. For all pracitical purposes they were identical.
 

Attachments

  • Screenshot (42).png
    Screenshot (42).png
    329.2 KB · Views: 2
I updated the above post...
 
Your welcome... Sent you a direct message...
 
Back
Top