Impulse TAP length

grizaudio

Registered
Thread Starter
Joined
Apr 25, 2021
Messages
2
Hi, Can someone please describe how/where TAP length is manually adjusted for impulse response export..
I would like to run a 1:1 ratio of Taps to Sampling rate, but can't see how to adjust for this.

Within rephase (the software I normally use for convolution) this process is a simple data entry cell, whereby any number of Taps can be entered.
Once calculated Rephase would provide a visual representation of how well the response curve has been modelled v's tap count.

Any assistance would be amazing, as I am having difficulty locating this functionality in REW.
Thanks Steve
 
Last edited:

grizaudio

Registered
Thread Starter
Joined
Apr 25, 2021
Messages
2
Ok, I got a hint that the new early access build allowed for tap management, and it does. A couple of comments:
1. The software doesn't provide a visual representation of response accuracy vs tap count
2. Tap count isn't customisable by sample rate, unless you export each individually
3. It would be great if tap count could be manually entered, so tap count could mirror sample rate
1682414515241.png
 
Last edited:

John Mulcahy

REW Author
Joined
Apr 3, 2017
Messages
7,342
The software doesn't provide a visual representation of response accuracy vs tap count
You can see the effect of different lengths by adjusting the windows on the Impulse graph, either through the IR windows dialog (which shows the sample count within the window) or by dragging the L and R markers at the top of the impulse graph. If you drag the markers you automatically get a response preview, or you can use the option in the graph controls to always show the windowed magnitude response.

Tap count isn't customisable by sample rate, unless you export each individually
No, it isn't, but then REW is primarily a measurement tool.

It would be great if tap count could be manually entered, so tap count could mirror sample rate
Downside would be losing the easy selection of binary powers, which are a little more efficient for the convolution engine to process.
 
Top Bottom