V5.20.14 early access build

[sm52]

I figured out when this situation occurs. If acoustic synchronization is set in the measurement window when measuring SPL, the repetition counter is reset to 1 and cannot be changed. If you then switch to impedance measurement, you can set the repeat count to any number, but it is reset after the impedance measurement or calibration step.
Regarding the error message, I cannot yet repeat it so that the message appears again. Everything is fine for now.

 

[John Mulcahy]

If acoustic synchronization is set in the measurement window when measuring SPL, the repetition counter is reset to 1

Yes, I have fixed that for the next build.

Regarding the error message, I cannot yet repeat it so that the message appears again.

I couldn't repeat that either, but I have added some code to guard against it.

 

[sm52]

I couldn't repeat that either, but I have added some code to guard against it.

I suspect a situation where REW loses connection with an external sound card connected via USB. Because some Windows process does not allow to process requests from usb at the required speed.

 

[JoinUs]

@John Mulcahy
Hello John, Is this possible function?
A*(-B)

Let's say A is 2, B is 3. (Just for example)
2 * (-3) = -6
2 * (1/3 which means 1/A) = 0.666666.....

Sometimes negative Multiplication results are needed.

In Equalizer APO, can do that.
And I can loop back the results and use the responses, but the process is a bit cumbersome.
So I wonder if this is possible within REW.

 

[jeppekristoffer]

1. Many times when i open a measurement with eq-filters in it(all pass filters) and when looking in the overlay tab "Predicted phase" and "unwrap" engaged i have to disengage/engage the filters for the phase change to take place.
2. And if the phase trace is wrapped and i press unwrap then not all traces selected unwrap. To be able to make them unwrap i have to switch to another of the overlays tabs and back to "predicted phase" tab and then all traces selected are unwrapped. The same applies when goind from unwrapped the wrapped.

Regards Jesper

@John Mulcahy please see if this can be replicated. Regards Jesper

 

[John Mulcahy]

Is this possible function?
A*(-B)

No, you would need to make a copy of B and invert its polarity.

 

[John Mulcahy]

Builds updated today (ea75) with these changes:

• Changed: Allow frequencies down to 1 Hz for custom dual and triple tone signals
• Changed: Show box volume with 4 decimal places on the TS params panel
• Changed: Show driver diameter with 2 decimal places on the TS params area calculation popup
• Fixed: Unwrap phase may not update some graphs
• Fixed: Scale IR data did not update the windowed IR if a frequency-dependent window was applied
• Fixed: Coherent averaging did not work well for tones that were not locked to the FFT length
• Fixed: Number of sweep repetitions for impedance measurements would be reset to 1 if the SPL measurement timing reference choice did not allow multiple sweeps

 

[sm52]

John Mulcahy​

In the latest version, phase graph smoothing in the SPL window and smoothing in the GD window do not work correctly. If you disable smoothing and then enable 1/48, everything is correct. But if you turn on anything other than 1/48 and ctrl 0, and then turn on 1/48, the phase is not displayed correctly. And GD.

 

[John Mulcahy]

if you turn on anything other than 1/48 and ctrl 0, and then turn on 1/48, the phase is not displayed correctly

Not correct how? What are you seeing and what are you expecting to see? Do you have a file that shows the problem?

 

[JoinUs]

Can add "Set Graph Axis Limits" Preset? 1, 2, 3 things.
Even with the same impulse, there are certain numbers that are used in various ways, such as a narrow range of 1-50 ms and a wide range of ~500 ms % units.
However, it is cumbersome to enter and change it again every time.
Only one limit setting is saved, so I want to be able to add it in various ways.

 

[sm52]

I took any graph, pressed ctrl 0, then ctrl shift 9, ctrl shift 7, ctrl shift 9. This is shown in pictures 1,2,3,4. Two right - smoothing 1/48.

 

[John Mulcahy]

I took any graph, pressed ctrl 0, then ctrl shift 9, ctrl shift 7, ctrl shift 9. This is shown in pictures 1,2,3,4. Two right - smoothing 1/48.

Thanks, I have fixed that for the next build.

 

[John Mulcahy]

Can add "Set Graph Axis Limits" Preset?

I'll add it to the list for future consideration.

 

[Draki]

I was watching some YT videos from the contributor OCA , and he is routinely using a REW second screen by (seemingly) clicking the "restore down" double square icon at top right.
How is it done? Obviously I can't replicate that, hence my question - you can tell I am not very savvy at this...

 

[John Mulcahy]

The second is probably the Overlays window.

 

[John Mulcahy]

The V5.20.14 early access build has now been replaced by the V5.30 beta release. That has the following changes since the last 5.20.14 build:

• Changed: RTA Coherent averaging is possible when playing a tone burst or CEA-2010 burst from the signal generator with an interval which matches the FFT length when FFT overlap is set to 0%
• Fixed: Changing smoothing from a fraction more smoothed than 1/48 to 1/48 did not apply the improved phase smoothing

Edit: The V5.30 beta release update notifications will be activated from 4th November.

 

[JoinUs]

The V5.20.14 early access build has now been replaced by the V5.30 beta release. That has the following changes since the last 5.20.14 build:

• Changed: RTA Coherent averaging is possible when playing a tone burst or CEA-2010 burst from the signal generator with an interval which matches the FFT length when FFT overlap is set to 0%
• Fixed: Changing smoothing from a fraction more smoothed than 1/48 to 1/48 did not apply the improved phase smoothing

Edit: The V5.30 beta release update notifications will be activated from 4th November.

Well done! Finally!

 

[serko70]

Dear John,

I bumped into a paper that describes a method to determine the point in time when the early reflections fully transition to late reverberation (mixing time) by analysing normalized kurtosis values. Calculation is free of room dimensions but takes room properties into account.

It seems relatively easy to implement and maybe(?) also used for more realistic clarity measurements instead of the defined 50/80ms.transition points in REW.

A novel and SOTA technique also uses it for truncation before inversion:

What do you think?

 

[John Mulcahy]

Like the Clarity measure, the paper is relevant for spaces that are very, very large by domestic standards (the "smaller" space considered in the paper was a 350-seat venue). Potentially more relevant to domestic spaces is the work of Alejandro Bidondo in characterising room acoustic "texture", examining the behaviour of early reflections and the transition to a diffuse field. See Texture of a room impulse response and Room acoustic texture: a methodology for its quantification

 

[serko70]

Like the Clarity measure, the paper is relevant for spaces that are very, very large by domestic standards (the "smaller" space considered in the paper was a 350-seat venue). Potentially more relevant to domestic spaces is the work of Alejandro Bidondo in characterising room acoustic "texture", examining the behaviour of early reflections and the transition to a diffuse field. See Texture of a room impulse response and Room acoustic texture: a methodology for its quantification

This is truly innovative work!
In my limited understanding, REW's ETC smoothing is already doing something quite similar to the decay cancelling method in the paper to generate echo density function. Are we going to be able to measure room impulse response texture some day?

 

[Lowclock]

Has something changed about the way "Normalize step to IR peak" works? I could've sworn it never used to behave like this:

 

[serko70]

"Response copy" of the attached measurement (a .wav import from rePhase) generates pre-post ringing in the impulse response in the V5.30 Beta:

 

[John Mulcahy]

Has something changed about the way "Normalize step to IR peak" works?

No. That response looks to have a DC offset.

 

[John Mulcahy]

"Response copy" of the attached measurement (a .wav import from rePhase) generates pre-post ringing in the impulse response in the V5.30 Beta

Please post about the 5.30 beta in the new thread. That is just the interpolator trying to make sense of the sample values (i.e. reconstruct the inter-sample behaviour in accordance with the sampling theorem). Looks a somewhat odd response to start with, time-reversed perhaps.