REW Beta Release REW API beta releases

[John Mulcahy]

For FSAF, I've noticed when using filters such as "measure with these filter" via the EQ window, that saving the excitation WAV file doesn't include the filtered result

Yes, that's because it is the unfiltered result which is used by the FSAF processing to extract the impulse response. I can add an "excfilt-" file in the next build that has the actual output data.

As a feature request, when loading audio data into REW, such as FSAF residual, LTI, MIC, EXC files, it would be great if there were a playback button in REW somewhere.

The residual can be played back using the system default WAV player via the button in the Distortion graph controls. If the option to save the other signals has been selected then the corresponding WAV files are available for playback. REW isn't a file player and I think it would be wrong to try and move it in that direction, there are many file playback applications.

 

[dcibel]

Yes, that's because it is the unfiltered result which is used by the FSAF processing to extract the impulse response. I can add an "excfilt-" file in the next build that has the actual output data.

It's no problem, I can get what I need from the LTI file, I was just curious.

 

[sm52]

If you open the Alignment tool on the All SPL tab, and click Align phase at cursor, the delay value is calculated correctly, and the slider line is shown incorrectly.

 

[John Mulcahy]

If you open the Alignment tool on the All SPL tab, and click Align phase at cursor, the delay value is calculated correctly, and the slider line is shown incorrectly.

I have fixed that for the next build.

 

[user44455555]

Is there in the API beta release a script possible or avaiable which set output to left speaker, do a measure, set speaker to right do a measure ?. so it can work automatic and get 2 measures 1 left, 1 right
also usefull a script that do ever 7-10 sec a measure, so can put speaker in hand and walk in the room with the speaker do measures on diffrent places and basstrap foam placement

 

[sm52]

John, Is it possible to add to Alignment Tool the ability to calculate the average delay in the specified (allocated) by the user of the frequency range? The same thing as Align Phase At Cursor, but for the frequency range. At my screenshot, I would like to allocate 400-550 Herz, and get an average delay. If you get delays at each frequency in this range, they will be very different. This is similar to a random result. You can limit a possible range, such as 300 Herz, if such work greatly loads the processor.

 

[achugh]

@John Mulcahy Feature request for taking measurement sweeps.
I have recently purchased Pioneer Elite VSX-LX805 AVR which is reviewed https://www.avnirvana.com/threads/elites-dynamic-vsx-lx805-av-receiver-–-a-full-review.14618/

In this AVR, when I have to take REW sweeps, the AVR has a mechanical relay which switches depending on the input signal to choose the right codec/decoder. By default when no audio signal is present, the AVR defaults to the default listening mode for stereo i.e. Ext. Stereo. However, when multi-channel PCM signal is sent; which is how I think REW sends the measurements sweeps when using JAVA drivers in exclusive AVR mode; the AVR now switches to Dolby Surround listening mode. This switch takes about a seconds to happen. On top of this switch if the correct room correction memory profile needs to be chosen that adds further delays which cannot be selected unless REW is sending the signal causing further headaches to choose the correct setup for measurements.

So my request is, is it possible for REW to send the PCM signal as a starting delay before the acoustic reference is sent followed by another delay before the requested channel sweep is sent?

Here is what's going on at the moment with this newer receiver.

When I click on Measurement button, the measurement window opens. At this time, REW is not sending any signal to AVR so AVR is defaulting to Ext. Stereo.
When I click measure with acoustic reference then REW sends a PCM signal to the chosen channel for acoustic reference which is CENTER channel in my case. When this PCM signal is sent, the AVR relay clicks and switches to Dolby Surround decoder. There is a small delay for this relay click to happen. Then REW takes I believe 0.5 to 1 second pause where no signal is sent to AVR so AVR relay clicks again to switch back to Ext. Stereo as the PCM audio stream is stopped. Now REW sends the measurement sweep for the requested channel say LEFT. So, the AVR goes through another relay click to switch back to the Dolby Surround decoder. All these relay clicks mess up the measurement and REW will then present an error dialogue that the acoustic reference delay is too high with a value going above 1 second in this error dialogue. This does not instill confidence in me that the measurement was done properly.

So is it possible to either sent some dither audio like PINK NOISE as multi-channel PCM signal with some delay before the actual signal is sent to the AVRs channel just so the AVR has time to mechanically click its relay and switch decoders properly?

FYI, this should not be a problem to this AVRs sister model which is Integra DRX 8.4 because it has a feature which allows an individual to chose your preferred decoder for any audio signal. This feature is NOT available for my AVR which is the consumer version of AVR. Integra brand is for custom installer and audio integrators and has features I did not need hence I chose Pioneer brand. But now I am having a hard time taking measurements normally. My previous AVR; Pioneer Elite SC-37; did not have this problem so I did not think that this new version will have this issue.

I hope you can help solve this issue for me and others using this product or similar products where mechanical relays are causing issues with switching between PCM and Stereo audio modes.

Thanks in advance for your help.

 

[serko70]

Dear John,


"Result spans combined frequency range" tickbox in A * B trace arithmetic operation seems to have lost its functionality recently. Could you please check?

Correct operation shows up in the measurements column:

 

[John Mulcahy]

Is there in the API beta release a script possible or avaiable which set output to left speaker, do a measure, set speaker to right do a measure ?. so it can work automatic and get 2 measures 1 left, 1 right
also usefull a script that do ever 7-10 sec a measure, so can put speaker in hand and walk in the room with the speaker do measures on diffrent places and basstrap foam placement

You don't need the API or a script for that, use the Measure dialog features.

 

[John Mulcahy]

Is it possible to add to Alignment Tool the ability to calculate the average delay in the specified (allocated) by the user of the frequency range?

The alignment tool uses the phase within a 1 octave span centred on the alignment frequency. If your measurements give inconsistent results for phase alignment the impulse alignment mode might be more consistent.

 

[John Mulcahy]

So is it possible to either sent some dither audio like PINK NOISE as multi-channel PCM signal with some delay before the actual signal is sent to the AVRs channel just so the AVR has time to mechanically click its relay and switch decoders properly?

The output option to fill silence with dither may help.

 

[John Mulcahy]

"Result spans combined frequency range" tickbox in A * B trace arithmetic operation seems to have lost its functionality recently. Could you please check?

Please attach an mdat file with representative data and explain what you are expecting to see in the result.

 

[sm52]

If your measurements give inconsistent results for phase alignment the impulse alignment mode might be more consistent.

Impulse alignment does not give the correct alignment (it gives close to the correct alignment), phase alignment does it better.

 

[serko70]

Please attach an mdat file with representative data and explain what you are expecting to see in the result.

#4 in the attached mdat is a A*B result I have done in the past (June 2025) using "Result spans combined freq range"
#5 is what I get when I do the same operation today. I cannot convolve a windowed response with a full range measurement lately.

 

[John Mulcahy]

#4 in the attached mdat is a A*B result I have done in the past (June 2025) using "Result spans combined freq range"

If I calculate A*B with 5.40 beta 89 (the version used for your #4 result) I get the same as the current build. That is driven by epl-trimmed, which has no content below 657 Hz as the IR window is so short. Could you have been using longer windows for the epl result in the past?

 

[serko70]

If I calculate A*B with 5.40 beta 89 (the version used for your #4 result) I get the same as the current build. That is driven by epl-trimmed, which has no content below 657 Hz as the IR window is so short. Could you have been using longer windows for the epl result in the past?

Maybe John, I am sorry I don't remember. I am sure I windowed L first and then generated EP and its 1/A and kept L * 1/A result full range but I can't replicate it now.

 

[user44455555]

When in All SPL call menu "clear trace selections" on left side(the measure list) the highlight of the measures are not cancel. also when checkmark a measure the left side stay not in sync. is it clear what i mean or should i do a video ?

 

[John Mulcahy]

When in All SPL call menu "clear trace selections" on left side(the measure list) the highlight of the measures are not cancel. also when checkmark a measure the left side stay not in sync.

The trace check boxes are independent of the measurement selections.

 

[user44455555]

The trace check boxes are independent of the measurement selections.

ok, i thought was forget to sync.I see the popup menu have in measure selections "clear selections", and "toggle selections" entry. so i use this now

 

[John Mulcahy]

Builds updated today (beta 101) with these changes:

• Changed: When the FSAF excitation for a file measurement is saved there is also an "excfilt-" file which includes the effect of any filters applied
• Fixed: Magnitude division results had a +3dB offset when using the dBFS graph axis
• Fixed: The Alignment tool fine delay slider behaviour for small negative delays was incorrect
• Fixed: Make sure the Filter tasks pane is configured correctly for the chosen equaliser when the EQ window is opened with no measurement loaded
• Fixed: Dutch & Dutch 8c discovery could fail on IPV4 networks

 

[serko70]

REW V5.40 Beta 101 running Azul Systems, Inc. JRE 1.8.0_462 64-bit on Windows 10 Language en, country GB, keyboard GB, windows-1252 Screen 3840 x 1600 at 96 DPI Running in C:\Program Files\REW


Message:
java.lang.IllegalArgumentException: Row index out of range
Stack Trace:

Spoiler

Row index out of range
javax.swing.JTable.boundRow(JTable.java:2146)
javax.swing.JTable.addRowSelectionInterval(JTable.java:2196)
roomeqwizard.fJ$1$1.run(y:1101)
java.awt.event.InvocationEvent.dispatch(InvocationEvent.java:311)
java.awt.EventQueue.dispatchEventImpl(EventQueue.java:758)
java.awt.EventQueue.access$500(EventQueue.java:97)
java.awt.EventQueue$3.run(EventQueue.java:709)
java.awt.EventQueue$3.run(EventQueue.java:703)
java.security.AccessController.doPrivileged(Native Method)
java.security.ProtectionDomain$JavaSecurityAccessImpl.doIntersectionPrivilege(ProtectionDomain.java:74)
java.awt.EventQueue.dispatchEvent(EventQueue.java:728)
java.awt.EventDispatchThread.pumpOneEventForFilters(EventDispatchThread.java:205)
java.awt.EventDispatchThread.pumpEventsForFilter(EventDispatchThread.java:116)
java.awt.EventDispatchThread.pumpEventsForHierarchy(EventDispatchThread.java:105)
java.awt.EventDispatchThread.pumpEvents(EventDispatchThread.java:101)
java.awt.EventDispatchThread.pumpEvents(EventDispatchThread.java:93)
java.awt.EventDispatchThread.run(EventDispatchThread.java:82)

 

[John Mulcahy]

Was that after dragging a measurement?

 

[serko70]

I think I was in the Overlays/Distortion window, selecting and unselecting measurements, when this happened. My 100 Beta also threw a Java exception shortly before. It might just be that my PC needs a clean install of REW, so I wouldn't worry too much about it.