REW Beta Release REW API beta releases

[Breeman]

Hi John,

I often find that I run out of characters when using more descriptive wording to name measurements in the window that pops up when right clicking on the left graph pane within the All SPL tab. Not sure what the current character limit is but can you please consider increasing it? Thanks.

 

[John Mulcahy]

I often find that I run out of characters when using more descriptive wording to name measurements in the window that pops up when right clicking on the left graph pane within the All SPL tab. Not sure what the current character limit is but can you please consider increasing it? Thanks.

The problem with that is it would mean the measurement selectors and all the graph legend entries would have to get wider to accommodate the longer names.

 

[Breeman]

The problem with that is it would mean the measurement selectors and all the graph legend entries would have to get wider to accommodate the longer names.

They do not have to be. The text field could be fixed however with the actual text scrollable to the right.

 

[darklight]

There have not been any changes to the alignment tool dialog. That dialog is owned by the All SPL graph and remains on top with viewing the All SPL graph (at least on Windows and macOS). It is hidden if you view another graph type.

Its weird, I installed the beta again and now its working as normal. I dont know what was up the first time i install that beta

 

[keantoken]

I would like to be able to "summon" the right and left window flags to my cursor by clicking just above the impulse response graph. In a very old version of REW I remember I could do this quite clearly, I used it all the time. But I think you told me it must have been a bug. Still, it was extremely convenient until it was fixed!

 

[keantoken]

When MTW is enabled, the measurement phase curve does not change.

When switching between two stepped THD measurements, the X axis does not scale correctly and doesn't respond to adjustment. If I click the scroll bar, then it will let me adjust the axis again. But whenever I switch to another stepped THD measurement it happens again and the X axis settings are lost. Measurements are attached in case it matters.

 

[keantoken]

During stepped level distortion measurement, REW was measuring the THD of my air conditioner at 27Hz rather than the fundamental at 1060Hz. This also causes the THD graph to look funny at the left hand side. I know this isn't an ideal measuring situation, but if we can be more tolerant of background noise, then we should be. I could enable the HP filter though.

 

[John Mulcahy]

During stepped level distortion measurement, REW was measuring the THD of my air conditioner at 27Hz rather than the fundamental at 1060Hz.

You can try setting the RTA option to get the fundamental from the generator, I'll put a proper fix in the next build.

 

[John Mulcahy]

When MTW is enabled, the measurement phase curve does not change.

I have fixed that for the next build.

 

[John Mulcahy]

When switching between two stepped THD measurements, the X axis does not scale correctly

I have fixed that for the next build.

 

[dcibel]

Hi John,
As a small feature request, could you make the SPL level calibration window open off to the side of the SPL meter display, instead of directly over-top. I have to move the window off to the side every time I open it.

 

[John Mulcahy]

could you make the SPL level calibration window open off to the side of the SPL meter display, instead of directly over-top.

That was done to try and discourage people from simply typing in the figure that was showing on the REW meter instead of the reference figure.

 

[achugh]

Hi John,

In the EQ panel, is it possible to keep the reference measurement the same on which the EQ filters are being tested? What I mean is, if I take a sweep and then use this measurement to open the EQ panel, now when I click the "Match response to target" button for REW to do its magic followed by "Measure with these filters"; a new measurement is taken which is expected but that new measurement is now the active measurement so in the EQ panel everything is reset or changed i.e. the filters that were used are no longer available since the reference measurement for the EQ panel is changed.

This behavior has changed because in the 5.30 one can manually adjust the filters and take another measurement as many times as needed where as in the current beta one has to keep on closing the EQ panel and open the EQ panel again using the right click option of the sweep that is being tweaked to get the best filters.

Is it possible to stay with the measurement in 5.40 in EQ panel like it was possible in 5.30?

 

[John Mulcahy]

Using Measure with these filters in 5.31.3 produces a new measurement which becomes the current selection. The EQ window shows the current selection. You can change the current selection using Alt+1..9 to select any of the first 9 measurements or Alt+up arrow/down arrow to select the previous or next measurement. Those shortcuts can be used from the EQ window.

 

[serko70]

Opening both windows on the same screen works great with such procedures when you need to move current selection regularly. I do multiple "offset t=0" adjustments much faster with Overlays/Impulse open on one side using "Apply & Next" while reading Info from different measurements in the main window with alt + arrow keys. For those measure with EQ like problems, maybe a "make this the permanent selection" item can be added to the right click menu.

 

[serko70]

API FR is only returned when Alignment tool is open:

And same fetch with Swagger after the tool is closed:

One would normally expect to get an aligned response in the API after changing delay b value in the tool.

 

[achugh]

Using Measure with these filters in 5.31.3 produces a new measurement which becomes the current selection. The EQ window shows the current selection. You can change the current selection using Alt+1..9 to select any of the first 9 measurements or Alt+up arrow/down arrow to select the previous or next measurement. Those shortcuts can be used from the EQ window.

Thank you John for sharing the shortcuts. The problem is not the shortcuts but the fact that one has to constantly change the current selection which was not the case. Maybe when "Measure with these filters" option is used can we NOT change the current selection. This will be very helpful where someone is manually adjusting the filters after REW has come with its suggestions.

Opening both windows on the same screen works great with such procedures when you need to move current selection regularly. I do multiple "offset t=0" adjustments much faster with Overlays/Impulse open on one side using "Apply & Next" while reading Info from different measurements in the main window with alt + arrow keys. For those measure with EQ like problems, maybe a "make this the permanent selection" item can be added to the right click menu.

Thank you for sharing this suggestion. I will give it a try. My laptop screen is on the smaller side (13 inches) so opening multiple screens side by side becomes a challenge. I need to look for a different laptop with 15 or 16 inch screen so that I can open side by side windows and can actually work/use them properly (I now wear eye glasses as age has caught up to me so I don't have the great vision I used to have in the past).

 

[John Mulcahy]

which was not the case

Not the case when? It is the 5.31 behaviour.

 

[John Mulcahy]

API FR is only returned when Alignment tool is open

You have to be careful mixing API and GUI activity. Closing the alignment tool GUI removes the alignment sum response.

 

[John Mulcahy]

Builds updated today (beta 59) with these changes:

• Added: Multi-input capture has options to align the individual measurements to the first input's measurement using cross correlation alignment and to vector average the individual measurements, both are only available when saving inputs to individual measurements
• Added: STI calculation shows STIPA(IR) and %ALcons figure
• Added: STI tool has a check box for whether to include masking in post-processing
• Added: The alignment tool has an option to remove any common time delay between the measurements being aligned so that the first IR starts at t=0, this may reduce phase wraps
• Added: IR text export header includes the data offset figure, that offset will be used if the file is later imported
• Added: Left and right window positions can be set by clicking in the area above the impulse graph
• Added: API Measure endpoints save-inputs-individually, no-overall-average, use-group-averages, average-umikx-modules, preserve-relative-timing, align-individual-measurements and vector-average-individual-measurements for multi-input capture options
• Changed: STI calculation shows whether post-processing or masking were applied
• Changed: STI results are shown to 2 decimal places
• Changed: StormAudio equaliser limits and filter types updated
• Fixed: On macOS and Linux spinners might not commit their edits on loss of focus
• Fixed: NPE would occur if the EQ window was opened with no measurement loaded and the filter adjust graph Y axis was dBr
• Fixed: The t=0 offset phase preview smoothing behaviour did not match the measurement phase smoothing
• Fixed: Calculate target level would not adjust the target by more than 100 dB from the current value
• Fixed: Stepped sine THD measurements should base the fundamental frequency on the test frequency
• Fixed: Applying MTW did not update phase
• Fixed: Stepped level measurement X axis limits were set incorrectly on the SPL & Phase graph

 

[serko70]

You have to be careful mixing API and GUI activity. Closing the alignment tool GUI removes the alignment sum response.

Dear John,


Thank you very much, the new additions sound great and looking forward to test them.
The Alignment Tool API will also not return aligned sum response by just changing delayB value. It needs to "Align IRs" after a delayB value is posted. I no longer need it, solved it another way but would normally assume it would produce a result based on the posted delay as the GUI tool does. In GUI, you don't need to align first to generate an aligned sum with the settings in place.
Lastly, I have a small suggestion. Just like the recent SPL Offset addition which can work with multiple measurements simultaneously, could we add a similar "response copy" option?

 

[John Mulcahy]

The Alignment Tool API will also not return aligned sum response by just changing delayB value.

It does for me.

 

[serko70]

It does for me.

Ok I will check again but I was getting null returns unless I post "Align IRs". Probably my code

 

[achugh]

I am using UMIK-1 set to 0.1db Volume gain in Windows Sounds settings as I cannot really set it to 0.0db. When I take REW measurements using 5.40 beta 59, I am getting "The measurement distortion is very high, 10.0%." error. When i click on "Check Levels" in the measurement window I get 82 db as the level. This is for subwoofer measurement. When I hear the sweep play from 0-24Khz the volume is low. Does anyone know what mistake I am making? I tried running REW with 1GB as well as 2GB of RAM as the setting during the installation when it asks how much memory to allow for use on Windows 10 Pro 22H2 build and that has made no change.

EDIT: Checking and unchecking "Apply cal files to distortion" under the Preferences > Analysis TAB is making a big difference. The screenshots below are when this checkbox is CHECKED. I have unchecked this checkbox and the error shown is reduced although not fully gone.

 

[serko70]

I am using UMIK-1 set to 0.1db Volume gain in Windows Sounds settings as I cannot really set it to 0.0db. When I take REW measurements using 5.40 beta 59, I am getting "The measurement distortion is very high, 10.0%." error. When i click on "Check Levels" in the measurement window I get 82 db as the level. This is for subwoofer measurement. When I hear the sweep play from 0-24Khz the volume is low. Does anyone know what mistake I am making? I tried running REW with 1GB as well as 2GB of RAM as the setting during the installation when it asks how much memory to allow for use on Windows 10 Pro 22H2 build and that has made no change.

EDIT: Checking and unchecking "Apply cal files to distortion" under the Preferences > Analysis TAB is making a big difference. The screenshots below are when this checkbox is CHECKED. I have unchecked this checkbox and the error shown is reduced although not fully gone.

View attachment 75637
View attachment 75633
View attachment 75634
View attachment 75635
View attachment 75636

You are measuring a subwoofer full range and REW detects very low output beyond 250Hz relative to noise floor and pops a warning. You can ignore it for subs or measure them with a limited frequency range.