REW Beta Release REW API beta releases

Builds updated today (beta 79) with these changes:
  • Added: Low input level warning messages can be disabled by not selecting "Warn if input levels are low" on the Measure dialog
  • Added: API measurement Protection options has a warnForLowLevels boolean
  • Fixed: Null pointer exception could occur with room sim measurements
  • Fixed: Graph capture title text was offset left
  • Fixed: Virtual balanced controls were enabled for mono inputs
 
Builds updated today (beta 80) with these changes:
  • Added: API Target level and match target frequency range have range checks
  • Fixed: When stepped IMD measurements are plotted against the input level (rather than the generator level) the IMD reference tone level was used for the X axis instead of the total input rms level
  • Fixed: API warnings for target level included invalid text about hitting OK to continue
 
Another persistent quirk I've noticed in the group delay plot view:

1747273162725.png


The trace says 1/12 smoothing (changed setup options from no smoothing default) and yet that is not reflected as such in reality.
 
Another persistent quirk I've noticed in the group delay plot view:
There's an issue in the current build where group delay traces for smoothed measurements are not being updated on IR window or t=0 changes, I have already fixed that for the next build. That might relate to what you are seeing, though I can't reproduce what you show above.
 
Hello John,

Could you check input and output assignment related to scope usage like shown here:


Scope is not working when I assign CH1 for out and CH8 for ref with my 8/8 USB audio interface.

Thanks for this great SW.
 
Scope is not working when I assign CH1 for out and CH8 for ref with my 8/8 USB audio interface.
Scope CH1 is whatever has been selected as the measurement input channel, scope CH2 is whatever has been selected as the ref input channel.
 
Scope CH1 is whatever has been selected as the measurement input channel, scope CH2 is whatever has been selected as the ref input channel.
Sure. It works with my dual-in / dual-out audio interface but does not with my 8 in / 8 out interface when its CH8 input is assigned as scope's CH2 input source.
 
Sure. It works with my dual-in / dual-out audio interface but does not with my 8 in / 8 out interface when its CH8 input is assigned as scope's CH2 input source.
If it's a Motu Ultralite mk5, try ASIO drivers. Mine has issues with Wasabi drivers.
 
It works with my dual-in / dual-out audio interface but does not with my 8 in / 8 out interface when its CH8 input is assigned as scope's CH2 input source.
Seems to work as expected on a 4-input interface I just tried.
 
If it's a Motu Ultralite mk5, try ASIO drivers. Mine has issues with Wasabi drivers.
No, its an ADA8200 connected over ADAT to an MCHstreamer. REW running on MacOS.
Edit: this interface is working nicely with 8 mics at once (REW Pro)
 
Last edited:
Builds updated today (beta 81) with these changes:
  • Added: Scope has a Save button that saves the visible portion of the current traces as measurements
  • Added: High distortion and low SNR warning messages can be disabled by not selecting them on the Measure dialog
  • Added: API measurement Protection options has warnForHighDistortion and warnForLowSNR booleans
  • Fixed: Group delay values for smoothed traces were not recalculated after IR window changes or shifting t=0
  • Fixed: StormAudio equaliser shelf shape was wrong
  • Fixed: Unsubscribing from an API endpoint prevented further subscriptions to it from working
 
Beta 82 uploaded with this fix:
  • Fixed: Missing bundle entries when switching between light and dark modes
 
One issue I have been fighting with REW with is EQ. Using API, REW refuses to EQ some areas, however if I use the "Match response to target" from the EQ tab, it correctly EQs the response. Is this a known issue or are there any workarounds?

This is an example but it happens in other cases as well - taking my averaged FR response + FDW 15 cycles and then generate min phase to produce the response I want to EQ. Manually EQ-ing brings down the hump at 235hz, but if I use API with the same settings, it leaves that hump alone.
1747751360703.png


This is the same response being EQ-ed by API. "FRo" is the response I am feeding to the EQ and "FRfinal" is what I am getting. I am only EQ-ing to 300hz and the responses are identical -
1747751765198.png


Same issue with my TFL response -
1747751825340.png


Messing around with the frequency range and/or boost values make the EQ algorithm even more unpredictable. Any help/advice appreciated.
 
Using API, REW refuses to EQ some areas, however if I use the "Match response to target" from the EQ tab, it correctly EQs the response.
Most likely is that different target level or match parameters are being used. Target level for a measurement is at /measurements/:id/target-level. You can read the match target settings the API is using from /eq/match-target-settings and change any or all of them by posting the parameter(s) you want to change back to that endpoint.
 
1747770770328.png


Hello John, I would like to report a bug.

As you can see, the generator has been set to output a 6.5 cycle sine wave tone burst. On the right is the oscilloscope showing the loopback (yellow) and the acoustic measurement (blue). The output is much longer than 6.5 cycles.

Thank you for adding the SAVE button in the scope!
 
Perhaps use the burst period setting to define the 6.5 cycles your want. I've used the burst mode before and it works fine when appropriately set up afaik.
 
Perhaps use the burst period setting to define the 6.5 cycles your want. I've used the burst mode before and it works fine when appropriately set up afaik.

You can see from my screen shot that the burst period has been set 128k samples. This exact setting was working earlier today on the previous beta version of REW.
 
Hi John,

I would appreciate it if you look into enabling one to zoom into the impulse responses when aligning the main loudspeker to the subwoofer in the alignment tool. This way, one can have better, real-time visibility into the alignment of the impulse peaks while doing fine delay adjustments, without resorting to Overlays.
Snap95.png
 
I would appreciate it if you look into enabling one to zoom into the impulse responses when aligning the main loudspeker to the subwoofer in the alignment tool.
That display is intended for use with the filtered IR used by the alignment tool, the range is automatically scaled to suit the filter frequency (i.e. the frequency at which they should be aligned). It wouldn't make sense to align the peaks of impulse responses that have different bandwidths as that would misalign them in time.
 
That display is intended for use with the filtered IR used by the alignment tool, the range is automatically scaled to suit the filter frequency (i.e. the frequency at which they should be aligned). It wouldn't make sense to align the peaks of impulse responses that have different bandwidths as that would misalign them in time.
Understood. The filtered IR feature in the alignment tool is a great feature by the way:T
 
Please return the possibility to obtain the basic T/S parameter, like 5.31.3 version
As in beta the measurements are pre-selected in the list and there is no free field, it does not allow the function
:T

5.31.png


beta 83.png
 
Please return the possibility to obtain the basic T/S parameter, like 5.31.3 version
As in beta the measurements are pre-selected in the list and there is no free field, it does not allow the function
I have fixed that for the next build, but it would also work if you had more than one measurement loaded since selecting a free air measurement clears the secondary measurement.
 
Hello,

scope is still not working here with my 8ch in / 8ch out audio interface. No trigger but instead this error message. (With my other 2 in / 2 out interface it's working correctly) Also measurements work nicely with cross correlation etc. with the 8 ch in interface…
 

Attachments

  • Bildschirmfoto 2025-05-24 um 11.31.23.png
    Bildschirmfoto 2025-05-24 um 11.31.23.png
    667.1 KB · Views: 9
Back
Top