REW Beta Release REW V5.20 beta 11

John Mulcahy

REW Author
Thread Starter
Joined
Apr 3, 2017
Messages
7,212
A few more changes than anticipated in 5.20 beta 11, along with another few bug fixes.

Java drivers now offer 192 kHz sample rate for all platforms, following some changes to improve behaviour at 192k. Make sure the interface is set to 192k in the OS configuration also.

There have been a few changes for the signal generator. The previous signal selection is now restored after checking levels, doing a soundcard calibration or making a stepped sine measurement. Triple tone signals have been added, and some TDFD (Total Difference Tone Distortion) presets have been added to the dual tone signals. Both dual and triple tone frequencies now have an upper limit of half the sample rate. The tone burst generator now has options to specify the period of the bursts in cycles, milliseconds or samples (binary powers to match FFT lengths). The minimum bandwidth for custom filtered random noise has been reduced from 1 octave to 1/3 octave or 10 Hz, whichever is greater. The minimum bandwidth for custom filtered periodic noise has been reduced to 10% or 10 Hz, whichever is greater. The dither bit depth choices have been expanded to cover all values from 16 to 25 bits, allowing the dither level top be scaled accordingly - REW's dither is triangular with an amplitude of 2 lsb peak to peak at the chosen bit depth.

On the Filter Adjust graph the Predicted response now uses the windowed rather than the full IR and reflects the equaliser selection's fixed sample rate rather than adapting to the measurement rate.

A few other minor changes:
  • The measurement name field width has been increased on wider screens
  • There is now a view preference to control whether phase wraps are shown with dashed lines or not at all
  • The dialog for exporting distortion data as text for stepped sine data has been revised to better match the source data
  • A message is shown if REW has changed the RTA window to Rectangular to allow multitone TD+N calculation, which happens if showing distortion on the RTA when using a multitone signal.
  • There is now an ASIO sample format display in the status bar. Note that the data bit depth is often lower than the sample format allows, REW also shows the detected bit depth for input data based on the data values received
  • The peak input SPL before clipping is shown in the main window status bar
  • Two more date format options have been added for measurement naming
  • The input device appears along with the input name in the title of the RTA, SPL meter, Scope and SPL Logger windows
  • The output device and output appear in the title of the signal generator window
  • In-band (22.4 Hz to 22.4 kHz) and out of band (> 22.4 kHz) levels have been added to the RTA levels panel
  • The zoom box is now included in graph image captures and the RTA levels panel appears in RTA image captures
  • An equaliser entry has been added for the miniDSP C-DSP 8x12 DL, providing 10 filters at 48 kHz

The bug fixes are:
  • When using Java drivers switching to an output device with fewer channels than the current output device would cause an exception
  • The Filtered IR graph could get stuck in a loop when changing the time reversed or zero phase setting
  • File playback did not accept sweep files that had the timing reference in one channel and the measurement sweep in the other
  • The measurement sweep end frequency was limited to 96 kHz even if the sample rate was above 192 kHz
  • The signal generator output level spinner was being updated twice if the "Full scale sine rms is 0 dBFS" setting changed
  • The signal generator output level peak value was incorrect if the unit was not dBFS and "Full scale sine rms is 0 dBFS" was selected
 

Breeman

Member
Joined
Jan 9, 2019
Messages
151
HI John,

I was making SPL measurements with this version and on two occasions I encountered an issue where within the measurement screen, the program would stall and freeze at 99% of calculating the impulse response. This issue occurred in only a specific microphone measurement position. I have attached a screenshot for your reference. The only way to get the program to work again would be to close and restart it however the same problem would occur repeatedly until I move the microphone slightly. It seems the program is unable to calculate the impulse response at some arbitrary microphone positions. I occurred twice while measuring the 2-way loudspeaker and a subwoofer separately and at two different measurement locations within the same space utilizing 16 microphone measurement positions I used so I could average them later.
Snap2.png
 

John Mulcahy

REW Author
Thread Starter
Joined
Apr 3, 2017
Messages
7,212
I'm guessing if you look in the log files you will see something like this:

Exception in thread "Thread-22" java.lang.NullPointerException
at roomeqwizard.IRData.offsetIRStart(Unknown Source)
at roomeqwizard.IRData.offsetIRStart(Unknown Source)
at roomeqwizard.IRData.offsetIRStart(Unknown Source)
at roomeqwizard.IRData.adjustStartToInterpolatedPeak(Unknown Source)
at roomeqwizard.SweepAnalyser.A(Unknown Source)
at roomeqwizard.wF.A(Unknown Source)
at roomeqwizard.wF.A(Unknown Source)
at roomeqwizard.wF.run(Unknown Source)

That has been fixed for the next beta, should release it in the next day or so.
 
Last edited:
Top Bottom