REW Beta Release V5.20 beta versions 19 to 28

[John Mulcahy]

As above, monitoring needs to be off. The output should be the signal from the computer and not any kind of mix.

 

[Jiffster]

There was no mic connected and I was sending the signal from othe output back to an input as stated in the calibration instructions.

 

[John Mulcahy]

Changes in V5.20 beta 22:

• Added an option to show the audio frequency bands above the graph for the SPL & Phase and All SPL graphs. The bands are:
  • Sub bass: 20 to 60 Hz
  • Bass: 60 to 250 Hz
  • Low midrange: 250 to 500 Hz
  • Midrange: 500 Hz to 2 kHz
  • Upper midrange: 2 to 4 kHz
  • Presence: 4 to 6 kHz
  • Brilliance: 6 to 20 kHz
• Allow Soundcard calibration measurement names to be changed
• Bug fix: Measurement trace on RTA window did not update when trace offset was changed

 

[John Mulcahy]

V5.20 beta 23 changes:

• Added an option to use CRLF as end of line when exporting measurements as text on macOS or Linux so the files have better compatibility with Windows applications
• Bug fix: The SPL measurement timing offset was also being applied to impedance measurements
• Bug fix: Prevent the frequency bands stripe text labels overwriting each other on narrow graphs
• Bug fix: Do not colour the frequency bands stripe in captured images if monochrome option selected
• Bug fix: Do not colour the modal resonance lines in captured images if monochrome option selected

 

[John Mulcahy]

Changes in V5.20 beta 24:

• Added 1/A and 1/B as trace arithmetic operations - results will only be valid for measurements made over the full frequency range
• Allow soundcard cal files for the Default output device (using the default device is not recommended)
• Allow 2 decimal places for the impedance measurement sense resistor value
• Use scientific notation for distortion percentages below 0.001% in the distortion graph legend
• Bug fix: Mic cal file entries were not created for devices with mono inputs
• Bug fix: The measurement trace on the EQ window should not update when the trace offset is changed unless "Add to data" is used

 

[sm52]

I have a version of REW_windows_5_20_beta7_with_JRE, and I’m not trying other updates because I don’t want to install Java on my computer. Can you make a version of REW_windows_5_20_beta24_with_JRE?

 

[John Mulcahy]

You don't need to install Java. If the REW installer cannot fiund a suitable Java runtime it will download a private runtime that only REW uses, exactly the same as the previous "with JRE" versions. The only difference is not including the runtime in the download to save bandwidth when it is not needed.

 

[thumprchgo]

It really should not be necessary to delete preferences for a new version and I don't recommend you do that.

For a long time, I would not delete anything REW beta xxx installed. Around this spring one of the betas did something that other betas were not clearing after updating. (Often a measurement would not "hear" the chirp at the start of a sweep, it would wait endlessly for the thing it never heard, really quite frustrating.) Searched up REW uninstaller app for my MacBook Pro, and used it. And the sound whacky disappeared. I took a screen shot of my prefs for REW and reconstruct from prefs screensnap after a new update In removing all the software I can choose to name the enclosing folder in my Applications folder for each beta version. REW-b25, kind of thing. I have had MUCH better luck with betas since...

 

[hibikijin]

Just a small detail regarding the Info window.. (Beta 24)

When playing from file (in Measure Playback From File), Excitation in the Info window shows whatever Level was set before when Playback was set to From REW which may or may not be the same as in the generated sweep file.

 

[John Mulcahy]

When playing from file (in Measure Playback From File), Excitation in the Info window shows whatever Level was set before when Playback was set to From REW which may or may not be the same as in the generated sweep file.

Thanks, I've fixed that.

 

[waudio]

My first post.
With latest REW beta (and the previous as well) when I click on the "Controls" button I cannot see the whole menu on screens with lower resolution (old Mac laptop). On better monitors all items in the menu fit fine in the frame. My question: Is it possible to add scrollbar in some of the next releases? Otherwise, excellent software and saves me a lot of the guess work by ear.

 

[John Mulcahy]

There are a few things that can be done to reduce the space used on a lower resolution screen: To make more horizontal space for the graph use the expand/collapse button above the measurement panels to reduce the width they take up and turn off the graph scroll bars using the button above the graph (the graph can be moved by right click and drag). To make more vertical space you can remove the labels under the toolbar and graph buttons by unchecking boxes in the View preferences and you can remove the toolbar by unchecking the Show toolbar box in the View preferences (everything in the toolbar has a corresponding menu entry and a keyboard shortcut).

If those don't work let me know the Mac's screen resolution and which menus cause problems.

 

[Jimmy Venema]

I've used REW, but I've installed a few new releases without using them. I'm now at version version 5.20 Beta 24, i use it on MacOS 10.14.6. I've used it today to update some documentation of a DIY speaker project. And I have found some issues with it.
If I used the Generate 1/A or 1/B it works the first and perhaps second time, but soon the Generate button stays greyed out and nothing is generated.
When using the capture button after using REW for .while, only first dialog pops up. After pressing the save button with the floppy logo, no dialog pops up to select the directory and file name. Thus essentially the capture can not be saved. Only when starting up REW freshly and make directly the capture first, then the second dialog pops up and the capture can be save. For the next capture you have to quit REW, start it up again and do the next capture, which is very annoying.
I uninstalled REW and reinstalled it freshly, but have the same problem.
Further I noticed, that at some moment the save all did not function. After pressing the button in the toolbar, the dialog for selecting the file name and directory does not pop up. It looks similar to the capture problem
I know that the 1/A function is rather new, but the version before that I've used without issues. So I no not know when the issues has creeped in, but it is rather recent.
The content of the log file for the generate issue was:
Sep 25, 2019 11:02:55 PM roomeqwizard.RoomEQ_Wizard main
INFO: REW V5.20 Beta 24 running JRE 1.8.0_202 64-bit on OS X 10.14.6, en, UTF-8
java.lang.NullPointerException
at java.util.Arrays.fill(Arrays.java:3092)
at roomeqwizard._G.A(y:1301)
at roomeqwizard._G.doInBackground(y:1847)
at javax.swing.SwingWorker$1.call(SwingWorker.java:295)
at java.util.concurrent.FutureTask.run(FutureTask.java:266)
at javax.swing.SwingWorker.run(SwingWorker.java:334)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:748)

 

[John Mulcahy]

For the 1/A problem I'm guessing the trace you were inverting only had magnitude data, I've fixed that for the next build.

There must be some trigger other than time for the file dialog failing to appear, I've generated and saved measurements and captures many dozens of times and not been able to reproduce that. If you have any idea of something you were doing or that was happening before the dialog stopped appearing I'd be pleased to hear it.

 

[Jimmy Venema]

I've restarted my mac and will try again. If there is a reproduction scenaria, I will make a screen recording.

 

[John Mulcahy]

Thanks. If it happens again can also try unchecking the box to use the macOS file dialog in the View preferences, maybe the Java file dialog will still work.

 

[Jimmy Venema]

After a restart of my laptop, I did not have seen any issue with a missing file dialog. So some strange interaction of macOS with java I guess.

 

[John Mulcahy]

Changes in beta 25:

• Added shelf filters with adjustable Q for the Generic and miniDSP equalisers. Use Q = 0.7071 for the maximum monotonic shelf slope.
• Change text colour for graph image comments on the graph depending on the graph background
• Added an SPL figure to the Measure dialog level check result
• When the ASIO secondary output is enabled select an output other than the primary output
• When filter numbers are shown above the EQ window filter adjust graph only show numbers for filters that have an effect
• Additional error reporting
• Bug fix: CTA-2034 limits should be displayed for fundamentals up to 5 kHz
• Bug fix: Zoom/measurement box on graph captures was drawn in the wrong position and did not use monochrome when the option was selected
• Bug fix: Exported impulse response WAV files were shorter than they should be if the peak was less than 1 second from the IR start
• Bug fix: Trace arithmetic 1/A or 1/B would fail if the trace being operated on only had magnitude data

 

[Breeman]

Changes in beta 25:

• Added shelf filters with adjustable Q for the Generic and miniDSP equalisers. Use Q = 0.7071 for the maximum monotonic shelf slope.
• Change text colour for graph image comments on the graph depending on the graph background
• Added an SPL figure to the Measure dialog level check result
• When the ASIO secondary output is enabled select an output other than the primary output
• When filter numbers are shown above the EQ window filter adjust graph only show numbers for filters that have an effect
• Additional error reporting
• Bug fix: CTA-2034 limits should be displayed for fundamentals up to 5 kHz
• Bug fix: Zoom/measurement box on graph captures was drawn in the wrong position and did not use monochrome when the option was selected
• Bug fix: Exported impulse response WAV files were shorter than they should be if the peak was less than 1 second from the IR start
• Bug fix: Trace arithmetic 1/A or 1/B would fail if the trace being operated on only had magnitude data

Thanks for this update John.

I just have the following suggestions for your consideration, when working with the waterfall graph

1. It would be preferable to have an option of changing the background color of the graph legend to plain white (highlighted with a green arrow).
2. A way to toggle the color gradient bar on the the right of the legend on and off would be nice too, especially when capturing the graph.
3. A pure black and white color scheme (without any shading) for the actual waterfal graph.
4. A way to toggle the time axis labeling on and off because when capturing the graph with lower resolution settings, these labels get embedded on top of the waterfal graph on the left hand side (highlighted with a red arrow).
5. Means to turn off the information box highlighted with a pink arrow.

 

[John Mulcahy]

I have added options in the next build to show or hide the time axis labels, settings label and colour bar. I have also amended the limits for the cubehelix colour scheme settings to allow min grey and max grey to both be set to 1.0, which will produce a white, unshaded colour scheme. I haven't put in a way to disable the depth shading on the walls and back of the plot, there isn't a convenient place for another control and it looks bad. Here is an example using the new options.

 

[Breeman]

I have added options in the next build to show or hide the time axis labels, settings label and colour bar. I have also amended the limits for the cubehelix colour scheme settings to allow min grey and max grey to both be set to 1.0, which will produce a white, unshaded colour scheme. I haven't put in a way to disable the depth shading on the walls and back of the plot, there isn't a convenient place for another control and it looks bad. Here is an example using the new options.

View attachment 25686

This is excellentThank you so much John.

 

[John Mulcahy]

Changes in V5.20 beta 26:

• Allow the default audio input to be SPL calibrated
• Show the phase trace in a different hue to the SPL trace and draw it dotted to better distinguish it from the SPL trace. Feedback welcome on how successful (or not) that is.
• Added options in the waterfall graph controls to show or hide the time axis labels, settings label and colour bar
• Upgraded to install4j V8 - that should fix the problem with window managers and similar applications on macOS
• Mac installers from this point onwards are notarized
• Attempted to set up file associations for mdat and req files on Linux - I don't have a test environment to verify this, so perhaps one of the Linux users could see if REW is launched by double-clicking an mdat file.
• Bug fix: Soundcard cal entries were not showing the sample rate
• Bug fix: IllegalArgumentException on sweep import
• Bug fix: Import sweep response accepted drop even if component was not enabled

 

[Breeman]

Changes in V5.20 beta 26:

• Allow the default audio input to be SPL calibrated
• Show the phase trace in a different hue to the SPL trace and draw it dotted to better distinguish it from the SPL trace. Feedback welcome on how successful (or not) that is.
• Added options in the waterfall graph controls to show or hide the time axis labels, settings label and colour bar
• Upgraded to install4j V8 - that should fix the problem with window managers and similar applications on macOS
• Mac installers from this point onwards are notarized
• Attempted to set up file associations for mdat and req files on Linux - I don't have a test environment to verify this, so perhaps one of the Linux users could see if REW is launched by double-clicking an mdat file.
• Bug fix: Soundcard cal entries were not showing the sample rate
• Bug fix: IllegalArgumentException on sweep import
• Bug fix: Import sweep response accepted drop even if component was not enabled

Wow, this is insanely good support. Just donated. Thank you John.

 

[Breeman]

Hi John,

While it is now easier to tell what is what, I feel the dashed phase response can look a bit odd sometimes; have a look at the attached screenshot for instance, the phase curve is now hardly readable. One may prefer to use the same colours for the SPL and phase curves for whatever reason. As in the screenshot, I may want to use pink for SPL curve and blue for the phase curve as an example,; currently, altering the SPL curve colour also changes the colour of the phase curve automatically in ways that a user does not have control over.

The idea behind the original request was more along the lines of having an ability to independently change the colours of either the SPL or the phase curves. This is because at the time, the colours were too closely matched making the reading or interpretation of the graphs a tad hard at times. Seeing that you have now made these changes, I should think an option to toggle between the dashed or solid curves will aid usability. The first price however and if at all possible, is still independent control of each curve colours, especially in the context of SPL and phase being drawn on the same screen.

Thank you for great support on this very useful tool.

 

[John Mulcahy]

I can ensure the phase trace isn't too light for the background, I already ensure it isn't too dark for the dark background. Separate control of the phase trace colour is awkward, currently only the measurement itself has a configurable colour and that colour is used wherever a trace from the measurement appears. There isn't a mechanism for defining individual trace colours on a plot and I don't want to spend time adding one while there are so many other things on the todo list.