V5.20.14 early access build

[John Mulcahy]

@John Mulcahy there seems to be a minor visual bug in the RT60 graphing:

It's present on both the Mac and Win ea36 builds.

Thanks, I've fixed it for the next build.

 

[pklose]

Further changes 25th February:

• Added: "Show ISO3382 average" option on the Overlay RT60 graph to show the average of the chosen ISO3382 parameter for the selected traces
• Added: Preferences menu entry to save preferences to a file

Honestly I am so happy! This is such an improvement to my workflow! Thanks so so much!

 

[serko70]

Java exception in version 36 during a very innocent EQ/"Match response to target" operation with a very short clipboard:


REW V5.20.14 running Azul Systems, Inc. JRE 1.8.0_352 64-bit on Windows 10 Locale: language en, keyboard GB, windows-1252 at 96 DPI


Message:
java.lang.ArrayIndexOutOfBoundsException: 0
Stack Trace:
0
roomeqwizard.hE.run(y:2245)





and this is still frozen on my EQ window:

The problem is the upper frequency range selected too close to lower range I think!

 

[John Mulcahy]

I have fixed that for the next build.

 

[John Mulcahy]

A few more updates today. There are two changes to provide a faster way to get response measurements:

• Added: A 64k sweep option for faster response measurements at sample rates of 48 kHz or lower
• Changed: Made the period of dither played before a measurement sweep optional with a check box on the Measure dialog to select it

The dither before a sweep may be necessary if the replay path needs to see some data before it becomes active, but for permanently active replay paths it can be omitted.

For people who often filter their impulse responses:

• Added: Fractional octave filter controls on the Impulse overlay graph to allow all selected impulse responses to be filtered at once

For those folk that like to stick to IEC263 aspect ratios for graphs:

• Added: The SPL & Phase, All SPL, Overlay SPL and Overlay Predicted SPL graphs have an IEC263 aspect ratio control to force them to a preferred ratio. If a ratio is selected the margins around the graph will be adjusted to achieve the specified dB/decade. At very narrow frequency spans the margins will be prevented from growing too large and the aspect ratio can no longer be maintained, the actual aspect ratio can be shown on the graph by selecting that option in the View preferences.

A further addition to preferences handling to accommodate starting with a known set of preferences:

• Added: REW will look for a file called initialprefs.txt in the log files folder on startup and if found will initialise all preferences from it. Alternatively a -initialprefs argument or a -Drew.initialpreferences.file roomeqwizard.vmoptions file entry can be used to specify a file with the initial settings for preferences. The initial preferences file will not be modified.

Other changes this time:

• Changed: Disable deletion actions while loading measurements to avoid measurements being deleted while still being loaded
• Changed: Use preferred octave fraction centre frequencies where appropriate when exporting distortion data as text
• Fixed: When using bars on the 1/3 octave RT60 graph the lines between the 50 and 63 Hz bins were not vertical
• Fixed: An exception could occur if match target was used with a very narrow match range

 

[thothsong]

It seems strange to constrain&clip the Controls box to the graph size for IEC263, when there's plenty of available blank space:

 

[John Mulcahy]

I can stop it being cropped, I'll do that for the next build. Letting it float right would be problematic because there are axis zoom buttons to the right when the graph has a right hand axis and it looks pretty bad when there is some margin to the right and the panel is partly over the graph and partly not.

 

[staticV3]

Would it be possible to add the ability to set a separate sample rate for the DAC and ADC, ideally using EXCL? I'd like to measure the DAC's response close to Nyquist, without the ADC's own Reconstruction filter affecting the results.
If I could set the DAC to play at 44.1kHz and the ADC to record at 96kHz or higher, then that would be super.

 

[onlyoneme]

Would it be possible to add the ability to set a separate sample rate for the DAC and ADC, ideally using EXCL? I'd like to measure the DAC's response close to Nyquist, without the ADC's own Reconstruction filter affecting the results.
If I could set the DAC to play at 44.1kHz and the ADC to record at 96kHz or higher, then that would be super.

At the moment probably only if the DAC is fed externally using REW's signal stored to a file.

 

[John Mulcahy]

In principle it would be possible to have different sample rates for separate devices, but it wouldn't be possible to make sweep measurements that way as the processing requires the output and input sample rate to be the same. Most straightforward would be playing a full range white random noise file at the desired replay rate on the DAC while capturing the DAC output on the RTA using the desired capture rate.

 

[staticV3]

Feeding the DAC externally is what I'm doing at the moment.
For some frequency response measurments though, I like to use the stepped sine method instead of the usual sweep or noise.
And for that, REW must control both DAC and ADC.

 

[onlyoneme]

Feeding the DAC externally is what I'm doing at the moment.
For some frequency response measurments though, I like to use the stepped sine method instead of the usual sweep or noise.
And for that, REW must control both DAC and ADC.

I know and I share the pain in case of stepped tests... This time I dream of client-server architecture added to the REW, to let control another REW instance.

 

[staticV3]

Also really nice would be the ability to load multiple measurements in the CTRL+O and CTRL+I dialogues:

CTRL+A to select all
CTRL+left click to select multiple
Shift+left click to select all in between

 

[dcibel]

For those folk that like to stick to IEC263 aspect ratios for graphs:

• Added: The SPL & Phase, All SPL, Overlay SPL and Overlay Predicted SPL graphs have an IEC263 aspect ratio control to force them to a preferred ratio. If a ratio is selected the margins around the graph will be adjusted to achieve the specified dB/decade. At very narrow frequency spans the margins will be prevented from growing too large and the aspect ratio can no longer be maintained, the actual aspect ratio can be shown on the graph by selecting that option in the View preferences.

Hi John,
Can the "as-plotted" view mode remain when selecting 25db/dec options for the preset view buttons?

 

[John Mulcahy]

Also really nice would be the ability to load multiple measurements in the CTRL+O and CTRL+I dialogues:

The measurement file dialog already allows multiple selection. You need to click on the files display to use selection shortcuts.

I'll allow multiple selection in the frequency response and impulse import file dialogs, but you can also drag and drop multiple files onto the main window to load them.

 

[John Mulcahy]

Can the "as-plotted" view mode remain when selecting 25db/dec options for the preset view buttons?

Meaning have the shortcuts operate as they did previously, changing the SPL range to fit the aspect ratio? Changing the SPL range generated quite a bit of push-back from folk that want to stick to an IEC 263 ratio, hence the more comprehensive solution in the latest build.

 

[dcibel]

Meaning have the shortcuts operate as they did previously, changing the SPL range to fit the aspect ratio? Changing the SPL range generated quite a bit of push-back from folk that want to stick to an IEC 263 ratio, hence the more comprehensive solution in the latest build.

I simply mean to treat aspect ratio and IEC263 display as two separate selections, so selecting 25dB/div in the view preferences doesn't automatically change the view mode to IEC263 constraints.For myself, making the most use of the screen space would be a priority, which would mean for 20-20kHz dispaly at 25dB/div aspect, horizontal width is maximized and vertical scale range is dynamic based on window size. Selecting +/- buttons on vertical scale breaks the aspect ratio, as does right click zoom, so one option that allows for maximum use of screen space would be to simply disable manual vertical scale adjustment for locked aspect ratio, change +/- button selection here to aspect ratio selection instead is a possibility.

 

[John Mulcahy]

I'll change the preset behaviour so that a preset's aspect ratio will only change the graph aspect ratio selection if the selection is not "As plotted", so presets can still be used to set an aspect ratio while using all the graph area (as they used to operate).

 

[serko70]

In SPL & Phase tab, "Fit to data" will lose impulse responses with a fixed 0dB SPL. The example below is a XO phase linearization pulse imported from rePhase but the behaviour is similar for a REW produced Dirac pulse.

What would be expected:



What "Fit to data" will do:



A very minor problem but makes it difficult to find and see the response again.

 

[John Mulcahy]

I'll apply a minimum span of 6 dB.

 

[John Mulcahy]

Builds updated today with the changes below. Note the preferences bug fix, the Java runtimes default system preferences implementation could ignore REW's attempts to use file-based preferences, that has now been addressed.

• Added: Multiple selection is now possible in the file dialogs for import frequency response and import impulse responses, though drag and drop may be easier.
• Added: The span and aspect ratio of the Impulse graph magnitude view can now be configured in the graph controls
• Changed: A frequency span preset aspect ratio will only change the graph aspect ratio selection if the selection is not "As plotted"
• Changed: Apply a minimum span for fit to data
• Fixed: The IEC 263 aspect ratio preference key was not unique for the All SPL graph so the chosen ratio could be forgotten
• Fixed: The Java runtime default preferences store could override the file-based preferences

 

[majorwong]

Builds updated today with the changes below. Note the preferences bug fix, the Java runtimes default system preferences implementation could ignore REW's attempts to use file-based preferences, that has now been addressed.

• Added: Multiple selection is now possible in the file dialogs for import frequency response and import impulse responses, though drag and drop may be easier.
• Added: The span and aspect ratio of the Impulse graph magnitude view can now be configured in the graph controls
• Changed: A frequency span preset aspect ratio will only change the graph aspect ratio selection if the selection is not "As plotted"
• Changed: Apply a minimum span for fit to data
• Fixed: The IEC 263 aspect ratio preference key was not unique for the All SPL graph so the chosen ratio could be forgotten
• Fixed: The Java runtime default preferences store could override the file-based preferences

Hi,John,where's the latest build version for download?

 

[dcibel]

It’s the first post in this thread.

 

[serko70]

John, I am trying the new 64k short sweeps and impulse responses are shifting from measurement to measurement by around 7mm. None of the 128k sweeps deviate from each other. Since 64k sweeps should be handling USB mic timing problems even better, I suspect something is wrong with the new 64k sweeps. It might be coincidence but as you'll see I took 11 consecutive 128k measurements with not one deviation whereas every other 64k sweep moved.

 

[Cristianolo]

I was wondering... is there any way to make the export process "easier"?
For example, when I have a series of measurements in a project, but I need to export 20 of them, today I have a few options like:
- export all and delete the TXT files I don't want;
- export one by one that I want.
Would there be any way to make it possible to select multiple measurements using CTRL/SHIFT, or even by checking a box, and using a command like "Export Selected"?
In the case of using CTRL/SHIFT, the measurements selected in the measurements panel may have a yellowish mask, as happens with the curve when pointing the mouse in the legend to one of them.
Too much work to implement?
Thanks.