V5.20.10 Early Access

The bug about signal generator square wave with the option « Band limits square wave samples »is not totally fixed.
Example with the followinq sequence with generator running
1/ « Band limits square wave samples » not selectionned and Duty cycle 50% --> Output signal OK :scope observation with 50% duty cycle signal
2/ changing duty cycle 15% --> output signal ok :scope observation with 15% duty cycle signal
3/ « Band limits square wave samples » selectionned --> output not ok : scope observation with signal duty cycle 50% and not 15%
 
I notice also an error with the following sequence
Open rew
Open scope
Open generator
Square signal and « band limit square wave sample » not selectionned
Running scope and running Generator --> OK
« Band limit square wave sample » selectionned --> REW error

Capture d’écran 2022-08-05 125506.jpg
 
I installed the latest build of REW. In the previous ones, when measuring in the car, I had a 'float' delay. For example, the first measurement gave -316 mm, the second -314 mm, and so on. But after 10 minutes it stopped. In the latest assembly, the delay began to 'float' faster and by a large amount. Maybe it's not related to REW. Could the clock speed of the CD player in the car be causing this? That is, over time, it became unstable.
 

Attachments

REW also stopped seeing sound card calibration files. Although they are on the preferences / cal files tab.
Highlighting the chart on hover seems too thick and bright to me. It is impossible to see other charts behind the selected chart. Maybe change to something in between the old way and the new one?
 
Last edited:
REW also stopped seeing sound card calibration files. Although they are on the preferences / cal files tab.
Highlighting the chart on hover seems too thick and bright to me. It is impossible to see other charts behind the selected chart. Maybe change to something in between the old way and the new one?
There is a new View preference to control whether highlight uses a thick trace.
 
I notice also an error with the following sequence
Open rew
Open scope
Open generator
Square signal and « band limit square wave sample » not selectionned
Running scope and running Generator --> OK
« Band limit square wave sample » selectionned --> REW error
Thanks, fixed for the next update (also fixes the duty cycle).
 
REW also stopped seeing sound card calibration files. Although they are on the preferences / cal files tab.
Can you explain what you mean by that? Stopped seeing the soundcard cal file where?
 
I installed the latest build of REW. In the previous ones, when measuring in the car, I had a 'float' delay. For example, the first measurement gave -316 mm, the second -314 mm, and so on. But after 10 minutes it stopped. In the latest assembly, the delay began to 'float' faster and by a large amount. Maybe it's not related to REW. Could the clock speed of the CD player in the car be causing this? That is, over time, it became unstable.
It doesn't look like it, the clock adjustment is zero for all the measurements (is clock adjustment enabled?) and there is no visible change in the shape of the IR to suggest clock drift.
 
I notice a bug in the square generator. I don't know if it is present before the new version 10.
When the generator is running, Actual frequency is shown and all is correct. But when we stop the generator, Actual frequency is maintained (and with an inconsistent value).

Capture d’écran 2022-08-05 182043.jpg
 
If you click on the info button of any measurement from post # 53, it says that there is no sound card calibration file.
Those were made with file playback, the soundcard Cal file is not used when the soundcard is not the source.
 
@John Mulcahy, I noticed that the graph display setting dropdown (SPL, dBFS, etc.) in the SPL graph window resets to the SPL setting if one switches to any other graph and then back to the SPL graph. Example, I set the SPL graph to dBFS then switch to the All SPL graph and then switch back to the SPL graph, the display setting has reverted to SPL instead of the selected dBFS. I ran the uninstaller and re-installed 5.20.9 and it had the same behavior. Behavior is the same on macOS and Win 10/11.
 
Last edited:
So it does, it's a legacy of a time when there were no axis choices. I've fixed it for the next update.
 
Not a big deal, and perhaps not new with this release, but I set a cursor in the All SPL window and it shows 120.0 Hz:
53734

I then take a Capture of it, and the resulting jpg shows 119.9 Hz:
53735

IWBNI they were consistent.
 
Your capture has a different width. The cursor is constrained to pixel precision, the reading is the closest value that aligns with a pixel. If you want an identical capture use the "Default" button to set the capture width the same as the source.
 
Is the new trace arithmetic functionality a better way to create a convolution file for room correction? Obviously, you can't do any hands-on changes as you can with EQ filters, but still?
 
Absolutely. Check the screen shot below, red line is the original measurement and highlighted line is after correction and no EQ filters, just the inversion. Note that you are correcting the full frequency band and there will not be any degradation in the sound as these are FIR filters:

53765


and "L inversed min phase" is the convolution file to generate the correction:

53766
 
Note that you are correcting the full frequency band and there will not be any degradation in the sound as these are FIR filters:
Have you listened to different tracks with your ears and are completely satisfied with the sound?
 
Have you listened to different tracks with your ears and are completely satisfied with the sound?
I have never before felt good about a correction if I tried to EQ above Schroeder's but with this technique it really is quite different. I almost always get better overall sound than with no DSP and I find myself just playing with the level of bass correction for optimization. My room and speaker positions are quite optimized so the difference is not huge but it's there.
 
Btw @John Mulcahy I don't think the term "Trace Arithmetic" is very good, sounds like it's about the visible trace only (like for example Trace Options). Wave or vector arithmetic sounds better ihmo.
 
If you want an identical capture use the "Default" button to set the capture width the same as the source.
In case my previous response was unclear, I can reproduce the discrepancy using the Default button, with All SPL, frequency range set to 80 to 200 Hz, image width at 1109, and cursor at 120.0 Hz. The captured image shows 119.9 Hz.
 
In case my previous response was unclear, I can reproduce the discrepancy using the Default button, with All SPL, frequency range set to 80 to 200 Hz, image width at 1109, and cursor at 120.0 Hz. The captured image shows 119.9 Hz.
You'll have to live with it.
 
I tried the inversion trick with an RTA measurement, but could not get it to work. Should it work?
 
Back
Top