OP
- Thread Starter
- #1,001
Probably not for a while. It'll require too many changes.
-
WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
Beta-test: DeltaWave Null Comparison software
OP
- Thread Starter
- #1,002
Preview version 2.0.13 is available for those who want to try it with OGG Vorbis file format: https://app.box.com/s/hlglgy3aj39rjfxu343fho4t0cdrq9v8
As it turns out, I added the support for this file format more than a year ago, but forgot to enable it
As it turns out, I added the support for this file format more than a year ago, but forgot to enable it
- Joined
- Feb 23, 2016
- Messages
- 23,118
- Likes
- 43,027
You mean you have been holding out on us for the OGG Vorbis format? Early on when storage wasn't as cheap that was my goto format for away from home use. Oh well not the worst mistake in the world.Preview version 2.0.13 is available for those who want to try it with OGG Vorbis file format: https://app.box.com/s/hlglgy3aj39rjfxu343fho4t0cdrq9v8
As it turns out, I added the support for this file format more than a year ago, but forgot to enable it
OP
Hello!
I own an E1DA Cosmos and want to measure the df using DeltaWave. However, every time when i try to write a signal from the DAP, it had significant delays that DeltaWave cannot resolve on its own and sync. Is there a described process or instructions available for testing standalone devices such as DAPs or CDs?
I own an E1DA Cosmos and want to measure the df using DeltaWave. However, every time when i try to write a signal from the DAP, it had significant delays that DeltaWave cannot resolve on its own and sync. Is there a described process or instructions available for testing standalone devices such as DAPs or CDs?
OP
- Thread Starter
- #1,007
If you have a large delay at the start of one of the waveforms (more than a few seconds) you can tell DeltaWave to ignore a number of seconds from that waveform:
You can see how much to trim by loading the waveforms without aligning them by pressing "Show'. As you change Trim settings, you'll see the time difference between them on the Original Waveform with the selected trim applied. You don't have to align them perfectly, just enough to bring them to have the start within a few seconds of each other.
staticV3
Grand Contributor
- Joined
- Aug 29, 2019
- Messages
- 11,680
- Likes
- 19,209
@pkane
Feature request for Deltawave:
Could you add an optional matching step that, besides matching timing and clock drift, also matches the steady-state frequency and phase response so that only the nonlinear distortion remains, which cannot be filtered out with a simple frequency/phase adjustment?
I think it would be a fantastic tool for stress testing Amps, headphones, and speakers with real music signals.
Feature request for Deltawave:
Could you add an optional matching step that, besides matching timing and clock drift, also matches the steady-state frequency and phase response so that only the nonlinear distortion remains, which cannot be filtered out with a simple frequency/phase adjustment?
I think it would be a fantastic tool for stress testing Amps, headphones, and speakers with real music signals.
OP
- Thread Starter
- #1,009
That is what the Non-linear EQ feature does. The feature computes the inverse impulse response, and then applies it to the comparison waveform. In effect, that should remove all the linear differences in phase and amplitude between the reference and comparison waveforms. The delta waveform/spectrum between the references and non-linear-EQ'ed comparison should then represent all the non-linear differences + noise.
staticV3
Grand Contributor
- Joined
- Aug 29, 2019
- Messages
- 11,680
- Likes
- 19,209
Awesome! I'll have a look.
OP
- Thread Starter
- #1,011
Just make sure to use long enough samples. 2-3 minutes at least, since computing the impulse response from noisy data requires more data to average. 10 seconds samples, like some have tried, will not work
splitting_ears
Active Member
Here's a weird quirk I've just experienced when trying to set custom zoom levels:
Or have I set something wrong?
Or have I set something wrong?
Last edited:
OP
- Thread Starter
- #1,013
Hi Cyril, no, the number you're entering is actually the min/max numbers of FFT bins (translated to Hertz below the entry box). I was a little lazy when I implemented this control and since the values in FFT bins were easier for me to use directly, I just left them like that. Use the value below the entry box to judge the actual frequency, min and max.
Last edited:
splitting_ears
Active Member
Oh, I see! I was surprised because I'm a bit more used to Multitone (which supports entering Hertz values directly) but I'll do as you suggest in the meantime. Thanks for the reply!
Just asked the help of portableapps.com for:
[FREEWARE] DeltaWave Audio Null Comparator | PortableApps.com
What is it? DeltaWave is an experimental software built to allow detailed comparison of two different captures of the same audio signal. It was inspired by the idea of Audio DiffMaker, but it is a completely new piece of software, written from scratch. No parts of Audio DiffMaker were used in...
portableapps.com
OP
- Thread Starter
- #1,016
Updated DeltaWave v2.0.16 is now available.
Changes in 2.0.16
- Fix: scaling issue with minimum-phase filter generator
- Change: filter corner frequency can now include fractional values and can be entered manually
manisandher
Addicted to Fun and Learning
Paul, I'm getting the following (consistent) response for the RMS of the difference of spectra: "NaNdB".
Any idea what's going on?
Any idea what's going on?
OP
- Thread Starter
- #1,018
Mani, with what settings and is it the same with all measurements?
manisandher
Addicted to Fun and Learning
Settings:
All filters OFF.
I've just taken 55 measurements (11 each for 5 different DACs). It's the same for all of them.
All the other measurements and all the plots are exactly as expected.
All filters OFF.
I've just taken 55 measurements (11 each for 5 different DACs). It's the same for all of them.
All the other measurements and all the plots are exactly as expected.
OP
- Thread Starter
- #1,020
Ah, that's the difference of the two spectrum plots, not the RMS of the difference. I know, confusing
That takes two averaged spectra and computes (in dB) the difference between those averages. The actual RMS of the difference is closer to the top of the report -- -73dB in your case. I'll check into why the difference of the spectra calculation isn't working right, as far as I know, it has been touched for years...Similar threads
