Version: Firmware 13.1.3 + DOS 2.2.4 Date: 2019-02-12

[David A]

I did the standard rew sweeps, with the mic close to the bass mid driver to reduce room interaction.  I then looked at all the measurements rew offers, and was amazed how close they all were, with no smoothing applied as well.  So lots of measurements, but no jitter test, but jitter was solved years ago in good designs, so any change there would likely be tiny.  

I would be surprised if there was an easily measurable change from a firmware that does not imply something was fixed or changed. But the implication above was something was measured, so it was worth looking into.  Sam could be a whole other thing, I didn't test that.

REW is good for measuring speakers and speaker behaviour in rooms but it's not good for measuring electronic components. No professional measures an amplifier's frequency response or distortion characteristics by measuring output through a speaker. Electronic equipment is measured using different gear and different tests to those REW offers.

I'll give you an example of why REW won't work for testing electronic equipment. It's not relevant in the RAAT/AIR case but it may give you an idea of why you would not want to use REW to measure electronic components and the signals they process.

Let's say you have an amp and you want to measure distortion. Measure it in the usual way and you get figures for the distortion level for different harmonics for harmonic distortion. Convert those figures to decibel values. The output of the amp when you do your frequency sweep is going to the combination of the level of the frequency signal and the level of the noise at the same frequency. The distortion is going to vary at different frequencies so you would expect to see this reflected in your REW measurements but your frequency sweep doesn't show the different levels of harmonic distortion at different overtones and it is only measuring one frequency at a time. The harmonic distortion levels are low, usually less than 1%. If the level of the frequency signal is x dB,  and the distortion is at the same level then the combination will measure 3 dB higher than the level of the frequency signal on its own but if the level of the distortion is 12 dB down from the level of the frequency signal, the combined level will only be 0.2 dB higher than the level of the frequency signal. If the SPL if the distortion is 12 dB dow on the test tone, the level of the distortion l n the combined signal would be around 10% of the level of the test tone in the combined signal. We can hear differences in amps with THD levels below 1% and the difference those levels of distortion would make to a frequency sweep are going to be much less than 0.1 dB in level. That's why no professional measures harmonic distortion in an electronic component by measuring speaker output, especially with REW.  The resolution of the test just isn't fine enough to reveal differences that people can and do hear in harmonic distortion, especially when speakers also produce harmonic distortion at levels much higher than those produced by electronic components and there's no way of telling from measurements of speaker output how much of the distortion you measure, if you can get a reliable measurement, is due to the speaker and how much is due to each of the electronic components in the signal path.

I'm not saying that REW is no good or that the tests you did aren't accurate to within REW's specification. I'm saying that the tests you did aren't the appropriate tests to measure things related to digital or electronic signals. They're appropriate for speaker output and room acoustics and they aren't appropriate for measuring some other aspects of audio equipment performance.