Destroying myths

http://archimago.blogspot.com/2018/12/measurements-raspberry-pi-3-b-as.html#more

I share a very interesting article.

Regard

I trust Archimago’s postings more than anyone in the audiophile world. Love his blog.

Thanks for sharing! Always a delight to read his blog…

Hmm I think he tries to get good results but uses the wrong approach.

I also think that many audiophiles fall victim to many kinds of biases, but so do scientists. We are only human after all.
That´s why measurements are normally repeated a lot of times under the same circumstances and repeated with better technology if available later on.

To be really objective you have to take every possible source of degradation into account and measure with appropriate equipment.
Since he is trying to measure very subtle differences it stands to reason that the measurement equipment is also capable of picking up those subtle differences.
It has a reason why professional engineering equipment is very expensive.

I have a few concerns about his measurement methods.
RM AudioAnalyzer is a very inaccurate measurement tool.
Most differences are masked by the capture device and even the digital capture controller (PC). Also Windows is the absolute worst OS for real-time measurements since it is itself only soft real-time capable. On top of that it is totally cluttered with competing services that prevent accurate timely and predictable measurements.
A properly configured true realtime OS like RTOS or Linux with the rt-preempt patchset would be much preferable. Even MacOSX would be better than Windows.

Also RM AudioAnalyzer tells you nothing about data-integrity. If for example the RAM, or any other digital component in the playback chain, is slightly defective or configured badly and introduces random errors RMAA would probably not pickup those differences.

Jitter and errors on the system-bus is also a concern which he has not taken into account. On a playback device, every transmission of false data due to noise induced errors will degrade the audio quality.

And lastly I would measure voltage noise of each power-source with an oscilloscope for hard data of its im-/purity.

What I do not like about this article is the dismissal of claims based on inaccurate measurements. It creates an illusion of seemingly scientific facts.
But they are not accurate facts, only measurements with improper tools.
Throughout history many discoveries could only be ratified years later when measurement equipment got better and more accurate.
The initial claims have not been false but just to hard to prove at the time of discovery.
Having said that, there are also a lot of esoteric garbage theories out there. :smiley: