Hi Chu,,,,,,,,,,,,,,,,,,,,,,,
Thanks for your questions. Some of your questions are already answered in this thread and most of the other questions are covered in detail in other threads on this forum. So, I'll just answer the "un-answered" questions.
All of the testing I do is for my own purposes to make the best products I can and to evaluate new products, parts, or technologies. Since the testing is for my own purposes, I follow whatever procedure suits me for the particular test. I get others, including Steve, to participate in some of the Subjective listening tests to broaden my information base. Steve has been doing more testing with me this year since my dear golden-eared friend, Don Schlicker, died in February. Sometimes I get results that are so dramatic or definitive to me that I don't bother with anyone else's participation. But, when I get test results where I can't tell the difference between A and B, or when I want to verify my results; then I get lots of additional participants. My results are "published" , if you will, in the form of my finished products. Other audiophiles evaluate my tests by auditioning my finished products and either keeping them or returning them for a refund. So, if I don't do quality tests, I get severely punished in my own wallet. ha, ha.
I'm certainly not going to publish too many details about how I do everything, because my competitors might be reading this? ha, ha.
Other than parts quality and less wire, I can't offer any improvement on the ABX box that you found and shared with the rest of us. Especially since it's the same basic circuit design I came up with 25 years ago. As I'm sure you already know, the volume pots at the A & B inputs are a super feature because they do 2 important things. They allow you to exactly level match A and B. ( I level match to within 0.1 dB.) And, they present the same constant load impedance to both A and B. So, the volume pots level the playing field in 2 ways. It's important to use quality pots so that the left & right channels are also level matched. If they're not matched, this could throw off the balance, the focus, and the sound stage effects. A low quality pot could possibly mask the subtle differences.
Several years back I built a special prototype of our Model #770 preamp that had dual circuits in it. One circuit was the reference and the other circuit was identical except for the parts. I substituted various brands & construction of capacitors & resistors in the parallel circuit. Then I could switch back & forth on the fly to instantly compare parts substitutions in the circuit.