Good points. I will do my best to clarify. Other than the note about the upcoming "new stepper controller" which is set to be an "Oak for steppers" all of my comments were directed to the $299 Acorn controller. FYI, all these controllers run the same onboard controller, the Beagle Bone Green.
The Acorn was introduced as a DIY step & direction version limited to 4 axis channels, 8 input, 8 outputs, a single 0-10v analog output and a single encoder input. OAk and Allin1 are for servos only. Popular demand has changed the add-on modules in the development schedule. Having just crossed the 2000 unit theshhold and because they seem to be paying attention to the user "want list", a few things are being shuffled.
I use refurbished Lenovo M92p loaded with WIn10Pro with my builds, always providing a setup computer with each control system. I have purchased them as low as $149, but have since started with a higher spec and now include SSD, extra RAM and Wifi due to customer preferrence. These cost around $250 with a Core i5 dual processor. Most DIY guys buy or build their own, as long as they meet the processor benchmark they run fine.
Like you I am not into the "well the machine did not run to my satisfaction, therefore. . . ." stuff. But it was much nicer than publicly stating that I took numerous videos of the same file being run on 3 different controllers, posted them privately on YouTube for the developers and was basically given no credibility for a year before I bailed on UCCNC. BTW, a portion of these files were also viewed by Gerry, mostly with the same conclusion.
When it comes to: "Does this certain controller really run a program 20-40% faster than others or not?" I have had USERS record their same production files at the same feeds and speeds and place them in a video side by side. An example here: https://www.youtube.com/watch?v=vUeE23IrZkk
If I am going to state something about a comparison I double check to be fair. I will almost always have a video. I have no need to BS or exaggerate, my customers buy from me because of my fair comparisons and NEVER selling them on anything. If I tell them a machine or controller will or will not do something, they can take it to the bank.
As far as most things CNC go, I am agnostic, I could give a crap about the brand. I go with what makes the machine happy. It is easy to tell when a machine is running smoothly, quietly, and I use controllers that make my machines look good. When it comes to controllers, I have built and sold over a hundred systems Running Mach (one, lasting less than a week), ShopBot, UCCNC, WinCNC and Centroid Acorn. Many of my systems are replacements and I get to run the OEM controller before I change it over, so add another half dozen (small machine) controller's to the list that I have a bit of experience with. A few more brands can be added that I was able to test in a former R & D position. These controllers all do basically the same things, the real difference is how well they do it.
You propose a bit of a test to pass around, I have been doing this on a daily basis for close to 12 years. Half of it on 2 CNC companies dime, the other half on mine. A year ago I had machines with 4 different controllers running in my shop, thankfully I am now only running two. It will stay that way for a while until I can find another controller that pushes a machine the way Centroid or WinCNC will.