Ok, this is in regards to the resolution issue; does tooth count matter?
Kremmit, you indicated the following tooth counts:
DOT spinner (original) has 128 teeth.
The Oscar DOT has 72, same as Blasteroids, Tempest, Oscar Vortex, and the 720 joystick.
The Pole Position ~ approx 80 teeth per one turn of the steering wheel.
Arkanoid has 24 teeth but 486 per single revolution of the knob.
According to Randy, resolution matters, so much so that you can never have too much only too little because you can always down sample your resolution by setting the sensitivity in MAME to some value below 100%.
Well...I own an Arkanoid spinner. If you can always down sample, why doesn't everyone use a geared spinner instead of a direct drive spinner?
For example, I could use my arkanoid spinner and then just set the sensitivity down as follows:
26.34% for DOT (128 teeth/486 teeth)
14.81% for Tempest et al. (72/486)
16.67% for Pole Position (81/486)
Comments??
Hey, real discussion!
First reason: If you've got an Arkanoid spinner, and have spent any time with a regular spinner, then you already know the answer. They
feel totally different. The Arkanoid spinner feels "gritty" due to the gears, as opposed to the silky-smooth feel of a regular spinner (in good condition). Also, the Arkanoid spinner has no "weight" to it. While super-long spin times may or may not be of value in gameplay,
some spin time certainly is, and the Arkanoid spinner pretty much stops turning the instant you take your hand off of it, no matter how hard you whirl it.
Second reason: Availabillity. Arkanoid spinners are in short supply, and the Happ/Wico replacement ones cost big $$ and the knob on them looks ugly as sin. Those are the only geared-spinner options I am aware of.
Third reason: Downsampling is much better than upsampling, but it's still not perfect. Unless your downsample ratio is an exact integer (1:2, 1:3, 1:4...), some rounding must take place, resulting in some uneven-ness. Probably not a big issue, but..
That said, your numbers
should give you roughly the same amount of on-screen movement per rotation of the spinner knob that you would have gotten with the original controller.
Of course, there's still the question of whether or not MAME is providing the correct amount of movement per rotation in the first place. The only way to verify
that is to test with a real arcade PCB and the original controller. There was a thread around here a few years back where folks did that for some games, you might be able to dig it up with some searching.