How to check paddle jitter without game

[oryan_dunn]

I just got a pair of paddles of ebay along with Super Breakout. The paddles were basically non-functioning, no matter how I turned it the bar was either to the left or to the right. I disassembled one of them, cleaned all the gunk off the black ring and contacts in the pot and reassembled. After that, it worked just fine in the game. I had seen some reports of people using this game or that, or saying game x wasn't a good game to test with. If you don't have a game, or want a more precise way to test your paddles, put your basic cart in and run the following program:

10 PRINT PADDLE(0), PADDLE(1) : GOTO 10

RUN

 

That will constantly print out the current value of the paddles 1 and 2 connected to joystick port 1. The values will range from 0 to 228, and the values should be steady without you moving the paddle, and should smoothly increase or decrease as you move them.

 

I've since cleaned the other pot, and both work quite well in Super Breakout. I think I may pick up the Castle Crashers game here soon (along with another set of paddles).

[Rybags]

Another method - gives a visual player bar moving with paddle 0, though the visible range doesn't cover the entire pot range:

 

 

10 GRAPHICS 0:POKE 53277,0
20 POKE 704,134:POKE 710,0
30 POKE 53261,255:POKE 53256,0
40 POKE 53248,PADDLE(0):GOTO 40

[+Stephen]

While on the subject of paddles and jitter, has anybody noticed that PAL machines seem to suffer from jitter more than the NTSC machines? I do mean when using the same set of paddles on each machine.

[Atari8bitCarts]

There is a paddle jitter test cartridge, with the image on Atarimania.

 

I have one of the original carts

 

http://www.atarimania.com/utility-atari-400-800-xl-xe-paddle-jitter-test_13692.html

[Rybags]

PAL uses slightly shorter duration scanlines though that shouldn't be a huge influence. The slower framerate might make it more noticable as there's 10 less samples taken per second.

I guess the ideal test would be to do a program that disables all interrupts then initiates a POTGO every 230 scanlines. That would come close to being a level playing field for both types.

 

Normal setup, difference is about 20% (59.92 vs 49.86 samples/sec)

230 scanline sample method, difference would be under 1% (68.34 vs 67.82 samples/sec)

[erichenneke]

While on the subject of paddles and jitter, has anybody noticed that PAL machines seem to suffer from jitter more than the NTSC machines? I do mean when using the same set of paddles on each machine.

Absolutely. I have both pal and ntsc 800xl machines, and all things equal the jitter on pal is much, much more pronounced.

 

Even more strange, if I use the pal machine with the syscheck2 running the bios off the syscheck device, then the jitter is much better and on par with the ntsc machine. As a result, paddle games are much more enjoyable on the pal machine when using syscheck as an alternate bios selector.

Edited February 18, 2017 by erichenneke

[Rybags]

Strange - maybe the addon creates some electrical stability so the samples don't drift as much.

 

Possible other ideas - NTSC machine framerate is closer to their native AC frequency, 0.129% slower with PAL being 0.279% slower. Could AC leaking into the DC supply or interference at different phases of the AC waveform cause the pots to function differently over time?

 

Or is it just that slower framerate makes it all more noticable? A good "scientific" test mechanism would be to use a known good DAC source where it could be calibrated against the Atari then varying voltages be fed in with changes occurring each frame, then the drift from expected value recorded.