Trying to write an AtariVOX driver.

[vdub_bobby]

At 60 characters per second, does that mean that if you used it for a game it would introduce some latency if you wanted to have it speak in sync with something happening on the screen in response to an event?

 

Wouldn't the effective speed be lower if you consider all the control information (pitch, speed, volume) that also needs to be transferred?

 

I'm just wondering whether it might be possible to write a display list interrupt routine that runs on every scanline.

849251[/snapback]

In my experience with the AVox, admittedly limited to the 2600, 60 bytes per second is plenty to keep the AVox speaking. I'm pretty sure that that is the frequency Man Goes Down uses and the speech syncs up pretty well with the onscreen action in that game.

For syncing to an animation of someone talking you might have to be more precise, I suppose.

[danwinslow]

You know, I am no 6502 assembly jeenyus...but if I understand whats going on here, this is a way to pump bits out of one of the joystick ports driven by vblank? Call me Capt. Obvious, but I just wanted to make sure. I could use this code for something I want to do, if you don't mind me borrowing it.

[Bryan]

You know, I am no 6502 assembly jeenyus...but if I understand whats going on here, this is a way to pump bits out of one of the joystick ports driven by vblank? Call me Capt. Obvious, but I just wanted to make sure. I could use this code for something I want to do, if you don't mind me borrowing it.

849337[/snapback]

 

Yes, with this driver you can write one byte per VBL = 60 bytes per second.

 

-Bry

[danwinslow]

Eh...ok 480 bits per second...hmm that seems kinda slow...thats 480 Baud, right? How many cycles does it actually take flat out to send 1 byte?

[Bryan]

Eh...ok 480 bits per second...hmm that seems kinda slow...thats 480 Baud,  right? How many cycles does it actually take flat out to send 1 byte?

849354[/snapback]

 

Actually, while it's sending it sends at 19.2K baud, but only one byte is sent per VBL. Since it takes 10 bits to represent 1 byte at with the start and stop bits, the effective rate is 600 baud.

 

-Bry

[danwinslow]

Ah, gotcha, very cool. So I assume you could theoretically send more..say 10 bytes per vblank? It looked like you were doing stage 1 vblank...so thats pretty time critical if I recall correctly. But anyways there's no real reason that this needs to be slaved to vblank for my purposes, so some other interrupt would probably do as well.