Cheetah Sweet Talker doc/test?

[stroebeljc]

Excellent point, Mark.
Yes, the address lines still need to be sequenced for all of the DRAM to refresh properly.
So, a risky move on Cheetah's part to take over /WAIT, even in the Speccy.

[1024MAK]

In the photos that Moggy posted of the Cheetah Sweet Talker, yes there is a connection to the /WAIT line.

The extra board does however look like it's picking up the /RD (read) control signal.

Will Rigters version doesn’t use the /WAIT signal. Instead it uses the /RD signal in combination with the 74HC138 and the Q1 transistor to provide a single status bit via the D7 data line. Hence the Z80 machine code IN a,(2F) instruction written on the circuit diagram.

Mark

[1024MAK]

The instructions for the ZX Spectrum version of the Cheetah Sweet Talker (available here or actually read the file attached by Moggy on the first page of this topic ) don't say anything about waiting between issuing data to the unit or mention any status bit.

Mark

[Fruitcake]

Firstly one working WAV and P file of "Chatbox" as promised.

Regarding the Chatbox program attached to this thread, is it exactly as on the cassette or have you modified it to work with your Sweet Talker since it does use the WAIT line due to the missing resistor and transistor?

Line 1010 performs PAUSE 1, and the instructions the program later displays (line 168) says to perform PAUSE 5 after RAND USR 16514.

Everything so far indicates that the interface asserted the WAIT line until the allophone finished playing and so there should be no need (in theory) for the PAUSEs, hence wondering whether these are always there or added by yourself?

[Moggy]

Firstly one working WAV and P file of "Chatbox" as promised.

Regarding the Chatbox program attached to this thread, is it exactly as on the cassette or have you modified it to work with your Sweet Talker since it does use the WAIT line due to the missing resistor and transistor?

Line 1010 performs PAUSE 1, and the instructions the program later displays (line 168) says to perform PAUSE 5 after RAND USR 16514.

Everything so far indicates that the interface asserted the WAIT line until the allophone finished playing and so there should be no need (in theory) for the PAUSEs, hence wondering whether these are always there or added by yourself?

No the program comes as is I would have thought and not altered by myself, it's something I picked up from the web somewhere or other purely for sytem11's benefit and isn't something I've used myself.

I just add pauses in my BASIC programs to make up for the lack of the resistor/transistor combo as without them the speech is too quick.
Since Charlie changed the port setting so as to allow a program to run at the same time, I find that running a very simple program whilst the unit is speaking does away with the need for pauses, although the Big Bang ROM allows the program to be a tad larger as it is faster than the standard ROM, it's a bit of a balancing act but fun to play with.