Uhrzeit unter Basic 10

Meinst Du TI$ ?

Setzen mit TI$ ="SS:MM:ss.ms" (Beispiel: TI$="19:25:30.7")

Ausgeben mit Print TI$

Der MEGA65 besitzt eine batteriegepufferte RTC, welche gesetzt werden kann.

Der c65 hatte im Original (iirc) auch keine batteriegepufferte Uhr.

In der MEGA65 GEOS Version haben wir schon ein Tool eingebaut, mit welchem die RTC gesetzt werden kann und dann auch angezeigt wird.

Hallo,

ich habe es gerade für dich gemacht, dass man einfach im MEGA65 Configure Utility, den RTC liesen und aktualisieren kann. Sowieso:

Ist noch nicht perfekt, aber es geht.

LG

Paul.

Hi mega65 ,

Paul, i have a question: so the RTC is completely independant from the C64/C65s internal Timer

(TI$/TI) and keeps its settings saved in the RTCs own small battery powered NVRAM?

Since, iirc, some C64 programs were using the (unprecise) TI$ for certain operations.

So the new RTC is an usefull addition to the MEGA65, to be able to have precise time dependant operations working ?

Or in other words, we now have TI/TI$ & a RTC ?

Yes, we currently have TI$ and the RTC. I am thinking about how I can automatically inject the current time into a running C65, ideally automatically. This just needs a Hypevisor scheduled interrupt "hyperrupt" that gives the hypervisor the chance to inject the value after the C65 ROM has started up and initialised TI/TI$. Then it would just transparently be there. You are welcome to file a github issue, so that I don't forget to do it.

LG

Paul

Hi Paul,

thanks and this is indeed a great idea. i've created an issue in Github, here:

https://github.com/MEGA65/mega65-core/issues/225

In Xemu (xc65 and xmega65) the actual time can be shown in BASIC 10 with PRINT TI$

I guess the emulator gets the time from the host system.

Yes, I believe so. We just need to implement a similar piece of magic for the MEGA65 to do this. I've got some ideas on this. It's just a matter of me having the time to attack it, along with everything else that needs to be done.

LG

Paul.

Zitat von FeralChild

I hope you all realize, that TI / TI$ is only updated during the interrupts, and as soon as you load a program from disk (not even mentioning tape) this "clock" will be late?

Yes, that's why i called it "unprecise". The RTC will be totally independant of it.

Like Paul said, our plan is to syncronise the TI$ with the RTC once at the bootup of the MEGA65 and then the user can do what he wants

with TI$ as usual.

If the User needs a precise clock, he can access the RTC ( MEGA65s GEOS is using the RTC i.e. already).

But thanks for the reminder anyway !

We could even periodically update it, but that is probably overkill.

But you are right that the C65's funny way of handling its internal drive causes a big problem for the TI/TI$ variables, as the CPU might be stopped for some time. That said, the DOS interface is so horribly innefficiently built, that apart from when seeking a new track or sector, the interruputs will be able to happen.

LG

Paul.

Yes, I quite agree that if we did it, it would be user-selectable.

I didn't realise that it was also a problem on the C64. I had figured that the IRQs would just be deferred, and get serviced after.

LG

Paul.

I'm just fixing the documentation in the MEGA65 user guide material now.

There is also 45 pages of "introduction to BASIC" which is coming along nicely now.

LG

Paul.