Creating a link from 6128 to 464 via printer port

[geebus]

Hi all, I was watching something a while back on YouTube about the oliver twins and they stated that they had a device going from their 6128 or 664 that went from the printer port to the spectrum so that if the spectrum crashed they didn't loose all their code etc.


I'm quite keen in trying to use my amstrads (464's and a 6128) in the way they would have been "back in 'day" when they were released. I am looking to start learning how to code on these machines using BASIC, ASM and possibly even C. So, I'm wondering if anyone has any knowledge on the creation of these devices. I'd like to send my code to a 464 to then test my code on. I am aware there are a few slight differences in the BASIC versions, so that will be a fun thing to learn.


Does the printer port just send out 7bit ascii characters in binary across the data lines to be interpreted by the device receiving it?
i.e. can I send raw parallel 7bit binary from the printer port of the sending cpc to the expansion port of the receiving cpc.
If so, this must mean that there is a way of having the memory on the receiving computer available to the expansion port (makes sense for ROMs and RAM expansions I suppose) and being able to write directly to it. Although that would then mean that the memory would have to be filled using compiled ASM? you couldn't just send raw binary data, or could you?
As you can tell i'm rather new to using these computers in a way other than: run"disc"


I'm just having some brain farts and thinking up experiments to have a play with my amstrads really.

So far I've only made simple things like a modern control pad and attached a 3.5" floppy drive to my 6128 (i'd like to work out how to do it to a 464 too but I realise that requires making a DDI device).


If anyone has some wisdom for me on what i'm looking to do, It would be great to learn from you.

[rpalmer]

geebus,
The device you maybe referring to is the vn96 device connected to the printer port.
It communicated via the printer data lines to send 1bit of data  (bit 0 if i recall) and receive 1bit of data using the status signal.It was a very simple device and could be used to communicate between 2 cpc's using a very simple protocol.

rpalmer

[LambdaMikel]

AFAIK, @bryce and @GUNHED  have re-created this.

[Bryce]

AFAIK, @bryce and @GUNHED  have re-created this.

Yes, I made a small batch of vn95 cards a while back, but they were completely hand made on stripboard.

Bryce.

[geebus]

Hi folks, thanks for the replies. I remember seeing the ones you made @Bryce But I didn't realise quite how much they could do.
I'll have a good look over them and read up the how to guide that comes with it.
Thanks @rpalmer for linking that in!

[Bryce]

To the technical side: The printer port can send 7bits, but the pins are output only. The port only has one input pin (Busy), so the connection is limited to one bit and it's as slow as a wet weekend.

Bryce.

[geebus]

So in theory, you could create a parallel transfer from Printer port to Expansion port? Or expansion to expansion?
That would (theoretically) speed things up 7 fold.

[Bryce]

Yes, but it would require quite a bit of additional hardware. It would probably be faster and more cost effective to connect the CPC's via bluetooth with one of the many expansions available.

Bryce.

[geebus]

yeah, I've seen the wifi units and thought that would probably be better. But i'm looking to keep things "period".
I'm more interested in learning how the chips interact so that I can learn from it. Find out where signals originate from and how to send said signals using software etc.
I would really like to be able to program a few things that utilise BB systems etc, but I need to start at the beginning!

[GUNHED]

geebus,
The device you maybe referring to is the vn96 device connected to the printer port.
It communicated via the printer data lines to send 1bit of data  (bit 0 if i recall) and receive 1bit of data using the status signal.It was a very simple device and could be used to communicate between 2 cpc's using a very simple protocol.

rpalmer

That's correct so far. However the protocol is anything but simple, in contrast, it's quit sophisticated. Regarding the hardware, it's needed. And it's quick, efficient and error prone. A masterpiece. The WizCats did a great job. 

[Bryce]

Error prone means it's likely to have errors? Is that what you meant to say, because the hardware is so simple shouldn't be able to cause errors.

Bryce.

[GUNHED]

You have a _NETWORK_ of CPCs, not a two computer connection (ok, you may more often have, but it's made for 5-10 CPCs). They may all differ where they are in their spot in time/CPU run. I talk about synchronization etc. There are great manuals in the internet, they can explain all this in great detail. 

[Bryce]

Ah, ok. So the architecture is error prone, not the hardware.

Bryce.