Multi-VIC systems?

[lance.ewing]

I've seen a few posts that have mentioned the availability of new old stock for the 6561-101, and indeed I've done some Internet searches and can see them advertised in a number of places. I don't have much of a need to get a new VIC chip, but given its possible to order a few, I started thinking about whether I might just do that and then play around with them on a breadboard, maybe even use the 6561 as the video chip for a home brew CPU that has been bubbling away on paper, in mind, and on breadboards for a few months.

This got me thinking: Has anyone done this before? i.e. tried to use the VIC chip outside of the VIC 20 for some purpose? Has anyone tried using it on a breadboard?

My thinking then roamed in to some fairly wacky realms. What would happen if I were to try using multiple VIC chips in one system? Would it be possible to mix the output of two chips to produce a single mixed video signal? I did a bit of Googling on this and it turns out that analog video mixers are insanely expensive, but a lot of that is to do with putting the two video signals in sync. So what if the two video signals are guaranteed to be in sync? If they're running as part of the same system, being driven by the same clock inputs, then surely the video output of each VIC is in sync? Could the signals be mixed more easily in such a scenario?

Take a look at this demo (particularly the video on that page):

http://gieskes.nl/visual-equipment/?file=schele-mixer

What would happen if we were to use a circuit like that with two VIC chips that are in sync? Perhaps we'd still need to use the LM1881 on one of the signals; not sure. Don't really have any experience with analog PAL/NTSC video signals. I'm sure that a lot of you do though.

What I'm trying to lead to is the possibility of using multiple VIC chips to produce more colours, by mixing their output. Would there be a relatively cheap way to achieve this?

Perhaps the whole idea is insane, but imagine the possibilities if there were a way to do this.

[Mike]

Not to put away the premise of this thread entirely... but if what you are looking for is what the VIC-I is capable of when some (inadvertently designed-in) restrictions on the VIC-20 mainboard are lifted, you could always take a look at my VFLI mod.

Just to "quote" some pictures from a posting in there:

[lance.ewing]

Yeah, those pics are amazing alright. I'm in complete awe. Imagine what you could do if there were more than one? If such mixing is technically possible at a low cost, would the VIC 20 expansion I/O capilities allow it to talk to multiple VIC chips mapped at different addresses? The other VICs would need their own video matrix and colour RAM, all addressable by the 6502.

[Mike]

Well, I made a quick hack of my VFLI converter for one of the images above to see how a hypothetical dual VFLI VIC-I output might look:



I changed the input size of the converter to 208x512 but just duplicated the pixel rows of the input image, so the converter would apply conversion and error diffusion always on two identical rows. The resulting 208x512 output then was re-sized by combining each of the resulting double pixel rows into one (taking gamma into account, etc., blah, blah). Finally, the image was resampled in IrfanView (once again, with gamma) for the correct pixel aspect ratio.

In principle, the output can consist of all possible 136 mixing colours, and this gives access to a dark green, etc. Also it is possible to overlay hires 8x1 attribute cells (FLI!) of one VIC with multi-colour 4x1 attribute cells of the other VIC.

Hmm ...

[ral-clan]

This got me thinking: Has anyone done this before? i.e. tried to use the VIC chip outside of the VIC 20 for some purpose? Has anyone tried using it on a breadboard?

I believe there was one arcade game released in the 1970s that used the VIC chip. Commodore's goal was to have the VIC chip used in many more, but that never panned out, and so we got the VIC-20 home computer that used all the left-over VIC chips. See link:
http://sleepingelephant.com/ipw-web/bul ... php?t=2236

[Boray]

I think I have read or heard somewhere that the word about that there was a lot of leftover chips is myth. Commodore/mos had the chip design over as such, and it was actually refined for the vic-20. I can't recall where I saw this.

[mrr19121970]

This can't be true. Maybe for the 1981 manufactured chips. The 'newest' I have are from 1984 which is some 3 years after the VIC-20 was on the market.

[Boray]

What do you mean? I mean the myth that they had a lot of unused vic chips in store before designing the vic-20.

[Mike]

Commodore actually was sitting on a *huge* stack of 2114 SRAMs, so Tramiel ordered the engineers to put as many RAM onto the VIC-20 mainboard as they deemed fit to reduce that stack in a sensible time.

The data sheet of the 6560/6561 VIC-I as found on zimmers.net dates from 2/80, well into the design phase of the VIC-20 (which was on market January 1981). Wikipedia tells something about improvements of the 6562 (sound generator) and 6564 (more colours) being fed back into the original design of 1977, before production of the VIC-I began.

The last units of the VIC-20 were sold in 1985, so it makes sense they stopped production of the VIC-I in 1984.

[ral-clan]

This can't be true. Maybe for the 1981 manufactured chips. The 'newest' I have are from 1984 which is some 3 years after the VIC-20 was on the market.

What do you mean by this? Are you saying Boray's post is not true or my post?

Seeing the manufacturing dates on the MOS 6562, it does make sense that they merely updated the unused/unsold design for use in the VIC-20 rather than that they had a huge stockpile of these.

[mrr19121970]

I meant yours, but on 2nd reading I guess I misunderstood what you wanted to say. I thought you are saying even vic20 every made was filled by the overstock of chips meant for arcade machines.

[Kakemoms]

Commodore actually was sitting on a *huge* stack of 2114 SRAMs, so Tramiel ordered the engineers to put as many RAM onto the VIC-20 mainboard as they deemed fit to reduce that stack in a sensible time.

The data sheet of the 6560/6561 VIC-I as found on zimmers.net dates from 2/80, well into the design phase of the VIC-20 (which was on market January 1981). Wikipedia tells something about improvements of the 6562 (sound generator) and 6564 (more colours) being fed back into the original design of 1977, before production of the VIC-I began.

The last units of the VIC-20 were sold in 1985, so it makes sense they stopped production of the VIC-I in 1984.

Yannes started the "MicroPET" aka the Vic-20 in May 1980, so nothing was on the drawingboard in early 1980.

[Mike]

Yannes started the "MicroPET" aka the Vic-20 in May 1980, so nothing was on the drawingboard in early 1980.

Of course YOU sat right beside Yannes, when that happened.

[Kakemoms]

Yannes started the "MicroPET" aka the Vic-20 in May 1980, so nothing was on the drawingboard in early 1980.

Of course YOU sat right beside Yannes, when that happened.

Well I am old, but not THAT old! Haha.

But seriously, do you have some inside information there?