FPGA replacement for VIC I chip?

[tlr]

To fit it in a reasonable amount of space (like the CPLD mentioned) you need to know enough to deduce most of the actual silicon implementation, not just the observable behaviour.

Really? I'd be interested to know why. If anything I would have thought that simulating the observable behaviour would be better, since you could take shortcuts that work with FPGA.

This is because the original implementation was most likely optimized to fit the most functionality possible in the smallest space possible. It's very likely that many observed phenomena has a common cause, i.e they are artifacts of what strategy was used to implement the intended behaviour of the original circuit.

If we try to implement what we can observe from our tests, observation for observation, the implementation will be a big mess of special cases.
If we try to deduce the actual original implementation, chances are that phenomena present in the original design that we haven't observed, will actually be present in our implementation as well.

The latter was occasionally the case when we implemented x64sc over x64.

[Kweepa]

Ah, I get where you're coming from. Thanks.

[groepaz]

If we try to implement what we can observe from our tests, observation for observation, the implementation will be a big mess of special cases.
If we try to deduce the actual original implementation, chances are that phenomena present in the original design that we haven't observed, will actually be present in our implementation as well.

this was/is also one of the main rules for how we work on the fpga core of the chameleon - it actually makes things simpler at the end, and the big advantage is that even yet undiscovered strange behavior will likely work.

[pjwhoopie]

Is the only replacement option for NTSC VIC I chip to salvage another unit?

[ral-clan]

Is the only replacement option for NTSC VIC I chip to salvage another unit?

No. As said earlier in this thread, there are quite a few sellers sellers on Alibaba.com (or perhaps AliExpress - although I couldn't find any there) that seem to have bulk quantities of new-old-stock VIC-I chips left over from the Commodore days (many of Commodore's factories were in Asia).

However, I find Alibaba, and even AliExpress a little strange to deal with and I'm not entirely comfortable risking my money there. Figuring out import fees, for instance is strange. You need to be a commercial importer to understand the intricacies.

http://www.alibaba.com/product-detail/- ... 0.0.GAAUTt

http://www.alibaba.com/product-detail/- ... 0.0.ZzGwMO

Perhaps there are others here on Denial who already have have bought from there and would be willing to advise.

It would be nice to be able to find a reputable source that individuals can purchase from in Asia as some of the sellers seem to have hundreds of these chips in stock.

[pjwhoopie]

Thanks. If anyone has bought some old Commodore chips, that would be great to hear.

[mrr19121970]

They have 'new old stock' here too, but at these prices you really have to love your vic to want to pay them.

As I mentioned recently to @eslapion I am not upto 13 VIC20s. I buy them generally for spares and repairs, but after seeing how badly they are mistreated I feel the obligation to treat them like injured animals. I buy repair parts (at top dollar) and get them like new again. Now I have the issue I've spent so much time and money on them I can't/won't be able to part with them. Soon I think I'll need a Greyhound bus too for storage, right @ken??

[Muzz73]

I bought 20 NTSC VIC chips (6560-101) from a surplus supplier in China a few years back - all were NOS and had never even been put into a socket before.

I believe that 17 were good and 3 duds, but I got them for like $2.80/ea. IIRC.

You can find them, if you peek & poke (not PEEK & POKE) around in the right places...

[eslapion]

This device has a VIC-20 core so there is obviously a VHDL model for the 6561.
http://www.fpgaarcade.com/replay/

However I suppose the output is RGB.

[groepaz]

there are at least 3 cores, the one that runs on MIST, the one that runs on Chameleon, and the one that runs on FPGA-Arcade.

however, none of them is accurate enough to serve as a source for replacing ICs, nor are they developed to work like that.

[eslapion]

... none of them is accurate enough to serve as a source for replacing ICs, nor are they developed to work like that.

I suspected as much.

Add to that the fact a real drop-in replacement would require both digital and analog technology. I guess quite more complicated than SwinSID.

I'm afraid for now the price tag you saw is the right one.

[mrr19121970]

I bought 20 NTSC VIC chips (6560-101) from a surplus supplier in China a few years back - all were NOS and had never even been put into a socket before.

I believe that 17 were good and 3 duds, but I got them for like $2.80/ea. IIRC.

You can find them, if you peek & poke (not PEEK & POKE) around in the right places...

This is a super price. The only PAL 6561-101s I've found they want:

A) 11 us$ each + 18 us$ postage + 7 us$ PayPal fees + import duty.
B) 9.5 us$ each + 35 us$ postage + import duty.

And then you don't know if you get the chips or if they work (dead or fakes). Thankfully PayPay should refund in an issue.

[groepaz]

Add to that the fact a real drop-in replacement would require both digital and analog technology. I guess quite more complicated than SwinSID.

no and no. its quite doable - and the VIC is actually a simple chip more or less.

[mrr19121970]

I will take some if you cab do it...

[groepaz]

i doubt that will happen in the near future, they'd still cost a lot more than a real IC when it is feasible, we'll do it i guess (although starting with VIA or CIA, because these are easier to get 100% right)