"2600 Plus" My modernized 2600 console project

[john_q_atari]

49 minutes ago, Atari2600PAL said:

Any possibility of buying a PCB and shipping to the UK please?

Sure. Keep in mind this is a NTSC design. I use the United States Postal Service. Please send me a PM and we can go from there.

[+Atari2600PAL]

9 hours ago, john_q_atari said:

Sure. Keep in mind this is a NTSC design. I use the United States Postal Service. Please send me a PM and we can go from there.

I hadn't considered the NTSC vs PAL issue!

 

Not very technically minded in this area, so will leave it for now. Thanks though

[grayescu]

If this could be made for cheap, it would probably sell really well in developed countries. 

[+Yurkie]

@john_q_atariwould you consider a go fund me to get a new unicorn 2600 chip manufactured?

 

Your new pcb redesigned for a new unicorn chip would be an entirely new 2600 and a great thing for the video game world.

[+Yurkie]

Multi-post

[+Yurkie]

Multi-post

[john_q_atari]

6 hours ago, Yurkie said:

@john_q_atariwould you consider a go fund me to get a new unicorn 2600 chip manufactured?

No I wouldn't want to take anyone's money. If there was an idea I wanted to take to production I would either do it with my own funds or make the business case to investors on why funding the effort makes good business sense. 

6 hours ago, Yurkie said:

Your new pcb redesigned for a new unicorn chip would be an entirely new 2600 and a great thing for the video game world.

Manufacturing a new TIA chip or a new "unicorn" TIA+6502+RIOT chip would require a large amount of time, money, and effort. It reminds me a little of the Flashback 2 ASIC so perhaps not entirely new. I don't think the market is big enough at a pricepoint that would be economically feasible to support the creation of such a chip.

[WhyLee commotari.club]

14 hours ago, john_q_atari said:

No I wouldn't want to take anyone's money. If there was an idea I wanted to take to production I would either do it with my own funds or make the business case to investors on why funding the effort makes good business sense. 

Manufacturing a new TIA chip or a new "unicorn" TIA+6502+RIOT chip would require a large amount of time, money, and effort. It reminds me a little of the Flashback 2 ASIC so perhaps not entirely new. I don't think the market is big enough at a pricepoint that would be economically feasible to support the creation of such a chip.

You are right. But a new unicorn chip would be the worst decision in my eyes.
I would like to have a graphic chip that is compatible with the TIA regarding the bus connection but with more features and a dram interface and a hdmi interface.
this way everyone could use this chip for upgrades. and maybe use a real 6502 and not the 6507 etc.

at the end we would need an fpga version first that would be developed in a joint effort and at some point a feature freeze. needs to have different resolutions in 4:3, 16:9, x/y size ratios that fit a 1920*1080 screen etc. maybe more layers, with a programmable charset or tiles for games, sprites,.......

but i have to say. if you look at ebay regarding prices, you will see that people pay a lot of money to have the old wouldgrain case. i think this says a lot. the pcb should be compatible to the old case. or someone makes a new case with cutouts for hdmi and other ports (old disk drives, tape drive, etc).
a lot of people want the 2600 six switch woodgrain as a kind of hipster decoration. there is no way around to keep that.

[azure]

On 2/5/2023 at 10:35 AM, Keatah said:

Despite what people claim (and like to think) FPGA isn't an exact clone of the original console's circuitry. It behaves differently, especially at power-on time. And collectorvision and analogue are using FPGA.

 

So you either get original chips. Or use ones that are "close enough".

I recently watched a lecture by USC professor Anthyony Levi on developments in non-destructive 3D x-ray scanning of microprocessors. His plan is to image modern chips so that their circuitry can be validated to do what they purport to do and checked for hidden backdoors. Once this technique becomes more commonplace and accessible to the mainstream it might become feasible to perfectly document the internal circuits of legacy chips. If it can be accurately documented and modeled in a hardware language, then it can be recreated. It would be significantly less labor intensive compared to sanding chips down and photographing layers. I think it'll become a boon for retro gamers and computer users, but it's still probably some time before regular people can get access to it, so it might be fantasy for the moment.

[john_q_atari]

.

Edited March 1, 2023 by john_q_atari

[Keatah]

1 hour ago, azure said:

I recently watched a lecture by USC professor Anthyony Levi on developments in non-destructive 3D x-ray scanning of microprocessors. His plan is to image modern chips so that their circuitry can be validated to do what they purport to do and checked for hidden backdoors. Once this technique becomes more commonplace and accessible to the mainstream it might become feasible to perfectly document the internal circuits of legacy chips.

Sure it would.

 

1 hour ago, azure said:

If it can be accurately documented and modeled in a hardware language, then it can be recreated. It would be significantly less labor intensive compared to sanding chips down and photographing layers. I think it'll become a boon for retro gamers and computer users, but it's still probably some time before regular people can get access to it, so it might be fantasy for the moment.

I'd guess another 6-8 years for this to become commonplace and accessible to hobbyists.

[+x=usr(1536)]

12 hours ago, WhyLee commotari.club said:

But a new unicorn chip would be the worst decision in my eyes.

I disagree.  Apart from the packaging advantages of a single-chip design, it's a fairly uncommon piece of Atari history and if it can be re-implemented (i.e., preserved) as a reproduction ASIC it should be.

 

This doesn't negate doing the same for the TIA, RIOT, 6507, GTIA, POKEY, etc., but 'we don't need this' is never a good perspective to have when it comes to this sort of thing.  All it ever results in is a lower quality of documentation and preservation.

[M-S]

 I don't think having a single chip would be the best option, it would be better to make it possible to have generic chips replacing the CPU and the RIOT chip, the only problem would be the TIA chip since there's nothing similar nowadays, although I think that the TIA is so simple that it would be possible to make a board that either makes it possible to use another 8-bit GPU instead of the TIA or replaces it completely. Something like the RGB mod or how people have managed to get Atari running on modiifed Apple computers.

 But I don't know if there really is a large public for that, I would be interested, nowadays we can build a NES from scratch with a new boards and chips, but most people just buy a dead NES or the original chips and install it.

 About it fitting on a woodgrain shell, I think this isn't the developer's objectivr, I know there's someone developing a board that fits, if anyone is interested.

Edited March 1, 2023 by M-S

[Keatah]

28 minutes ago, M-S said:

 I don't think having a single chip would be the best option, it would be better to make it possible to have generic chips replacing the CPU and the RIOT chip, the only problem would be the TIA chip since there's nothing similar nowadays, although I think that the TIA is so simple that it would be possible to make a board that either makes it possible to use another 8-bit GPU instead of the TIA or replaces it completely.

The TIA is rather intricately tied to the 6507 through the software and timing. Not sure any other then-contemporary chip could be made to work, not without circuitry and software mods becoming way more complex than just TIA + 6507 alone. But yes. It would be best to have separate parts.

 

28 minutes ago, M-S said:

Something like the RGB mod or how people have managed to get Atari running on modiifed Apple computers.

How would these Apple II systems have been modified to play 2600 games? That's something I want to see.

[WhyLee commotari.club]

6 hours ago, Keatah said:

The TIA is rather intricately tied to the 6507 through the software and timing. Not sure any other then-contemporary chip could be made to work, not without circuitry and software mods becoming way more complex than just TIA + 6507 alone. But yes. It would be best to have separate parts.

 

The 6507 is just a downgraded 6502. It really has a 6502 silicon inside with a different metal layer on top.
With a little bit of different wireing a 6502 cpu can be used and you instantly have the benefit of adressing 64k. that would allow to use additional RAM/ROM as has been done in the 7800.
I have to say that it is appealing to me to develop the 2600 a little bit in a similar direction like the 7800.
maybe not 100% maybe yes. i don't know if 7800 compatiility is desired.
for me, advancing the system is very appealing. like adding ram/rom/audio chip (in the from of a micro controller))

 

[+x=usr(1536)]

3 hours ago, WhyLee commotari.club said:

I have to say that it is appealing to me to develop the 2600 a little bit in a similar direction like the 7800.

Out of curiosity, what direction(s) do you see that going in?

3 hours ago, WhyLee commotari.club said:

maybe not 100% maybe yes. i don't know if 7800 compatiility is desired.

When you say, "7800 compatibility," are you referring to software compatibility or something else?

3 hours ago, WhyLee commotari.club said:

for me, advancing the system is very appealing. like adding ram/rom/audio chip (in the from of a micro controller))

Which I'm not arguing against, but whether or not that additional hardware is useful comes down to whether or not software makes use of it.

 

IMHO, the Harmony and Harmony Encore have been successful for three reasons: one, they're completely plug & play and don't require any additional time, money, and/or effort to install; two, they provide flashcart functionality in addition to beefier hardware; three, their design has created a decent-sized userbase for which writing software that requires the beefier hardware is viable.

[Keatah]

I'm happy enough with Harmony being the awesome expansion as it is. It passed the test of time and is now established and accepted seemingly by all 2600 gamers. And there are loads of tricks and things to do (on Harmony) that haven't been done yet.

 

To start over with another platform, with its own enhanced feature set, could potentially mean 5 more years while that something is fleshed out and developers learn the lay of the land. It would also introduce fragmentation and not likely receive ports of Harmony games anyways.

 

There comes a point where you enhance something so much that it is not the original thing. Harmony doesn't violate that principle.

[M-S]

23 hours ago, Keatah said:

The TIA is rather intricately tied to the 6507 through the software and timing. Not sure any other then-contemporary chip could be made to work, not without circuitry and software mods becoming way more complex than just TIA + 6507 alone. But yes. It would be best to have separate parts.

Probably it wouldn't be possible to use something like a NES GPU, but as far as I know the Atari RGB mod already replaces most of the TIA, it just needs it to receive some signals, it would probably take a lot of work to make something that could read it, the audio and inputs would be easy to read and the video would be harder, but I think it's the most plausible solution.

23 hours ago, Keatah said:

How would these Apple II systems have been modified to play 2600 games? That's something I want to see.

I really don't know the specifics of it, there's a homebrew computer based on Apple called Baffa and there's a fork called Baffatari. Maybe it's actually using the original TIA.

 

[Keatah]

On 2/28/2023 at 8:53 PM, M-S said:

Something like the RGB mod or how people have managed to get Atari running on modiifed Apple computers.

That really isn't a modified Apple II at all. It's really a whole separate homebrew computer - and a fork of that on top of it.

 

Or more specifically it's just a bus. And the parts of an original Atari 2600 are plugged into it. Any electronics enthusiast can repackage a vintage console as electrically simple as a 2600.

 

Because it runs the Apple 1 Woz firmware doesn't make it an Apple II by any stretch of the imagination.

Edited March 2, 2023 by Keatah

[Koopa64]

Now that is a cool project, a brand new 2600 compatible PCB.

 

Also neat to have it be upgraded to a 6502 processor for extra addressing, like the 7800. I guess if one were just wanting the CPU and not add support for the MARIA chip, perhaps the result would instead be a sort of Super 2600.

[WhyLee commotari.club]

22 hours ago, x=usr(1536) said:

When you say, "7800 compatibility," are you referring to software compatibility or something else?

well, if someone wants to start with the 2600 hardware and wants to make one step or more further to a bigger/better 8-bit gaming console, you need a system for bank switching or managing the chips selects.
so that you can switch into a 2600 compatibility mode, where the cartridge is visible at multiple address ranges.
and switch back, so that the cartridge is only visible at one single address range. and the other address ranges are mapped to ram/rom/other chips/whatever.

so by designing such a bank switching, it would make sense to look at the 7800, how they did it. because they had the same problem when they tried to make a better 8-bit coinsole but staying compatible with the 2600.
if the mechanism that they implemented for the 7800 is a good one, then it would make sense to use this and at the same time you get a single piece of compatibility with the 7800.

after doing this, you could then look at ROM concepts. it is not difficult, you can look at the 7800 regarding the address space for the internal ROM.
then you could come up with a chip select / bank switching which still could be compatible to implement a ROM similar or the same like the 7800.

and so on and so on.

at some point when you want something that the 7800 does not have, you have to break compatibility in some way but could still check if there is a configuration possible that allows 7800 compatibility.
an example would be an advanced stereo audio chip. you would need an address space for it, that was not reserved in the 7800. so you need a kind of configuration where you enable the chip select for a certain address range or disable it to be in 2600/7800 mode again.

at the end you might be 80% compatible to the 7800 and just have to implement the graphic chip.
if clever designed, you could have two big pin headers on the PCB around the old TIA chip and you could pull out the TIA chip and put a graphic daughter board over it, that supports HDMI, emulated 2600 TIA and the 7800 graphic chip with added HDMI.

so, every add-on to the 2600 could be desined in a way that would be possible with the 7800 as well to not disrupt a possible upgrade path to the 7800.

good idea?

[+x=usr(1536)]

35 minutes ago, WhyLee commotari.club said:

good idea?

It's not that the ideas are bad, but rather that we've seen how hardware upgrades play out time and time again.

 

Everything depends on price, accessibility, desirability, and software support.  Price is obvious: keep it as low as possible.  Accessibility refers to how difficult performing the upgrades would be on a scale of 'Harmony Encore' to 'building a 1088XEL from scratch'; the more difficult the upgrade, the less broad appeal it would have.  And, finally, desirability and software support feed off of each other: you can have the world's most desirable upgrade, but if the software isn't there to support it, it's difficult for someone to justify the investment.

 

As an academic exercise, or something to tinker with?  Absolutely.  But it's difficult to see where the support would be for such a series of upgrades, particularly when the 5200 and 7800 already exist.

 

Like I said: the ideas aren't bad.  It's just not clear how well they would be accepted or used.

[+wongojack]

I would love to build this as a kit even if it meant I ordered the parts from Mouser myself.  Following thread.

[M-S]

I hope bumping this thread won't cause any confusion with the similarly named 2600+, maybe this should be renamed, but how is this project going? Have you already achieved your goals with it? Is it still only sold through individual orders?

[BIGHMW]

6 hours ago, M-S said:

I hope bumping this thread won't cause any confusion with the similarly named 2600+, maybe this should be renamed, but how is this project going? Have you already achieved your goals with it? Is it still only sold through individual orders?

I think you shouldn't have any problem with namesake as long as you use the word "plus" to describe your "2600 Plus" as opposed to using the plus sign like in the console being sold soon.