Atari Ethernet Adapter / Blank Cart

[mistrtonee]

Hello All,

 

I'm considering a hefty undertaking as an electronics project to do on the side as I am an electronics geek. I'm very familiar with the Atari 2600 and its inner workings and think I could pull this off.

 

A blank cartridge, which plugs into the Atari 2600 and has ethernet on the other side. This could be used to play ROMS from a server, either those downloaded, or for developers testing newly written code. The cool thing is that it's done via ethernet.

 

It would be possible that I could allow the cartridge to operate in two modes.. download mode, and server mode, so that the Atari 2600 could actually act as a server. I am not sure if I would really want to deal with mapper support, but it could be done also.

 

Depending on the features, this device would cost about $100 - $200 in materials to produce, but would be very cool.

 

My question to you all is, how many of you would buy one for, say $100 - $200, depending on the features. I may mass produce these if I get the initial prototype all ironed out. Please reply and let me know. With enough interest I may mass produce the final device. At some point I will create a mailing list with all the replies to this post if you are interested. Please indicate if you are interested in the mailing list on this reply also.

 

Thank you,

 

Tony Y. - Computer Engineer and Atari fan

[joshk]

I would be possibly interested following the features and game support. I missed out on a cuttle cart.

[Goochman]

$120 would be my max output for such a device though I would be interested.

 

Might be nice if cart had RAM on board to hold game in memory so PC with ROMS doesnt need to be on all the time to play a game

 

How will you handle Bankswitched ROMS and such, or will this be like the Supercharger stuff where only 4/6/8K ROMS will work?

[mistrtonee]

I could handle the bankswitching for most of the carts if I place a PLD or FPGA chip on the cartridge. Depending on the complexity, this can easily increase my cost by $30, plus it may make it difficult to fit everything in a nice small package.

 

I definitely see your point.. supporting only smaller non mapped carts presents a problem for gameplay despite making a nice developer tool. I was thinking of providing a massive amount of SRAM.. maybe 128KB or so, and placing a small amount of hardware at the upper address lines to allow customizable bankswitching schemes. There will be a microprocessor on board the cart, so if I'm lucky it will be fast enough to do everything in software. Basically I could provide bankswitching support for everything, but it's a matter of cost. Ultimately, the best compromise may be to support the common mappers and let the hackers out there support the less common ones.

 

The cart may be able to power itself off of the Atari 2600.. That is something I will have to determine. If not, it will need to have a separate plug-in and will maintain its program. It will definitely use RAM.

 

The coolest possibility is that these carts could be designed to switch between download mode and ethernet interface mode. Basically they could act like a NIC for the Atari.. you could develop new software for it that can talk to the internet. I'm not sure if I will design it to do that yet.. Unless it is something people would really want to use. It could open up a new era of multiplayer 2600 games of course :-)

 

I think $120 could be a reasonable price depending on the features. I won't be gouging anyone.. just trying to recover the cost of the materials and time I spend prototyping and hand soldering each device.

 

I will probably set up a group specifically relating to this project somewhere on the web if enough interest is shown. Stay tuned to this forum for now for updates.

 

Tony Y. - Computer Engineer and Atari Fan :-)

[ubersaurus]

Sounds like a cuttle cart with an ethernet cable instead of an audio one...still cool stuff though.

[Goochman]

Support for all the bankswitched roms is prob unrealistic - for some of th emost popular would be nice -

 

Ok,

Heres one more thing to throw at you - What are the chances of making it a PCMCIA slot vs an ethernet one?

 

Why?

What if I have a wireless network and want to hook my 2600 up that way vs a cabled ethernet one - You dont have to supply the card(to reduce costs)

 

Just something tho think about - Even a 'cabled' one would be great and I have CAT-5 jacks all over my house so no prob for me

[mistrtonee]

I can definitely see the merits of PCMCIA, especially in terms of cost. I think that a PCMCIA interface would provide a separate, equally functional, but very different product from what I intend to produce.

 

The ethernet functionality will be an integrated part of the cartridge. Part of the point is to allow the Atari to directly talk to a network without any intervention on the part of a laptop. Believe it or not, I can add ethernet functionality for less than you'd pay for an old clunker of a 486 laptop. I'm looking at $50 for all the ethernet stuff.. and just some misc. simple circuits to tie everything together. I may release an 802.11b cart later should one of my key suppliers for the project develop new hardware to easily allow me to support that. I have emailed back and forth with these particular suppliers and 802.11b is something in their product pipeline.. it's just a matter of time. I figure I should develop this version now and maybe the 802.11b is something that could be done later. Until then, I think the prospect of some elite h4x0r out there creating an Atari web server is pretty cool :-) Granted of course, the supported functionality of the device will be to download code to the atari via ethernet.

 

Keep sending ideas if you have any.. part of my hope is to get some brainstorming going as well so this project can be as useful as possible to everyone. If enough people like the PCMCIA idea, maybe it's something I can look into and work on in parallel or later on.

 

Tony Y. - Computer Engineer and Atari Fan :-)

[+Kaboomer]

I'd like to buy this cart in a heartbeat! Put me on the list!

[oesii]

I think your price of about $120 for something like you are proposing sounds good. I could easily stretch an ethernet cable to my atari but I would also love something to work with my wireless Airport network.

 

So that's one more vote for a wireless option.

[+Kaboomer]

Oh yeah....I'm just about to get a wireless PCMCIA card for my computer now that I'm getting a WiFi switch in anticipation of a network to add my online PS2 capability so I'd prefer the wireless cart also!

[krisjohn]

You should provide a piggy-back socket at the top so people can plug in a Compumate keyboard. Include Telnet/SSH on the board and your Atari 2600 could be a terminal for your Linux box. Lynx or Elm on a 2600, yum.

 

Wireless sounds funky too -- I should have 802.11b running some time soon.

[mos6507]

The problem with 2600 I/O peripherals is a lack of available CPU time.

 

In order to make this valuable as an in-game peripheral the interface has to really offload the CPU almost completely. That means a really large FIFO buffering system and maybe two reserved bytes of address space. Every time the 2600 reads from one of the address locations, the FIFO auto-increments. Every time it does a "fake" write to the memory location (ala Starpath RAM) and it pushes the byte onto the outgoing FIFO buffer. Something like that where at no time is the 2600 stuck not being able to read or poll the interface fast enough. Remember that there are no interrupts on the Atari. It has to get to the data when it feels like it.

[mistrtonee]

I see many of you are interested in wireless connectivity. Unfortunately that is something I am unable to offer right now given my resources. What I would be most likely to develop is:

 

- Ethernet enabled atari cartridge with RJ-45 connector

- Simple interface for assigning IP address

(worst case is a bunch of dip switches. I'm thinking more along

the lines of a simple atari app on the cart that runs at startup and

serves as a config utility)

- Powered by:

- self powered by Atari OR

- worst case: powered by extra plug-in transformer

- 128 or 256K of ram on board for storing downloaded programs while

running

- simple mapper hardware support of some sort.. at least for the basic

mapping schemes used in the majority of carts

 

The cost for parts is something like this:

 

$50 - ethernet functionality (including ethernet chip and microcontroller)

$10 - SRAM for storing downloaded ROM

$30 - hardware for mapper support

$10 - Old cart, etc.. to be cannibalized to make device

 

The mapper support hardware might be a little cheaper, but everything else is my cost up above. It will probably cost around $100 to produce in small quantities and once perfected, it will take me a few hours to make each one.

 

I think $120 could be reasonable if I can figure out ways to keep the costs down. There is no forseeable way I could get 802.11b to work practically and inexpensively right now. However, given what I am prepared to design and build, you'd have to admit it could be very cool.

 

If you're interested let me know if it's worth $120 - $150 (max) to you.. I need to see that people will actually use this thing if I make it. Otherwise I'll just hack one together for myself rather than mass produce it. I would really like to provide these to people who would be interested if I get it to work.

 

Tony Y. - Computer Engineer and Atari Fan :-)

[Goochman]

Based upon part costs I would suggest selling it for $150 - your time investment is worth something!

 

for $150 based upon the description I would be interested

[CrazyImpmon]

$10 for a donor cart? What are you buying... Gremlins? I have about 20 or 30 Atari carts that are missing the ROMs I could just ship over. I have all the screws to hold the cart together. All you'd need is to delabel them all. (A bottle of Goo Gone is only like $5 at most stores)

 

Put me down for interested as I missed out the Cuttle Cart and I already have ethernet network in my house including a spare cable in my bedroom where my 2600 is. (FWIW the network is also connected to the internet via a gateway PC so if a 2600 web browser can be made I could surf the web with clunky graphic and such. )

[Andrew]

Even though I have a Cuttle Cart, I'd be game for this as well.

[Skywalker]

If you get your prototype to work ..... count me in.....

[+Kaboomer]

Count me in...even w/o the wireless capability!

[mvigor]

Add me to the pre-order list!

[Rob Mitchell]

Gee .. All I want is a "cuttle cart" with all game compatibility .. and a RS232 port .. for connection to my portable HP200LX DOS palmtop. On this device code can be edited, compiled and transfered over to the Atari.

 

Ethernet? That still keeps the Atari tied down ..

 

Rob Mitchell, Atlanta, GA

[Ushabin]

Network connectivity for the 2600 sounds awsome! Add me to the "I'll buy that" list!

[CrazyImpmon]

@Rob-

 

Actually ethernet is less restricted than serial. With serial, you're limited in speed and you can only go a short distance. With ethernet, you could go 10m/sec or faster and there exists a 100' long cable so your computer and the 2600 doesn't even have to be in the same house. Just IMO

[Rob Mitchell]

@Rob-

 

Actually ethernet is less restricted than serial.  With serial, you're limited in speed and you can only go a short distance.  With ethernet, you could go 10m/sec or faster and there exists a 100' long cable so your computer and the 2600 doesn't even have to be in the same house.  Just IMO

 

But speed is not an issue with the small file size of Atari BINs .. and I have all these Atari ROMs in my DOS palmtop and no way to load them up in my Cuttle Cart.

 

Rob Mitchell, Atlanta, GA

[Goochman]

You could put a wireless hub near the 2600 and then use whatever machie in the house that has wireless access.

 

Serial is too much of a pain IMHO - Its really best for DOS based devices. I havent found many good serial tranfser programs that work well under Windoze where most of the users of this prob will want to run.

 

If you already have a cuttle cart get an iPOD to transfer your 2600bin 'music'

[NoahsMyBro]

I'm definitely interested, but $120 is the high end of what I could afford to spend on this. That's not to say I wouldn't spend the $120, but I'd have to think about it, and I couldn't go higher.

 

Having said that, I didn't get a Cuttle Cart and would like to get one of these, as it sounds like it would support that functionality, and then some.

 

Wireless is a non-issue for me. I can run ethernet anywhere I need/want to, I currently have no wireless hardware, and I don't have any current intentions of getting wireless. It's very cool, but not secure enough/mature enough yet for me.

 

Just my two cents,

Steve