Atari 130XE Remake

[santosp]

Hi David,

Missing chroma connection? I think that this is happen in some models in XL line, not in XE.

 

Buying one of my new boards only makes sense if you have a Atari XE "as an components donor". You will have to transfer from there all the logic ic's, custom ic's, connectors, switch, etc. I have a list of all the passive components needed to buy as to assemble the new board, based on which you will have to make a list of materials, from any supplier do you prefer. I provide this to everyone who buys a new board. 

 

A new pre order thread have started here.

 

Edited April 24, 2021 by santosp

[Atari_bbs_fan]

Hmmm.. I swear I've read that the 130XE also had the "disconnected chroma" signal to the connector issue but you probably know far better than I. 

 

As to the "component donor" board, I get it.  I only have one working 130XE (my original one from years ago) so I wouldn't want to part that out.  Since I find all this fun, maybe I can find some dead 130xes on eBay to fill the need.  Maybe people here on the list have some better recommendations and even might be willing to sell some of their dead old stock so I can start my own dead old stock.  :-)

 

--David

[+mytek]

2 hours ago, Atari_bbs_fan said:

As to the "component donor" board, I get it.  I only have one working 130XE (my original one from years ago) so I wouldn't want to part that out.  Since I find all this fun, maybe I can find some dead 130xes on eBay to fill the need.  Maybe people here on the list have some better recommendations and even might be willing to sell some of their dead old stock so I can start my own dead old stock.  ?

Hear you go: http://best-electronics-ca.com/custom-i.htm

 

Since you will need all the Atari VSLI chips, this will easily meet the minimum order requirements, while not being too much for Brad to deal with.

 

[BillC]

On 4/24/2021 at 1:03 PM, Atari_bbs_fan said:

is there anything else missing that would need to be scrounged up from an old board?

I think that the cartridge/ECI connectors would be easier to re-use than get replacements for, same for the other external connectors(power switch and port/SIO port/joystick ports), the keyboard connector could also be re-used.

[Wilheim]

On 4/24/2021 at 4:17 PM, santosp said:

Hi David,

Missing chroma connection? I think that this is happen in some models in XL line, not in XE.

 

Buying one of my new boards only makes sense if you have a Atari XE "as an components donor". You will have to transfer from there all the logic ic's, custom ic's, connectors, switch, etc. I have a list of all the passive components needed to buy as to assemble the new board, based on which you will have to make a list of materials, from any supplier do you prefer. I provide this to everyone who buys a new board. 

 

A new pre order thread have started here.

 

I found some 65xe without chroma signal. To fix that, you have to add a 6k8 resistor at R201.

[santosp]

1 hour ago, Wilheim said:

I found some 65xe without chroma signal. To fix that, you have to add a 6k8 resistor at R201.

In this case is missing the component and not the implementation in the printed circuit.  This maybe happen by mistake in assembly line.

[Wilheim]

On 4/27/2021 at 12:23 PM, santosp said:

In this case is missing the component and not the implementation in the printed circuit.  This maybe happen by mistake in assembly line.

Maybe is a mistake, maybe not. Anyway, what I mentioned is something I looked in some not rare cases.

Edited April 29, 2021 by Wilheim

[Wildstar]

On 2/7/2021 at 3:02 PM, BillC said:

There are 65C816 adapters that add the extra circuitry to support the HALT signal, which I believe is the only significant difference between the 6502B and 6502C, this circuit was used on the early 6502 CPU boards for the 400/800. This circuit would allow using a 6502B instead of the 6502C/Sally, the WDC W65C02S should also work and is still being manufactured, and allow a possible upgrade path to the WDC W65C816S. Both the 6502B and W65C02S are available much less expensively than the 6502C.

 

I have seen a youtube video where they unsuccessfully tested a 65C816 adapter with a 130XE, they attributed the failure to the FREDDIE chip since the developer had it working on a non-FREDDIE machine.

The 6502C is NMOS and was manufactured by Commodore / MOS. It has the so called "illegal opcodes" and so forth. The 6502C is closer to the 6510 that is in the Commodore 64 than the WDC 65c02 parts found in the Apple IIe and the 8 bit mode of the 65c816 that was used in the Super Nintendo. Now, some can say we are splitting hairs but demos like used in the Commodore 64/128 used the "illegal opcodes" that were removed in the CMOS parts. These changes may cause programs that used those opcodes to behave differently and possibly crash/lock up. Therefore, it is nice to always have a fully compatible 6502C which is probably very similar to the 6502, 6502A, and 6502B. They may even behave like the 7501 (HMOS-1) in the Commodore Plus/4. These are MOS/CSG parts. Now, the currently produced 65c816 is nice for accelerators like the SuperCPU for C64/128 and on an Atari to benefit from the 16-bit operations and other benefits of the 65c816. However, you'll still want a 6502C or 6502 as found in the PET on an FPGA (which they do exist) that can be put into a drop in replacement. It can even be properly mapped to the cpu "socket" pinout. I know, you may actually direct solder to the PCB.

 

One potential benefit of an FPGA solution is you might even possibly it running at 100+ Mhz. Possibly even 100x faster clock frequency to that of the normal ~1.8 Mhz clock rate. So more room to do math and other number crunching function that you might not be needing or doing at the normal bus operations speed. 

[ijor]

On 5/3/2021 at 5:21 AM, Wildstar said:

The 6502C is NMOS and was manufactured by Commodore / MOS.

Just for the record because this doesn't alter your main point. But this is not accurate. Sally was not, or at least not only, manufactured by MOS. There are Sally chips made by Syntertek, Rockwell and others. I don't recall seeing one actually made by MOS/CSG. I think Atari intentionally avoided ordering chips from CSG, but somebody else might confirm if there was no MOS Sally at all.

 

I don't recall if it is know exactly which company designed (designing is not the same as manufacturing) Sally, although it is a very minor and almost trivial modification to the original 6502 NMOS design.

Edited May 4, 2021 by ijor

[Wildstar]

20 minutes ago, ijor said:

Just for the record because this doesn't alter your main point. But this is not accurate. Sally was not, or at least not only, manufactured by MOS. There are Sally chips made by Syntertek, Rockwell and others. I don't recall seeing one actually made by MOS/CSG. I think Atari intentionally avoided ordering chips from CSG, but somebody else might confirm if there was no MOS Sally at all.

 

I don't recall if it is know exactly which company designed (designing is not the same as manufacturing) Sally, although it is a very minor and almost trivial modification the original 6502 NMOS design.

Ok... you're right.... in terms of foundries manufacturing. Syntertek, Rockwell, and others were licensees of MOS/CSG which designed the 6502 core. This is not unusual for custom microcontrollers, SoCs, etc. where a licensee uses the IP of (in this case..... MOS/CSG) and added the the additional things like maybe the HALT pin or whatever). Rockwell and others later, also licensed the CMOS 65c02 and 65c816 from WDC. Around 1985 and afterwards, there was a shift away from MOS/CSG to WDC but it wasn't really until Commodore's bankruptcy that these companies ended their NMOS line. Now, of course, any of these companies and/or their successors (if they have the equipment) can manufacture the NMOS based 6502s because patents on them expired and they can do this without paying royalties. However, many have moved over to CMOS process because is other factors which would be beyond the point of this topic. At some point, I think Atari went through licensees because the supply of parts directly from Commodore (MOS / CSG) might have become expensive at some point. Rockwell and Syntertek was a "second-sourced" for various 65xx chips. The Sally (C014806) was a custom 6502 chip where they added the HALT line and second R/W line so it worked with the video circuitry but these changes were likely done by the licensees but they are not officially marked 6502C. The Sally probably is a customized 6502C with the mentioned modifications to work with the ANTIC chip (reference to the "liz" project) versus the standard 6502C that is made by Commodore. MOS/CSG 6502C is clockable to 4 MHz while the 6502B (found in some earlier model Atari 8-bits) was clockable to 2 Mhz. The SALLY (which we should call it other than 6502C because there are actual MOS/CSG chips called 6502C), is most likely manufactured by licensees. Not sure the story behind the design. 

 

It was likely custom designed for Atari by MOS/CSG under contract by Atari and tested by their engineers and MOS/CSG & MOS's "second source" licensees manufactured it for Atari. Atari didn't do the actual manufacturing of the chip and because the core itself is MOS's IP, this custom derivative was likely designed by MOS/CSG for some amount of money and MOS produced some of them and so did MOS's licensees Rockwell & Syntertek did as well to supply the demand need of Atari. By 1981, Commodore's MOS Technologies (later CSG) had a already a big draw on their foundry for not only the products made by Commodore but also others and so they already had in place licensees that served as second source foundries for the 6502 IP which Commodore received royalties from including the Sally which Commodore received royalties from it. 

 

 

[ivop]

1 hour ago, Wildstar said:

reference to the "liz" project

Darn, there already was a Liz Project? That would have been the name of choice for my next hardware project

 

About the C, MOS indeed had 6502 A/B/C which meant 1/2/4MHZ. The 6502C version never really got any foothold. With "our" 6502C, i.e. Sally, C means Custom. It's a 6502B with extra transistors for the /HALT line. The early Atari 400/800 models used a 6502B, with the HALT circuitry external to the CPU.

Edited May 4, 2021 by ivop

[santosp]

As I had promised in the past I did a alternate dram module for the base 64K memory, using 8 dram ic.

You can order here. Note that I have not test it.

[Mathy]

Hello Panos, guys

 

How difficult would it be (for the user) to add an extra 512kB to the motherboard?  The old fashioned way, not via U1MB!

Is there enough room, height wise?

 

Sincerely

 

Mathy

 

[wildstar87]

10 hours ago, Mathy said:

Hello Panos, guys

 

How difficult would it be (for the user) to add an extra 512kB to the motherboard?  The old fashioned way, not via U1MB!

Is there enough room, height wise?

 

Sincerely

 

Mathy

 

Are you talking about an additional 512k above with the Hias expansion on board does?  Otherwise there is that way to expand 512k on board.

[Stephen J Carden]

On 4/27/2021 at 12:23 PM, santosp said:

I would like one as well

 

Stephen J. Carden.

[Mathy]

Hello wildstar87

 

7 hours ago, wildstar87 said:

Are you talking about an additional 512k above with the Hias expansion on board does?

 

Yes

 

Sincerely

 

Mathy

 

[wildstar87]

Ah ok, yeah I have no idea. lol

 

I think Hias mentioned that the current GAL didn't have any more i/o pins or something like that.  Just talking out my ass here, but maybe you could add in another Hias upgrade and link it somehow?

[+MrFish]

On 4/22/2021 at 1:35 PM, santosp said:

The test period ended with success. All stages work properly!  Now I can ship the rest of boards in any interested member.  I have 6 more pieces available. 

Anyone who needs one, let me know in a personal message. 

 

Received my unpopulated (except for UAV smd's) board this morning in Ohio. Took 19 days to arrive from Greece.

 

Edited May 13, 2021 by MrFish

[santosp]

40 minutes ago, MrFish said:

 Took 19 days to arrive from Greece.

From the day of shipping needed 6 days to arrive in US. All the rest of time, was spent for the transit  inside your country. ?

I think this have to do because of Covid-19 limitations.

[+MrFish]

23 minutes ago, santosp said:

From the day of shipping needed 6 days to arrive in US. All the rest of time, was spent for the transit  inside your country. ?

I think this have to do because of Covid-19 limitations.

Yes, mostly held up here in the U.S.

 

No complaints. It's a little longer than I expected, but I was in no rush. I just posted the info so others can have some idea about when theirs might arrive.

 

[+ZuluGula]

@santosp, from what I see with tracking number, my package also got delivered today. I'm still at work so I haven't see it, but I suspect that might be some issue with the way you addressed it or your post office software did it. It took the package six days to arrive to USA, Chicago Illinois post office which is about ~25 miles away from my home. There were no updates for next nine days and then the package showed up in Queens, NY (about 1000 miles away). When I saw this, I contacted USPS and I was told that this package has wrong US address format and they had some problem with delivering it. They told me to wait. From there, it bounced between couple more pots offices in my area and I suppose, it's at my home now. I will check what's up with the address and will let you know later today.

[santosp]

@ZuluGula

I used the info from the exact point you sent me as address in the Google maps. You are the only one who use this way to send me his postal address, but I guessed that was right! ? At this time at the tracking system look that your parcel received by someone individual in your home.

[+ZuluGula]

Yes, my wife probably signed for it. I send the Google map link because I moved to new place and street name is a little complicated. But from what I can see, others also had to wait longer for their package, maybe @MrFishcan tell us what was going on with his parcel since it arrived in US?

[santosp]

From what I can see, using your tracking system, nothing unusual except that it was remains at Chicago for 10 days! For me isn't strange because we are in pandemic.

I hope that this is understandable from all.

 

USPS ALERT

Edited May 13, 2021 by santosp

[+ZuluGula]

I'm not complaining to you. 

I contacted USPS after they send the package from Chicago, Illinois to New York state, which is about 800 miles away. The person on the phone told me that there's an issue withe the way the package was addressed.