800XL PBI 5V fix?

[bfollowell]

I can't seem to find anything about the 800XL PBI 5V fix. Looking at the pinout, it appears that the 600XL has 5VDC ran to pins 47 & 48 of the PBI. So, is the fix just as simple as soldering a connection from a 5V source to those two pins of the PBI? Is there a best or ideal spot to pull that power from? It probably isn't needed all that much but is there any reason to not do this?

 

Thanks.

[+x=usr(1536)]

Not really, but is there a specific reason you'd want to add +5V to the PBI?

 

IIRC, the only reason that +5V was present on the 600XL was to provide power for the 1064 RAM expansion.  All other PBI devices were expected to be powered separately from the computer itself.

 

If you do plan on adding it, be sure to compare the 600XL and 800XL schematics if you haven't already.  It looks as though pins 47 & 48 are just tied together and fed from a common +5V source on the 600XL - easy enough to replicate on the 800XL.

[+MrFish]

12 minutes ago, x=usr(1536) said:

All other PBI devices were expected to be powered separately from the computer itself.

 

There are quite a few modern PBI devices that expect 5 volts from the PBI port. RAM 320XL and the Turbo Freezer 2011 would be two examples. RAM 320XL came with a cable with connectors for pulling 5 volts from a joystick port; but that's a rather cobbled looking solution compared to adding one small internal wire to the 800XL motherboard.

 

[bfollowell]

  

21 minutes ago, x=usr(1536) said:

Not really, but is there a specific reason you'd want to add +5V to the PBI?

 

 

Now that I think about it a little more, I guess not. I mean, I do now because my 800XL is dead in the water and I've been tinkering and troubleshooting a lot, then going back to check with my SysCheck card. Having 5V on the PBI connector would allow me to stop using the attached cable to steal power from joystick port #2. I'm worried I'm going to snap off the pin every time I connect it. Still, once I get my 800XL going again, I seriously doubt that I'll ever use the SysCheck, or any other PBI device on it.

 

4 minutes ago, MrFish said:

 

There are quite a few modern PBI devices that expect 5 volts from the PBI port. RAM 320XL and the Turbo Freezer 2011 would be two examples. RAM 320XL came with a cable with connectors for pulling 5 volts from a joystick port; but that's a rather cobbled looking solution compared to adding one small internal wire to the 800XL motherboard.

 

 

Yeah, but once I get my 800XL revived, I plan to install a U1M, so I won't need any PBI memory upgrades and I really have no need for the Turbo Freezer.

 

The more I think about it, I'll probably just leave things as they are and just be really careful when plugging and unplugging the little power cable on joystick port #2.

 

Edited November 13, 2020 by bfollowell

[Mathy]

Hello guys

 

1 minute ago, MrFish said:

RAM 320XL came with a cable with connectors for pulling 5 volts from a joystick port.

 

tf_hh's SysCheck also comes with a wire to pull 5VDC from the joystick port.  (Oops, bfollowell typed faster than I did.)

 

BTW if somebody comes up with an idea to make this cable easy to plug and unplug from the PCB (not the joystick port) please tell.  I prefer to take the 5VDC from the motherboard, but sometimes it's nice to be able to plug the wire back in to check out a "strange(rs)" XL.

 

Sincerely

 

Mathy

 

 

[bfollowell]

1 minute ago, Mathy said:

BTW if somebody comes up with an idea to make this cable easy to plug and unplug from the PCB (not the joystick port) please tell.  I prefer to take the 5VDC from the motherboard, but sometimes it's nice to be able to plug the wire back in to check out a "strange(rs)" XL.

 

 

That's a good point I hadn't thought of and probably and much easier fix than permanently modifying my 800XL. It would be easy enough to just use a jumper wire with alligator clips and connect one end to the end of the attached blue wire from my SysCheck and connect the other end to one of the 5V legs of the power connector.

 

[Mathy]

Hello bfollowell

 

I guess you didn't really understand what I was saying, but if it inspired you in some way, that's nice.  I'll upgrade both my 800XLs at some point in time to where they will have 5VDC on pins 47 and 48, so I don't need the extra wire (and don't need to somehow @#$% the connector on the end of it into the joystick port to then find out that I connected it to ground instead of 5VDC).  But there will always be situations where it would be nice to use the hardware on for instance somebody else's 800XL that doesn't have 5VDC on pins 47 and 48 of the PBI.

 

BTW did I mention that I don't like wires hanging from devices?

 

Sincerely

 

Mathy

 

 

[+MrFish]

I just opened up my 800XL and snapped a shot. The 5 volt mod is the yellow wire, obviously.

 

 

[+MrFish]

16 minutes ago, Mathy said:

BTW if somebody comes up with an idea to make this cable easy to plug and unplug from the PCB (not the joystick port) please tell.  I prefer to take the 5VDC from the motherboard, but sometimes it's nice to be able to plug the wire back in to check out a "strange(rs)" XL.

 

Yeah, I'm not saying the joystick connection for 5 volts isn't useful, I just prefer the internal mod. I often used the joystick connector when I would get a new 800XL and wanted to test the RAM 320XL out on it.

[+x=usr(1536)]

1 hour ago, MrFish said:

 

There are quite a few modern PBI devices that expect 5 volts from the PBI port. RAM 320XL and the Turbo Freezer 2011 would be two examples. RAM 320XL came with a cable with connectors for pulling 5 volts from a joystick port; but that's a rather cobbled looking solution compared to adding one small internal wire to the 800XL motherboard.

 

 

Sure, and I'm not in disagreement with you on the above points.  However, I was referring to the PBI as implemented by Atari, not modifications made to it after the machines were out the door.

[+MrFish]

2 minutes ago, x=usr(1536) said:

Sure, and I'm not in disagreement with you on the above points.  However, I was referring to the PBI as implemented by Atari, not modifications made to it after the machines were out the door.

 

Yeah, it's a kinda weird situation. It's obvious that it's useful to have it on all PBI's, because external power is often not needed/necessary (the Atari power supply provides enough power). Maybe it was a means of insuring that any PBI devices would have whatever power they needed, because they would be responsible for supplying it (externally).

 

[+x=usr(1536)]

3 minutes ago, MrFish said:

 

Yeah, it's a kinda weird situation. It's obvious that it's useful to have it on all PBI's, because external power is often not needed/necessary (the Atari power supply provides enough power). Maybe it was a means of insuring that any PBI devices would have whatever power they needed, because they would be responsible for supplying it (externally).

 

 

It never made sense to me either that Atari didn't keep +5V on all versions of the PBI, if for no other reason than consistency's sake.  Even the ECI on the XEs had it, IIRC.

 

About the only reasonable ideas I have on this are:

• Preventing 1064s from being plugged into and activated on 800XLs, then having to deal with the 'where's my 128K of RAM?' support fallout when they stayed 64K machines.
• Enforcing external power supplies to keep demands on system power low (i.e., not letting 3rd parties get lazy about where they pulled power from for their devices)

Really, though, there's just no way of knowing, and it's not like it's out of character for Atari to have made strange development decisions ?

[+MrFish]

16 minutes ago, x=usr(1536) said:

Even the ECI on the XEs had it, IIRC.

 

Yeah, thankfully they had the sense to include it on the XE's. Maybe because you couldn't plug a 1064 into it. Haha...

 

[+MrFish]

1 hour ago, MrFish said:

Maybe it was a means of insuring that any PBI devices would have whatever power they needed, because they would be responsible for supplying it (externally).

 

47 minutes ago, x=usr(1536) said:

• Enforcing external power supplies to keep demands on system power low (i.e., not letting 3rd parties get lazy about where they pulled power from for their devices)

 

This is basically what I was saying in the quote above. Some developer might have perfect success using their device powered on the PBI when it's the only add-on device doing so; but as soon as one or more other devices get plugged in expecting to receive power from the Atari, all bets are off. Devices could draw power from the machine on the SIO port too (unless you're an unmodded 1200XL), which adds another ingredient to the cauldron. Developers have no way of knowing how many other devices someone might have hanging off the machine, and could easily create devices that were completely void of consideration for what else might demand power from the machine (especially back in the "wild west" of computing 80's). So, by Atari forcing them to get their PBI devices powered elsewhere, it could avoid a lot of potential user frustration (and support frustration for Atari).

 

Edited November 14, 2020 by MrFish

[+MrFish]

15 minutes ago, MrFish said:

Devices could draw power from the machine on the SIO port too (unless you're an unmodded 1200XL)...

I should say: SIO, Cartridge, and Joystick ports.

 

[+kheller2]

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

 

It never made sense to me either that Atari didn't keep +5V on all versions of the PBI, if for no other reason than consistency's sake.  Even the ECI on the XEs had it, IIRC.

 

About the only reasonable ideas I have on this are:

• Preventing 1064s from being plugged into and activated on 800XLs, then having to deal with the 'where's my 128K of RAM?' support fallout when they stayed 64K machines.
• Enforcing external power supplies to keep demands on system power low (i.e., not letting 3rd parties get lazy about where they pulled power from for their devices)

Really, though, there's just no way of knowing, and it's not like it's out of character for Atari to have made strange development decisions ?

 1064s can’t plug into 800XLs, the opening is different. 

 

Perhaps Atari added the 5v line to 600XLs.  Easy enough to validate.  There are several documents on pre finalized PBI design notes.   

[+x=usr(1536)]

4 minutes ago, kheller2 said:

 1064s can’t plug into 800XLs, the opening is different.

 

IIRC, the opening is the same.  But plugging a 1064 into an 800XL is going to block all sorts of ports that won't be inconvenienced on a 600XL.

 

Quote

 

Perhaps Atari added the 5v line to 600XLs.  Easy enough to validate.  There are several documents on pre finalized PBI design notes.   

Yes - the 5V line was (as previously stated) specifically added to support the 1064.

Edited November 14, 2020 by x=usr(1536)

[+MrFish]

4 hours ago, kheller2 said:

 Perhaps Atari added the 5v line to 600XLs.  Easy enough to validate.  There are several documents on pre finalized PBI design notes.   

 

So, I guess the 1090 (where all upgrades would live) was the main goal of the PBI, and it couldn't be powered by a host machine, so no reason to have 5 volt power on the PBI. But the 600XL (economy machine) needed a simple (relatively cheap) memory upgrade, which required 5 volt power; and external power would have been rather silly. It all makes sense, if the 1090 gets released.

 

[+x=usr(1536)]

5 hours ago, MrFish said:

 

So, I guess the 1090 (where all upgrades would live) was the main goal of the PBI, and it couldn't be powered by a host machine, so no reason to have 5 volt power on the PBI. But the 600XL (economy machine) needed a simple (relatively cheap) memory upgrade, which required 5 volt power; and external power would have been rather silly. It all makes sense, if the 1090 gets released.

 

 

All of this makes perfect sense.

 

Then the XE range is launched, and +5V is present on the ECI.

 

I swear that I will never understand some of Atari's design decisions - it's not so much that they don't make sense in retrospect, it's that they didn't make sense at the time.

[+mytek]

6 hours ago, MrFish said:

So, I guess the 1090 (where all upgrades would live) was the main goal of the PBI, and it couldn't be powered by a host machine, so no reason to have 5 volt power on the PBI. But the 600XL (economy machine) needed a simple (relatively cheap) memory upgrade, which required 5 volt power; and external power would have been rather silly. It all makes sense, if the 1090 gets released.

But when looked at from today's viewpoint it was a short sighted decision, and presumes only one way of doing business - the 1090. From a financial point of view adding 5V to the 800XL PBI would have not changed the cost what so ever - so no momentary benefit to come from leaving it out.

 

The same mentality played into the decision to leave out the PBI on the 1200XL (which was present in the 1400 base design) - in order to save a few pennies in the short term, but to cut off the possibility to make more money after the fact by selling PBI peripherals to existing owners.

 

Although a bit off topic, I'm still amazed that Atari chose to essentially repackage the same hardware no less than 8 times, with but minor differences in capability (400, 800, 1200XL, 600XL, 800XL, 65XE, 130XE, XEGS). Who does that? And what a nightmare it has created for makers of upgrades, with nothing in the same place on the various motherboards. Crazy shit.

 

[+x=usr(1536)]

8 minutes ago, mytek said:

But when looked at from today's viewpoint it was a short sighted decision, and presumes only one way of doing business - the 1090. From a financial point of view adding 5V to the 800XL PBI would have not changed the cost what so ever - so no momentary benefit to come from leaving it out.

 

Agreed on both counts, but I suspect a couple of things were at play here:

1. Atari wasn't concerned with what the aftermarket did (or did not do) with the PBI
2. IMHO, Atari never had a solid plan for PBI-compatible peripherals

WRT the second point, sure, they had the 1090 in development, and absolutely should have got that to market.  But this is a company that has a history of sitting on potential goldmines (*cough*7800*cough*) and only seeing them for what they are, not what they could be.

 

8 minutes ago, mytek said:

The same mentality played into the decision to leave out the PBI on the 1200XL (which was present in the 1400 base design) - in order to save a few pennies in the short term, but to cut off the possibility to make more money after the fact by selling PBI peripherals to existing owners.

 

Along similar lines, I'm honestly surprised that the 600XL was brought to market in the form that it was.  Having an entry-level offering makes sense, but in 1983 16K of RAM on-board was...  Antiquated.  Most machines released around that timeframe had gone to 64K on-board by then, though there were some still being made with 32K or 48K out of the box.

 

Not to mention that with the 1064 being Atari's only PBI-based peripheral, it had a relatively small target market.  If it had been designed to also include 800XL compatibility, that could have seriously broadened its appeal.  But no +5V on the 800XL's PBI killed that possibility.

 

8 minutes ago, mytek said:

Although a bit off topic, I'm still amazed that Atari chose to essentially repackage the same hardware no less than 8 times, with but minor differences in capability (400, 800, 1200XL, 600XL, 800XL, 65XE, 130XE, XEGS). Who does that? And what a nightmare it has created for makers of upgrades, with nothing in the same place on the various motherboards. Crazy shit.

 

Yeah, Atari did a fine job of squeezing as much life as they could out of the A8 range.  13 years on the market with only minor changes is pretty impressive by any measure.

[+MrFish]

17 hours ago, MrFish said:

This is basically what I was saying in the quote above. Some developer might have perfect success using their device powered on the PBI when it's the only add-on device doing so; but as soon as one or more other devices get plugged in expecting to receive power from the Atari, all bets are off. Devices could draw power from the machine on the SIO port too (unless you're an unmodded 1200XL), which adds another ingredient to the cauldron. Developers have no way of knowing how many other devices someone might have hanging off the machine, and could easily create devices that were completely void of consideration for what else might demand power from the machine (especially back in the "wild west" of computing 80's). So, by Atari forcing them to get their PBI devices powered elsewhere, it could avoid a lot of potential user frustration (and support frustration for Atari).

 

3 hours ago, mytek said:

But when looked at from today's viewpoint it was a short sighted decision, and presumes only one way of doing business - the 1090. From a financial point of view adding 5V to the 800XL PBI would have not changed the cost what so ever - so no momentary benefit to come from leaving it out.

 

Considering the ideas in the post of mine that I'm quoting here, it wouldn't have been short-sighted. It wouldn't have left the 1090 as the only way of utilizing the PBI for upgrades either; it just dictated that devices pull power from elsewhere.

 

Do I agree with this philosophy or their decision? It's hard to say. We can't rewrite history and see how things would have turned out if the 1090 had been released. From the modern standpoint, 5 volts on the PBI makes perfect sense. But considering the remedy is so simple to implement, it's hardly a realistic point of concern -- more of an annoyance.

 

[+MrFish]

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

All of this makes perfect sense.

 

Then the XE range is launched, and +5V is present on the ECI.

 

I swear that I will never understand some of Atari's design decisions - it's not so much that they don't make sense in retrospect, it's that they didn't make sense at the time.

 

The thing about the XE range is that we're dealing with a completely different company by that time.

 

[+x=usr(1536)]

4 minutes ago, MrFish said:

 

The thing about the XE range is that we're dealing with a completely different company by that time.

 

Absolutely.  I just find it interesting that Atari under Tramiel was every bit as capable as Warner-era Atari of making decisions that were, to be charitable, perplexing.

 

That's not to say that the Tramiels' decisions were all terrible; they weren't, and some are understandable given the state of Atari when they acquired it.  But more than a few babies were thrown out with the bathwater.

[Mathy]

Hello guys

 

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

Yeah, Atari did a fine job of squeezing as much life as they could out of the A8 range.  13 years on the market with only minor changes is pretty impressive by any measure.

 

It also meant that they only had to develop software for one machine, instead of a couple.  Guess what would have happened if they would have had to write different version of their software for 400, 800, 1200XL, 600XL, 800XL, 65XE, 130XE and XEGS?

 

Sincerely

 

Mathy