Jump to content

Photo

D1xx on both XL and XE computers


172 replies to this topic

#151 mytekcontrols ONLINE  

mytekcontrols

    Stargunner

  • 1,627 posts
  • Location:Santa Rosa, CA

Posted Sun Jan 15, 2017 2:09 PM

since we're on the theoretical, could we use the clock on the bus and multiply it to a usable frequency for the pair? they would be sharing the same clock and would code/decode at the same fixed rate.... it could be interesting to play with and who knows what unused portions exists on other chips in whatever the device is being cooked up... there are other chips but I chose that one because plenty have played with them and the later uses included digital implementations as well, looking a few years back you're likely to find some very simple uses all the way to... let's see what this little chip can do crazy stuff....   :)   It was rattling around in my head I know we used them in some projects many years ago....and it wasn't at all difficult or pricey to do.... like so many things it took forever to get it from the fog of my distant memory to light... so for my part I apologize for taking so long to get my grey matter to move on it... and likely longer to remember what we did... :o

 

looks like there are better options for slightly more money these days... all pretty much built in 1 wire solutions..... in any event neat stuff to delve back into but I'm just playing about now :)   btw they lie.... I mean 1 wire active 1 wire ground  that's two but ground already exist still.... amazing bidirectional wonder the more you want the pricier they can be!

 

I would love to implement a shared audio aspect on the signal assigned PBI audio input (that would be so cool!), but only if it were able to be done with a minimum of components since my 1088XEL motherboard project's goal is to build something with a very small footprint. The nice thing about this approach, is that it would still be compatible with the original mono usage, and obviously not require any other spare pins.

 

Your idea of using the system clock is a good one, and likely PH2 would serve that purpose well (great idea!). I think this would be fun to experiment with and see what is possible. Any suggestions on a good dual channel sample and hold amp that would suit this implementation? Or an all in one chip that also includes the demux aspect as well?

 

- Michael



#152 mytekcontrols ONLINE  

mytekcontrols

    Stargunner

  • 1,627 posts
  • Location:Santa Rosa, CA

Posted Sun Jan 15, 2017 3:02 PM

Dropcheck you made an interesting statement, and something that you alluded to sometime early in this thread as well...

By the way the internal bus pin out on the 1090XL/1090XLR is not nor can it be a pin for pin match to either the PBI or ECI bus.  There will be additional signals and power feeds present.  So no your bus is not compatible with the 1090XL internally  or the 1090XLR either externally or internally.  :(

 

So if I now understand this correctly, are you saying that if someone has an existing XL PBI device it will not be able to be plugged into the 1090XLR as is and work? And your box will not have anyway to directly plug into the standard XL PBI connection? (it sure sounds like that from what was said)  Just checking, because if this is true then I shouldn't even be here, since my 1088XEL project was about supporting the 'standard' XL PBI, and maybe tossing in a few extra bits that wouldn't break compatibility with pre-existing PBI devices but possibly still prove useful. I guess in the same way that you don't understand my stereo proposal, I haven't fully understood your project. I guess that's only fair  ;)

 

Hmmm... I just picked up on this when I reread your post

It's really amazing how circular logic is used to justify a decision.  We just decided to standardize four signals on four unused pins of the 600XL/800XL PBI and 130XE ECI.  How can we know how many PBI devices will be made that use any or all of those pins when it's only been a couple days? And to use that uncertainty to justify going back on that agreement is nuts. Because you don't know.

 

I underlined the part that is simply not true. To the best of my knowledge at the time, I thought I was fully supporting all of the signals that we had decided to go with through something called the configuration block on my 1088XEL. And in my defense, I am not the first to offer a way to 'configure' something in different ways. So as I now have it speced, YES you can route those signals we discussed as part of the PBI bus, and at the exact locations we agreed upon. Now where I may have erred is not realizing that this wasn't all of it, and that there were more changes to come beyond the 50 pin XL PBI connector as suggested in your other comment I included at the beginning of this post.

 

Obviously I haven't helped, and it looks like you and I are not heading in the same direction anyway. So I will peacefully leave this topic and let you carry on :)  

 

I sincerely look forward to seeing what the 1090XLR ultimately evolves into.

 

- Michael


Edited by mytekcontrols, Sun Jan 15, 2017 3:03 PM.


#153 _The Doctor__ OFFLINE  

_The Doctor__

    River Patroller

  • 2,247 posts
  • Location:10-0-11-00:02

Posted Sun Jan 15, 2017 7:01 PM

but but but nooooooo,   I was so proud of us all discussing and coming up with a good solution and it's, well, very good! Please don't walk away from each other when such wonderful ideas have been discussed and only got derailed on a possible or preliminary plan... Let our better selves prevail and continue in the spirit in which has been serving us so well. I think you are both wonderful creative and curious people, These are the traits of all good engineers. I don't want two awesome people having a bad time of it. I ask that a few good minds join to help in creating the stereo audio solution. I so want BOTH projects to be a success. This would be a glorious combination in so many ways!

 

I've got Bob the Builder on the brain!

Yes we can! :)


Edited by _The Doctor__, Sun Jan 15, 2017 7:22 PM.


#154 mytekcontrols ONLINE  

mytekcontrols

    Stargunner

  • 1,627 posts
  • Location:Santa Rosa, CA

Posted Sun Jan 15, 2017 8:11 PM

I would still like to pursue the idea that you inspired on time multiplexing of the ANALOG-IN line to be able to convey stereo audio through the single pin-49 connection. If nothing else it would be good to know if an easy to implement working solution could be developed that could then be put in our bag of tricks for use in whatever.

 

And I still intend to provide HALT, $D1xx, and RD5 on those XL PBI pins we discussed. But it's looking like there will be a divergence in what the 1090XLR expects to see from a physical interface perspective. I spent the time today reviewing the only 1090XLR related topic I could find besides this one (link: here), which I should have done before entering the discussion in this topic, and what I saw was an interface much more like an XE Cart/ECI, so nothing at all like the original 1090XL, which was obviously meant to work with the XL computers at the time. I personally like the XL's PBI way of doing things, so that is what I intend to support on my 1088XEL project.

 

Of course anything I eventually implement will get documented and made available for anyone to use. So if I can get a one-wire analog stereo solution to work, it will be made public, as well as implemented in the 1088XEL bus.

 

- Michael


Edited by mytekcontrols, Sun Jan 15, 2017 8:14 PM.


#155 Dropcheck OFFLINE  

Dropcheck

    Stargunner

  • Topic Starter
  • 1,044 posts
  • Location:Stigler, OK

Posted Sun Jan 15, 2017 9:40 PM

Dropcheck you made an interesting statement, and something that you alluded to sometime early in this thread as well...
 
So if I now understand this correctly, are you saying that if someone has an existing XL PBI device it will not be able to be plugged into the 1090XLR as is and work? And your box will not have anyway to directly plug into the standard XL PBI connection? (it sure sounds like that from what was said)  Just checking, because if this is true then I shouldn't even be here, since my 1088XEL project was about supporting the 'standard' XL PBI, and maybe tossing in a few extra bits that wouldn't break compatibility with pre-existing PBI devices but possibly still prove useful. I guess in the same way that you don't understand my stereo proposal, I haven't fully understood your project. I guess that's only fair  ;)
 
- Michael

 
   You can't plug an existing PBI device that only follows the 600/800XL PBI bus pin out (or the 130XE ECI) into the 1090XL and have it work.  You never could.  The external PBI bus connection is radically different than the internal one as far as signal pin out because of the need to handle different power needs of the various planned internal cards and additional signals used to allow multi-cards on the internal bus.  The same basic signals are present on the internal bus, but not with the same pin out.  Since the 1090XL was never more than a prototype, most PBI device developers simply used the PBI pin out on the 600/800XL computers and if they supported multiple devices at all created their own proprietary internal bus structure on their board aka the Black Box.  That's why you rarely see multi device PBI boards.  Most are single function and sometimes don't even allow pass thru of the PBI bus from the computer.  Nearly all do not use more than the +5V available on the bus, sometimes creating a somewhat uneasy power draw on the computer.   
 
    The 1090XLR is designed to handle multiple devices with differing power needs and to tie them all together will have a different internal bus pin out.  It physically will have a 50 pin external connector that can plug into a 600XL/800XL PBI connector directly, but it will be expecting HALT, D1XX and RD5 on pins 33, 37 and 39.  The ECI2PBI adapter that will be supplied will allow a 130XE to connect through the adapter to the 1090XLR and it will expect CASinh/ExtNB on Pin A of the ECI side.  Not all the signals from the 600XL/800XL PBI or 130XE ECI will be used on the internal bus.  There's no need for the +5V on pins 47/48 since the 1090XLR will have it's own power supply.  CAS/RAS on pins 41 and 44 will not be used because there won't be any cards developed using that obsolete memory refresh method.  Pins 19 and 42 will also be re-purposed since there's only so many GND pins needed.  On the 130XE ECI I'll loose the +5V and GND on pins 7 and H.  No need.  By freeing up those pins a second bus can be created for possible additional microcomputer additions.  I plan to have +5V and +12V power plugs for each slot to handle power for hard disks or other heavy power users.  +3.3V will be supplied to each internal slot for more delicate hardware.  A max of 5 internal slots will be available, with a basic I/O package for serial and parallel ports on the back as part of the main board.   Potentially I plan to make these ports jumper-able, thereby freeing up one to three device IDs for the 2nd bus.  There will be active cooling of the power supply and to some extent the card slots. 
 
   These are some of the planned features that I'm reasonably sure will make the final production model.   But....
 
    Existing PBI devices that follow the 600XL/800XL or 130XE computer connector pin outs will not work in the 1090XLR, just like they would not have worked in the 1090XL, without a board pin out redesign, ie they will have to conform to the new internal bus pin out, connecting to the various pins for the signals they need.  Also there will be some height/width/length restrictions.  Most boards already easily meet them.  While this sounds bad, those PBI devices are no longer going to need enclosures, won't have to have different versions to connect to the different computer models, won't have to worry about adequate power supply, simplified address decoding, may be able to reduce part counts, etc.

 

    I won't require boards to use D1XX if they want to implement their own address decoder, but they can not conflict with cards that do. True PBI devices will have to follow the rules for actual PBI devices, ie will have to supply their own device handler in onboard ROM, but then they had to before.  Anything else is just a board that is attached to the port because it provides an easy access to address and data lines, not because it is a real PBI device. 



#156 mytekcontrols ONLINE  

mytekcontrols

    Stargunner

  • 1,627 posts
  • Location:Santa Rosa, CA

Posted Mon Jan 16, 2017 12:24 AM

Yes I see now after studying the documents that even the original 1090XL would not support existing PBI boards. But (and I'm not telling you what to do) is there a valid reason for not supporting existing boards? After all you are not recreating an exact clone of the 1090XL, but taking many liberties to enhance it. So that would mean that you are not bound by the original 1090XL specification. And it would just seem logical to support the many devices that are already out there, since it would give people a stronger reason to buy what you are making. I'm sure you have your reasons... I was just curious as to what they are  :)

 

- Michael



#157 Dropcheck OFFLINE  

Dropcheck

    Stargunner

  • Topic Starter
  • 1,044 posts
  • Location:Stigler, OK

Posted Mon Jan 16, 2017 7:14 AM

Yes I see now after studying the documents that even the original 1090XL would not support existing PBI boards. But (and I'm not telling you what to do) is there a valid reason for not supporting existing boards? After all you are not recreating an exact clone of the 1090XL, but taking many liberties to enhance it. So that would mean that you are not bound by the original 1090XL specification. And it would just seem logical to support the many devices that are already out there, since it would give people a stronger reason to buy what you are making. I'm sure you have your reasons... I was just curious as to what they are  :)

 

- Michael

 

You can't support all the existing hardware cards that hang off the 600XL/800XL PBI connector in their existing physical forms.  For one some of the pcbs are at right angles, some are straight, some require connector mods to pick off signals not originally present on the bus, some are not true PBI devices and are simply sucking the convenient signals off the bus, what's worse is that some real PBI devices all have the same device ids,  AKA the Black Box, try fitting that into a space half that sprawling size with vertical edge card slots.  :(

 

Then you have the reality that the existing 600XL/800XL pin out puts a strangle hold on new hardware and new hardware demands.   Who needs CAS/RAS now?  Who needs 8 GND pins?  Two should be sufficient for almost all modern hardware now.   But those pins are still taken up.  Why the 8 GND pins?  Some say that it allowed the connector to handle more power draw from the computer, but really 8?  I doubt the Atari power supply or the motherboard could handle what power that implies.   It doesn't have some signals that it really should have and maybe were planned, but Atari engineers and the marketing department couldn't agree on what to support. 

 

 Despite the idea that the 130XE is the ugly step child of the 600XL/800XL and should not be seen, it is quite popular here in the states, and there are ECI based devices developed for it.  It doesn't use the same pin out. or physical connection, but it is the same computer inside.  It has additional signals not present on the 600XL/800XL connector, some are missing.  How can you support it's ECI devices in their current physical form in one expansion box? 

 

 Atari missed the boat when they did not follow through with a standardized multi-card design expansion box.  Unfortunately the 1090XL is not a standard, it's just a prototype that never was formalized.  Instead various developers had to improvise to use the 600XL/800XL or 130XE pin outs and quite literally almost all treated their device as the only one on the connector or that was ever going to be on the connector.  You can't standardize that without changing their physical construction. 

 

 I do in fact provide signal support and will adhere to the ECI/PBI device premise.  It's just literally impossible to physically put all the current PBI/ECI devices in one enclosure and have most all of them fit or work using the original 600XL/800XL PBI pin out.   For the most part conversions will be simply physical in nature.  Assuming the devices were correctly coded and designed in the first place.  The ones that want to use the same device ids will take a bit more effort.  :)

 

Otherwise every PBI/ECI device is an island unto itself.  Which breaks the fundamental premise that multiple devices can co-exist on the same bus at the same time. 


Edited by Dropcheck, Mon Jan 16, 2017 7:17 AM.


#158 mytekcontrols ONLINE  

mytekcontrols

    Stargunner

  • 1,627 posts
  • Location:Santa Rosa, CA

Posted Mon Jan 16, 2017 9:34 AM

Yep everything you said makes good sense, and you are right that it would become quite the monstrosity with some of the odd shaped and large boards that were made hanging off the bus. And of course the male vs female connection issue, and not to mention the XE vs XL differences really makes a mess of things. And yes the RAS and CAS signals are quite useless, because anybody that is going to hang a expansion box off of their A8 has either exypanded the ram internally, or if they wish to do so through the PBI wouldn't just be going for the 64k of the 1064 module. So realistically PBI ram expansion will be utilizing SRAM thus not requiring row/column addressing. So yes I agree with all of this when all aspects have been fully considered.

Also I just had a thought, since you will not necessarily be shooting for compatibility connection-wise with the existing PBI devices on the market, this will require a whole new set of products to be developed specifically for the 1090XLR. Now keeping what I just said in mind. That means modern day devices, likely using modern technology will need to be developed. Where will those new 1090XLR products come from? I'm thinking the same people that are presently developing current PBI devices, since it will just need a new PCB for the most part. Who are those people, well I know it's not me, and unless you plan to do so, it probably isn't you either. So I guess what I am really saying is, shouldn't the perceived need for extra signals be driven by a request from the current crop of builders, and not from you or I? And with that in mind, what have they asked for? Or are they curiously absent from this discussion? Whatever the case it would be great to hear what they think.

- Michael

Edited by mytekcontrols, Mon Jan 16, 2017 9:50 AM.


#159 Timothy Kline OFFLINE  

Timothy Kline

    Moonsweeper

  • 323 posts
  • Location:Perry, Michigan USA

Posted Mon Jan 16, 2017 9:53 AM

Also I just had a thought, since you will not be shooting for compatibility connection-wise with the existing PBI devices on the market, this will require a whole new set of products to be developed specifically for the 1090XLR. Now keeping what I just said in mind. That means modern day devices, likely using modern technology will need to be developed. Where will those new 1090XLR products come from? I'm thinking the same people that are presently developing current PBI devices, since it will just need a new PCB for the most part. Who are those people, well I know it's not me, and unless you plan to do so, it probably isn't you either. So I guess what I am really saying is, shouldn't the perceived need for extra signals be driven by a request from the current crop of builders, and not from you or I? And with that in mind, what have they asked for?
 

 

I've been following Dropcheck's development with tremendous interest because, as I understand it and perceive it from an end-user's vantage point, it seems that Dropcheck is trying to develop a standardized platform for future development which also offers current PBI developers an opportunity to revisit their work in much the same way as Candle is with the updated Incognito.

 

Dropcheck's platform provides developers with a framework and looks like its job will be to handle the grunt work of mediating the communication between the 1090XLR and its devices, and the Atari home computer it's connected to. Currently, the available PBI devices share the PBI and/or ECI interface, but little else, giving way to differently-shaped PCB boards which can provide their own challenges when it comes to issues like support the PCB card/device, separate connectors so that it can be used on more than just an 800XL or a 130XE, hindering a wider adoption which might be reached in a 1090XLR environment.

 

As to the types of modern-day devices, I'd say the sky is the limit once integration with the PBI is achieved. Printer buffers, hard drives, SD drives, bluetooth, and on the list could go. ICD at one point felt confident about being able to add 80-column video through the PBI, so we could feasibly see the development of add-in video cards along the order of the VBXE, complete with the appropriate connections at the rear for VGA, etc.

 

Of course, I could just be talking about of my hat here. ;)

 

--Tim



#160 _The Doctor__ OFFLINE  

_The Doctor__

    River Patroller

  • 2,247 posts
  • Location:10-0-11-00:02

Posted Mon Jan 16, 2017 10:11 AM

If the agreed on pin/signals remain, almost anything could live in the xlr box, it's a thing of beauty!

I also liked the idea of having a couple of jacks in the box to take audio or video within the box to allow for a sound card mixer or possible gen lock card... nice to have if they are there.  Of course a general purpose grommet hole could do the same.....having one or the other already present make for a standard professional enclosure



#161 mytekcontrols ONLINE  

mytekcontrols

    Stargunner

  • 1,627 posts
  • Location:Santa Rosa, CA

Posted Mon Jan 16, 2017 10:36 AM

 

I've been following Dropcheck's development with tremendous interest because, as I understand it and perceive it from an end-user's vantage point, it seems that Dropcheck is trying to develop a standardized platform for future development which also offers current PBI developers an opportunity to revisit their work in much the same way as Candle is with the updated Incognito.

 

Dropcheck's platform provides developers with a framework and looks like its job will be to handle the grunt work of mediating the communication between the 1090XLR and its devices, and the Atari home computer it's connected to. Currently, the available PBI devices share the PBI and/or ECI interface, but little else, giving way to differently-shaped PCB boards which can provide their own challenges when it comes to issues like support the PCB card/device, separate connectors so that it can be used on more than just an 800XL or a 130XE, hindering a wider adoption which might be reached in a 1090XLR environment.

 

As to the types of modern-day devices, I'd say the sky is the limit once integration with the PBI is achieved. Printer buffers, hard drives, SD drives, bluetooth, and on the list could go. ICD at one point felt confident about being able to add 80-column video through the PBI, so we could feasibly see the development of add-in video cards along the order of the VBXE, complete with the appropriate connections at the rear for VGA, etc.

 

Of course, I could just be talking about of my hat here. ;)

 

--Tim

 

Hi Tim,

 

Yes we are certainly in agreement with all of this. And I don't like beating a dead horse (but I will anyway  :) ), I think it's very important to also follow suit with what signal requirements have been deemed necessary by the current crop of PBI Device developers, . So can we please get some feedback from people like Simous and Candle (and others) that have actual experience at designing and implementing these devices in real life and in real situations. I for one have never attempted to develop a PBI device, so my street cred is very low in this area. Of course I can make educated guesses, but since I am learning as I go here, they aren't worth as much as coming from a PBI developer.

 

 

If the agreed on pin/signals remain, almost anything could live in the xlr box, it's a thing of beauty!

I also liked the idea of having a couple of jacks in the box to take audio or video within the box to allow for a sound card mixer or possible gen lock card... nice to have if they are there.  Of course a general purpose grommet hole could do the same.....having one or the other already present make for a standard professional enclosure

 

I agree that the signals we have thus far settled on would certainly be good from a non-decodable aspect (HALT), offering convenience ($D1xx), and  to support an existing recent product release (RD5). Since we have agreed to free up CAS and RAS, that now gives us two more signals for possible re-assignment. What will those be? What is missing that makes the most sense?

 

On the audio scene, I'm still looking into ways to stereo multiplex, but still be fully compatible with the original Audio-In signal provided by the PBI. So if it's possible to do that (which I think it is), it shouldn't impact on what is being talked about here, so long as the 1090XLR directly passes that signal through to each PBI slot and into the computer. I see that as something that would obviously require a decoder to be installed inside of the computer end of things and doesn't concern the 1090XLR so long as the trace exists. However I do like the idea of providing for audio/video connectors to be installed within the 1090XLR, but since the possibilities are endless as to what someone might like to do, if there is an enclosure planned, it might make sense to design it with a removable sub panel in the back that can be utilized specifically for customized holes. That way if you change your mind, just buy another blank sub panel.

 

- Michael


Edited by mytekcontrols, Mon Jan 16, 2017 10:43 AM.


#162 _The Doctor__ OFFLINE  

_The Doctor__

    River Patroller

  • 2,247 posts
  • Location:10-0-11-00:02

Posted Mon Jan 16, 2017 10:48 AM

I still like ras and cas staying.... just saying. That sub panel would work for the back +1 :)



#163 mytekcontrols ONLINE  

mytekcontrols

    Stargunner

  • 1,627 posts
  • Location:Santa Rosa, CA

Posted Mon Jan 16, 2017 10:59 AM

I still like ras and cas staying.... just saying. That sub panel would work for the back +1 :)

 

Yeah I was reluctant to see those go as well :( , but I've been thinking long and hard for a possible alternative use for those and haven't been able to come up with anything that makes sense. Do you see an application other than the established DRAM usage?

 

- Michael

 

OK I really should say this before someone thinks otherwise. This is Dropcheck's project, so anything I am saying is nothing more than hot air and speculation on my part. She is ultimately the one who will decide upon what this becomes  :-D


Edited by mytekcontrols, Mon Jan 16, 2017 11:58 AM.


#164 Dropcheck OFFLINE  

Dropcheck

    Stargunner

  • Topic Starter
  • 1,044 posts
  • Location:Stigler, OK

Posted Mon Jan 16, 2017 2:58 PM

I think there are several reasons why there seems to be few ECI/PBI developers and fewer still joining this conversation. 

 

1.  Maybe not the most important reason, but who wants to design an enclosure for his/her own device.  Just stick it inside the computer and be done with it. ;)

 

2.  A true PBI device is not just hardware bits and pieces, it also is low level device handler programming.  As has been said, the hardware is easy, the software is not. 

 

3.  Two different/competing connection methods for the two models of computers.  Makes non-standard a catch word. :?

 

4.  Limited resources available for understanding and creating new ECI/PBI devices.  No true standardized test boards to use, everything is cobbled together ad hoc.  Some developers only have an 800XL or 130XE not both. Very few true existing devices are open, ie no source information to recreate them or modify them or learn from them.  :(

 

5.  Knowledge of or interest in this avenue of development.  Has anyone contacted Simius or Candle or other developers?  Has this thread been brought to other Atari forum websites?  I'm kinda guilty here, but if there is a demand for this, is there users who are asking developers to develop?  We (I mean developers) all have our own pet projects and schedules etc, but we do respond to the prodding of users.  If only to feel the satisfaction of fulfilling a need (even a perceived need).

 

I could go on.  But in the absence of their participation, all I can go on is the assumption of the basic signals and needs I would expect if I could do all the development myself.  Granted that might be incomplete.  :)



#165 mytekcontrols ONLINE  

mytekcontrols

    Stargunner

  • 1,627 posts
  • Location:Santa Rosa, CA

Posted Mon Jan 16, 2017 3:53 PM

I think there are several reasons why there seems to be few ECI/PBI developers and fewer still joining this conversation. 
 
1.  Maybe not the most important reason, but who wants to design an enclosure for his/her own device.  Just stick it inside the computer and be done with it. ;)

 
Yep that's probably very true, but because there are sooo many cool gizmos, starting to run out of room  :grin:

 

2. A true PBI device is not just hardware bits and pieces, it also is low level device handler programming. As has been said, the hardware is easy, the software is not.


It may be considered easy, but then if the hardware doesn't deliver what you need, the software work around becomes an even bigger issue. This is the whole point of what you are trying to do by adding more signals to the bus  ;) 

 

3. Two different/competing connection methods for the two models of computers. Makes non-standard a catch word. :?


Yep this is a big time problem, but as I now understand it the 1090XLR will steer it towards a common bus.

 

4. Limited resources available for understanding and creating new ECI/PBI devices. No true standardized test boards to use, everything is cobbled together ad hoc. Some developers only have an 800XL or 130XE not both. Very few true existing devices are open, ie no source information to recreate them or modify them or learn from them. :(

 
Yeah as I've found out on my own, you really have to scour the web to find anything of actual use, and even then a lot is left to your imagination to figure out. Not good  :spidey:  It would be great to have a website setup to consolidate and index what is out there  :idea:

 

5. Knowledge of or interest in this avenue of development. Has anyone contacted Simius or Candle or other developers? Has this thread been brought to other Atari forum websites? I'm kinda guilty here, but if there is a demand for this, is there users who are asking developers to develop? We (I mean developers) all have our own pet projects and schedules etc, but we do respond to the prodding of users. If only to feel the satisfaction of fulfilling a need (even a perceived need).


You shouldn't feel guilty at all. Those people have been active very recently on A.A., and I'm pretty sure they are aware of this topic. If they haven't jumped in, it's likely that they are very busy in their lives with other things and/or observing where this will go before commenting. But it sure would be great if they did  :grin:

 

BTW, I apologize for anything I may have done to make this road bumpy. My only excuse is I'm getting older, and suffer from some medical issues that make me anxious at times. Believe it or not I also worry about not living long enough to finish what I have set out to do. I know pretty silly huh?  :P  Anyway I'll try to calm down, and would still like to participate in these discussions. And who knows if in the end the new specs look doable on my end, I'll try to incorporate them into the 1088XEL interface as well so that perhaps there will be at least one system that doesn't require modification to interface with what you are creating  :)

 

- Michael



#166 gozar OFFLINE  

gozar

    Dragonstomper

  • 736 posts
  • Location:Ohio

Posted Mon Jan 16, 2017 6:41 PM

There is also another route. Combine a new motherboard with the 1090XLR into a micro-atx form factor, kind of like what the C64 reloaded project did 3 years ago with the C64 main board. There are tons of 600XLs out there for donor chips (6502, Pokey, Antic, GTIA, etc.).

 

 

If you have to modify the 800XL or 130XE to use the 1090XLR, then you've added another hurdle for the users to leap over, and there are quite a few Atari users that don't like modifying their machines (me on the other hand have an addiction to it :-). 



#167 Kyle22 ONLINE  

Kyle22

    River Patroller

  • 2,804 posts
  • Location:McKees Rocks (Pittsburgh), PA

Posted Mon Jan 16, 2017 7:39 PM

My only concern is that if you add signals to unused pins, what happens when someone later plugs in a legacy PBI device that grounds (what it considers to be) unused pins?



#168 Dropcheck OFFLINE  

Dropcheck

    Stargunner

  • Topic Starter
  • 1,044 posts
  • Location:Stigler, OK

Posted Mon Jan 16, 2017 11:16 PM

My only concern is that if you add signals to unused pins, what happens when someone later plugs in a legacy PBI device that grounds (what it considers to be) unused pins?

 

If someone knowingly modifies their ECI/PBI connector on their computer to make it compatible with the 1090XLR and then disconnects that device and plugs in a legacy device that does something with those pins instead of ignoring them, honestly how is that my fault.  :?

 

They've made several conscious decisions to do what they've done.  I have no control over what they do with their equipment.  And never will. 

 

 I don't have any way of knowing what would happen.  Since we aren't talking about power here, probably nothing.  ie doesn't work.  Doubt it would fry anything.  :?



#169 Dropcheck OFFLINE  

Dropcheck

    Stargunner

  • Topic Starter
  • 1,044 posts
  • Location:Stigler, OK

Posted Tue Jan 17, 2017 12:20 AM

There is also another route. Combine a new motherboard with the 1090XLR into a micro-atx form factor, kind of like what the C64 reloaded project did 3 years ago with the C64 main board. There are tons of 600XLs out there for donor chips (6502, Pokey, Antic, GTIA, etc.).

 

 

If you have to modify the 800XL or 130XE to use the 1090XLR, then you've added another hurdle for the users to leap over, and there are quite a few Atari users that don't like modifying their machines (me on the other hand have an addiction to it :-). 

 

I repeat forever begin.  I am not going to recreate the computer itself.  This is strictly going to be a physical framework for expansion of it's capabilities.  ;-)

 

Some people are confirmed modders.  Some just want to turn on the switch and have it work.  I get that.  But you know, I get so tired of hearing this rationalization for not doing something.   No one is going to force these individuals to mod their computer.  No one is going to force them to buy this product.  They have a perfect right to go on about their business never giving a second thought to this product.  :D

 

But you don't get something for nothing either.  You don't drive your car forever without putting gas in it, or airing the tires up or the thousand other things you have to do to drive a car.  ! 

 

If you want to use this product and benefit from it, then you will do what is necessary to get it to work, whether you do it yourself or have it done by someone else.  And lets face it, we're talking about soldering three jumper wires on the 600XL/800XL and 1 on the 130XE. Hardly a major hurdle.  :)

 

I'm reminded of my days as a tech support person providing support for internet access over the telephone.  We'd have someone call in complaining that they could not get on the internet.  I learned real quick that you can't assume anything.  Nearly 90% of the time the problem was the person did not know how to operate their computer.  It ranged from the simple act of turning the monitor on so they had a display to trying to teach them how to use the mouse to control what they were seeing on the display.  It rarely was an actual internet service problem.  But they wanted that internet access, because their granddaughter was on it all the time, or their son sent them video chats every week, or they had to find out what was happening on their best friend's Faceprint page. They were rarely young people, sometimes they were old, but mostly they were people that just didn't interact with technology very much. But people generally will do what is necessary to get what they want.  :grin:



#170 _The Doctor__ OFFLINE  

_The Doctor__

    River Patroller

  • 2,247 posts
  • Location:10-0-11-00:02

Posted Tue Jan 17, 2017 1:55 AM

My only concern is that if you add signals to unused pins, what happens when someone later plugs in a legacy PBI device that grounds (what it considers to be) unused pins?

That's why we were so careful to keep what is there and observe the behavior of PBI on all the machines as well as the devices we could find to plug into it. If kept the way it was last described... all the add ons of the past still work and all the new will work too! If we only add what was agreed on ;) and not take away the other stuff



#171 mytekcontrols ONLINE  

mytekcontrols

    Stargunner

  • 1,627 posts
  • Location:Santa Rosa, CA

Posted Tue Jan 17, 2017 10:25 AM

Although it would be nice to reassign something useful to the obsolete RAS and CAS pins on the XL PBI, that mod would require cutting traces which is a much more permanent change. And I could see a lot of people not wanting to do that to their precious Atari's. So those pins are probably not going to be good candidates  :ponder:

 

- Michael



#172 Mathy OFFLINE  

Mathy

    River Patroller

  • 2,328 posts
  • Location:Heerlen, NL

Posted Tue Jan 17, 2017 6:19 PM

Hello Lenore

 

If someone knowingly modifies their ECI/PBI connector on their computer to make it compatible with the 1090XLR and then disconnects that device and plugs in a legacy device that does something with those pins instead of ignoring them, honestly how is that my fault.  :?

 

They've made several conscious decisions to do what they've done.  I have no control over what they do with their equipment.  And never will. 

 

 I don't have any way of knowing what would happen.  Since we aren't talking about power here, probably nothing.  ie doesn't work.  Doubt it would fry anything.  :?

 

 

I don't want to discus if this would be your fault or not, as I haven't formed an opinion on that (yet?).  But what might happen is that people might think they might fry their computer or their PBI device when they change pin allocations (even if at the moment these pins are free).  That might keep people from buying your product (or any other that requires a modification).  Maybe it's worth looking into making the mods such, that they are (kinda) bullet/fool proof.

 

Sincerely

 

Mathy



#173 mytekcontrols ONLINE  

mytekcontrols

    Stargunner

  • 1,627 posts
  • Location:Santa Rosa, CA

Posted Tue Jan 17, 2017 6:37 PM

I don't want to discus if this would be your fault or not, as I haven't formed an opinion on that (yet?).  But what might happen is that people might think they might fry their computer or their PBI device when they change pin allocations (even if at the moment these pins are free).  That might keep people from buying your product (or any other that requires a modification).  Maybe it's worth looking into making the mods such, that they are (kinda) bullet/fool proof.

 

Hi Mathy  :)

 

If I may interject... a possibility would be to connect some low ohm resistor in series with the 'new' assignments going to the reserved pins. I'm thinking of 100 ohms, which should protect the HALT, $D1xx, and RD5  outputs from a complete short if those reserved pins were grounded on the PBI Device, while still passing enough current to do what is needed on the receiving end. No need to protect the PBI device since the only connection they would ever make is either going to be ground, or maybe an input (RD5 sense on IDE Plus 2.0 Rev E).

 

- Michael


Edited by mytekcontrols, Tue Jan 17, 2017 6:41 PM.





0 user(s) are browsing this forum

0 members, 0 guests, 0 anonymous users