Jump to content
mytek

1088XEL Alternative Mother-Board Project

Recommended Posts

Regarding the 0.3" vs 0.6" wide RAM chips, if there is room on the board what about putting pads in for both(the 0.3" socket footprint inside the 0.6" socket footprint). Either socket can then be installed as desired.

 

I know you already have the part Michael, but it would save someone else in the future who can't get the .3" wide part from needing an adapter(or having to modify the layout of the PCB before having them made).

 

Hi Bill :)

 

I will take this into consideration, but no guarantee that it'll actually happen. Not that it is a bad idea (on the contrary it is a good suggestion), but it might come down to space constraints. keep in mind that the previously posted image of the board lay-out is far from complete, and missing all of the discrete components that will be required. So even though there presently looks like there would be room for what you are suggesting, this might not be the case in reality.

 

- Michael

Share this post


Link to post
Share on other sites

i've noticed that you have RAM chip select connected to PHI2 line - there is considerable delay in data out to chip select time, and i wouldn't go that way, but if you have your math done here, and feel line on the safe side - it's up to you

i would rather use that chip select as extsel/casinhibit and use proper oe and we strobes

 

I am revisiting this suggestion by candle (where he does make a good point).

 

Here is an excerpt from an IBM Static RAM datasheet (notice the last sentence)...

When OE is high, the data outputs (DQs) are always tri-stated.
When OE is low, the outputs are active, that is, data, if
available, can appear at the output pins.
During a read operation, this signal is used to prevent data
from appearing at the output until needed. Prior to a write
operation, OE is sometimes used to tri-state the data bus
to avoid data contention. OE is ignored during a write operation.
OE is an asynchronous signal. It can be switched at any time
and the SRAM will respond to the signal. Use of the output
enable, although recommended, is optional.

----

Here is my circuit...

 

WwnipVK.png

 

 

So if both chip enables are selected, the SRAM will be put on the bus (/CE1 = Low and CE2 = High). For a Write, the address and data will be stable when PHi2 is high. For a Read, address and data will be stable just before the falling edge of PHi2. And of course the 6502's R/W signal will simply tell the RAM which operation is desired. Since I have the buffered PHi2 (B02) feeding the RAM's CE2 chip select, the operation won't commence until B02 has gone high or been high previous to coming out of the setup time. Here's a couple of timing diagrams showing the relationship between the various signals.

 

 

The 6502 Write Timing

6502WriteTiming.png

 

The 6502 Read Timing

6502ReadTiming.png

 

 

So the reason I chose the method that I did, is because of its simplicity of only requiring one single gate while still providing for all contingicies (factoring in /CASinh and /EXTsel). To also select the /OE line when reading, would have likely required a minimum of 3 gates. However to be on the safe side I will thoroughly test the method that I am using for reading and writing to the SRAM chip, just in case that particular chip has a variance in setup timing that pushes it out of the envelope.

 

- Michael

Edited by mytekcontrols

Share this post


Link to post
Share on other sites

well, i'm using phi1early (shortened by 70ns version of phi1) to catch valid state of RW signal as i found it is not stable within a cycle - same goes for address bus

but it might be me using very fast logic able to respond to such transients

  • Like 1

Share this post


Link to post
Share on other sites

well, i'm using phi1early (shortened by 70ns version of phi1) to catch valid state of RW signal as i found it is not stable within a cycle - same goes for address bus

but it might be me using very fast logic able to respond to such transients

 

I'm also using very fast logic on the PHi2 buffer, so all things considered I might be fine as well. But I'll know for sure when I breadboard this and see what happens.

 

Hey just a slight change of topic... did you ever consider actually finishing and releasing that MOBO design you did back in 2011? It looked very cool :thumbsup:

 

- Michael

Share this post


Link to post
Share on other sites

main reason it get at stall was the cost i had to swallow to get it done - doing pcb in the country was out of question (because of the cost) and China fabs were out of my reach

casewise it was not an issue - first thing i did was to recreate it in solidworks so it would be possible to simply print it - technology was expensive, but still there, and it seemed that outside my country everything was way ahead in these terms

i don't know if i have solidworks models for the case anymore, but all remaining bits are still on my disk

the other reason was i really wanted to have 816 there, and i had no working solution i could be sure of, so i've split the design into cpu card and the motherboard just to be able to change it later, but this brings another can of worms to the design goals

from where i stand now it will be another two or three smaller projects as milestones before i could return to that board, but when i do it should be a gamechanger (if anyone would still care ;) )

  • Like 3

Share this post


Link to post
Share on other sites

main reason it get at stall was the cost i had to swallow to get it done - doing pcb in the country was out of question (because of the cost) and China fabs were out of my reach

casewise it was not an issue - first thing i did was to recreate it in solidworks so it would be possible to simply print it - technology was expensive, but still there, and it seemed that outside my country everything was way ahead in these terms

i don't know if i have solidworks models for the case anymore, but all remaining bits are still on my disk

the other reason was i really wanted to have 816 there, and i had no working solution i could be sure of, so i've split the design into cpu card and the motherboard just to be able to change it later, but this brings another can of worms to the design goals

from where i stand now it will be another two or three smaller projects as milestones before i could return to that board, but when i do it should be a gamechanger (if anyone would still care ;) )

 

I think people would care, as evidenced from the feedback I get on this topic. And I've witnessed people jumping in willing to pay a lot of money for some of the new upgrades that have been coming out lately. High on the popular list seems to be video upgrades, accelerators, SSD's and/or Flash carts. And also the U1MB is extremely popular as well. If you had told me 20-25 years ago that this would still be going on, I'd have called you crazy :grin: So yes I do believe if you built it, they would come and buy it.

 

Speaking of solid works one of these days I need to sit myself down and learn that. I got a free copy last year by mistake when it was sent to my business as an update for whatever business had leased the shop before me. I tried to make contact with the original owners but they were no longer around, and when I contacted the company that sent it, they said to keep it. So it was free to me (pretty sweet).

 

- Michael

Share this post


Link to post
Share on other sites

Well I scratched my head and rattled my brain, and was able to steal a gate from another section to use in the SRAM circuit.

 

lRlgwnJ.png

 

 

So now we have a gated /OE signal which should keep our new A8 board very happy :)

 

------------------------

 

And here is my solution (inspired by a suggestion from Kyle) for a IRQ electronic switch for the 2nd Pokey chip interrupt connection.

 

RHVNC33.png

 

 

IRQ2-EN goes to the S1 output on the U1MB where it can be controlled (turned ON or OFF) via the U1MB setup menu (S1 High = 2nd Pokey IRQ enabled).

 

------------------------

 

See I do listen to everyone's suggestions, process it, and if it makes sense then try my darndest to integrate it if at all possible ;)

 

- Michael

Edited by mytekcontrols
  • Like 4

Share this post


Link to post
Share on other sites

Perhaps It would be good if you would put a shift register over s1/s2 signals and have It expanded to 8 bits of Control. I'm sure we can handle It in software

  • Like 2

Share this post


Link to post
Share on other sites

Perhaps It would be good if you would put a shift register over s1/s2 signals and have It expanded to 8 bits of Control. I'm sure we can handle It in software

 

Now you're trying to make more work for FJC ;)

 

Although I did use both S0 and S1 for my own purposes, the other device control bits are still available for their designated uses (M0 mono/stereo switching was utilized, but only for its intended purpose). If S0 or S1 are needed for some other upgrade, I am not aware of what that would be :?

 

- Michael

Share this post


Link to post
Share on other sites

Question to anyone. Are there any games or applications that used more than one light gun?

 

- Michael

Share this post


Link to post
Share on other sites

vaporware, real soon now light gun games...... I for one would love a game where you compete at the same time with someone.... should I put the light pen in port one or port 2 or the light gun in port one or port two... and how about those apps that use a drawing pad and a pen at the same time..... yeah I think we should keep support for at least two.. although I still love the 800 and 5200 for the 4 ports they offer and is one of the reason the family and some friends still find the old machines pretty cool. My one daughter thought it would be cool if the software allowed for 4 light pens so she could pick them up like real pens/tools to use... she wan't to be a graphics artist and dislikes the selection methods on most platforms just some thoughts

Edited by _The Doctor__

Share this post


Link to post
Share on other sites

Michael, need i to remind you that i wrote original incognito bios? :) s0 and s1 pins were reserved for my siplestereo upgrade and this is their original function, but i see no reason to add expander there

 

light guns are fun, but unless someone makes one that actually works with lcd screen they are doomed

  • Like 2

Share this post


Link to post
Share on other sites

Michael, need i to remind you that i wrote original incognito bios? :) s0 and s1 pins were reserved for my siplestereo upgrade and this is their original function, but i see no reason to add expander there

 

light guns are fun, but unless someone makes one that actually works with lcd screen they are doomed

 

Candle yes don't worry I definitely know you are the creator of the original hardware and firmware design for that project (and many more). But unless I'm mistaken FJC continued to develop the firmware beyond it's original capabilities, and it is usually what people now upgrade to. Funny thing is I just bought two more U1MB boards from Lotharek recently and was surprised to see that they came with your firmware version instead of FJC's. Although Lotharek does have a link to FJC's U1MB webpage on his site. I remember using the original firmware on my first U1MB, but found FJC's version to be a lot more versatile, so that is what I go with now. Not sure why new boards are not getting this version flashed instead (copyright issues???).

 

Speaking of your simple stereo board, what were the functions associated with S0 and S1 control? BTW, you probably noticed that I created a more recent stereo upgrade that I patterned after what you did with simple stereo having Dual Pokeys, electronic stereo/mono switching, and a PS2 keyboard adapter (TK-II-STEREO). The only thing I left out was Covox.

 

Yep my feelings exactly on the light guns. The reason I was asking is that I'm only thinking of supporting one such device on Joystick port 1. With the recent changes, I stole the AND gate I was using to distribute both trigger inputs to /LP on Antic. So instead of adding another gate package or using mickey-mouse logic (diode/resistor) to reinstate that, I figure that I would just limit it to one, especially considering its limited usefulness.

 

- Michael

Share this post


Link to post
Share on other sites

vaporware, real soon now light gun games...... I for one would love a game where you compete at the same time with someone.... should I put the light pen in port one or port 2 or the light gun in port one or port two... and how about those apps that use a drawing pad and a pen at the same time..... yeah I think we should keep support for at least two.. although I still love the 800 and 5200 for the 4 ports they offer and is one of the reason the family and some friends still find the old machines pretty cool. My one daughter thought it would be cool if the software allowed for 4 light pens so she could pick them up like real pens/tools to use... she wan't to be a graphics artist and dislikes the selection methods on most platforms just some thoughts

 

Can you name a few games that use more than one light gun?

 

- Michael

 

EDIT: Drats I found one that uses a light gun on joystick port 2 (Gangsters). Well I guess I'll be adding a couple of diodes :grin:

Edited by mytekcontrols
  • Like 2

Share this post


Link to post
Share on other sites

S0/S1 had nothing attached to them for some 5-6 years in 99 per cent of cases, so I think - unless you're Apple Inc. - it's reasonable to let the owner use the hardware as they see fit. New BIOS allows complete redefinition of M0/M1/S0/S1 in software using plug-in architecture, and this has so far included Rapidus (core enable/disable), VBXE core flash protection, among other things.

  • Like 2

Share this post


Link to post
Share on other sites

I have crt monitors, tube tV, and projection televisions, of course I have more modern LCD's as well.... I love me some gun fun..... to be honest I really don't care for long term lcd viewing it seems to bother me... even when using the 240 rate versions... it has a sort of after effect... a flicker of the real world to me.... I really would love for a race the scanline mode for LCD's, you would think it would be a firmware issue but it who knows....

Edited by _The Doctor__
  • Like 1

Share this post


Link to post
Share on other sites

S0/S1 had nothing attached to them for some 5-6 years in 99 per cent of cases, so I think - unless you're Apple Inc. - it's reasonable to let the owner use the hardware as they see fit. New BIOS allows complete redefinition of M0/M1/S0/S1 in software using plug-in architecture, and this has so far included Rapidus (core enable/disable), VBXE core flash protection, among other things.

 

Thanks for that clarification :)

 

 

I have crt monitors, tube tV, and projection televisions, of course I have more modern LCD's as well.... I love me some gun fun..... to be honest I really don't care for long term lcd viewing it seems to bother me... even when using the 240 rate versions... it has a sort of after effect... a flicker of the real world to me.... I really would love for a race the scanline mode for LCD's, you would think it would be a firmware issue but it who knows....

 

I hear you. And yes it's really too bad that non-CRT's don't allow for such a simple way to know positionality of a light gun (or pen). But it's really much more than a firmware issue, and way more to do with the mechanics of the situation. The only solutions that I know of have been demonstrated by the likes of Wii with their infrared sensor bar (although ultrasonics would also work).

 

- Michael

Share this post


Link to post
Share on other sites

Is there some reason for not connecting /OE to /CS? That's what I always do.

 

Bob

 

 

Well I scratched my head and rattled my brain, and was able to steal a gate from another section to use in the SRAM circuit.

 

 

 

 

So now we have a gated /OE signal which should keep our new A8 board very happy :)

 

------------------------

 

And here is my solution (inspired by a suggestion from Kyle) for a IRQ electronic switch for the 2nd Pokey chip interrupt connection.

 

 

 

 

IRQ2-EN goes to the S1 output on the U1MB where it can be controlled (turned ON or OFF) via the U1MB setup menu (S1 High = 2nd Pokey IRQ enabled).

 

------------------------

 

See I do listen to everyone's suggestions, process it, and if it makes sense then try my darndest to integrate it if at all possible ;)

 

- Michael

Share this post


Link to post
Share on other sites

Is there some reason for not connecting /OE to /CS? That's what I always do.

 

Bob

 

Technically no reason why what you suggest wouldn't work (/OE to ground also would likely be ok). However from a perfectionist stand-point, and to assure that all is well timing-wise as in no fighting over the data bus, it is more proper to have the outputs enabled only during an actual write. And since I was able to shuffle things around with my existing glue logic to achieve this, so much the better. Think of this as insurance, probably not needed, but nice just in case.

 

- Michael

Share this post


Link to post
Share on other sites

 

Technically no reason why what you suggest wouldn't work (/OE to ground also would likely be ok). However from a perfectionist stand-point, and to assure that all is well timing-wise as in no fighting over the data bus, it is more proper to have the outputs enabled only during an actual write. And since I was able to shuffle things around with my existing glue logic to achieve this, so much the better. Think of this as insurance, probably not needed, but nice just in case.

 

- Michael

 

Opps I meant to say 'it is more proper to have the outputs enabled only during an actual READ'.

 

- Michael

Share this post


Link to post
Share on other sites

Since most PC's now days have their power controlled via a momentary push button switch, I thought it only fitting to do the same. So the following circuit will be applied to the 1088XEL motherboard's power control section.

 

wmYMii9.png

 

The latching MosFet/Transistor circuit came from this article: Latching power switch uses momentary pushbutton EDN 1/20/2014 by Anthony Smith

 

Note: That particular MosFet has an ON resistance of 0.07 ohms. So for a 5 V input and a 1 amp load, we'll end up with approximately 4.95 volts coming out.

 

 

- Michael

 

EDIT: I'll probably need a Schottky diode in series with the output going to the load to insure the MosFet shuts off.

Edited by mytekcontrols
  • Like 2

Share this post


Link to post
Share on other sites

You're opting for pc case, why not to use pc Power supply? It has all the logic Inside already, well, almost

Share this post


Link to post
Share on other sites

You're opting for pc case, why not to use pc Power supply? It has all the logic Inside already, well, almost

The 'latching' logic would normally be on the Mini-ITX motherboard (pretty much the same as on an ATX motherboard).

 

The PSU you are talking about has a green sense lead that when pulled to ground turns on the main power. However this is not a momentary flip-flop action, thereby requiring that the green lead be maintained to ground for the supply to stay active. And in order for the latching logic on a PC motherboard to work, there is standby power of 5 Vdc @ 1-2 amps which is always available from the PSU via an independent connection from the other high current 5 and 12 Vdc switched power connections.

 

Bottom line is that the normal ATX, Mini_ATX, or even Mini-ITX PSU is total overkill for the power concerns of the 1088XEL, so I was opting for the use of a cell phone charging wall wart instead. However I could have my latching circuit also turn on a separate NPN transistor that could be used to activate a standard PC PSU as well for those that still wish to use that option (gives you 12 Vdc). Of course the amperage coming out of these guys could really fry something if not careful, so a fuse would be in order.

 

I'll look into providing that NPN switch as well.

 

Funny thing is our 5 Vdc power supply needs are entirely met by the standby power of a ATX style PSU thereby never actually requiring that the main supply circuits be powered up. Back in the pre SSD days where mechanical floppies and hard drives were the norm, that power was needed. Now days not so much, unless you want that nostalgic feel.

 

- Michael

Edited by mytekcontrols

Share this post


Link to post
Share on other sites

I don't like tons of boxes and cables lying around my desk, so for me yet abother psu wgere there is one slready in the Case is to write at least awkward. I could use that additional spersge to run internal floppy Drive Or whatever. These xases do contain 5.25 Drive bay, dont they?

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...

  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...