• Please review our updated Terms and Rules here

Language 16K RAM card or Saturn 128K RAM card access advice welcome

george

Experienced Member
Joined
Apr 1, 2011
Messages
410
Seeking for an apple2 experts advice here. Is there any software except for the original Saturn RAM card's tests that rely on or check for the *double* consecutive read accesses to C0x1 or C0x3 or C0x9 or C0xB in order to write enable the (language) RAM card? I am asking this because I've designed a 128K RAM card that accepts only a single such access and all software I tested worked except for the Saturn specific RAM card test that complains with a message "RAM WRITABLE WITH ONE POKE" which strictly speaking is incorrect since they mean a PEEK.
 

David_M

Veteran Member
Joined
Oct 1, 2016
Messages
647
Location
Melbourne, Australia
A read or write will activate the soft latches so peek or poke will work.

There is a page in the Saturn card documentation that details how the soft switches work, I used it to design the Satrun128 Reboot card.
 

Anthony Adverse

Experienced Member
Joined
Oct 18, 2018
Messages
120
Is there any reason there couldn't be a saturn 256? Or is it just pointless, or exceed the number of banks swappable?

A
 

george

Experienced Member
Joined
Apr 1, 2011
Messages
410
A read or write will activate the soft latches so peek or poke will work.

There is a page in the Saturn card documentation that details how the soft switches work, I used it to design the Satrun128 Reboot card.

This page does not describe all peculiarities of the saturn card's control. My main question was about the *double* access. What if single access was sufficient? This would have saved one output pin in the 22V10 if used to emulate the saturn card. On the other hand the saturn card differs from the language cards because a write access to C0n1 also works in Saturn for enabling the write access to the card's RAM. Moreover the Saturn testing software specifically checks if this is possible and refuses to detect the card at all if only read (like in language cards) is required at an odd C0nx. Another difference -- at power-up unlike the language cards the Saturn card is disabled for writing.
 
Last edited:

george

Experienced Member
Joined
Apr 1, 2011
Messages
410
@Anthony, the mechanism that Saturn card uses limits the number of 16K banks to 8 due to the lack of more free addresses at C0nx space. On the other hand what is the point of having hardware that is not supported by any software?
 

David_M

Veteran Member
Joined
Oct 1, 2016
Messages
647
Location
Melbourne, Australia
@Anthony, the mechanism that Saturn card uses limits the number of 16K banks to 8 due to the lack of more free addresses at C0nx space. On the other hand what is the point of having hardware that is not supported by any software?

I created a Saturn card last year and some lowlife in Russia has already ripped off the design and is selling poor quality pcbs.
Why redo it again? It would be far more useful to find another card to redesign.

This is my current pcb alongside the rom card i designed.
20181013_094041.jpg
 

Anthony Adverse

Experienced Member
Joined
Oct 18, 2018
Messages
120
@george All to true, but I had wondered why they stopped at 128.... interconnectivity doesn't mean interoperability :)
 

groink

Experienced Member
Joined
Mar 3, 2014
Messages
171
Although the Russian is producing PCBs, isn't it still useless without the GAL/PAL? Or, did someone release it as well?
 

george

Experienced Member
Joined
Apr 1, 2011
Messages
410
@george All to true, but I had wondered why they stopped at 128.... interconnectivity doesn't mean interoperability :)
Because back in 1982 this was enough and expensive enough ;)

@ David your design has stability issues (according to the corresponding topic in the applefritter forum and not only) and wastes the 22V10 while adding 3 more '74 ICs to it. The PCBs of the Russians are not that bad (if they are correct?) but are expensive for the purpose. Don't plan to try them. I have my own ideas. I thought that your project was an open project since is a bit crude so far...I thought maybe you had released it exactly because it has problems in order for others to improve and fix them.

@ groink David himself has published two versions of his GAL firmware in the applefritter forum. I personally have never investigated them since I have an original saturn card (as a model for my project now).
 

David_M

Veteran Member
Joined
Oct 1, 2016
Messages
647
Location
Melbourne, Australia
Because back in 1982 this was enough and expensive enough ;)

@ David your design has stability issues (according to the corresponding topic in the applefritter forum and not only) and wastes the 22V10 while adding 3 more '74 ICs to it. The PCBs of the Russians are not that bad (if they are correct?) but are expensive for the purpose. Don't plan to try them. I have my own ideas. I thought that your project was an open project since is a bit crude so far...I thought maybe you had released it exactly because it has problems in order for others to improve and fix them.

@ groink David himself has published two versions of his GAL firmware in the applefritter forum. I personally have never investigated them since I have an original saturn card (as a model for my project now).

There are no stability issues other than faulty RAM chips. As "crude" as it is I was able to make it work with just the page from the saturn manual... you apparently can't. So forgive me if I don't place any value on your opinion.
 

george

Experienced Member
Joined
Apr 1, 2011
Messages
410
There are no stability issues other than faulty RAM chips. As "crude" as it is I was able to make it work with just the page from the saturn manual... you apparently can't. So forgive me if I don't place any value on your opinion.
The topic in applefritter only (I heard about FB too) is full of reports that your ram card not stable on different machines. And I may know why. But as you say this has no value to you and your ram card will remain buggy. I refrained from saying this but no more -- just one look at your amateurish PCB design with the track thickness of the power supply and many other issues I am no longer wanting to tell you since you treat all remarks with silence, hostility and contempt and haughty. BTW these your words tell me that you are not fully aware of what is going on with your design:

As part of the process of designing the quickloader replacement, I saw that they combined the Q3 signal with R/W. They did it in a roundabout way using a latch but a simple OR gate will get the same result. The effect of combining the two signals delays the write signal being asserted from the beginning of the CPU phase of the clock cycle to the middle of the CPU phase. This allows extra time for the data on the data bus to become stable before it is written to the RAM.
 

george

Experienced Member
Joined
Apr 1, 2011
Messages
410
You make a lot of assumptions, unfortunately most of them are wrong.

My assumptions are right and unlike you I would have listen to your logical arguments that you never provided. When I wrote you in another forum that only a 22v10 should be enough to control the SRAM chip I meant it. You opposed stubbornly to my observations and I decided to design the RAM card myself...

http://www.ebay.com/itm/264218272308
 

David_M

Veteran Member
Joined
Oct 1, 2016
Messages
647
Location
Melbourne, Australia
My assumptions are right and unlike you I would have listen to your logical arguments that you never provided. When I wrote you in another forum that only a 22v10 should be enough to control the SRAM chip I meant it. You opposed stubbornly to my observations and I decided to design the RAM card myself...

http://www.ebay.com/itm/264218272308

How is your butthurt coming along. I hope it is healing now.
 

george

Experienced Member
Joined
Apr 1, 2011
Messages
410
How is your butthurt coming along. I hope it is healing now.

I guess this attitude leads to your design flaws, but it could be something general -- for instance Australian law that kills cats...
 
Top