Not quite... The end of the ROM based PLA for the 64

[banman]

Hi eslapion,

Above all, stay away from any logic ICs sold on eBay coming from China. Very often, the chip inside isn't at all what it says on the package. Then you ask yourself why it is you get inconsistent results.

That statement I absolutely agree on....

levoman if you are at all concerned about copyright issues I found this interesting link.


https://en.m.wikipedia.org/wiki/Video_game_clone


Personally I don't think there's anything to be concerned about copyright.

[MCes]

Eslapion,
I explicitly agree with you about chinese market, and I can note that you implicitly agree with me that 74LS279 is not a great way to find the spikes... (for time reaction, level sensed...).

As you just done, I suggest to engegnerize a cartridge for spike testing with a dual D type Flip Flop 74(xxx)74 that can be fitted with the fastest genuine 74(xxx)74 that is possible to find from a serious seller (all of them have obviously the same pin configuration...)

[levoman]

Very interesting debate going on here!

Just a quick question for eslapion , did you ever test a clone cartridge when you were testing all these PLA substitutes?

[eslapion]

I can note that you implicitly agree with me that 74LS279 is not a great way to find the spikes... (for time reaction, level sensed...).

No I don't. Commodore 8 bit computers use either NMOS, HMOS or TTL-LS technology and anything that affects the circuits used in these computers will affect a 74LS279.

If you use faster chips such as 74F, 74S or 74AHCT circuits to detect glitches then the glitches you find may not necessarily affect a C64, C128 or VIC-20 computer. In fact, they may not be glitches at all as these computers only operate at 1/2MHz and operate with slow rise/fall time signals.

Just a quick question for eslapion , did you ever test a clone cartridge when you were testing all these PLA substitutes?

If I test a clone cartridge, I make sure I use a computer that's fully working and reliable, including the PLA.

If I test a PLA, I make sure the rest of the computer is fully working and reliable, including whatever cartridge may be plugged in.

I never test whatever part A to test whatever part B. Only one 'device under test' at a time.

This being said, I did use a Super Zaxxon clone - previously checked and known good - to check a PLA substitute.

[MCes]

If you use faster chips such as 74F, 74S or 74AHCT circuits to detect glitches then the glitches you find may not necessarily affect a C64, C128 or VIC-20 computer.

So you are agree with me: if you want detect ALL possible cause of problems you have to detect all of them: 74(fast family)74 !

If you want also detect the QUALITY of the PLA solution that you are testing (PLA replacements haven't a 30 years old technology) : I want to see if the lines sensed have any type of problem, elsewhere the replacement is not a good replacement independently if the situation will be (or not) destructive for well working of the machine.

...they may not be glitches at all....

In this case your glitcher tester has failed.

[levoman]

Thank you for your reply eslapion but did yhe cartridge fail the PLA substitute? All i am trying to find out is if the clone cartridge will actually fail PLA substitutes that are not truly compatible.

[banman]

Hi levoman,

Do you, by chance have access to an oscilloscope?

[levoman]

yes i have a 20mhz scope. But i would like to Know the answer , otherwise building a cartridge to test would be useless.

[banman]

Hi levoman,


I totally understand your concern.....



Your oscilloscope will be plenty fine for viewing the PLA glitch.

If you made the cartridge you could use your oscilloscope and do a side by side comparison of the 2.

The worst case being your version of the Zaxxon cartridge won't detect the PLA glitch.
Then again that's what science is all about, and you'll know conclusively does it work or doesn't it.

I heard it's a pretty neat game. That will be one good outcome in either case. I would very much like to build your version of the Zaxxon cartridge.


By the way did you check your PM?

[levoman]

Thanks for the PM banman.

My postulation is this :-
The original circuit had two different roms ( i assume mask or fuse proms ) with different delays with a '74 to change address and it has been packed into one rom with one delay , also extra logic has been put in to solve the single prom addressing issue ( adding more delays ) . Will it still do the job? or would it be better to create a cartridge with 1 8k rom and one 32k rom and stick with the original circuit so that both can be on board and still be link selected for normal/super.

[eslapion]

if you use a genuine 74S74 the nS become 6,
and if you use a genuine 74F74 the nS become 4.

If you use faster chips such as 74F, 74S or 74AHCT circuits to detect glitches then ... they may not be glitches at all ...

Then...

In this case your glitcher tester has failed.

Nope. A 74F74 is designed for signals used in faster computers.

If you want also detect the QUALITY of the PLA solution that you are testing (PLA replacements haven't a 30 years old technology) : I want to see if the lines sensed have any type of problem, elsewhere the replacement is not a good replacement independently if the situation will be (or not) destructive for well working of the machine.

Then you should use an oscilloscope.

Thank you for your reply eslapion but did yhe cartridge fail the PLA substitute? All i am trying to find out is if the clone cartridge will actually fail PLA substitutes that are not truly compatible.

The Super Zaxxon clone is just as good as the real cart at detecting problems with a PLA. It works the same way as the genuine cart except for the way the data is arranged in the EPROM and it uses the same chip to perform the same function. Therefore, it has exactly the same type of sensibility.

[levoman]

I agree with you eslapion , however if you said that the speed rating of the logic is not related to inputs but how fast an output changes related to the input. Propogation delay!

[MCes]

I close here my possibility to help more then how I already have done, I'm reading too much nonsense concept and Also into the "lockdown" I have more interesting activities to do that write more about this.

I wrote what I wonted to tell, anybody is free to ignore (or not) my words.

Bye!

[eslapion]

I agree with you eslapion , however if you said that the speed rating of the logic is not related to inputs but how fast an output changes related to the input. Propogation delay!

There are 2 different aspects you must consider when deciding if a particular type of technology chip is adequate for a particular digital system.

The propagation delay is one of them, the other is the slew rate. For example, PLAnkton works well in a C64 which is decades older than the XC9536XL used on PLAnkton because this particular IC has a slew rate limiter build-in. In other words, you can forcefully slow down the rise/fall time of the outputs so they are compatible with older chips and the type of copper traces (larger than modern ones) found on older circuit boards.

If you feed a signal with a low slew rate to a circuit such as a 74F which is designed for faster machines, it could interpret a very small amount of noise as transitions between low and high logic states.

The XC9536XL also has hysteresis on all inputs so it's not sensitive to noise whether it receives a signal with a high or low slew rate.

See: https://en.wikipedia.org/wiki/Slew_rate
and
https://en.wikipedia.org/wiki/Schmitt_trigger

Using inadequate technology can also produce noise of a wide range of frequency noise and cause unpredictable erratic behaviors.

https://interferencetechnology.com/what ... generated/

I close here my possibility to help more then how I already have done, I'm reading too much nonsense concept and Also into the "lockdown" I have more interesting activities to do that write more about this.

Read this and it will all become clear: https://www.microsemi.com/document-port ... t-app-note

See figure 3

if an input signal becomes too slow (i.e.low slew rate), then noise around the (any fast circuit) input voltage
threshold can cause multiple state changes

The slew rate of signals coming from NMOS and TTL-LS circuits is too low for 74F technology ICs so you have exactly the same problem.

[MCes]

We are speaking about a circuit that has to find all spikes onto 2 lines, without reasons the lines have to not change state, otherwise could be generated a problem.
It implies that after the reset routine, during the test time we must note that the state does not change, so the correct output transition between the 2 logic states (H to L and then L to H) will having a slew rate of ZERO: for a corret behaviour no transition have to happen!

if an input signal becomes too slow (i.e.low slew rate), then noise around the (any fast circuit) input voltage
threshold can cause multiple state changes

The slew rate of signals coming from NMOS and TTL-LS circuits is too low for 74F technology ICs so you have exactly the same problem.

So:
you are telling that IF a potential problematic spike is generated (spike = 2 fast state changing= slaw rates presence) my solution will find all of them, also the fast one.
you are also telling that without a spike (=without states changing =without presences of slew rates) my solution has not problem.

I have a news for you:
This is exactly what is wanted!

(Don't worry, plankton's behavior isn't mentioned in this discussion ...)

Another time I have to read a lot of fake technical words used only to mask the situation of the conversation.

I'm sorry only for who will be fished....
https://www.youtube.com/watch?v=E49YsPLWmbE

Good night
http://sleepingelephant.com/ipw-web/bul ... f=1&t=9220