I've never been so thankful and so mad (peevish, really) at the same time! (well, not in at least, like a whole week, or so!). I just found the RES pin/sig in the MOS technology manuals (circa 1976). Thanks Garth ... I was all prepared to brag about my good work ethic... oh well ... next time I have a silly question, I will just go find the answer myself.

I think the next step for me is to shop for EEPROM, pick one, arbitrarily, based on what I think, maybe, I might need, and then price it, download the datasheet, and study the pin diagram, logic levels and signal speed i.e. freq.

Actually, I do have one more stupid question (for my computer project; not for the board game); Can one use multiple oscillators on the same board? That wouldn't make sense would it? Or , maybe, if some peripheral chip was running at higher freq., a different crystal might run that; values might be stored in that in the peripheral chip, to be accessed later by the uproc.? is that possible? or would the universe align against me? spurious charge? spurious chip activities, etc.?

Just my paranoia, about things unknown. I think I have seen frequency divider circuits in one of my Forest Mimms radioshack handbooks? And my Wakerley Digital design textbook had a couple pages on fan out/clock fan-out. hmmmmm.....

Also, I think it was your site that told me how important IRQ and NMI are, and the more I look at the pin diagrams and the manuals, the more I see that this is true. I almost left these out of the 6502-opoly game. Can't do that now.

Also, I read a bit of that 1976 manual (was it the N-MOS version?Must I flip around, in my head, any of the voltage values....probably not... I think they will maintain for the CMOS/modern version W65C816?), and I noted the importance of keeping RES low, on uproc and peripheral chips (and presumably RAMS and ROMS?) so that the uproc isn't "confused" while the system starts up. Especially important for input/output devices, while clock and output lines stabilize.

OK, I just checked my pin diagrams; RAM doesn't seem to have a RES pin, and voice synthesizer chips will probably interface through I/O IC. Anyhoo, SP0256 has a RESET and a SBYRES pin, and SC-01 seems to have neither. extransoues information.

Cheers, thanks for the help. I will try to not screw-up the game (or my chips/computer!). It should be fun and informative when I am done.

It is quite common to have more than one oscillator on a board, for example to run the processor at one speed, like 4MHz, 10MHz, whatever, and then the UART may need 1.8432 MHz or other odd frequency to derive accurate baud rates.

The CMOS 6502 (ie, 65c02) wasn't out until about 1982. I think the '816 came out about the same time.

I would look at the EEPROM data sheets before choosing one, to make sure it runs on the power supply voltage you want, that it's fast enough for what you want, make sure it has a parallel interface (serial is fine for after the computer is booted up and can run software to access the serial device), etc..

I don't know what you mean by "uproc." That's a term I've never heard. The I/O ICs usually have a reset input to initialize the registers. It's often not a big deal though, because the processor comes out of reset with the interrupt-disable bit set so interrupts from the I/O ICs won't affect anything, and then the processor can set the I/O ICs up in software before clearing the interrupt-disable bit (I). In the case of NMI which is edge-sensitive and not level-sensitive, you might have to be a bit careful because of the non-maskable nature of it, but if the IC produces an NMI before the processor sets up the IC, the edge will probably be before the processor comes out of reset anyway, and get ignored. In any case, don't let RST\ come up until after the power supply and clocks are stable.

_________________
http://WilsonMinesCo.com/ lots of 6502 resources
The "second front page" is http://wilsonminesco.com/links.html .
What's an additional VIA among friends, anyhow?

I presume by "uproc." he means microprocessor, the way it's sometimes abbreviated "uP".

KC9UDX: That's why I post on the "Noob" page. You are right. "uP." is better than "uproc.". I have been poking around on your other pages and I am realizing this is a BIG world! You folks are neck deep, and I am dipping a toe into the pool.

Looking for EEPROM today, and thought I would poke around and look what part #'s and manufacturers you guys use. I'll probably pop on over to the GARTH primer and a few other sites recommended to me upon my joining your discussions a week ago (or thereabouts). I won't forget the 6502-o-poly (and maybe a x86-o-poly; though you purist won't want to hear that, and I just might "screw up the story" of that device too! Care must be taken when trying to understand someone else's world, whether its 6502 or x86!).

The family of 65xx is a somewhat amazing phenomena to me. Difficult to grasp all of the details of one uP. and its function with various other hardware, nevermind discuss the whole family! I must simplify; and become more precise in my usage.

I own the W65C816s and I should take pains to only learn its function and use and programming and interface.

Garth; thanks for your most recent post. I will cogitate before response. The post previous to that, I have thought about. Should my ROM (whether EEPROM or EPROM, no matter, for this question) cover, at the very least, those RESET values (i.e. FFFC and FFFD)? I assume so, since RAM will be empty on startup. And, so, for the simplest of programs on a 65xx system (Did I note that the 65816 has different reset vectors?), one might place two addresses at the reset vectors, in ROM) and point to a small program in ROM. A simple program, not unlike the one I wrote (borrowed from Zaks) for ADC?

I checked it with the 6502 virtual (on-line) resource, and it produced valid code; BUT, when I posted to this site, I mapped it into 0000 region and put it in machine code (immediate addressing instead of zero page, which, in retrospect, would might have been more frugal? or maybe not?). Next time I (1) write some code and (2) put it through the emulator, you can bet I will be thinking about RAM, ROM, and RES/startup.

Thanks again for you time, and your forum; you needn't reply unless you sense a "train-wreck" in my logic. (P.S. at first glance, your most recent post has good suggestions regarding oscillators and power supply of ROMS. Generally, several different oscillators might be used (I should have looked at any old circuit board to verify that this is true! Silly of me not to know!). But good point, I hadn't really thought much about, was keeping all the power supplies the same; I guess I sort of thought about it, since I know several instantiations of CMOS and/or miniaturization, have seen logic levels change, and thus, power consumption and operating voltage, etc. .... thanks for the reminder... and I will read that post again and think some more about it.)

The 65816 (Eyes and Lichty) manual mentions 00FFFC and 00FFFD; but this is wrapped up in something I don't quite understand yet; native mode and emulation mode.

Also, I note, other interrupts have dedicated addresses, and I should probably endeavor to understand them too.

I think I will go to your (GARTH) primer now, and EEPROM shopping second; The ICs on the address bus lines (74139s? I think?) are interesting and unknown to me? I have seen some references in other people's posts to a "memory map", and perhaps this decoding circuitry (74139s?) will let me understand what my map should look like.

(hmmm... I guess I am just now noticing the use of the "bank" term; just a few days ago "page" took on a new meaning! I guess the two leading Hex digits are "banks"? Bank 00: Page FF: FC,D).

To reply to things without quoting it all:

There are exceptions, but when you're starting out, just go with the assumption that page 0 and page 1 are always RAM, and at least page FF is always ROM.

FFFC and FFFC hold the reset vector on both the 6502 and the '816, the reason being that the '816 boots up in 6502-emulation mode, so it can be used like a 6502. This was done for backward compatibility, so computers like the Apple IIgs could be used like a regular Apple II if you didn't have software that's particular to the '816. There is one reset vector, and it takes two bytes, since it's a 16-bit address. (On the '816 which has 24 address bits, the reset vector always points to a reset routine in bank 0, so there's no need to specify the bank.) The reset routine typically does the initial setup of I/O, variables, etc. and then dumps you into the initial program. That could be the real program the computer was made to run if it's for a dedicated purpose and it's in ROM, or it may be a BIOS to load an OS, or it may be a language the user can start using immediately to give commands without first loading anything further from mass storage, or something like that.

There is no benefit to putting code in zero page ("ZP" for short). ZP is faster for data access and also gives more addressing modes. It's kind of like having 256 bytes of virtual processor registers. If data is in ZP, accessing it is faster; but wherever the data is, putting code in ZP will not benefit anything. (Again there are exceptions, and I have put a short piece of self-modifying code in ZP for increased efficiency by making variables reside in operands of instructions where I wanted double indirection, but it's just that-- an exception.)

Regarding the 74xx138 or '139: They are often used for how they divide up the address map; but the 6502 primer shows a way that offers lower propagation delays and minimal parts count. The beginner tends to come up with fine-grained address-decoding schemes which are logically correct, but he doesn't realize that the chips are not infinitely fast, and that your maximum clock speed may be severely limited if a signal has to get through a large number of cascaded gates for whatever scheme he has in mind. There is no need for fine-grained address decoding for someone who's new to the field. Having 32KB of ROM and 16KB of RAM (or vice-versa) and setting apart 16KB space for I/O is usually more than adequate.

For EEPROM, I would use something like the 28C256. For EPROM (which requires the UV eraser), I would use something like the 27C256. (Note the second digit in both.) Both options are shown in the circuit for a basic 65c02 computer at the top of the circuit potpourri page of the 6502 primer, at http://wilsonminesco.com/6502primer/pot ... ml#BAS_CPU .

"Banks" on the '816 are 64K long. "Pages" are still 256 bytes long. There are 256 banks in the 816's 16MB address space, and there are 256 pages per bank, and 256 bytes per page. With an 8-bit bank byte, there's a total memory map of 16MB, 256 times as much as the 6502's memory map. However, you can gain a lot of benefits from the '816 even if you don't latch, decode, or use the bank byte. The 65802 (which is not in production today) was basically that, an '816 made to go into a 6502 socket, and the bank byte was not used. Still tons of benefits. Do keep an eye on the '816; but the '02 hardware will undoubtedly be simpler to start. (As far as software goes, you can initially use the '816 exactly like a 6502, and move into using the additional capabilities little by little, at your own pace.)

_________________
http://WilsonMinesCo.com/ lots of 6502 resources
The "second front page" is http://wilsonminesco.com/links.html .
What's an additional VIA among friends, anyhow?

When Garth mentioned that startup mode of the Apple II gs this is what he referred to - the 65c816 can run in "native" mode, i.e. with a 24-bit address space, additional 16-bit instructions etc., or it can run in "emulation" mode, where it tries to behave exactly as a 65C02. Then there are some intricate in-between modes where you can set up only some registers to be in 8-bit mode. And some '816 features are available even in 65C02 full "emulation" mode.

Thanks for understandable and complete answers. I am converging upon an understanding of this (these?) machines and have started thinking about PCBs. Perhaps I'll start making my pattern on my CAD program?

Have studied a bit of address decoding from my "odds and ends" computer library. Barry Brey Assembly (x86/Intel) book had some interesting pieces on address decoding, and involved SIMM and DIMM cards. It covered the 74139 and 74138 fairly completely, but with some peculiar attention to x86 architectures.

Also had a look at Marvin Dejong's "6502 microcomputer and projects" book. I forgot I even owned that book! He helped, a little, with "memory map"/Address decoding. I have a 6800/68000 textbook (motorola), an Intel programmer/developer handbook (somewhat useless, it seemed, at first glance, to "Hardware People"), and a hardware design book--I forgot to check this one thoroughly-- by Wakerly; these should provide some insight too.

Interesting to note that, though different uP. chips and architectures (i.e. computer designs/capabilities) its all the same stuff; just attached in different ways and in different amounts.

" (Again there are exceptions, and I have put a short piece of self-modifying code in ZP for increased efficiency by making variables reside in operands of instructions where I wanted double indirection, but it's just that-- an exception.)"

Intriguing notion; I will keep it in mind for when it applies to my coding needs/abilities. For now, consider it memorized "Zero page is for fast read/write data".

I'm not sure I like absolute addressing, as it seems the use of two bytes to specify one 16 bit address (or, in the 816, 3 bytes) is inefficient. I will avoid it if I can; immediate addressing seems best to me... is this a bias of mine? probably, since I have no experience coding.

I am thinking I will do much hand coding; much assembly writing, followed by machine language verification. I know what you will say; "inefficient", but I prefer frugal, fastidious code, and a "nuts and bolts" knowledge of what I am doing.

As I might have indicated earlier, I tried to learn x86 from an old pdf "Art of Assembly" and Brey's book, and found both to be overemphasizing "MASM" and "TASM" and a hundred other things that were inaccessible to me (i.e. wouldn't run on my machine, or did not--for the uninitiated--have a usable interface).

Also, I now have the time to slow down and think about these things. Its been on my "bucket list" to learn how these magical machines function, and I feel a bit closer to that goal today/this-past-week (only 35! I have time to learn the rest!).

"With an 8-bit bank byte" ... and Peter Piper picked a peck ... too.

Thanks for EEPROM and EPROM numbers. Still didn't get to your primer, but I have been reading the various manuals and pdfs, pin diagrams and mnemonics. Found a few more links on the forum and got a few more "Assembly 6502" books, (mostly KIM manuals, but one called "Machine code for Humans"; a few books and articles on "6502 and commodore 64") and I will look at those when I get a chance.

I think I should probably design for the use of the 65816. Or, at least, design with expansion in mind. I don't imagine I'll make a second board; I suppose I could breadboard (a 6502 design), but my previous experience of not knowing the "breadboard frequency barrier", and mistakenly trying to surpass it, and getting jittery performance as a result, sours that idea. I should "build to last" and build with expansion in mind. Maybe, just buy extra chips, and have the decoder circuitry already in place on the PCB board; then just add chips and code and voila! upgrade to '816!

I wonder what my need for I/O truly is. I suppose a place to type and see what's stored in memory is useful, but I have no delusions of fine graphics or "websurfing" or even word processing. I have scientific applications in mind, so, suffice it to say, I/O will still be a focus of mine. Suffice it to say, phoneme concatenation will occur!

I saw on ebay some USB EEPROM writers. They all seem to use the word "flash", though perhaps this is marketing (designed to sell to a generation that thinks "USB-jack" = "Flash"?) and has no real meaning about the technology contained on the EEPROM chip? If not that way, how does one program an EEPROM? I suppose I can consult the Wakerly book again ...

Ooops, I've run at the mouth.... errr ... keyboiard again.

Cheers, and thanks

It's your choice of course, but I think it would be good for you to build it first on something that easier to make corrections on, until you understand things better. I like wire-wrap. There's a chapter answering wire-wrap questions and doubts in the 6502 primer, at http://wilsonminesco.com/6502primer/WireWrap.html . (Someone plagiarized it and translated it into Russian too.)


Be sure to see the address-decoding page of the 6502 primer, at http://wilsonminesco.com/6502primer/addr_decoding.html .


Absolute addressing is the only way to do a lot of things. Immediate is when the operand of and instruction is the data itself, rather than the address of the data location. For example LDA #$71 (ie, immediate addressing mode) puts $71 (that very value) in the accumulator. It would not normally make any sense to have a store-immediate instruction, especially if the instruction is in ROM, or an LSR immediate, etc.. LDA $71 (ie, ZP addressing mode) puts the contents of address $71 (in ZP) into the accumulator. LDA $3871 (ie, absolute addressing mode) puts the contents of address $3871 in the accumulator. Then of course there are the indirects, the indexeds, etc..


As I put in the article, Assembly Language: Still Relevant Today, in a microprocessor class in 1982, we had to learn a processor's internal registers and its way of doing things and its instruction set, and write programs and assemble them by hand (on paper) and then type the machine code into the computer to try it. There was an assembler onboard, but the teacher wanted us to really understand the lowest levels of the machine, including the assembly process, to layer knowledge upon knowledge, instead of starting directly at higher language levels that would have insulated us from the nitty-gritty and kept us from learning what really goes on down there. That was good, and I'm glad I had that exposure at the start. It's also good to move on, after you get a good grip on those basics. I like to promote macros a lot in assembly language. They are not the place for the beginner to start; but it would be foolish to commit to never touch them when you're ready to and need to use them to improve your programming.


Do go through the primer. Its pages are organized in logical steps. I wrote it because much of that material was not covered adequately (or at all) anywhere on the web. It is mostly about designing and building your computer though, and not much about software. There are loads of books available for free download on beginning 6502 programming; unfortunately, most of them are on the NMOS 6502 and completely ignore the CMOS's extra instructions and addressing modes. The Eyes & Liechty manual on WDC's site is excellent. I'm not aware of any better manual for this.


It is nice that the pinouts are so similar, so you can make a board with some jumpers to accommodate either one.


Follow the recommendations in the 6502 primer chapter on construction for good AC performance and the links there, including to forum topics, and you won't have any problems.


The circuit potpourri page of the 6502 primer has a lot of helps specifically for scientific applications. This is what I use my workbench computer for. See my "Introduce Yourself" post for how I got there, and that really is what I do, the first post of the topic.


They should specify what they can program. The EEPROM family you're mostly interested starts with "28" (possibly after some leading letters telling who the manufacturer is). Flash is a type of EEPROM that erases and programs portions in whole blocks at a time to save time. It may be nearly as fast to program 1K bytes at a time as to program a single byte. This is extra helpful when you're regularly writing whole files.

_________________
http://WilsonMinesCo.com/ lots of 6502 resources
The "second front page" is http://wilsonminesco.com/links.html .
What's an additional VIA among friends, anyhow?

OK, here are a few more picture; and maybe an outline of what is next.

I have decided to use die photos (colorful ones; vintage ones!) as the backs of the "community chest" and "chance" cards.

If anybody out there knows--for a fact--of some open-source (i.e. free to use, under all circumstances, even commercial and without attribution) die-shot photos of FAMOUS microprocessors, be sure to send me a link (in a PM if necessary or wise) or file. I have found wonderful photos of every chip I could ever want or need, under most circumstances (light field, dark field, DPCM Differential? Phase? Contrast? Microscopy?).

A few years back, I found this "Silicon Zoo" webpage; but now there are two. The fsu.edu page is the one I knew first, and the cartoons are the best! Chip art is a wonderful thing!

MY copy of Technopoly will use ANY AND ALL IMAGES THAT I CAN GLEAN, as a single version of technopoly is a WORK OF ART AND MY FREEDOM OF EXPRESSION CANNOT AND SHALL NOT BE STIFLED! However, if I ever make one--or 10,000 or 10 million--of these for sale, it would be nice to have permission, so I don't get sued. And, better even than permission, is to just use those that are already acknowledged as free.

http://micro.magnet.fsu.edu/galleria/index.html (scroll down; left side of webpage has cartoon names, click on some of those; also see the "chip shots" at this link; http://micro.magnet.fsu.edu/chipshots/index.html)

The newer webpage; or rather, the one I have just recently found, is http://siliconzoo.org/ , and the chip art/photos are found on the right hand side of the screen, four links (RFID, RFID etc., MegaMOS, etc.) or so, with many simple chip photos. I don't know about the legal status of these, but I probably won't use too many of these photos.

Maybe I will call "Comm Chest", "Vintage Processors" and "Chance" will be called "The Silicon Zoo". That seems smart enough for today. I have been creative enough for the week!

Here are the photos of the corner squares. If anybody wants to help, I need a suggestion for the "Free Parking" , i..e "Free Programming" square. So, players of the game technopoloy will have to obtain either (1) a ROM-chip (i.e. card, from "chance" or "comm chest") or (2) Land on "Free Programming" in order to initiate their program.

The game is designed (rather, I intend to design it) so that one must play for an hour or two before a viable program can be made. Hopefully, after the first easy program, the game can proceed in such a fashion that one or two other programs can be made, quickly, before people get bored and wander off (I figure 3 hours is tops, for holding 4 or 5 peoples' collective attention spans).

Hopefully, in this way, simple addition, subtraction, multiplication and division, can be learned. There will be templates, to follow, but ample room for free expression and creativity. After playing the game a few times, perhaps people can even get good and do their own, tailor-made programs?! If they lack that sort of creativity, inclusion of some advanced program templates might also help keep the players' interest.

So, is there a vintage" or "classic" "ROM programmer device" that I could get a picture of? Something very "1980's" that every smart technophile of a particular age will immediately recognize? that is what I need, and I need a free photo (or, at least, one simple enough for me to trace and make into my own "original artwork"). That way, when somebody lands on free parking, they won't put money in the meter, but, instead, "borrow a friends ROM programmer", thus "Free Programming" square!

Also note the attempt to build RAM envelopes, to slide the data and code into, and, also the section of ROM with the reset vector. I might have to alter my ideas again, and make the "RAM envelopes" into "ROM envelopes" since the programs will not be complex enough to really use a lot of RAM.

I'll have to think it through a little more; I will need ROM envelopes, to hold the program; RAM for the temporary data; and then maybe an Input/Output section, to show people how to use the microprocessor get the answer "out of the computer".

The envelopes I made suck; they stick together. Too much glue. I think I will buy little 2 in X 3 inch manilla envelopes (like the paperboy uses when its time to pay him for delivery). That will work better.

Maybe I could transistorize and LED the game? Get some pole switches, surplus; bag of 100 QTy or so, and then have lots of fun throwing the switches to make the program run (incidentally, I have given my TTL computer (7400 series, 74181 and etc.!) project to PandaPro and we are going to try to make it run, this time! I was bummed by this initial failure, some ten years ago!). Probably better than a "bank of envelopes".

(oops! OK, so you can't see the envelope bank" for RAM and/or ROM? It is off the picture, I think. Picture 16 (next post), you can just see the edge of white, blue and red paper that will have sleeves to represent memory addresses; Op-code cards, one chosen every turn (before or after rolling dice? I dunno?) will slide into these spots when the player is ready to run a program.

OK, that's enough for today. I "run at the mouth" (as usual); i.e. "run at the keyboard"....

Three great places for die photos:
Pauli Rautakorpi on wikimedia (all good licenses):
https://commons.wikimedia.org/wiki/User ... _this_page

Visual 6502 project (not just 6502!) for example (but I'm not sure about licensing...)
https://www.google.co.uk/search?q=visua ... t&tbm=isch

Also John McMaster's site, but again I'm not sure about licensing:
https://www.google.co.uk/search?q=site: ... g&tbm=isch

Oh, and another, zeptobars
https://www.google.co.uk/search?q=zeptobars&tbm=isch

Please notice, school is a place where you are rewarded. ten times the value of the dice is earned, while stuck in school. I intend to design the game so that the players going around the board have trouble making money; they buy an invention or theory or discovery or two, but then are cash poor; being in school early on allows (10 a chance to make money and (2) an opportunity to buy the properties that the other players can't afford. (this is a lot like real life; except college students are generally poorer than people on the outside; i.e. "established people".).

As time goes on, the "people on the outside" make more money, and the properties get more expensive, so, being in school is less advantageous. Luckily, doubles gets you out of jail,... errr, I mean, school!.

Here is pic 16, I think you can see the RAM/ROM envelopes on the right hand side; the left hand side ( I JUST FIGURED IT OUT!); the left hand side I was reserving for stack, with right hand side being regular RAM; Now, i think it would be better for right hand side envelopes to be the ROM, beginning with the reset vector, and a few slots beneath for the program, and then, the left hand side envelops can be the RAM, with a short segment of stack, there, too!

Good ideas, as I write!

Cheers!

THANK YOU, THANK YOU, THANK YOU, Ed.

Paul R.'s page is PERFECT!

And it would be easy to attribute to him, since he is "one-stop-shopping" (a brief note at the bottom of a page somewhere and I would be legally covered! I.E. "All die shots are the work of Paul R. and not affiliated with the makers of technopoly , etc. etc. legalese, blah blha blah!" ).

Also, regarding the 6502; since IT IS, after all, 6-5-0-2-opoly, I was thinking of using a large die shot, the size of the whole board, and making it a "watermark image", i.e. pastel. And that way, the whole board for the board game would be one big 6502 die shot. I will look for one that is (1) high res and (2) free to use, alter, share, and/or distribute.

Again, if attribution is all the photographer wants, since it is just one photo (6502 high res photo) I could easily meet that particular requirement.

I am only trying to avoid making a huge list of a million different photographers. Thanks again; your help is/was invaluable/incalculable!

This will help me get the professional and/or artistic vision that I want, and still keep me within the "social-circle" of quasi-legal, quasi-respectable people.

(I guess it is delusions of grandeur, to think anybody else would want a 6-5-0-2-opoly; but since I also want the game to teach the history of computing, I could really make different versions, too! x86. 6800. Z-80, etc. ; Just change the die shot in the background, and a few of the particulars of the instruction set and buses and support/memory chips; but one thing a t a time! I'll get the one game done, and then see if it is adaptable to other vintage tech!)

Here you go, CC-BY-SA licensed:
https://commons.wikimedia.org/wiki/File ... S_6502.jpg
(By "org" here on this forum. It's a 6502 die layout overlaid with transistor-level schematic.)