Preserving Great Tech For Posterity — the 6502

trebonian writes "For great old hardware products like the MOS 6502 (used in the Apple II, the C64, the Nintendo NES), the details of the designs have been lost or forgotten. While there have been great efforts to reverse engineer the 6502 from the outside, there has not been the hardware equivalent of the source code — until now. As Russell Cox states: 'A team of three people accumulated a bunch of 6502 chips, applied sulfuric acid to them to strip the casing and expose the actual chips, used a high-resolution photomicroscope to scan the chips, applied computer graphics techniques to build a vector representation of the chip, and finally derived from the vector form what amounts to the circuit diagram of the chip: a list of all 3,510 transistors with inputs, outputs, and what they're connected to. Combining that with a fairly generic (and, as these things go, trivial) "transistor circuit" simulator written in JavaScript and some HTML5 goodness, they created an animated 6502 web page that lets you watch the voltages race around the chip as it executes. For more, see their web site visual6502.org.'"

Comments

Not to split hairs. . .

[DancesWithRobots]

But wasn't the C64 processor a 6510? I could be wrong.

Re:Not to split hairs. . .

[Amorymeltzer]

Yes, but the difference ain't much.

Re:Not to split hairs. . .

[Anonymous Coward]

Yes, Jeri Ellsworth made an FPGA of the C64 a while ago, but they're very similar.

Re:Not to split hairs. . .

[Anonymous Coward]

But wasn't the C64 processor a 6510? I could be wrong.

Same CPU "core", with some additional peripherals. Kind of like a 6502 together with part of a 6522. The thing is, the C64, didn't really use the stuff in the 6510. Looks like they were going to, and then changed their minds.

Re:Not to split hairs. . .

[Arthur Grumbine]

Yes, but the difference ain't much.

According to my rough calculations the difference is at least two more than the difference between an 8080 and an 8086.

Re:Not to split hairs. . .

[hlavac]

Not true. In C64, the on-CPU 6510 specific I/O accessible at addresses $00 , $01 has been used for switching the Basic ROM, character ROM, attribute color memory and I/O mapped area on/off in the CPU address space, allowing you to access to the parts of RAM normally obscured by them.

Re:Not to split hairs. . .

[djdarling]

The first thing I did when I read the article was search for 6510 to see if someone flagged this. :) Being a former C64 owner this is what I remember as well. IIRC, the 6502 was in the Commodore family though -- in the Vic-20.

Re:Not to split hairs. . .

[stretch0611]

Yes, The C64 used a 6510. Its predecessor, the Vic-20 used a 6502. FWIW, I believe the C128 used a 2mhz 8502 which was compatible.

The brand-new http://www.commodoreusa.net/CUSA_C64.aspxCommodore64 uses a Dual Core 525 Atom processor and Nvidia Ion2 graphics chipset.

Re:Not to split hairs. . .

[commodore64_love]

6510 has some extra registers to control external peripherals like disk drives and cassete recorders. It's a difference but nothing that really matters. In fact the original Pitfall was ported directly from the 6502 Atari to the 6510 Commodore with few changes.

TRIVIA: The 6502 (and its variants) were owned by the COMMODORE Semiconductor Group. It's why the commodore computers were cheaper than any other computer - they could get the CPU for free and charge the competition (Atari, Apple) more. The Commodore 6502 and its 16 bit upgrade(65816) were used for:

Apple I/II
Apple IIgs (65816)
Atari 2600 Video computer system
Atari 5200 Supersystem
Atari 400/800/1200 computers
Nintendo ES
Super Nintendo (65816)
Commodore PET/VIC/64/128
and a whole bunch of lesser-known items like calculators, watches, and so forth. Eventually most of these companies moved from the Commodore 6502 to the Motorola 68000 (32 bit) or PowerPC (also 32 bit).

Re:Not to split hairs. . .

[toejam13]

That's actually my problem with the MOS 6510 - it was TOO similar to the MOS 6502. There were a bunch of bugs in the old 6502 core that had been around since it was released: weirdness with undefined op codes, wrapping index with direct page ops, unknown state of D flag after interrupt, unable to BRK after interrupt and so on.

The WDC 65C02 fixed all of these issues and then some. In fact, once you've gotten used to programming on a bug fixed 6502 variant (WDC 65C02, Rockwell 65C02S, Hudson 6580, WDC 65816 and CSG 65CE02), it is really painful to go back.

I would have really killed for one of the Rockwell 65C02 chips in the C64. In addition to the bug fixes, extra ops (DEA/INA, PHX/PLX, PHY/PLY, TSB/TRB), it also allowed you to relocate zero page and stack page to places other than page $00 and $01. Imagine being able to write your own interrupt routines using your own private stack and liberal use of zero page ops. GEOS would have run light years faster because you wouldn't need to save and restore the contents of the stack on a task switch.

Re:Not to split hairs. . .

[badkarmadayaccount]

I wonder if such optimizations would be applicable to modern CPUs.

Painful

[WolphFang]

Oh what a pain it was to program that chip when I had a VIC-20!

Re:Painful

[oldhack]

Way simpler than 8080/Z80.

Re:Painful

[morgan_greywolf]

I wrote a couple of non-trivial programs assembler programs on the TRS-80, which uses the Zilog Z-80 processor at 1.78 mHz. Wasn't too bad, it has more registers than 6502. The only bad thing was the lack of a multiply instruction.

Re:Painful

[biryokumaru]

Wow, I knew old computers were a lot slower than modern ones, but 1.78 thousandths of a cycle per second? Jeez!

Re:Painful

[oldhack]

6502 made up for lack of registers with extra addressing modes, but they were relatively uniform across the instruction set, making the instruction set simpler and more orthogonal, whereas many x80 instructions designate specific (and different) registers for operand, making the instruction set rather ugly. You could see the nasty root of x86 instruction set even then, although, as you said, it wasn't so bad because 8080/z80 was much simpler.

6502 didn't have multiply either, and it's true z80 was more powerful in several ways.

Re:Painful

[Opportunist]

Dude, balancing a hyperactive Husky on a flagpole was simpler than programming the 8080.

And more rewarding too.

Re:Painful

[morgan_greywolf]

I agree that the 6502 was more orthogonal -- I wrote several assembler programs on the Apple II as well. OTOH, I never thought the x86 instruction set was that bad. For better or worse, it appears that we're stuck with it, anyhow. :)

Re:Painful

[Anonymous Coward]

Funny story, I was working for a major semi manufacturer writing tests for the PCI controller integrated on a processor. We had an external clock for the PCI supplied by a signal generator.
The signal generator lost power during a move to another bench and afterward all test failed. Looking at a logic analyzer, set to sample at the expected rate showed all signals static.
I set a trigger for the first transition and finally, 33 mS into the run I got a trigger. The signal generator was set to mHz, Not MHz. The display on the signal generator used a very poor 5X7 font
and mega and mille were indistinguishable.

Re:Painful

[Anonymous Coward]

Universal truth, Any text following the phrase "Funny story" is 100% guaranteed to be as funny as soya-milk is tasty.

Re:Painful

[PatPending]

It's like most things--one gets used to it.

I spent several years writing nothing but 8080 assembly language in the early 1980s using an assembler running on a DEC VMS computer. Pretty much all I had in my office then was a phone, a couple of reference books from Intel, a set of hardward schematics, and a VT-100 terminal. I got damn good at writing "structured" assembly code (lots of single-purpose functions). My last project there amounted to 9k of 8080 code written in three weeks for a real-time monitoring system for industrial machines.

Re:Painful

[Zedrick]

What assembler code did you do for the Vic 20 that was so difficult? I don't have much experience of pure 6502-coding myself, but I've done plenty of 6510-coding on the C64 and don't find it very painful at all.

Links, please!

Re:Painful

[Pflipp]

And we were happy!!

Re:Painful

[TheRaven64]

The 6502 is also interesting from an historical perspective because, along with Berkeley RISC, it was one of the inspirations for the ARM architecture. ARM was created by a small team at Acorn to replace the 6502 in their product lines. Being able to easily port code from the 6502 was considered a requirement, but the design would probably have been similar to the 6502 anyway because that was the architecture that the team was most familiar with.

ARM isn't a simple extension to the 6502, as the 8086 was to the 8080, but it was designed to be the chip that people using the 6502 wished they were using (within the constraints of what was actually possible in the '80s).

Re:Painful

[muchtooold]

And we were happy!!

You were lucky!! I know it's a bit late but somebody moderate immediate parent up - funny - immediately

Re:Painful

[Anonymous Coward]

1.78 thousandths of a cycle per second?

Yes... but they put that cycle on your desk. Before that, my 'computer time' included physically mailing the punchcards to Toronto.

Re:Painful

[Joce640k]

6510 is exactly the same as 6502 from a programmers point of view.

Re:Painful

[Dr. Hok]

The code itself was not usually difficult, but writing it was.

Assembly coding for the VIC-20 went roughly like this: First you wrote the program on a piece of paper (what is a "text editor"?), then you tried to figure out what the instructions were in decimal, then you wrote a BASIC program which POKEd the values into memory and executed it with a SYS. Oh yes, and first you had to think of reducing the memory available to the BASIC interpreter with another two well-dosed POKEs, which I knew by heart but forgot over the past decades. Otherwise the basic interpreter would overwrite your assembly code and make it crash.

10 POKE 3657,123
20 POKE 3568,67
[...]
4500 SYS 3567

At least that was how I wrote machine code until I found out there is some software called "assembler" with some kind of editor.

Re:Painful

[19thNervousBreakdown]

Most things are like balancing a hyperactive Husky on a flagpole?

Re:Painful

[Anonymous Coward]

Another big advantage was that the 6502 didn't divide the clock down to create a slower instruction cycle, so 1MHz got you about 300,000 IPS. Not needing to distribute fast clocks all over the board made it much easier to work with.

Having coded in 6502 for quite a while, it's amazing the tricks that have been found to compensate for the simplified instruction set.

Re:Painful

[tehcyder]

Universal truth, Any text following the phrase "Funny story" is 100% guaranteed to be as funny as soya-milk is tasty.

It's also a universal truth that there are no universal truths.

Re:Painful

[WolphFang]

Heh... I extended BASIC with graphics instructions, including line drawing functions (remember... no multiply or divide... remember HARDWIRED 256 byte stack.. only one "var" to do math in...). In 4K. Oh.. and I loved 'turbotape" :)

Re:Painful

[mcgrew]

I have to disagree. I had a TS-100 that used a Z-80 and a TRS-80 MC10 that used a 6502. The Z-80 was far easier to write and hand-assemble machine code for.

Not all of that was the chip, of course. Some of the TS-1000's ease of use came from its structural simplicity (to read the keyboard or write to the screen, you simply used the address space as if it were memory).

Re:Painful

[mcgrew]

The only bad thing was the lack of a multiply instruction

Not a problem, there were shift instructions and add instructions. All you had to do to multiply A*B was to put in a small loop that added A and decrimented B if B was >0. You could speed it up using shifts and adds.

Re:Painful

[Megane]

Nope, the MC-10 used a 6803, which was a 6800 variant.

Re:Painful

[FreonTrip]

Save that bit about death and taxes. Prove me wrong, people - I'd love to find release from one or the other!

Re:Painful

[FreonTrip]

Happy? Hell, we were grateful!

Re:Painful

[mcgrew]

That's how I wrote machine code on the TS-1000, except I skipped the BASIC program and its POKES and did the poking by hand. I first put in a REM statement, poked the values in, put "save" after the REM and zeroed out the "rem" before the SAVE.

This way, as soon as the program was loaded from tape it ran, and the user didn't even have to enter a RUN statement like you usually had to do with a BASIC program.

Well, ok, there was a BASIC program.
10 REM
20 POKE n, 0
30 SAVE

Where n was the address holding the REM token.

AFAIK there was no assembler for the TS-1000. IIRC my fully functional (and fun!) battle tanks game weighed in at less than 500 bytes. I was pretty proud of myself!

Re:Painful

[I8TheWorm]

That TS-100 was my first computer, and looking back, had it not been for the simplicity of programming the Z80 relative to both the 6502 and the 8080, I probably wouldn't be a programmer today (I was 14 at the time).

Re:Painful

[Rockoon]

..and the 8086 wasnt designed to be compatible with the 8080 either.. well, sort of..

It was designed to be source-code compatible (assembler source code) so that an assembler for the 8086/8088 when given 8080 code, could still produce 1 opcode per 8080 instruction.

Re:Painful

[mcgrew]

Hmmm, I believe you're correct. Been a while.

Re:Painful

[agbinfo]

That's like saying that ADD wasn't necessary because you could use AND and NOT and some some conditional loops. Yes it works but it's still a problem. It requires more instructions, it's is slower and it's harder to test for overflow conditions.

Re:Painful

[rayd75]

Fortunately, I had the Epyx Fast Load cartridge as a semi-permanent part of my C64 setup so my "assembly" was done using its debugger to hand-key opcodes in hex. I probably never broke a few K in code but I did write a terminal emulator that would keep up with a 2400 baud (and probbaly faster) modem. To this day, I judge technical manuals by the Commodore 64 Programmer's Reference Guide. It was like the O'reilly In a Nutshell book of its day... Easy to understand and to the point with no word count fluff to derail your understanding.

Re:Painful

[bored]

Even on some modern arches there are a fair number of cases where shift/add combinations are faster than using the hardware multiplier. Also take a look at GNU MP.

Re:Painful

[bored]

Speed it up?

Ha, you change the algorithm from O(2^N) to O(N) (where N=number of bits in the number) by using shift, instead of the first grade method of multiplication (repeat sums). No one in their right mind does repeated sums.

BTW: when one of the values is a constant you obviously don't need the loop, instead the compiler just emits a couple shifts and adds right into the instruction stream. This still happens today, as some arches are faster doing a shift/add for small constant values, than invoking the full blow multiplier.

Also, Big Number libraries often still do shift/add for certain operations/variable sizes because it ends up being faster than calling the hardware multiplier. I've got a little C++ template bignum library of my own creation that uses the multiplier for small bignums (512 bits or less, think column multiplication with base 2^64) and then switches to shift add based algorithms because they are simply faster.

Re:Painful

[agbinfo]

That's an interesting trivia. Is this for special cases only - for example several multiplication requests and not enough resources - or is it generally true?

In any case, it doesn't make the lack of a multiplication a non-problem; It just makes this an additional problem.

Re:Painful

[mcgrew]

Ha, you change the algorithm from O(2^N) to O(N) (where N=number of bits in the number) by using shift

As I said.

BTW: when one of the values is a constant you obviously don't need the loop, instead the compiler just emits a couple shifts and adds right into the instruction stream.

We're talking about assembly, not C++. There is no compiler in assembly, there's an assembler, and it directly translates the human-readable asm to machine code.

If you know assembly, you can assemble the code by hand. It's tedius, but doable.

Re:Painful

[bored]

BTW: when one of the values is a constant you obviously don't need the loop, instead the compiler just emits a couple shifts and adds right into the instruction stream.

We're talking about assembly, not C++. There is no compiler in assembly, there's an assembler, and it directly translates the human-readable asm to machine code.

Maybe we misunderstood each other, I didn't see any mention of language, only instruction set. As someone who has written tens of thousand of lines of assembly professionally, and maybe just as much as a hobbyist dating all the way back to a 6502, I thought we were talking about the lack of a multiply instruction. That affects anyone writing code, not just assembly programmers, because hardware multiplies tend to be much faster than generic software algorithms for the common case of generic two variable multiplies (neither a constant). As far as coding in assembly (6502), multiplying two non constant variables is as easy as loading them into a couple registers (or 0-page for 16-bit) and jsr'ing to the multiply routine. There was no reason to rewrite it all the time. Heck with a decent macro assembler (ORCA/M), optimizing constant multiplies was as easy as using a macro, again not requiring any real thought.

BTW: The macros for constant multiplies obviously aren't (usually) going to be 1-1 mnemonic to machine code. So, in that regard assembling things by hand, is nearly as much a PITA as compiling something by hand. I can take a fairly short C routine and generate machine code with that too. Nothing special there, its just tedious.

I just replied to your comment because you provide a repeat addition algorithm, instead of the just as simple, shift/add algorithm. This made me blink, because for most cases the shift/add is so much better its not even worth talking about repeat sums, unless for some reason you don't understand shift/add.

So for those people who have never written a multiply using shift/add, a basic one is:

while multiplier is !=0, check low bit of multiplier, if set, add multiplicand to result. Shift multiplicand left, multiplier right, repeat.

That algorithm works in C, unlike many of the roll through carry ones, which only work in assembly. Its actually fairly useful even today, if you want a fairly small integer larger than the word-size of the target machine. Aka 128 bit multiply on 64-bit arch.

I used to write 6502 assembly...

[Super Dave Osbourne]

I learned it while on vacation in europe in 1981 prior to my even knowing BASIC or FORTH on the Apple ][ line of computers. It was the most important step in my otherwise stellarly mediocre life as a senior software engineer with NeXT and Apple. Its great to see others taking a lasting approach to the chip that made the most impact on me and others in my industry. Thank you! BTW, did you know there is at least one logic bug in the CPU? :) Its fairly well known now, find it and you will have a bit of history on your hands.

Re:I used to write 6502 assembly...

[Anonymous Coward]

You mean the ROR bug that was in the pre-6/76 steppings?

Re:I used to write 6502 assembly...

[Super Dave Osbourne]

That bug is a bit less known, I meant the paging flaw with hex.

Re:I used to write 6502 assembly...

[sconeu]

I still have my MOSTEK 6502 manual -- 8.5x11 softcover.

Re:I used to write 6502 assembly...

[Anonymous Coward]

Hmm. What was that one?

Re:I used to write 6502 assembly...

[khellendros1984]

I know there's one that if you're using an indexed address mode, and the index would cause the address to cross to the next page of memory, it loops back to the beginning of the page.

Re:I used to write 6502 assembly...

[bennomatic]

That rings a bell. By "index address" are you referring to the relative address mode used by the BNE/BEQ commands that only allowed you to jump 128 bytes forward or 127 bytes back? And what was the size of the page in question? I remember on the C64 that it seemed to be built as four main chunks of 16k... was that the page size, or are we talking about something much smaller?

Man, every time the C64 or the 6502 (or 6510) chip comes up, I get a wave of nostalgia.

Re:I used to write 6502 assembly...

[Anonymous Coward]

Mmmmmmm. I hope it didn't have serious effet on real world 6502 usages, like: :-)

Re:I used to write 6502 assembly...

[White Flame]

It's not just indexing, it happens with indirect JMP as well. JMP ($01FF) would read the low byte from $01ff and the high byte from $0100 instead of $0200.

Re:I used to write 6502 assembly...

[Anonymous Coward]

To the beginning of the next page or to the beginning of the "old" page?

Does that pertain to both X and Y indexed addressing modes?
Can it index backwards (negative offset)? What happens then?

Re:I used to write 6502 assembly...

[Anonymous Coward]

Well - the first use of a 6502 I made was with a self-build board. The schematic was in a magazine for electronics called "Radio Bulletin". It was a interesting project at that time. I still have some pages stored.

After that it was the Acorn Atom - a amazing piece of clever hard- and software. It contained a BASIC that could be mixed with machine code (in mnemonics like STA, RTS etc.). It was the forrunner of the BBC-a and BBC-B computers that where really workhorses. And if you did hot have enough power with that BBC computer, you could add a box with a second processor (That could be a 6502C, Z80 or even a ARM RISC) that was connected wit a fast "tube" connection. It was a time that was really exiting and innovative. Compared with today developments where going at incredible speeds. Today there is only one base hardware design left (even Apple cracked down) and a few OS versions. I must say it is far less exiting - even dull...

Re:I used to write 6502 assembly...

[Super Dave Osbourne]

I still have the 6502 for Apple source assembly and the Inman and Inman book (soft cover as well, but 5x7 or whatever sizing) that I took to Europe with me. Its just hard to let go of this stuff.

Re:I used to write 6502 assembly...

[Blymie]

Just the thought of Super Dave discussing the 6502 has me ROTFL!

Rock on SuperDave!

Re:I used to write 6502 assembly...

[uglyduckling]

Yup - inline assembler. It was awesome, you could take a basic program and carefully rewrite key parts in assembler, it would compile and then run the code in place.

Re:I used to write 6502 assembly...

[rjmx]

As I recall, it was an indirect jump when the first byte of the operand was at the top of a (256-byte) page (i.e. at address xxFF). The second byte of the operand (the high byte of the target address) should have been taken from the first byte of the next page (xy00), but, due to the bug, it was taken from the first byte of the same page (xx00).

A later CMOS version of the 6502 (the 65C02) fixed this bug.

Re:I used to write 6502 assembly...

[rjmx]

(I remember seeing one game that used this to confuse would-be crackers).

Re:I used to write 6502 assembly...

[WolphFang]

X and Y registers!

Re:I used to write 6502 assembly...

[Anonymous Coward]

And broke several games that I wanted to play :(

Re:I used to write 6502 assembly...

[Anonymous Coward]

Mostek never made the 6502. I know, I worked for them from 1976 to 1980. MOS Technology made the 6502. Mostek
  did manufacture (as a second source) the Fairchild F8, Zilog Z80, Motorola 68000, and had a license to manufacture
  the 8086.

My first assembly language program was written for the 6502 for a microprocessor class I was taking at college. It
used a DAC to generate a triangular wave output and used self modifying code to start the output at the zero crossing.

Re:I used to write 6502 assembly...

[call -151]

Indeed there were some pretty nasty obfuscation efforts as part of copy protection- code that ran in the stack, code that relied on processor quirks/bugs, tricky self-modifying code. It was a good way to really learn assembly.

Re:I used to write 6502 assembly...

[Anonymous Coward]

Are you talking about the one where if you, say, JMP ($3AFF), it will jump to the address with the low byte in $3AFF and the high byte in $3A00, not $3B00 as it should?

Re:I used to write 6502 assembly...

[nomorecwrd]

I still have "Mapping the Atari"
A complete and comprehensive list of every memory address on that machine.

Oh my gawd! I think I'm gonna cry.

Re:I used to write 6502 assembly...

[sconeu]

I went to UCSC with Inman Jr. (Kurt Inman). We were in the CS program together from 83-84.

Re:I used to write 6502 assembly...

[sconeu]

Sorry, MOS Technology. Call it a seniorI didn't have the manual in front of me when I typed that.

The manual is the:

MCS6500
MICROCOMPUTER FAMILY
PROGRAMMING MANUAL
JANUARY 1976

Second edition
(c) MOS TECHNOLOGY INC 1976

Brings back memories

[Sooner Boomer]

The 6502 was what I learned computer design, assembly, and machine code on. First was a Buck Engineering trainer, then a KIM-1 with an add-on board. I learned the value of coffee when coding. Finally I built my own system with an S-100 style archetecture. Memories indeed.

Re:Brings back memories

[Anonymous Coward]

You forgot the "get off my lawn."

Re:Brings back memories

[jcr]

I learned FORTH on a KIM-1. Who else remembers the classic article (I think it was in Byte) "How to write a compiler in 50 words or less"?

-jcr

Re:Brings back memories

[narcc]

I learned FORTH on a KIM-1. Who else remembers the classic article (I think it was in Byte) "How to write a compiler in 50 words or less"?

You're thinking of "Forth Extensibility: Or How to Write a Compiler in Twenty-Five Words Or Less"

It was in Byte -- August, 1980 (The Forth Issue)

Archived here, with others, as a gigantic PDFs

Seems their server runs on a 6502

[Anonymous Coward]

Slashdotted already.

Re:Seems their server runs on a 6502

[Metrathon]

No, the website runs on a javascript emulator of a 6502.

Re:Seems their server runs on a 6502

[Ethanol-fueled]

That's no excuse. The T-800 series Terminators and Bender both ran a 6502 at their core and performed quite well.

Re:Seems their server runs on a 6502

[Pentium100]

No, the website runs on a javascript emulator of a 6502...

... on a 6502

Re:Seems their server runs on a 6502

[ae1294]

No, the website runs on a javascript emulator of a 6502...

... on a 6502

but will it run GNU/lunix... from the cloud?

Re:Seems their server runs on a 6502

[93 Escort Wagon]

No, the website runs on a javascript emulator of a 6502...

... on a 6502

... running GNU/Hurd

Re:Seems their server runs on a 6502

[TheRaven64]

Wow, Apple product placement was really obscure in the '80s.

Re:Seems their server runs on a 6502

[mcgrew]

No, the website runs on a javascript emulator of a 6502... ... on a 6502

but will it run GNU/lunix... from the cloud?

I imagine it's pretty cloudy in that slashdotted server's server room, what with all the smoke and all.

Re:Seems their server runs on a 6502

[ae1294]

No, the website runs on a javascript emulator of a 6502... ... on a 6502

but will it run GNU/lunix... from the cloud?

I imagine it's pretty cloudy in that slashdotted server's server room, what with all the smoke and all.

... and they can't open a window because it's located in the basement...

Re:Seems their server runs on a 6502

[badkarmadayaccount]

Under Xen, with OSOL dom0.

C'mon. It's a cool page

[msobkow]

Just admit it -- someone really spent a lot of time on this idea, and it's cool that it works at any speed at all. I've only got an old P4, and it still ticked along quite nicely. Much faster and you wouldn't see anything but flicker-flashing anyhow.

"Machine Language for Beginners"

[Petersko]

It was 1983, my Commodore Vic 20 (bought delivering newspapers) was soon to be replaced by a Commodore 64 (bought the same way), and nobody understood my fascination.

Except, apparently, for Richard Mansfield, whose book I devoured. I remember trying to figure out how the heck to get the opcodes into memory. I had nobody to teach me what peek and poker were about, so it took a while.

It's also possible to say that the 6502 and 6510 were perhaps the very last processors that I understood in real, intricate detail. Once I hit the 286 it might as well have run on magic pixie dust. I can't remember ever masking interrupts on an x86. I've only written in languages at the level of C or higher ever since., and I've never embedded assembler to fine-tune performance.

My geek level has diminished.

Re:"Machine Language for Beginners"

[Brett Buck]

I had the same sort of thing with the PDP 11/04, and later, the late and totally unlamented Teledyne 1750a processor. Both of those date me appropriately.

Re:"Machine Language for Beginners"

[Hylandr]

/Signed.

- Dan.

Re:"Machine Language for Beginners"

[olsmeister]

Ditto. I read that book backwards and forwards and could do just about anything in ML on the C64. I remember writing floating point math routines because the instruction set had nothing built in.

Fast forward to my first PC, a 486SX. I learned x86 assembly, but never felt the same kind of complete and utter control over the machine, probably because by that point in my life I didn't really have the time to dedicate to really immerse myself in it.

However it gave me a great intuitive understanding of what is really going on behind the scenes, even though my present life has taken me in a direction that doesn't allow me to do much programming anymore, and I still do have that book, I just saw it in a box a month or two ago hunting for something else (along with a stack of Compute!'s Gazettes). Good times.

Re:"Machine Language for Beginners"

[oldhack]

Same here. Once they started throwing in real/protected modes and MMUs, I knew I was far from the "metal".

Re:"Machine Language for Beginners"

[fwarren]

Fast forward to my first PC, a 486SX. I learned x86 assembly, but never felt the same kind of complete and utter control over the machine, probably because by that point in my life I didn't really have the time to dedicate to really immerse myself in it.

No it was not the level of immersion. A computer like the C64 had 20K of ROM, hence a kernel that never changed. Always had a 6510 and a VIC chip and 64K of RAM. You could learn every byte inside out. Because the platform did not change over time, there were many volumes of literature written about the internals of this machine.

Your 486 had a bios made by one of several different BIOS vendors, one of several revisions of that BIOS, for the particular chipset on your board. You had an audio card made by one company, a video card made be another company, an ATA controller made by yet another company.

Who else really had your setup? No one wrote a complete manual for what you had. By the time you could figure out the BIOS and figure out what it took to POST the hardware you have, it is time to buy a new computer.

Nowdays there are many levels of abstraction, the one thing it robs us of is the ability to understand exactly what our machines are doing.

Re:"Machine Language for Beginners"

[Gaygirlie]

I too bought my C64 the same way, delivering newspapers and ads :) Had a helluva time with it, too: played a few games on it, but mostly taught myself BASIC and then moved on to assembly. Could spend whole nights up just coding and learning. It really kindled my interest in computers, their internals and programming, and my life would have been quite different without those first contacts with a C64.

Though, it died then suddenly one day, got an old 286 from my uncle, and that's where my story differs from yours: I immediately went on to teach myself x86 assembly :)

Re:"Machine Language for Beginners"

[bennomatic]

I loved the Compute! book series. ML for beginners, Mapping the C64... Good times, good times.

Re:"Machine Language for Beginners"

[93 Escort Wagon]

I had nobody to teach me what peek and poker were about...

I know it's rather too late to be useful - but I think I found your problem.

Re:"Machine Language for Beginners"

[Opportunist]

You should have moved on to Amiga and its Motorola 68000. That processor sure was MUCH easier to understand than the segment/offset mess the X86 architecture was in.

The 68000 was the last processor I truly understood. It was pretty straightforward too. Very streamlined and quite powerful.

Re:"Machine Language for Beginners"

[Opportunist]

And today, you're even further away from the silicon. Your instructions ain't the same anymore when they finally get processed.

Sometimes it really seems like a miracle that something sensible comes out of it at all, considering the amount of layers between you and the ALU(s).

Re:"Machine Language for Beginners"

[Anonymous Coward]

Consider what this means for computer security. You have no idea what's running on your computers.

Re:"Machine Language for Beginners"

[rdg55]

You're living my life! I had paper route in a 110-unit high-rise condominium in Puerto Rico. It was great, except for Sunday when half the residents had Sunday-only deliver of the NYT. I had to sub-contract my little brother for help on Sundays. I remember coding a Mandelbrot(sp) Set visualizer in assembly language using a library of large precision (not floating point) routines. It was great fun to zoom in to a new area and watch as each dot in the raster image was rendered. Then it occurred to me to do a more modern pre-rendering by doing only every 4th pixel on every 4th row first, and letting the neighboring pixels inherit that color until they were themselves properly computed. Then offset by 1, and repeat, and watch in minute-by-minute amazement as the details revealed themselves. I also remember trying to code Super Mario... I was only able to render level 1-1 with SMOOTH left-right scrolling, with no cycles left for ANY game logic. I don't know how Nintendo did it. The feeling of KNOWING what the machine was doing is indescribable. I often wish I had the time and a few extra IQ points to understand the 8086... but alas, both resources are limited. Good memories.

Re:"Machine Language for Beginners"

[Anonymous Coward]

I had a similar experience, except we had an Apple IIGS. It had a 65816 which was essentially a 16 bit version of the 6502. It was fully backward compatible with the 6502. I wrote everything in assembly language and knew all the ins and outs of the machine. My first PC was a 486DX-33, obviously a big jump from the Apple. I only wrote C and C++ from then on, never even learned assembly for x86. I'm not sure if its because things got too complicated to bother learning or if the new processors were just fast enough that it didn't seem necessary to optimize everything. These days I still write lots of code, but I don't seem to care much about what the hardware or O/S is doing. I just want it to work. Maybe it's just old age.

It's hard to believe the 6502 only had 3500 transistors. Transistor counts are approaching a million times that these days.

Re:"Machine Language for Beginners"

[Anonymous Coward]

I still have "Machine Language for Beginners" & "The Second Book of..." on my bookshelf. Read the book's soft copy at this URL: http://www.atariarchives.org/mlb/

Re:"Machine Language for Beginners"

[mcgrew]

I had nobody to teach me what peek and poker were about

I believe that's referred to as "strip poker" ;)

Re:"Machine Language for Beginners"

[trackedvehicle]

It's also possible to say that the 6502 and 6510 were perhaps the very last processors that I understood in real, intricate detail. Once I hit the 286 it might as well have run on magic pixie dust. I can't remember ever masking interrupts on an x86. I've only written in languages at the level of C or higher ever since., and I've never embedded assembler to fine-tune performance.

My experience, also. While I did experiment with interrupts with x86, I never did much with assembler, on them, and the segments, near and far stuff, definitely discouraged one from getting too deep into it.

Not so fast...

[jimmydevice]

"the details of the designs have been lost or forgotten. While there have been great efforts to reverse engineer the 6502 from the outside, there has not been the hardware equivalent of the source code — until now"
The schematic for the 6502 has been available for years on the net. Printed on one sheet of photo paper at 1200 dpi, every transistor is visible. It's quite amazing.

Re:Not so fast...

[Anonymous Coward]

You mean Balasz' schematic. That's actually also reverse engineered, and has quite some mistakes in it.

Re:Not so fast...

[NixieBunny]

This schematic also looks like it was reverse-engineered from the chip. Not that I would expect the original schematics to be available to the world, since all that stuff is considered to be quite proprietary by the manufacturers.
It would be a lot more useful to see the original logic design.

Re:Not so fast...

[jimmydevice]

I had assumed it was the schematic from a working chip and was most likely from MOS or maybe Western Design Center.
Color me enlightened.

Re:Not so fast...

[mysidia]

Not that I would expect the original schematics to be available to the world, since all that stuff is considered to be quite proprietary by the manufacturers.

Now why would they consider schematics for a chip no longer produced to be so proprietary?

When do the lawsuits against these guys start rolling in for publishing images of copyright-protected visual arrangements of wires and transistors?

Re:Not so fast...

[NixieBunny]

By the time it's not so proprietary, it's obsolete and of no interest, so they toss it in the dumpster.

Re:Not so fast...

[JWSmythe]

When do the lawsuits against these guys start rolling in for publishing images of copyright-protected visual arrangements of wires and transistors?

    You know, you'd think that wouldn't you. But leak one drawing of one chip for an old government "communications" satellite that never existed, and WOOSH... Swooped up by three guys in a black van, a bag over your head, and then you wake up in a cell in some facility with guys in suits asking lots of questions, but never answering questions like "Where am I?" and "Who are you guys?" Just remember, don't ask either of those, nor the magic question "What agency are you from?". You may not have noticed the electrodes taped to your testicles, but you sure will after they push that little red button in front of you. It seems they just want answers, but the answers you give aren't ever enough. Oh, and "your momma" is never the right answer either. It's apparently worse than most expletives.

    Seriously, copyright is nothing to mess with. And more importantly, watch out for black vans, no matter how innocently they may be following you. If you've been posting stuff like that, they aren't innocently following you.

    Why's that van been parked outside all night? Gotta run, someone's at the door.

Re:Not so fast...

[arivanov]

Yep.

And if for whatever reason the schematic cannot be reproduced the schematics for the East German and Russian "illegal" clones can be reproduced without any trouble (sorry, it's been a while so I have forgotten the actual chip numbers).

Re:Not so fast...

[cheekyboy]

Finally someone can make a 350,000 core CPU using the 6502 specs running at 1000mhz it would fly.

Re:Not so fast...

[Kazymyr]

According to the wikipedia article on the 6502, it is still manufactured for embedded systems. Thus the diffusion masks must still exist. In which case the design hasn't been "lost"

Either that, or the wikipedia article is wrong.

Re:Not so fast...

[ishobo]

Available from Western Design Center, started by Bill Mensch, the person who co-designed the 6502 with Chuck Peddle. Both also helped design the 6800. WDC ha seen selling the 6502 based products since the 1979. Both the Apple IIe and IIc used WDC product, the 65C02. All the second source products over the years have been lisenced from WDC. WDC has been able to sell products based on the original 6502 design because they co-held the patents with MOS.

It is still sold in its original 40-pin plastic DIP. Verilog soft cores are available too.

Re:Not so fast...

[mysidia]

Swooped up by three guys in a black van

That's why we've got a city ordinance banning black vans.

They don't dare enter; they'd be pulled over and ticketed. And you know how much people in black vans like to be pulled over and have to show identity?

Re:Not so fast...

[hierofalcon]

But the 65802 and 65816 are much better if you're really going to program anything 6502 based. It's too bad they never officially came out with their 65832. That would have been an awesome chip. We might still be using them in our embedded designs if they had. As it is we stopped with the 816. People lavish praise on the new stuff, but having even the weirdest addressing modes execute in 7 cycles and many only take 2 cycles with no wait states is awesome if you are doing basic control operations. On the 65816 you can move zero page around as well as the stack so you can have multiple zero pages (one for each task you're running). We developed our own assembler, linker, and ANSI C compiler cross-development environment that runs on Linux to support our embedded controllers along with a multi-tasking priority based small O/S, along with a 65816 simulator that let us test our software on Linux before putting it in the board. Lots of these are still in use in the oil field at least, along with many other embedded markets.

Re:Not so fast...

[JWSmythe]

And you know how much people in black vans like to be pulled over and have to show identity?

    From what I understand, not very much at all. Of course that leads to newspaper blurb the next morning "Officer Shot During Routine Traffic Stop". If you bother to read down (barely anyone does) there is a mention that there were no witnesses. In the continuation on page E68 towards the bottom you'll notice "dash camera footage is missing"

    A few days later, you'll also notice some random citizen missing, last seen somewhere in the vicinity of the previous story. There was your witness. And their little dog too.

    Never, not ever, will you see the story about you, the poster of the reverse engineered chip for the officially non-existent satellite. If you do show back up, your story will be considered as plausible as an alien abduction or fairy circle disappearance. Well, that is if you tell anyone. Which is better, to tell the world "I was abducted by a mysterious government agency" which no proof can be provided, or STFU? Oh, the later. If you keep talking about "the incident", you'll either be disappeared, or end up in a padded room. The padded room is more useful as incentive for others to STFU though.

    Makes you wonder how many conspiracy nuts are just plain nuts, or have something to say that's just too wild to believe. I'd say the later has a percentage lower than 0.01%. ... and on that topic. :)

    I got my favorite conspiracy crap newsletter this morning. The lunatic signed me up because I made a press contact years ago, even though I've mocked them more times than I care to count. There's a news story about Tampa International Airport closing a runway to renumber it because the normal natural magnetic shift has changed the magnetic heading just enough to necessitate it. I did find it on a "good" conspiracy site. He runs interesting news, and people comment on it. I found the truth behind it, and posted it there. The truth is that it was one degree away from needing a change, and it's finally passed that last degree. They're closing the runway because they have two, and it won't effect normal flight traffic this time of year. This morning, the nuts conspiracy newsletter attributed it to government conspiracies, the end of the world, and the idea that the world will be shattering on Dec 21, 2012 because of distortion of the magnetic fields caused by the sun and the universe as a whole. {sigh} The truth may be boring, but when the conspiracy is so nuts, it just gives me a headache.

    Oh, gotta run. The black van is here. The guys are picking me up for work. We have to have a friendly talk to someone posting on Slashdot this morning by the name of "mysidia". Oh good, that's you. Can you confirm where you'll be, so we can save some time. It's always a pain when we have to track your cell phone to pick you up. :)

Re:Not so fast...

[confused one]

They're splitting hairs. The later designs are 65C02 done in a different process with new(-er) masks.

Re:Not so fast...

[greed]

I always wondered about cycle time comparisons with the 6502. If memory serves, it used a split-phase clock; which effectively creates a clock multiplier without having to deal with high frequencies. That is, 01 and 02 were the same frequency as 00, just offset by 'X' degrees in the cycle.

So, where the 6800 had multiple input clocks per operation, the 6502 could make use of multiple clock edges within a single master clock cycle... a bit like some of the double-pumped and double-edged clocks we've got on SDRAM today.

Which is still great, as we all know power consumption goes up with the square of the clock speed (for any given circuit; hence the need for narrower features at high frequencies). So by cleverly spreading the "parts" of an opcode across the several clock phases, you got enough time for the gates to go valid... but without having a 4 MHz master oscillator.

Or memory doesn't serve, 'cause I left the VIC-II disabled too long and so DRAM refresh didn't happen and it's all gone now....

Re:Not so fast...

[mysidia]

In the continuation on page E68 towards the bottom you'll notice "dash camera footage is missing"

Always have a hidden backup camera with satellite uplink and an assistant at remote location sitting at keyboard ready to push a red "post to youtube" panic button. Citizen cell phones with streaming video are also a good idea.

And make traffic stops near "safety" / "traffic camera" points. Destroying dash cam footage just adds 'defacing government property' to the list of criminal charges.

Re:Not so fast...

[badkarmadayaccount]

For deterministic execution, use a basic RISC core, why go to all the trouble of CISC?

watch the voltages race around the chip

[Anonymous Coward]

Voltage doesn't "race around," current does--well, it flows.

Re:watch the voltages race around the chip

[Anonymous Coward]

Well, the current doesn't, either, technically; the orthogonal EM field transfers energy along the conductors.

Re:watch the voltages race around the chip

[Hodapp]

In some sense, it does. What they're talking about, I believe, is watching it propagate through gates at each clock cycle.

Re:watch the voltages race around the chip

[WolphFang]

THANK YOU!

There are already schematics on google from 2008

[theblondebrunette]

A simple google search for "schematics of 6502" reveals the link above. Also, there are books out there, in Bulgarian, with the schematics and design of Pravetz 82, a 6502 clone.

Sounds familiar

[serutan]

An animation with voltages racing around the chip as it executes?

Where's my light-cycle?? I must save my User from the MCP!!

Re:Sounds familiar

[Anonymous Coward]

An animation with voltages racing around the chip as it executes?

Where's my light-cycle?? I must save my User from the MCP!!

There was a program (we didn't call them "apps" then!) called, IIRC, "The Visible 6502" that would, using Apple ][ HIRES graphics, show the execution of whatever 6502 machine code you wanted to run against it. This was like 1978 or 79. Great fun to watch the little sparkly bit zoom around to the instruction decoder, ALU, address and data latches, etc.

I'm sure I've got it on a 5 1/4 floppy somewhere...

Watch this...

[steeleyeball]

Ok what we really need is the Video chip schematics and logic.... I would love the 6560/6561. I would very much love a Vic 20 / C64 based watch...

Re:Watch this...

[Anonymous Coward]

I'd rather have the schematics for the SID. Yum, chiptunes.

For justice!

[binarylarry]

John Titor can now leap home.

Re:For justice!

[morgan_greywolf]

I thought John Titor was looking for an IBM 5100?

Yes, but can it run Pitfall?

[itomato]

Yep.. Space Invaders, too.. SIGGRAPH talk: http://www.visual6502.org/docs/6502_in_action_14_web.pdf

Re:Yes, but can it run Pitfall?

[Anonymous Coward]

The page mentions the 2600. Yet they still fail to mention the XL and XE series. Which in my opinion were much better products by Atari. Sigh...

Re:Yes, but can it run Pitfall?

[alteveer]

I was at the SIGGRAPH talk and it was really interesting to watch that chip function (obviously a bit slower than what it would natively). The speaker went through how they laid out all of the circuits based on the different chips they had stripped and photographed, and then he showed it in action. He started by filling and emptying single registers and then he started doing some ops on it. I wish I could provide some video of it in action, but I have a feeling SIGGRAPH wouldn't like that very much.

Download it an try it for yourself, it is really hypnotic to watch.

Re:C'mon. It's a cool page

[aliquis]

Why? It's a shitty card and it's not like there are any games which need any weird surround effect features of it in Linux anyway. And Creative drivers and support isn't the best around even in Windows.

So why?

What about the ASUS Xonar?

Or maybe some generic USB sound interfaces work? I don't know if they can do higher bit-rates that way but if they can I assume the require very little work and may still have a nice sound quality. Eventually suck for latency though?

Yep. I like it.

[PopeRatzo]

This is cool.

It'll be even better when all the traffic dies down to the point you can load all the images.

nontrivial!

[sillivalley]

The 6502 as used in the Apple ][ had some interesting quirks -- such as dummy read cycles that appeared on the bus when executing indexed operations. Woz used these dummy memory cycles in designing the original Apple ][ disk controller to whack the disk controller state machine. Undocumented at the least! Some of the Apple ][ disk copy protection schemes (particularly for games on 5 1/4 inch floppies) also relied upon undocumented behaviors in the processor, such as some of the "unused" opcodes.

Re:nontrivial!

[NixieBunny]

Woz wasn't the only person to abuse the 6502. Don Lancaster took advantage of the address bus behavior or the chip when executing NOPs to make his cheap video display (the TVT 6-5/8) board. He basically used the processor as a 16 bit counter that could be loaded under software control during the horizontal blanking time. So the CPU chip became a video chip.

Re:nontrivial!

[Framboise]

I remember distinctly that the Apple ][ would execute self-modifying op-codes in the first 256 byte memory page.

Re:nontrivial!

[call -151]

That would have been as part of various copy-protection schemes, not under general purpose computing. The general OS (the monitor, Integer Basic, and later the Disk Operating System) were pretty understandable to disassemble and were quite well-constructed. The source code (from Woz) for the Sweet 16 interpreter was part of the original Apple ][ manuals and also quite good to read and document, although I don't know how many people made much use of it.

Re:nontrivial!

[Nethead]

Now I have to go dig under the stairs for my copy of "Son of Cheap Video".

Re:nontrivial!

[metaforest]

Using the funky behavior of bus accesses on the Apple ][ it was possible to synchonize video-page flips. The apple ][ did not have a VerticalBlank interrupt or a flag to indicate that Vertical Blank was active.
The apple //e (and later systems) did have a Vertical Blank Flag bit in the I/O pages, but it was not as useful as being able to read the video latch data via the video switch bank.(This was a timing bug in the Apple ][ that allowed the video latch data to be snapped by executing an immediate Read (LDA, LDX, LDY) on Write-Only I/O switches.) Also: Most of the 'undefined opcodes' on the 6502 were actually very useful. They were very consistent across the entire 65xx line. I published a list of them back in 1985 or 1986 on USENET. Nick Westgate published a similar list after the 65816 had been out a while. Many of the most useful OPs combined common 6502 idioms. It is unknown to me if those were an artifact of the instruction decoder logic or if those OPs were designed in, but undocumented. I am inclined to believe they were undocumented since most of them appear in the 6502, 6507, 6502A, 6510, 65C02, and 65802. Some of them are also active in the Nintendo NES but I am not sure how many. I saw documents reporting a few of the OPs but never had time to explore the undefined ops on that implementation when I was a game dev for Atari/Tengen.

Re:nontrivial!

[metaforest]

Just a side note. Over clocking a 68000L8 at >=24MHz causes the address and data bus to act as synchronous counters. I am inclined to believe that this was a test mode designed into the 68000. I discovered it by accident while chasing down (one of many) design flaws in a 68000 motherboard designed by Granger & Assoc. The clock chain was borked thus overclocking the CPU by a factor of 3. To me it this behavior from the CPU seemed very useful.... it was also VERY consistent across many members of the 68K family.

Really lost?

[ArchieBunker]

You mean to tell me that the companies who designed or licensed this chip plus the chip fabs themselves have no information on how they were manufactured?

Re:Really lost?

[Mashiki]

I suppose it's possible. I mean we did loose the ability to make things like blue glass, green glass, and concrete at least one in the world. Losing schematics is also possible, I remember hearing a few years ago that some company was looking for a particular tube that RCA used to make, because RCA lost the diagram on how to manufacture the tube, whether or not it was ever true I have no idea. It was a passing fancy at the time for me, but even I have electronics that I use today that have tubes in them still.

Re:Really lost?

[Tumbleweed]

You mean to tell me that the companies who designed or licensed this chip plus the chip fabs themselves have no information on how they were manufactured?

Well, sure, but they had them backed up on their Apple /// machines, and ... well ...

Re:Really lost?

[man_of_mr_e]

Yeah, I find it hard to believe too. The 6502 was heavily licensed to other companies, and was probably produced in more variations than anything short of the ARM. I would find it very difficult to believe that there aren't plenty of photolithographs stashed away in various files. Now, maybe an actual design document indicating what everything does might be missing, but there should be plenty of film available to reverse engineer from.

Re:Really lost?

[JWSmythe]

    That isn't all that uncommon. I guess no one recalls the 2009 news stories about the classified "fogbank" material in nuclear warheads. When the gov't wanted to upgrade some nuclear warheads, it was discovered that the material wasn't documented, and no one remembered how it was made.

    I couldn't find anything on the lost vacuum tube design, but I wouldn't be totally surprised. There have been plenty of things over the years that were superseded by something better, and the predecessors were forgotten about. Trade secrets often include keeping very little documentation about them, so they won't be accidentally released (or stolen).

    I've worked with quite a few things over the years where a working copy existed, but you couldn't replace it. Sometimes it took a virtually clean room reverse engineering to replace it. Sometimes it was hardware. Sometimes it was some application (like a server-side app) where it had been compiled once a decade ago, and no one had a clue where the source was. Well, they *did*, it was just "oh, well Bob did it on his computer. He left 8 years ago. We don't know where to find him", and his desktop had been reformatted, handed down through several other people, and finally retired to the trash a few years before.

    For stuff like vacuum tubes, and ancient apps that don't run on modern hardware, it's usually worth doing it fresh. For things like classified materials that keep nukes from spontaneously exploding, that knowledge can be virtually irreplaceable. There are various lost arts, that if you have to fall back to raw materials, we'd be screwed. I like using the loss of civilization and modern technology as an example. If you, I, and a few thousand skilled Slashdotters were dropped on an deserted island (or an alien but Earth-like) planet, how long would it take for us to raise technology from nothing to build a working computer? Assuming an abundance of easily identifiable raw materials, like "hey, that's iron ore", maybe we could build a forge, and make some decent hand tools. We may be able to build primitive radios, but I doubt we'd ever get far enough fab the first solid state chip, much less a CRT or LCD screen, before the first generation died off of old age. The second generation may have a clue, but no true memory of the technology in working form.

    Consider other "high tech" items. I love the car analogies, so we'll go in that direction. :) Could we build a car? I know quite a few of us know a lot, but could we actually design and fabricate one from scratch, even with another car as a template. Sure as hell in such a scenario we'd never build the first ECM. Could we get as far as a primitive carburetor? Who am I kidding, we'd probably get stuck on something like how to make tires. :) That's assuming no meeting-hungry managers or aspiring politicians were in the aforementioned group. If that happened, we'd end up stuck in endless meetings forming committees to discuss every feature, and we wouldn't get the first item put down on paper because how to make paper and a pencil would be stuck in committee for the rest of our lives.

    The idea of "lives" may be rather short anyways. How many of us are hunter/gatherers and farmers? Not enough to start from scratch.

    Kinda ruins those fun aspirations of colonizing alien planets, doesn't it? 1,000 space nerds fly to and are dropped off on a habitable planet circling one of the closest stars, just to die of starvation before establishing a society capable of self sustaining itself, because we don't have the skills to bring our technology up to "modern" levels.

Re:Really lost?

[TheRaven64]

There's a post a few after yours explaining how this chip was designed. A modern chip is designed on computer, and all of the steps from the HDL down to the final masks will be backed up and archived. The 6502 schematic was hand-drawn on a really large piece of paper, then onto Rubylith for production. The schematic may have been lost long before they stopped producing it, because they just needed the mask.

Once they stop producing them, there's no commercial incentive to keep the old designs around, so they get lost.

Re:Really lost?

[SmilingSalmon]

There is an interesting discussion in Matt Ridley's new book, The Rational Optimist about some cultures losing technology over time. One of the reasons he gave was that their society was too small -- not enough individual members and they were cut off from others by geopgraphy so they couldn't trade with a wider circle of people. He posits that you need enough members in your society so that you have enough people to specialize in all the needed arts. You also need more than one person practicing each art. You wouldn't want your only knowledgable glass blower to die by accident and leave you with no one who knew how to blow glass.

I'm not sure if a "few thousand" individuals would be enough to support even a radio-level of technology. Remember, you need people to grow food, make shoes, sew clothes, prepare meals, collect trash, raise the next generation, and hundreds of other activities that aren't considered high-tech but are needed to support the circuit designers, light bulb makers, wire drawers, and other high-tech practitioners.

Re:Really lost?

[bertok]

Reminded my of stories where immortality is possible for future humans through high technology.

A real risk for such beings is the possibility of becoming marooned on a lifeless rocky planet, possibly forever, or close to it.

Could you get off such a planet, given essentially limitless time and no other resources other than yourself? (Assume that you cannot use your own body parts or fluids or whatever, due to the requirement of conserving the special high-tech / organic matter of your own body.) Could you ever build a spacecraft or even some sort of powerful beacon?

For these 'boot-strap' scenarios, we take a lot for granted. The Earth is extremely resource-rich, and not just in raw materials, but pre-processed ready-to-use items. Wood is actually a very complex fibre-reinforced composite with an excellent strength-to-weight ratio, and it also doubles as a convenient fuel. Animals have skins, bones, and tendons that can be made into clothes, tools, and fasteners. Even if one has no need to breathe, the atmosphere is still useful, because it is highly reactive. How far could we really get technologically starting from nothing without a reducing atmosphere? If there's no oxygen, there's no fire, and without that, most primitive technology is out the window. How would one refine ores into metals?

Re:Really lost?

[Magada]

Robinson Crusoe. IN SPACE! Take your stories and make them into a movie script. You'll make millions in Hollywood with such a tried-and-true recipe.

Re:Really lost?

[sartin]

That's assuming no meeting-hungry managers or aspiring politicians were in the aforementioned group. If that happened, we'd end up stuck in endless meetings

Don't worry, we'll them all away in a B-Ark. The Golgafrinchans dumped them on us, we'll make them somebody else's problem.

Re:Really lost?

[confused one]

Given the ISA and the required signal timing I can create an equivalent for you. When do you need it by?

My point is some of those are reverse engineered and/or modifications on a design provided under license. To be fair, they're splitting hairs, referring (I believe) to the 6502 and not the later 65c02 implemented in the newer process (with new masks). fwiw, you can still buy 65c02 from Western Design, which was started by one of the designers of the 6502.

Re:Really lost?

[Anonymous Coward]

The whole "recreate technology after being dropped on a primitive planet" theme is the plot of "Space Prison", by Tom Godwin. http://www.gutenberg.org/files/22549/22549-h/22549-h.htm

Re:Really lost?

[m50d]

I think you underestimate us. As long as you have someone who knows each step in the process, you're fine. IIRC My school used to make solid-state chips; sure, we wouldn't get these 65nm processors, but a bunch of transistors in a package is straightforward enough. Making a car would be no problem; consider the conditions they were made under during WW2. I think I remember reading about an enthusiast constructing a flintlock musket from scratch (going so far as to smelt the ore himself), so that level of manufacturing is achievable. Some manufacturing processes would be beyond us just because of scale, but we'd find ways around them - probably bizarre compared to history, but they'd work. Just our knowledge of the scientific method, and basic QM would make so many things so much easier.

Re:Really lost?

[smellsofbikes]

It's easy, with a reasonable number of people, to recreate 1870's technology. The problem once you get into 1930's tech is just the sheer number of man-hours it took to build the infrastructure required for decent development. I've read estimates that it took 50,000 man years of effort to turn Fleming's discovery of penicillin into mass-production of penicillin, and most of that wasn't basic research, where having foreknowledge saves you time, it was applications development: breeding hyperproductive strains, building bioreactors. Likewise, the tech required to build a factory capable of producing silicon with 99.9995% purity, needed for semiconductors to actually work, is hundreds of man-years of effort when you already have the equipment, and tens of man-years of effort go into building each of those pieces of equipment.

Sure we might have the knowlege, but you'd need a million people with that level of knowledge, to get to 1950's-era tech, and that's presupposing you conveniently manage to skip some horrible disease that wipes out half your population before you manage to develop antibiotics. Even getting *farming* productivity high enough so that half your people can spend their time developing equipment, rather than farming to keep from starving, is going to take several plant/harvest seasons.

BTW it's fairly easy to make muskets, and even engines from cast iron -- although smelting iron from ore is an extremely time-consuming process, and getting iron pure enough to be useful is even more difficult. Again, you run into dependency issues: good iron/steel is dependent upon a supply of high-purity oxygen, for refining the carbon out of the smelted iron, and high-purity oxygen is dependent upon having vacuum technology and fractional distillation technology, which is dependent upon having lathes and mills and a way to make vacuum seals, and so forth. I've made small engines from aluminum I've cast myself, and made gun parts and the like, and all the steps are quite easy, if you already have the tools and raw materials. I've even made a metal lathe from aluminum I've cast myself, which is getting very close to from-scratch, except for the whole refining-aluminum thing. (It would be possible in cast iron but it'd take a lot longer.) The issue is how very fractal it is, and how very much time each step takes, and when one of those steps would be 'feeding everyone' and another would be 'keeping everyone healthy' (and building, say, sewer systems so everyone doesn't die of cholera) you're already spending a decade just getting to the point where you can start building stuff beyond staying-alive-tomorrow stuff.

Re:Really lost?

[mcgrew]

I mean we did loose the ability to make things like blue glass, green glass, and concrete at least one in the world.

Well, loosing this information is a GOOD thing. Or did you mean "lose"? If so, what you wrote was completely different than what you intended to write.

If I loose my knowledge, everyone benefits. If I lose my knowledge, only my enemies benefit.

Re:Really lost?

[metaforest]

One of the more interesting (and flat-out-fun) projects I had over the years was reverse engineering a production product where the source code for the app that ran on it had been lost (years before my arrival to the team.) There was a 32K binary image to be burned into EPROM. There was a schematic of the hardware, and the end-user documentation. From those items I produced compilable, fully documented source code and a theory of operation document.

Re:Really lost?

[JabberWokky]

The idea of "lives" may be rather short anyways. How many of us are hunter/gatherers and farmers? Not enough to start from scratch.

If you're talking about humans in general, plenty. Alas, the hunter/gatherer population of the world don't have many specialized roles such as engineers or mechanics. They tend to be really busy hunting and gathering in their forest or on their island.

Well, you'd also have Ted Nugent with a bow running naked through the woods, so that's a plus.

Tickled to see this

[bfwebster]

I have fond memories of the 6502. I co-designed and did most of the coding for a computer game for the Apple II (Sundog) and so did a lot of 6502 assembly coding. A few years later, I taught assembly language coding to CS students at BYU, and we use 6502-based systems there as well (which was a vast improvement over the IBM 360 assembly + JCL on punch cards that I had to learn on a decade earlier as an undergrad myself).

Re:Tickled to see this

[slashqwerty]

People still remember your work. I just read another comment about Sundog (on an Atari)...

Re:Tickled to see this

[TheLink]

I vaguely remember Sundog... My cousin used to play it on his computer decades ago.

I'm curious: how do you feel about the emulator versions of Sundog being easily downloadable on the Internet?

Re:Tickled to see this

[sunspot42]

Hey Bruce! I wasted many hours playing Sundog on the ST. Thanks!

You should make a version for the iPad!

Re:Tickled to see this

[Anonymous Coward]

That was a brillant game. I got a 1040ST just to play this - thank you!

Re:Tickled to see this

[wlad]

Sundog :) Ooh that reminds me of good times when I was young, playing it on my Atari ST, long long ago. I loved that game, it had so much detail. It deserves a place in the same fame as the Elite series.

Re:Tickled to see this

[I8TheWorm]

How did that compare to your work on the 68000 when you ported to the Atari ST? I only ask because I bypassed the 68000, and some previous commenters mention it's ease of use, especially compared to the x80 and x86. My only assembler work was on the Z80.

Re:Tickled to see this

[Anonymous Coward]

Note to Sundog fans: There is a (somewhat stagnant) effort to build an open source clone here, started by Mr. Webster:

http://www.silvae.net/sundog/index.htm

Re:Tickled to see this

[Anonymous Coward]

And I have fond memories of Sundog! Thank you for creating a wonderful game! I'm going to go have a burger now...

Re:Tickled to see this

[Anonymous Coward]

You are a god. Sundog was awesome. Er, sorry, that's all I had.

Re:Tickled to see this

[Anonymous Coward]

You are *that* Webster aren't you? I will send you one of these fan emails one of these days (or one of these years). I really will :)

Sundog was, and still is, one of the most amazing game ever made. No single game made me dream more than Sundog.

At one point I had that theory that you guys at FTL had some machine really faster than light and somehow made it to the future thanks to that machine. Future where you "stole" technology and came back and used that "stolen" technology-from-the-future for fun: to create the amazing Sundog.

This game is one of a kind and people who have never played it have simply missed one of the most important game of all times :)

Re:Tickled to see this

[bfwebster]

The Atari port happened after I left FTL - and, I'll note, it was a great improvement over the Apple version - but I did do a fair amount of 68000 programming over the next few years. A great processor, and it's a shame that the IBM PC reference architecture wasn't based on it.

Re:C'mon. It's a cool page

[Cwix]

Shit man, I gave up on that at least two years ago and moved on. Go get a new sound card.

Best chip in the world

[liquid stereo]

Nothing but love for this chip. It was the first thing I started programming as a geek-in-training, back in1981. I'm now a well-adjusted professor using using computational science, numerical simulation, and some of the fastest supercomputers in the world, to solve challenging problems. I owe it all to the folks that developed the 6502, and my mom who bought me an Atari 400, then an 800XL. Cheers!

Re:Best chip in the world

[PatPending]

I'm now a well-adjusted professor using using computational science, numerical simulation, and some of the fastest supercomputers in the world, to solve challenging problems.

+1 unintentional irony

Re:Best chip in the world

[nomorecwrd]

The first assemby language I ever learn!!

Back in the days, I teach myself ASM 6502, that, together with the book "Atari Mapping", and I was able to do fun stuff my "other" nerd friends woudn't even dream of, all on an Atari 800XL.
Want more colors than the 4 you can get on this graphics mode? Bam! here you go! 256 colors (but only up to 4 by scan line, changing the color palette on DLI... but that was the magician's secret :-) )

I was 15 by that time, and this little Atari, and it's simple, easy and straight forward assembly language began to forge the computer engineer I am today.
Atari inc. has a very special place in my nerdy heart.

They lost the blueprints? I'll go get em

[Anonymous Coward]

I happen to live near the former GMT Microelectronics fab in Audubon, Pennsylvania where those chips were produced. I've been waiting for a good excuse to get my buddies together to do some urban spelunking in that building.

Re:They lost the blueprints? I'll go get em

[PatPending]

The building is For Sale or Lease according to the sign out front when Google drove by.

Re:They lost the blueprints? I'll go get em

[Anonymous Coward]

The problem with the site is that it requires several million dollars in EPA compliance work to clean up the water table surrounding the plant, which is pretty severely contaminated.

Its been a real hot issue in the plant's township for a few decades now.

Re:They lost the blueprints? I'll go get em

[PatPending]

Roger your intel, Alpha Charlie. Right. Black-Op, Sunday, 0900 zulu, rendezvous Alpha Charlie's place. Bring standard issue gear. And, oh, by the way: don't drink the water. Whiskey Tango Foxtrot, out.

Re:They lost the blueprints? I'll go get em

[monkyyy]

heres a helpful song to remember
http://8bc.org/music/PDF+format/HOAP+-+Report+on+Planet's+Drinking+Water+332-A+-+Birthday+Song/

Just imagine ....

[martyb]

Just imagine a beowolf cluster of these? No, let's have some fun with math, instead.

FTFS:

'A team of three people accumulated a bunch of 6502 chips, applied sulfuric acid to them to strip the casing and expose the actual chips, used a high-resolution photomicroscope to scan the chips, applied computer graphics techniques to build a vector representation of the chip, and finally derived from the vector form what amounts to the circuit diagram of the chip: a list of all 3,510 transistors with inputs, outputs, and what they're connected to.

Okay, bear with me here:

• 45 Years Later, Does Moore's Law Still Hold True? ("Intel has packed just shy of a billion transistors into the 216 square millimeters of silicon that compose its latest chip... which linked article goes on to report: "The quad-core desktop Sandy Bridge die clocks in at 995 million transistors." )
• Researchers Claim 1,000 Core Chip Created (By using FPGAs, Glasgow University researchers have claimed a proof of concept 1,000 core chip that they demonstrated running an MPEG algorithm at a speed of 5Gbps.)
• Intel Talks 1000-Core Processors ("An experimental Intel chip shows the feasibility of building processors with 1,000 cores, an Intel researcher has asserted. The architecture for the Intel 48-core Single Chip Cloud Computer processor is 'arbitrarily scalable,' according to Timothy Mattson. 'This is an architecture that could, in principle, scale to 1,000 cores,' he said. 'I can just keep adding, adding, adding cores.'")

So let's perform a few calculations, shall we? There are 995,000,000 transistors in the Sandy Bridge Quad Core die. There are 3,150 transistors in the 6502. That means that within the space of the Sandy Bridge chip, there could be, instead, 315,873 complete 6502 cores!!

But wait, it gets better! Back in its day, IIRC the 6502 ran at what, 1MHz? 2MHz? With today's technology, we could run each of these cores at least one-thousand times faster than the original! That's like having another thousand times as many 6502 cores.

So, finally, in the space of just one Sandy Bridge Quad Core die, we could have the processing equivalent of over 300 Million 6502 cores!(*)

(*) Okay, granted, it would take a not insignificant amount of space on die to connect all these together, along with a metric ton of lines for sending data and address info to/from each 6502 processor. Nevertheless, I'm just boggled to see how far we've come from the chip that was in the first computer I ever bought!

Re:Just imagine ....

[NixieBunny]

Just imagine if the same effort that went into minimizing the 6502's transistor count back then were to be applied to a modern CPU. It would take decades from concept to tapeout!

Re:Just imagine ....

[liquid stereo]

Its really incredible stuff. Personally, I think today's chips are pretty wimpy. Give me a Mips R8000 or Alpha 21264! But seriously, I think the lack of diversity with regards to chip design has been fairly detrimental to advances in computing.

Re:Just imagine ....

[Tumbleweed]

That means that within the space of the Sandy Bridge chip, there could be, instead, 315,873 complete 6502 cores!! ...
With today's technology, we could run each of these cores at least one-thousand times faster than the original! ...
So, finally, in the space of just one Sandy Bridge Quad Core die, we could have the processing equivalent of over 300 Million 6502 cores!(*)

That would make for one bad-ass game of LoadRunner!

Re:Just imagine ....

[jimmydevice]

Adding 64K of local memory, interface and interconnect would approach 1M transistors / core, and we're back to 1000 processors.

Re:Just imagine ....

[Anonymous Coward]

That would make for one bad-ass game of LoadRunner!

You fall into a stream, but somehow end up at the top of a precipice. Mwahaa

Re:Just imagine ....

[Anonymous Coward]

God this gives me a mental woody; tyvm for the imagery!!!

Re:Just imagine ....

[Mathness]

It doesn't follow that fitting ~300k 6502 cores onto the Sandy Bridge equals the same in processing power, only that they could fit onto it.

Additionally you should also consider the clock cycles of instructions as there have been significant improvements (as well as "special" sets like SSE) which also increase the computing power.

Re:Just imagine ....

[renoX]

I think that the 6502 had no memory cache, so 'memory contention' would be .. interesting .. to say the least.

Re:Just imagine ....

[nomorecwrd]

But you had two registers (X and Y), and a stack accumulator (A) on chip! :-D

So many great memories!! (yes, nerdy memories, but hey!! this IS slashdot and this why I'm here for)

Re:Just imagine ....

[Anonymous Coward]

So, finally, in the space of just one Sandy Bridge Quad Core die, we could have the processing equivalent of over 300 Million 6502 cores!(*)

Maybe then I could get the Gettysburg game I had for my Atari 800XL to complete a computer-player turn in something less than 3 days...

YMF262

[Anonymous Coward]

do that one! make it a device driver even!

The MAME ROM dumping project has been doing this

[NotInfinitumLabs]

They've been doing stuff like this for a long time: http://guru.mameworld.info/decap/index.html

A blank space for the electrical outlet...

[jtara]

Did you know that the very first 6502 layout had an unused space reserved for an electrical outlet? No, not an electrical outlet on the chip, silly! An electrical outlet on the wall of the designer.

I was writing the software for a firmware-based gasoline pump based on the 4040 when the Mostek 6502 was announced. The pricing and power were a huge breakthrough - we could now afford to use an 8-bit processor instead of 4-bit, and the chip was way easier to interface. We arranged a visit to Mostek and came back with a prized pre-production chip with the lid soldered on. We met Chuck Peddle, and he showed us a prototype of the KIM development board. We also took back with us a 9-track tape of a 6502 assembler (written in Fortran) for installation on the local university's timesharing system, which at the time sold time to the public.

We also met the chip designers and saw the original hand-drawn layout for the chip. No automatic routing software - just drawing on a huge sheet of paper on one wall of one of the guy's offices. There was actually an area of silicon that could not be used because there was an electrical outlet on the wall that they needed for something in the office, so they just didn't draw on that part. The finished design was then rendered in Rubylith, and we were shown a "cell library" which consisted of a set of large drawers with various circuit elements pre-cut in Rubylith.

Since the KIM was not yet available, we built our own development system - first wirewrapped, and later a set of circuit boards, using 44-bit edge connectors and defined a bus We let my friend Rene use the circuit board layouts we had, and he did some additional boards himself, laying down black tape on mylar. We were given a monitor program that would allow us to load paper tapes produced on a TTY connected to timesharing system and do some debugging.

It was a really fun and easy chip to program, and I worked on several other firmware projects using the 6502 over the next few years. I did some 6800 as well, but always preferred the 6502.

Re:A blank space for the electrical outlet...

[Anonymous Coward]

You met Chuck Peddle and Bill Mensch?

I really, REALLY envy you. Those guys created the building blocks of my world.

Re:A blank space for the electrical outlet...

[fzammett]

Ditto. Those two are on my VERY short list of people I'd most like to meet. They probably had more of an impact on my life, albeit indirectly, than any two other people save my parents.

Re:A blank space for the electrical outlet...

[CaptKilljoy]

> There was actually an area of silicon that could not be used because there was an electrical outlet on the wall that they needed for something in the office, so they just didn't draw on that part.
Being curious, I just had to look at the actual chip layout in the online simulator to see if this was true. You can see a pair of rectangular voids left of the center, near the edge, the lower of which seems to have the right dimensional ratio to have been a space for a wall socket.

It's fascinating to hear the lost bit of engineering history that explains something would otherwise have forever remained a mystery. Somebody should forward this anecdote to the visual6502.org guys.

Re:A blank space for the electrical outlet...

[PatPending]

Did you know that the very first 6502 layout had an unused space reserved for an electrical outlet? No, not an electrical outlet on the chip, silly! An electrical outlet on the wall of the designer.

It was (don't know if it still is) common for the layout "artist" to include a graphic of his choice on the die.

Re:A blank space for the electrical outlet...

[Inda]

It's common in all design work.

The strengthening ridges behind the rear number plate of the BMW X5 are in the shape of the designer's initials (or they were when I left the project just before the tooling was made).

Re:A blank space for the electrical outlet...

[anegg]

I have a KIM-1; haven't fired it up in a while... When I was a freshman in college, my KIM-1 wasn't working. I sent it in to MOSTEK for repair, but received it back as undeliverable. Luckily, a very smart high school student who lived in the college town was able to troubleshoot the hardware and replace a discrete component for me. Good times...

Re:A blank space for the electrical outlet...

[Anonymous Coward]

> You can see a pair of rectangular voids left of the center, near the edge, the lower of which seems to have the right dimensional ratio to have been a space for a wall socket.

I can't see it here http://4.bp.blogspot.com/_Ltj3p6TIy1Q/TJAYg7AutHI/AAAAAAAAAAY/8cFyDQU2yuA/s1600/6502_pad_annot_07.png or here http://2.bp.blogspot.com/_Ltj3p6TIy1Q/TJAZcx-UMSI/AAAAAAAAAAg/X9m4EnXyaSs/s1600/aligned_top_900h.png

Re:A blank space for the electrical outlet...

[noidentity]

Did you know that the very first 6502 layout had an unused space reserved for an electrical outlet? No, not an electrical outlet on the chip, silly! An electrical outlet on the wall of the designer.

So in this case, the map really was the territory, in a sense. :)

Re:A blank space for the electrical outlet...

[NotQuiteInsane]

I've exchanged emails with Bill Mensch... Well actually it was more like:

I bought a couple of W65SC02 chips. Wired them up, and two of them wouldn't do as they were told -- the logic analyser said the data was good, but the 6502's ALU wasn't returning the right results for some reason.

Anyway, Bill gets hold of my original email, then proceeds to send me a rather long list of things to check. Never did get those two chips working, but if that isn't customer service, I don't know what is!

Also: the "missing" ROR instruction isn't a bug. It was intentionally left out of the CPU, until -- in Bill's words -- "with market feedback, we realised we needed to add it and then modified the design."

Still used today

[chaoskitty]

I have a Commodore A2232 seven port serial card in my Amiga 4000 in my datacenter which provides serial consoles to a number of other machines. While other multiport serial cards have RISC processors or large buffers, this card is simply a 3.58 MHz 65CE02 which polls each port and puts incoming characters into its 16k of memory, which the Amiga can access directly. It's a beautiful example of simplicity at work.

45 years from now...

[PatPending]

45 years from now, perhaps some of you "youngsters" reading this will be trying to resurrect this site from a roomful of forgotten, dusty hard drives.

Nah...

We still do it

[Sir Holo]

Amazing bits of history in this thread...

We pros still de-lid and polish back circuits for reverse-engineering. These days, instead of an optical microscope, you need an SEM to do it. The principle is still the same.

Delicate work, but quite amazing results.

Re:We still do it

[Sulphur]

Amazing bits of history in this thread...

We pros still de-lid and polish back circuits for reverse-engineering. These days, instead of an optical microscope, you need an SEM to do it. The principle is still the same.

Speaking of SEM: TI made a video of an IC running in a SEM.

It showed the current so to speak.

I can't remember if it was the SEM video or Making of a Microchip (MMC01).

Is there any code that can damage the 6502?

[jameskojiro]

And if you use it on the simulator what happens?

Re:Is there any code that can damage the 6502?

[SnarfQuest]

Obviously, it will toast your host machine. Haven't you watched Star Trek? All computers will explode if they are given a problem they cannot solve!

Re:Is there any code that can damage the 6502?

[Anonymous Coward]

Physically damage, no. But there are unused opcodes that will crash some versions of the 6502 (there are several versions) hard, needing a reset or powercycle to get it running again.

If the emulated cpu is one of those versions, it should be able to see how it locks up.

The emulator used in VICE (the commodore emulator) stops with the message "cpu jam" if it hits one of those opcodes. Though not really comparable to this emulator, as vice doesn't emulate the chip at a transistor level.

Re:Is there any code that can damage the 6502?

[phozz bare]

Not sure about the 6502 in general, but the version used in Atari computers actually does have a (thankfully well hidden!) Self Destruct Vector. As the machine reached the end of its commercial life, Atari Corp. released the access procedure, perhaps as a sort of farewell present.

A program demonstrating its use was released, with full source code, in ANALOG computing's April 1986 issue (article here). It was aptly named "Paperweight". You can download binaries from the net but I'd rather not give a link here for obvious reasons. However, for those curious to see what a typical execution (note the humorous pun!) looks like, it's on youtube.

Oh, it works very well on an emulator! Not sure if this simulator could do it though, I doubt they have a rising smoke effect in the javascript...

;)

Re:Is there any code that can damage the 6502?

[Anonymous Coward]

There were certain things you could do on Commodore PETs that could zap the monitor, but it involved screwing the video controller [sic] with some peeks and pokes.

Re:C'mon. It's a cool page

[Anonymous Coward]

I've got to assume he was being facetious.

jewelry shop

[Anonymous Coward]

Jewlery shop online could provid all kinds of beads and other findings for fashion jewlery making, which will save money and time. But the most important thing is that you should choose a top jewlery shop which could supply good service.

http://www.arlyse.com/

Re:C'mon. It's a cool page

[JWSmythe]

    You know what's funny, that (the parent's argument) was the song of the Windows fan boys for so long. Now I'm somewhere that I have an abundance of old hardware, and a mix of new and old operating systems. I'm finding it easier to throw Linux on a box, than to pray that Vista or Win7 will work. CPU and memory wise, sure it'll work. But dear god don't try to find drivers for some fairly standard old video card (like a 4Mb to 32Mb PCI card), sound card, or printer. There are three categories for old systems. "Good enough for Win7", "Wow, a great Linux machine", and "don't bother". Why bother with a 1Ghz machine, when I have stacks of 2.0Ghz to 2.8Ghz machines to use. Oh, and if anyone is interested, I stumbled on a stack of probably a dozen 200Mhz Pentium CPU's. I have no idea what to do with them, but they'll probably end up in my own personal museum. :)

    We just played a little game with one of our techs. "Can you get Win2k3 to install" on some old Dell servers with the original Win2k3 license stuck on the case by Dell at the factory. Linux? 5 minutes from an install CD I made, or 15 minutes from the distro original CDs. Both are current. Win2k3? 3 days of head pounding, calls to Dell support, downloading and running BIOS updates, and some mystery driver emailed and being told "try this". 24 work hours for 1 server, versus 0.25 work hours. Poor guy, he was a huge Windows fan. By the end of it, he looked like he was going to personally go and bomb Redmond. :)

    The best work around I've found is to run Linux on them, and then provide any pesky Windows needs with a virtual machine under VMWare.

    I inherited several old printers, which were wonderful old workhorses of their time. My mom still on her WinXP machine, because it still runs any app she wants, and is fine for CPU and memory (2.8Ghz, and I upgraded her memory for 2GB last year). I tried to plug it into my Win7 Home Premium laptop (USB to Parallel converter required for both machines). Nope, sorry, no Win7 drivers available anywhere. No kludges other than "get a new printer". Yet still, it's supported perfectly well under Linux. heh.

    I actually haven't had a problem with a new run of the mill system, or even most exotic hardware, for years under Linux.

Bill Mensch and WDC

[Anonymous Coward]

Western Design Center will sell you a 6502 core or a chip... check it ->

http://www.westerndesigncenter.com/wdc/index.cfm

Re:C'mon. It's a cool page

[Opportunist]

Go ahead and do it, it's not like you're not allowed to.

Aww. Can't do it? Boo-hoo. But pooping on someone else's effort is much easier, ain't it?

Sorry, but it pisses me off to no end, every time someone accomplishes something, someone who obviously CANNOT has to stand up and go "What a waste of time, why couldn't they have done X?" Well, maybe because they don't give half a shit about X? And considering the crappy state the Creative-official one for Windows is in, I'd guess it is not trivial to write one without all specs available...

Can't people appreciate what others do even if it ain't their pet project?

ZX Spectrum ULA

[Alioth]

The same thing has been done to the Sinclair ZX Spectrum ULA.

Little remains of the original Ferranti technical documentation of the Ferranti ULA (which was used for the ZX Spectrum, the 5C- and 6C- series, the same ULA technology was used in the BBC Microcomputer and other European machines during the 1980s). At the time Ferranti was the leader in custom logic until they sat on their laurels and let other companies transistion to CMOS and field programmable logic and eat their lunch.

The guy who did the reverse engineering (Chris Smith, a friend of mine) has written a book about the Spectrum ULA which details the Ferranti ULA and how it was used in the Spectrum. It has quite a lot on the Ferranti ULAs, including how they were made, the process for making the IC, etc. as well as its particular implementation in the Spectrum. This was also done by de-encapsulation (which I think involves fuming nitric acid, rather than H2SO4).

Chris's work can be found here (if you want to buy the book, he gets more of the money if you order it through his website rather than Amazon, but it's also available under the GFDL):

http://www.zxdesign.info/book/

I think it's very important that this kind of thing is preserved, the early personal computers are a bit like the equivalent of the early Industrial Revolution textile machinery, but fortunately can be preserved at an individual level rather than needing a huge rich museum.

Re:C'mon. It's a cool page

[Corporate Troll]

Just to add in a little anecdote of mine. I recently upgraded my brothers PC, in the sense that we replaced motherboard/CPU and RAM. The rest stayed the same. As he now has 16GB RAM, we decided to shell out for Win7 Pro 64-bit instead of reusing the OEM XP license (which would be borderline in legality).

Anyway, that machine has a Linksys WMP54G (version 1.0) wireless PCI card. Silly me, expected it to work by default in Windows as the card was a bit older and my experience from XP was that "if it existed prior to release, there is a driver". Well, no.... *sigh*

Linksys website, no avail. Tried their customer support which said "Not Supported". Call me old-fashioned, but a NIC should be one of the things that is least-"obsoleteable". Luckily, I found a forum post (look for the post by "skinnypirate") that this particular card uses a RALink Chip and you can use their drivers.

Now, I freely admit I didn't try Linux on this machine... (Gaming machine) Perhaps it wouldn't work either... Who knows...

As for Win7 obsoleted scanners... Before throwing them out, consider the following first: VueScan (Not affiliated, just a very happy customer). It's an amazing little piece of software that seems to operate pretty much any scanner you throw at it. My dad has a SCSI Dia/Negative-Scanner with advanced functions for rewinding film and stuff like that. On Linux, XSane doesn't even detect it... VueScan sees it and all functions can be used. Same for my wifes cheap Canon LiDE20. Not supported on Mac OS X 10.6.x at all... VueScan sees it and it works perfectly. Old SCSI AGFA SnapScan 310? No problem... (XSane does this one too though.)

I've never used in on Windows though... Only on Linux and Mac OS X. I have no reason to believe that the quality of the software is less on Windows. You may argue that the software is more expensive than a new cheap scanner. True, but once you have the software, you'll never use another one ever again.

win2k3 installed on 900mhz celeron, 1hr, BAZINGA

[cheekyboy]

Dude, I used generic as basic as you can get crap machines 900mhz found in a dumpster, with 512meg ram, and its 40gig HD reformatted, win2k3 using 'googled results' licence worked quickly. Less than 90meg ram on boot used. Sp2 doubled that.

But it worked great, so either your hardware was junk or your dude tried too hard.

Wot no BBC Micro?

[mustrum_ridcully]

Thought I should mention the BBC Micro (aka Acorn Proton), manufactured by Acorn Computers (RIP) who gave the world the ARM microprocessor, as it also used the 6502. BBC micros equipped with a 6502 2nd processor were actually used to develop the first generation of ARM processors. So yes the humble 6502 is a pretty important processor, if Acorn had used the 6800 or 8088 then we might not have the ARM processor today.

Re:Wot no BBC Micro?

[Stooshie]

Yay, BBC Micro! My first computer. I learned Assembly on that. I haven't done assembly for years. Most languages we write in these days must be L5 or even L6 languages. I found myself having to really dredge my memory to understand some of the code in that article.

Re:Wot no BBC Micro?

[Alioth]

I'm actually using a BBC Master now to do some data preservation of a project I did at school in the late 1980s. I've got the data off the SJResearch MDFS-formatted floppies, now just need to get it onto the USB stick (yes, my Master has a USB interface!) I'm actually writing a program in 6502 asm to do the job.

Re:Wot no BBC Micro?

[Anonymous Coward]

Acorn is not RIP. They renamed to Element 14, and are still doing business licensing ARM, afaik.

Re:Wot no BBC Micro?

[Anonymous Coward]

I transferred all my BBC micro code and games to my PC a couple of years ago using a serial cable. It was a little unreliable so I wrote some 6502 assembly for the first time in years to CRC the data. I was surprised at how well it still worked: out of more than 100 floppies only four had bad sectors.

Changing the subject: a few years ago I read about someone creating a working BBC micro clone as a system-on-a-chip but I've lost the link. Have you ever heard of this?

CPU speed

[s-whs]

Ca. 2000 I did a speed test with a few machines: Pentium pro 200, AMD K6 233, and my old BBC micro, running a program to find all the possible permutations of the upper layer of a rubik's cube with a given number of quarter-turns. I did this already in 1984/1985 but couldn't get very far, I think 11-12 speedwise was the maximum. In the end I got the result that all permutations can be reached in 19 quarter turns. This took about 60 days of CPU power, which I split over a few machines. Then I found a bug in the program :-( This made all results worthless so I programmed in code to check if that the cutoffs I made in the program, were correct. This resulted in a speedup that eventually made the program run in ca. 2 days on the ppro, instead of 60. With all the low level optimizations I'd done the only speedup was a few percent! This showed me once again that algorithmic improvements are what you need to spend most of your time on...

Allright, the interesting bit was how much faster a pentium pro 200 was (which was slightly slower than the K6-233): About 500 times faster than the 2 Mhz 6502 (this is without any cutoffs, which I did with handmade + large computer tables). This was not that impressive but of course, all the program did was move bytes around in a fairly small set of datalocations (for which I used zpage on the 6502). I did just program the ppro/k6 using gcc, so no assembly as I never really got comfy with x86 assembly language and using C seems more appropriate anyway as that's what you program these CPUs in in most cases (I'm not sure if I checked the output of gcc -S to see what quality asm gcc produced).

An AMD 3000 (ca. 2.2 GHz) that I bought several years ago is about 10 times faster than the ppro200, so ca. 5000 times faster than the 6502 at 2MHz.

Of course you have to keep in mind the small memory of the 6502, it will get slower once you allow it to acces more memory, and for the large cutoff tables (the computed ones were many megabytes) it would need a super-6502 that has such large memory addressing ability which would be slower. So the factor for the algorithmically optimized search program on a ppro vs. a super-6502 would be higher than 500.

Re:CPU speed

[Anonymous Coward]

Pentium pro 200, AMD K6 233, and my old BBC micro,[....]I did this already in 1984/1985

Me thinks you made a boo boo. No Pentium Pro in 1985.

Re:CPU speed

[gander666]

He did say that in circa 2000 he did the work with the PPro, but that he had done some work in 1984/1985 that wasn't fruitful (didn't get far). Probably due to limitations in the available HW.

Re:CPU speed

[drfreak]

According to this article, you just might be on to something...

Re:win2k3 installed on 900mhz celeron, 1hr, BAZING

[Dr_Barnowl]

It's not about the hardware being unsuitable - it clearly was, since the machine came with win2k3 server out of the factory (or it wouldn't have an original license sticker on it). It's about it being a royal pain in the ass to actually install Windows on some hardware.

The major issue with systems like these is probably drivers for the disk controllers. You obviously can't install an OS without using the disk controller. Everything else you can install afterwards. But what if your OS install disk doesn't have a driver for the disk controller in your machine?

For Windows, the answer has historically been to bang F6 at the appropriate moment during installer boot and shove a floppy disk containing a special "OEM" driver in. What's that? Your machine no longer has a floppy drive because they are an obsolete technology? It doesn't work with your USB floppy drive because there is only a limited set that are supported? Then you are left cooking your own install disk with the OEM drivers on it, using nLite or similar.

Linux tends to solve this problem by just having drivers integrated into it. One of the advantages of the Linux driver development model is that because the source is available for most drivers and most devices in a similar class need similar driver code, adding a new driver typically just involves adding either a very small abstraction layer or sometimes just a row in a list of supported devices, so the basic kernel can have support for a vast selection of devices in a relatively small space which happily fits on an install CD. Drivers only get thrown out of the main kernel when they get extremely obsolete AND a maintenance headache, so Linux tends to support more and more hardware each year.

With the Windows proprietary model, of course, everyone has their own super-secret-sauce driver, so it's impossible to fit them all on the install disk. Older drivers must be frequently dropped from the disk image or it won't fit in 700MB. Hence OEM floppies and banging F6. Drivers for ubiquitous consumer hardware are probably more likely to be on there than obscure server hardware, so I'm not surprised you had a different experience.

I suspect the GP poster also encountered a lot of problems finding device drivers for the other components in his server, like the network adapter.

Re:C'mon. It's a cool page

[UnknownSoldier]

Yeah, it is cool.

The funny thing there was an old Apple ][ program that would let you single-step programs and it would visually trace how the bytes moved from memory to/from A/X/Y and the status flag.

Can't remember the name of it though! Something along the lines of "Inside the 6502" Can't find it on Asimov either :-(

Commodeore PET

[merde]

The first machine I ever programmed was a Commodore PET - a 6502 with 16k memory and mono video.

I then spent hours and hours and hours playing the newly programmed version of Spacewar against my friends...

Damn - those were the days. I don't have time for games any more :(

Awesome!

[chiark]

To me, Chuck Peddle is an absolute inspiration. He's not done the easy thing, or the materialistic thing, but the right thing many many times in his career.

These were the early days of the computer revolution, and I strongly recommend Brian Bagnell's book, Commodore: A company on the edge, to anyone remotely interested in the era... It's a healthy dose of realism and a perfect antidote to historical revisionism that seems to be coming from a couple of areas in the States...

The guy is a hero, as were the small teams laying the foundations that, ultimately, means we all have more interesting jobs. No article on the 6502 should fail to mention Chuck Peddle and the team at MOStek, which ultimately became part of Commodore... History tells us that what becomes part of commodore burns brightly, but briefly...

Get that book, it's great.

Do a discrete version?

[mangu]

Considering that a 1000-pack of transistors can be bought for $50, the total price for the transistors today would be less than the $300 a 6800 cost in the late 1970s, even unadjusted for inflation.

So, here's a challenge for that badass electronics hobbyist: build a working 6502 out of discrete transistors.

Re:Do a discrete version?

[Megane]

Yeah, it could end up something like this.

More orthogonal??

[Joce640k]

No increment/decrement accumulator instruction. Addressing modes that only worked with the X or Y register.

I'm not saying the Z80 was more orthogonal but the 6502 was a shining beacon. If it appears more orthogonal it's because it only had three registers so there's not much room to build non-orthogonalities into it.

Re:C'mon. It's a cool page

[Anonymous Coward]

So Motorola were blatantly profiteering before some employees decided to do another verison themselves - and a giant leap in technology was created (as it was 10 times cheaper).

I think Bill did something similar didn't he? Or we'd all be paying 10K for a Mac by now.

The same needs to be done with drugs... those GlaxoProfitKlien's of the world that are profiteering from human suffering.

The 6502 was never lost - WDC owns the core

[carlhaagen]

Not sure why the original poster claims that the 6502 core design was lost in time. WDC (Western Design Center http://www.westerndesigncenter.com/wdc/ ) bought the core design and the rights to license it many many years ago. These are the guys behind a few variants building on top of the 6502 as well, like f.e. the 65816 used in the Super Nintendo.

Re:The 6502 was never lost - WDC owns the core

[ishobo]

That is not correct, they never bought a license. WDC co-held the design with MOS. WDC has been selling 6502 based products since 1979.

Re:The 6502 was never lost - WDC owns the core

[Anonymous Coward]

that's not what he said, though.

"...bought the core design and the rights to license it..."

Re:There are already schematics on google from 200

[Arlet]

The guys that are doing the 6502 simulator have been long aware of this. This is just an independent effort, with some overlap, and some novel stuff. For instance, the schematic doesn't show you the exact area of the transistors, which can be important. By comparing the two you can also identify some mistakes that were made in either process, which is never a bad thing.

6502 still available, in silicon and cores

[ishobo]

There is nothing to preserve, the 6502 is still being manufactured, available in silicon and cores (both hard and soft).

http://www.westerndesigncenter.com/wdc/

Re:6502 still available, in silicon and cores

[Arlet]

Sure, but that's not the same silicon anymore. The original NMOS 6502 had certain quirks and bugs that changed or got fixed in later versions. For nostalgic reasons, people still like to preserve the design of the very first 6502.

processor volumes by family across the years?

[niks42]

I guess that the 6800 and 8051 families have won the race for the most chips ever shipped now - they get used in smartcards- but I don't know what the estimates are for how many of each processor family have ever been shipped. Anyone know? I've tried googling for obvious search terms .. thought I would ask /.

Re:C'mon. It's a cool page

[Anonymous Coward]

"Can you get Win2k3 to install" on some old Dell servers with the original Win2k3 license stuck on the case by Dell at the factory. Linux? 5 minutes from an install CD I made, or 15 minutes from the distro original CDs. Both are current. Win2k3? 3 days of head pounding, calls to Dell support, downloading and running BIOS updates, and some mystery driver emailed and being told "try this". 24 work hours for 1 server, versus 0.25 work hours.

That's not exactly fair.

To be fair, you should use a linux CD that is at least as old as Win2k3 or the server, or install the latest Windows Server (maybe not possible if you can't find the x86 demo, or the processor isn't x64).

What I find is that old machines come in two categories:

1. The machine is OLDER than the OS you are installing, so the OS has had time to add support for those devices. Think SATA AHCI support and RAID support in Windows XP vs. Windows 7.

2. The machine is NEWER than the OS you are installing, so the OS probably DOESN'T have support for those devices. So you'll need to "get" the drivers from Dell, or wherever. If the machine wasn't particularly popular, or you had something stupid like a GMA500, you'll need to track down this stuff the hard way as many websites will no longer list those old drivers.

What you did was take a machine and try #1 with linux and #2 with Windows. Perhaps reverse, and try again.

I'd like to see you try and get linux working (any release prior to the machine's manufacture date) no problem on this machine, and install the Windows 2k8 x86 demo. I bet it will be the reverse.

Re:win2k3 installed on 900mhz celeron, 1hr, BAZING

[domatic]

DriverPacks are your friend here. Of course if you slipstream 2 GB of drivers into a Windows install CD then you'll want a DVD drive in the machine.

Re:win2k3 installed on 900mhz celeron, 1hr, BAZING

[VGPowerlord]

The major issue with systems like these is probably drivers for the disk controllers. You obviously can't install an OS without using the disk controller. Everything else you can install afterwards. But what if your OS install disk doesn't have a driver for the disk controller in your machine?

Most of the time, the generic disk controller drivers are "good enough" to get the base OS installed. Of course, if it was a server circa 2003, it likely used SCSI disks, and yes, there are about a billion and a half different SCSI drivers... although, to be fair, EIDE/SATA with RAID has the same problem.

Even worse, if you had an IDE (or SATA, iirc) drive >137GB, WinXP (and likely 2k3) wouldn't recognize the disk even WITH appropriate drivers unless your disc had SP1 or newer on it. Either that or the installer would crash, I can't remember which.

For Windows, the answer has historically been to bang F6 at the appropriate moment during installer boot and shove a floppy disk containing a special "OEM" driver in. What's that? Your machine no longer has a floppy drive because they are an obsolete technology? It doesn't work with your USB floppy drive because there is only a limited set that are supported? Then you are left cooking your own install disk with the OEM drivers on it, using nLite or similar.

Yes, when you deal with older OSes, you have to deal with old limitations as well. Windows Vista, Server 2008, 7, and Server 2008 R2 all support CD/DVD and USB flash drives for loading OEM drivers.

With the Windows proprietary model, of course, everyone has their own super-secret-sauce driver, so it's impossible to fit them all on the install disk. Older drivers must be frequently dropped from the disk image or it won't fit in 700MB.

Why 700MB? Windows discs are DVDs now, bumping that to 4.7GB.

Linux tends to solve this problem by just having drivers integrated into it. One of the advantages of the Linux driver development model is that because the source is available for most drivers and most devices in a similar class need similar driver code, adding a new driver typically just involves adding either a very small abstraction layer or sometimes just a row in a list of supported devices, so the basic kernel can have support for a vast selection of devices in a relatively small space which happily fits on an install CD. Drivers only get thrown out of the main kernel when they get extremely obsolete AND a maintenance headache, so Linux tends to support more and more hardware each year.

No, the problem here is that Microsoft only writes generic drivers and drivers for its own hardware (mice, game controllers, etc...). For specific devices, it relies on the manufacturers to provide drivers... and most of them don't provide older device drivers for newer OSes.

There *are* exceptions. nVidia and ATI (now AMD) are two. The catch here is that, at least with nVidia, their "Universal" drivers only go so many generations back. Right now, they go back to the GeForce 6 line, which were released in mid-2004.

Am I the only one?

[killmenow]

"the details of the designs have been lost or forgotten. While there have been great efforts to reverse engineer the 6502 from the outside, there has not been the hardware equivalent of the source code — until now"

Heeere we are!!! Born to be kings! We're the Princes of the Universe...

Come on...am I the only one who thougth of that?

Terminator and other pop culture references

[adam100]

A well-known bit of trivia (if it's well-known, is it still trivia?) is that the original Terminator movie used 6502 assembly instructions in the "Terminator View" shots. Other pop culture references to the 6502 are listed in the section "6502 in popular culture", on the page http://en.wikipedia.org/wiki/MOS_Technology_6502. The Atari 400 (and 800, and others) also used the 6502.

Re:Terminator and other pop culture references

[Anonymous Coward]

"trivia" has nothing to do with known or not known. Trivia is something of small importance.

http://www.google.ca/search?hl=en&biw=1766&bih=893&defl=en&q=define:trivia&sa=X&ei=RyonTf_nJsH88Aa6rJyWAQ&ved=0CBUQkAE

Re:C'mon. It's a cool page

[orange47]

yeah, right, go find linux drivers for tv card like theatrix..

Re:win2k3 installed on 900mhz celeron, 1hr, BAZING

[Anonymous Coward]

Dude, linux is garbage and there really is no comparison to Microsoft alternatives if you remove the factor of cost. I've tried to install linux on lots of computers over the years, and there is always something that doesn't work. Then fucking hours get spent reading forums and following ambiguous how to documents.

Windows has always worked on this hardware with easy to find drivers that are easy to install and don't require a kernel compile. Windows 7 has the fucking drivers built in for all of this.

Did you really just state that linux has built in RAID controller drivers for every controller out there? That's fucking laughable. I guess you said it "tends to", whatever that means.

Also, once you get the Windows Server and linux servers installed, what do you have with the linux server? With linux you have no directory service, no way to apply anything similar to group policy, no good messaging service that compares to Exchange.

Linux is a kludge of inconsistency and bloat. Yea, bloat. Have you seen your repositories lately?

Now I've gone easy on you not comparing linux server to Microsoft Server Core. Holy shit, linux is a fucking bloated kids toy compared to Server Core.

Re:C'mon. It's a cool page

[Hatta]

There are four categories for old systems. "Good enough for Win7", "Wow, a great Linux machine", "don't bother", and "Commodore 64? SWEET!".

FTFY

your favorite CPU sucks

[Thud457]

there, now that that's been covered, can we move on now?

going way back to where there were different architectures.
Oh wait, I guess we're x86-64 vs ARM these days.

One other 6502 of note: The Atari 800

[EQ]

I loved the assembly language and the ease of programming this chip's design made possible back in the day. I still recall how we crammed more routines into the 800 by deactivating the ROM during the vertical blank interrupt, and hit the RAM below it for sorting and running tightly coded (in assembler) routines, mainly for drawing and moving sprites. Funny how much we could accomplish in only 32K, 48K or 64K of RAM in those days.

Re:C'mon. It's a cool page

[JWSmythe]

    I haven't run into a C64 in years, so that category was dropped. It was actually "Sweet, I used to own one of these...." TI-994A, Apple ][e, C64. If I actually see one in person somewhere, I'll be excited. The last one was in the early 90's, sitting in the back of an old nonfunctional truck, in a barn, with a bunch of other scrap. I figured after almost 20 years, it was time to let it go.

Re:C'mon. It's a cool page

[Anonymous Coward]

please don't talk to me about vmware running on Linux. Every time you update your linux box's kernel, you'll break vmware. Sometimes easy to fix, sometimes a real pain.

Funny thing about nostalgia for old systems...

[mswhippingboy]

Every time a story like this is posted, everyone begins reminiscing about the "good ole days" when they had to enter their programs using toggle switches or paper tape.

Fortunately, most of the platforms of old are still around and anytime I get to feeling nostalgic about a particular machine I used to work on, I just fire up one of the many emulators available. After about 10 minutes of playing around with it and the reality of how much dealing with the limitations sucked, I simply shut it down and get back to my current 8Gb Quad-core machine and I'm happy again.

For those of you who haven't checked these out and are feeling nostalgic, here's a few links that may bring back some memories...

http://applewin.berlios.de/ (Apple IIe emulator)
http://www.discover-net.net/~dmkeil/ (Various TRS-80 emulators Model I, III/4, Coco 1,2 & 3)
http://www.altair32.com/index.htm (Greate Altair 8800 emulator - complete with front-panel)
http://www.viceteam.org/ (Various Commodore emulators - C64, C128, VIC20, PET)
http://fms.komkon.org/Speccy/ (Sinclair ZX)

For the greybeards out there
http://www.hercules-390.org/ (IBM System/370/390/z emulator)
http://www.ibmsystem3.nl/emulators.html (IBM System 3 emulator - anyone remember this baby!)

And for the "whitebeards"...
http://members.optushome.com.au/intaemul/Emul1401.htm (IBM 1401 emulator - Autocoder anyone?)


Go ahead, get it out of your system so you can stop pining for the "good ole day", that were, in truth not really as great as they seem in restrospect.

Enjoy...

Re:C'mon. It's a cool page

[JWSmythe]

    Actually, I don't believe I made such a stupid mistake. No, I totally agree with you. I gave up on VMWare several years ago, using VirtualBox instead. The reason was exactly as you said. I wanted a new kernel for some new feature, and VMWare would only work on something several versions behind. That was back when it was Innotek's Virtualbox, before it was bought and then absorbed in a merger. :)

    The only thing I've had a problem with lately was getting OSX to run right in a VM. Well, running wasn't very hard. Getting it to dynamically resize the screen just plain doesn't work (with good reasons). The whole reason was for my girlfriend, so my machine would look like her Mac, except when I switched to the other OS's. Once I had it working I was like "what now?" It reminded me of installing BeOS. :)

Re:There are already schematics on google from 200

[Anonymous Coward]

Apple II clone actually

Re:C'mon. It's a cool page

[I8TheWorm]

A lot of that has to do with faulty manufacturer drivers, the cause of a great many BSODs (I'm sure most /.ers have seen at least the comments from the NT4.0 code). MS completely revamped their drivers and signing for Vista/Win7, vastly improving the stability of their OSs.

And to be fair, I'm not a Microsoft fanboy, my favorite OS right now is Ubuntu. "It just w..." oh wait, I think that comment is trademarked now.

Re:C'mon. It's a cool page

[Chaos Incarnate]

Now, I freely admit I didn't try Linux on this machine... (Gaming machine) Perhaps it wouldn't work either... Who knows...

Wireless card in Linux? Nope, odds are that it wouldn't work either. :)

The hidden language of...

[Anonymous Coward]

Totally digging the Javascript sim. I've been waiting for something like this ever since I read Charles Petzold's "CODE: The Hidden Language of Computer Hardware and Software"
http://www.charlespetzold.com/code/

Re:C'mon. It's a cool page

[Corporate Troll]

Been told RALink is Linux-Friendly. I booted it up in Ubuntu before installing 7, but I had it wired up and didn't think of checking. If I find time this weekend, I'll try to test it. The machine is still at my place, as I had some trouble installing a game (defective CD-Rom it seems)

The worst, I had was an Atheros based chip, but on all machines I had wireless problems with this worked. (Obviously, if I didn't have problems, it was supported) Granted, it was always on Ubuntu which make this stuff braindeadly easy. The other distribution I use regularly is Debian, but rarely on desktop/laptop computers.

Re:C'mon. It's a cool page

[Hatta]

They're still as much fun as they were 20 years ago. Modern computers are boring, so I'd say even more so. Computing was once an end in itself, these days it's just a tool.

BBC Master with USB

[bagofbeans]

Care to share details of the USB implementation?

And you may be interested in John Kortink's MMC card storage system for BBC B and Master (at http://web.inter.nl.net/users/J.Kortink/home/hardware/gommc/index.htm )

Re:C'mon. It's a cool page

[lwsimon]

There are still exceptions. I have a Toshiba notebook that refuses to work right under Linux. No ACPI, flaky wifi driver, etc, on a 2010-era machine. Deplorable.

Re:There are already schematics on google from 200

[Anonymous Coward]

My hovercraft is full of eels.

Re:C'mon. It's a cool page

[ncc74656]

Been told RALink is Linux-Friendly.

I have an 802.11b/g USB dongle kicking around somewhere that uses one of their chips; getting it working was almost trivial.

The worst, I had was an Atheros based chip, but on all machines I had wireless problems with this worked.

I've had fairly good luck with their chips. First one was a CardBus 802.11a/b/g card I used with one of my notebooks that didn't come with WiFi; I continued using it with a newer machine whose built-in Broadcom WiFi didn't start working right in Linux until more than a year after I bought it. Most recently, I picked up one of these for a MythTV frontend. 802.11n still doesn't work (not even with the carl9170 driver in the just-released Linux 2.6.37), but it's worked well enough at 802.11a to stream HD MPEG-2 (as well as less bandwidth-intensive formats) for the past few months. At some point, 802.11n will probably start working. It'd be nice to have it now, but it's not been a deal-breaker.

Nostalgic feeling != great tech

[Kaz Kylheku]

The 6502 was a piece of crap.

A great piece of tech, introduced just four years later in 1979, was the Motorola 68000.

This is probably the only processor worth studying, from that era.

Let's look at the 6502: eight bit registers only, and just three of them. 256 byte stack. 16 bit address space. Invalid opcodes execute strange actions instead of vectoring to an exception. Synchronous interfacing with no recovery: if a peripheral doesn't produce data within the clock period, the processor simply reads garbage!

By contrast, the 68000 gave us 32 bit data and address registers (well not all address lines useable in the first generation). Wonderful instruction set. Decent interrupt and exception handling for operating system development. Terrific interfacing with handshaking protocols for slower peripherals.

The only thing that's amazing is some of the cool software that was produced for computers based around chips like these.

What's also amazing is how long that 6502 lasted: how a processor introduced in1975, it continued to be used right up to the mid 80's for home computers. Unthinkable today.

What's amazing is the incredibly retarded follow ups too, like the 65C816 with its segmented memory model, well into the era of decent microprocessors.

The 6502 belongs in the same historic trash bin as the Intel 8086/88 and its progeny.

Re:Nostalgic feeling != great tech

[metaforest]

I'll point out that the 65816 had to be binary compatible with the 6502.

Also I'd point out that the 68k line had some similar constraints to be binary compatible with the 68000.

You are correct.... the 68K was an awesome CPU in it's day. However: the 6502 was MUCH easier to write code for. Writing a boot kernel for the 6502 took less than a day. Writing the same for a 68K took a week. Why because the 68K required a lot of crap set up in your boot code to get a sane execution environment. With the 6502, the environment was mostly sane before the first instruction was executed.
Hardware development for the 68K required a lot of glue logic to deal with the async I/O bus. Almost no glue logic is required to set up a 6502 bus.

Comparing the two CPUs is not even fair. They had very different markets and design goals.

Re:Nostalgic feeling != great tech

[bored]

I know your probably trolling, but I would just like to point out a better comparison to the 6502 would probably be the 6800 (or 8080), not the 68000. When you do that you find that they are very similar in their limitations, as well as their release date

Back in the 70's computer designs were accelerating a lot faster than they are now. When you didn't have infinite transistors for the logic aspects of a CPU design things got cut. The 6502 (and 6800) ran in about 4k transistors, compared with the 68k which was about 68k transistors, or about 4 generations newer. See http://www.classiccmp.org/pipermail/cctalk/2001-May/170288.html. When you compare the 6502 to a CPU of its generation it comes out looking really good. By the time the 68k was out, the world had changed. Of course the 6502 (production in 1975), vs the 68k (production in 1980) game can also be played the other way. Compare the 68k with the ROMP, or if your willing to wait another 5 years the R2k.

Re:C'mon. It's a cool page

[smithmc]

Cool? It's stupid. It's someone who had too much time on his hands and a healthy dose of autism.

Given your lack of sensitivity, one wonders where you might fall on the A-A spectrum...

s/senior/senior moment. /

[sconeu]

I need to learn to preview

Re:C'mon. It's a cool page

[NotQuiteInsane]

One of the (many) nice things about Linux is the way the printing system is designed. For CUPS, you generally only have to write a format converter (which takes a PNM and spits out printer control codes). CUPS deals with the issues of talking to the printer, and rasterising incoming PostScript vector data into PNM data.

Plus most of the major format converters are OSS/FS -- HPLIP (the PCL converter), Gutenprint (nee Gimp-Print), and a bunch of add-ons (like the P-touch label printer driver). Write a format converter and a quick PostScript PPD and you're in business. Porting to a new platform is often as simple recompiling, assuming the code is bit-width agnostic (32/64/128bit clean).

On the Windows platform, it seems the driver interface changes with every major version iteration, for better or worse... Win2k drivers could occasionally be coaxed into working on XP 32-bit (and vice versa), but that's pretty much the end of it. WinXX drivers run in kernel-space, even if all they're doing is printer data generation, where they don't actually *need* any hardware access which couldn't be provided in a better way by a HAL...

6502 Multi-Core

[transami]

I always wondered what it would be like to stick 1,024 of these on one die for parallel processing.

Re:C'mon. It's a cool page

[benhattman]

Can't people appreciate what others do even if it ain't their pet project?

Do people need to appreciate what others do for hobbies? Personally, every time I see a story on /. about resurrecting some ancient hardware I wonder why. The whole "I'm rocking old hardware" hobby barely more valuable to society than if someone spent a week playing solitaire.

I don't see a problem with putting the story on /. Some people will appreciate seeing the story. And I don't think there's a ton of value in posting to the story that you think they wasted their time (like the GP did). But even granting all that, why should I be inclined to appreciate something I have no personal interest in?

Re:win2k3 installed on 900mhz celeron, 1hr, BAZING

[toddestan]

Working with Dell server-class hardware from years past can be extremely infuriating. It's a bunch of propriety, non-standard stuff that Windows doesn't recognize out of the box, and Dell seems more interested in selling you a new server instead of helping you get that dual Xeon 1.5 running again. So yeah, good luck finding the drivers and utilities you need for it.

On the other hand, everything supports your standard home/office Dell PC. I put Vista on a 2004-era Dell P4, and didn't even have to install a single driver.

citation needed?

[dwater]

So,someone can now fill in a citation for wikipedia?

"
The 6502 design, with 3,150 transistors,[citation needed] was...
"

Re:Watch this...youtube video...

[steeleyeball]

http://www.youtube.com/watch?v=EJcbxrdErkY a Sidmulation and capacity for glue logic...

Re:There are already schematics on google from 200

[badkarmadayaccount]

Bulgaria uber alez! :-) Feelin' proud.

Re:Yep. I like it.

[badkarmadayaccount]

Subscribe to the slashdot newsletter, and just batch through the links whenever - you'll find out everything has been uns-slashdotted.

I wish they'd do the same with the Amiga chipset

[Anonymous Coward]

It would be nice to have full diagrams of the workings of the Amiga custom chips. In fact it would help nostalgic hardware developers reverse engineer the entire system to make compatible system so I could run all my old games in hardware not emulation. 8^)