EEVblog #1308 - 1970's Intel MCS-85 8085 Design Kit

[EEVblog]

Will a 1970's era Intel 8085 design kit power up after 40 years?
A look at the Intel MCS-85 System Design Kit and some vintage computer and processor history.

[retrolefty]

Nice visit down microprocessor history, well done Dave. I was a Z-80 fanboy during the day.

[DrG]

Enjoyed watching.

FYI: The Jan, 1979 Byte magazine features the first of a three part series titled, Build a Computer Controlled Security System for your Home, by Steve Ciarcia (who else?) beginning on p. 56. It used the SDK-85 kit that you were programming. At the end of the article the SDK-85 kit is listed as costing US$250.00.

https://archive.org/details/byte-magazine-1979-01/page/n57/mode/2up

[johnlsenchak]

Intel System Development Kit
https://en.wikipedia.org/wiki/Intel_System_Development_Kit

[wilfred]

I'd beg to differ that the 8008 was an 8bit version of the 4004. The 8008 was independently developed as a TTL CPU by Computer Terminal Corp in Texas. They then sought to have it built and approached Texas Instruments and Intel to make a microprocessor of equivalent function. One turned out too expensive and late, and the other too slow.

A story that is fascinating. https://en.wikipedia.org/wiki/Datapoint_2200


P.S. I am enjoying the electronics videos of late.

[drvtech]

I started work in 1977 and our state-of-the-art system was an MCS-80 Kit (based around the 8080 of course). We had an ASR33 Teletype and we had our own boot loader which we typed in to make it read a program from paper tape. We could also dump to the tape punch. We stored our programs in TTL SN74S472 (I think) bipolar PROMS - one time programmable as they used physical fuses internally. Occasionally we could patch them so long as the patch required adding '1's to the program as they started off all zeroes. Initially we had to take a paper tape over to a nearby company that had a programmer but one of my early jobs was to design and build a programmer for these chips so we could program in house. Later we moved on to the Z80 as our core device as migrating the 8080 code to the Z80 was trivial. About the same time we moved on to 2716 EPROMs (which needed multiple supply rails) This certainly brought back a few memories - thanks Dave.

[firewalker]

I have this one:



Alexander.

[artag]

8085 was intel's offering for 8080 upgrades. Like the Z80 it had a less awkward power/clocking requirement than the 8080 and, unusually, an on-chip UART.

But it didn't have the Z80's extra registers.

I'm not sure how early in the Z80's life it happened but another huge advantage of the Z80 that intel failed to see coming was multiple sources. Zilog, of course, but also Mostek. Were there others ?

[pickle9000]

I do not miss the "Good ol days".

[David Hess]

8085 was intel's offering for 8080 upgrades. Like the Z80 it had a less awkward power/clocking requirement than the 8080 and, unusually, an on-chip UART.

The improvement in the power requirements cannot be overstated; the 8085 only required +5 volts while the 8080 also required +12 and −5 volts.

The first programming language I learned was 8080 machine code.  I did assembly by *hand* on paper forms designed for this purpose.

[Fungus]

If you put ABCDEF on a single row then where do you put the zero?

"Review the Assembly Language Reference and the MCS-80-85 Manual before attempting to write programs"

Good advice 

[drvtech]

The first programming language I learned was 8080 machine code.  I did assembly by *hand* on paper forms designed for this purpose.

Such luxury - we used graph paper. Got very proficient at remembering the op-codes!

[coppice]

A lot of the dates shown for these early parts must have been when design began, because they were definitely not being talked about at those times when you met with the vendors.

[Fungus]

The first programming language I learned was 8080 machine code.  I did assembly by *hand* on paper forms designed for this purpose.

Such luxury... you must have worked at a large company to get perks like that.

I used to do 6502 and Z80 hexadecimal by sight (...on a CRT I could see more than one byte at a time, unlike the board in the video).

My first machine code games were written using the hex monitor thing on a Commodore PET after I figured out that pokeing the screen in BASIC was never going to cut it.

[David Hess]

The first programming language I learned was 8080 machine code.  I did assembly by *hand* on paper forms designed for this purpose.

Such luxury - we used graph paper. Got very proficient at remembering the op-codes!

It was just a preprinted form with spaces for address, opcode, operands, and the hex translation which you had to do yourself.

I used to do 6502 and Z80 hexadecimal by sight (...on a CRT I could see more than one byte at a time, unlike the board in the video).

I learned 8080 hex pretty quickly but documenting the assembly was good practice.  I still remember some like C3 for JMP.

My first machine code games were written using the hex monitor thing on a Commodore PET after I figured out that pokeing the screen in BASIC was never going to cut it.

I used a Prolog PROM programmer to burn the EPROMs directly by typing the hex code into the number pad much like Dave shows.

[floobydust]

This stuff was all very expensive back then, especially after you added RAM and support IC's. A CPU alone was $100 in today's dollars. I thought muxed address/data bus needs more clock cycles per instruction (CPI), so the Z-80 won over that.

My first 1978 dev board was a RCA CDP1802 COSMAC ELF 256 bytes of RAM and CMOS, from Quest Electronics. It's the weirdest CPU architecture I've ever seen and still have no idea how to code for it, just brutal.
Later I went Z-80 and still have the Assembly Language Programming by Lance Leventhal, programming cards and books. That CPU is a great teacher, it was so much fun knowing "BC" stood for byte-counter...

My buddy sold his KIM-1 for $2,000 on eBay, someone has one for sale $9,995 right now. I don't see that much nostalgia for these.
Byte Magazine archives

[coppice]

If you put ABCDEF on a single row then where do you put the zero?

Lots of microprocessor development kits had 4x4 grids for the hex keys. People were not so used to numeric pads in those days, and the two forms of keypad they were starting to use - DTMF phone and calculators - had completely different layouts. So, arranging a hex pad as an extended numeric pad didn't make as much sense then.

[bitwelder]

At the end of the article the SDK-85 kit is listed as costing US$250.00.

Which according some inflation calculator on the 'net should be equivalent to about US$ 980 in 2020.

[coppice]

At the end of the article the SDK-85 kit is listed as costing US$250.00.

Most microprocessor evaluation kits were around that price, which was quite a lot in those days. It was probably around cost price. Some competitors offered $100 kits for their microprocessors, which were probably shipped at a loss, when they were desperate to get some traction. A LOT of microprocessors appeared between 1974 and 1977, many of which people have long forgotten. Most struggled to get any kind of traction at all, and died quickly. Everyone was pushing hard to get their stuff into actual engineer's hands.

[jgruszynski]

I still have one of these.  I bought it while I was working at Intel.  It was US$100 IIRC with employee discount.

That job involved writing production test code in 8085 assembly in 1982.  If you have any Intel 27128A EPROMs - all were tested with my code.  That job was technically an internship between my junior and senior year of engineering school (EE) but it was "real serious work" not some make work stuff.  This was in "Santa Clara 2" back when the corporate HQ was still in Santa Clara 1 (which connected to SC2).  Now Intel HQ is in a new building on Mission College Blvd about 1.5 miles from SC1/SC2.

There was no mass storage with the MCS85 boards.  You either had to enter your entire program in hex when you powered up or you wrote an EPROM.  The marketing theory was that you'd use one of the Blue Box Intellec systems to generate code and EPROMs.

The "Mass Storage" solution required the Blue Box Intellec, which was only floppy based when I used it to develop production test code. 
We had two systems at work: one with one floppy and one with two floppies.  The latter was the one we fought over because you had to use one 175K disk for your OS (and as you ran the OS it would swap-in code from the floppy) and you could use the 2nd floppy for your code.  When you used the 1-floppy version, you had to constantly swap floppy disks on demand alternating between the OS disk and your code disk.   The operating system was CP/M though Intel marketed it as "ISIS".   The work flow was:

• Boot ISIS with system disk in floppy
• Load your program in edlin
• (on single floppy system: get prompted to put system disk in)
• (on single floppy system: system thinks a bit)
• (on single floppy system: get prompted to put data disk in)
• (on single floppy system: system writes data to your disk)
• (on single floppy system: get prompted to put system disk in)
• Edit code
• Save your code
• (on single floppy system: get prompted to put system disk in)
• (on single floppy system: system thinks a bit)
• (on single floppy system: get prompted to put data disk in)
• (on single floppy system: system writes data to your disk)
• (on single floppy system: get prompted to put system disk in)
• Exit editor
• Invoke assembler on your code
• Wake 5-10 minutes
• Invoke EPROM writer command on your binary code
• Wake 5-10 minutes
• Take EPROM to tester and insert EPROM into ZIF socket for programming
• Debug test program on wafer in wafer prober
• Find a problem with test program on wafer in wafer prober
• Think about what could be wrong/Study code/Diagnose problem
• Throw EPROM into the EPROM eraser
• Go to step #2 unless program works
• program works and is complete

It seems insane now but compared to what the MSC-85 board required and what my previous non-floppy CP/M system from the 1970s required, it was luxurious. I haven't really used by MCS-85 in part because even the Intellec solution is painful.  In "theory" you can edit memory from the hex keypad but you can't insert code - only overwrite a memory location so generally you have to "know" the exact sequence.

One thing with this job was I quickly memorized the hex codes for most of the 8085 instructions.  That's because the tester had an instruction pointer that you could single step forward for debugging but the only display was a pair of hex LEDs for address and another for instruction data, so you'd have to know the instructions only from that.  I got pretty good at reading 8085 hex.  I still remember a few like 0xCD which is CALL.

[rrinker]

This stuff was all very expensive back then, especially after you added RAM and support IC's. A CPU alone was $100 in today's dollars. I thought muxed address/data bus needs more clock cycles per instruction (CPI), so the Z-80 won over that.

My first 1978 dev board was a RCA CDP1802 COSMAC ELF 256 bytes of RAM and CMOS, from Quest Electronics. It's the weirdest CPU architecture I've ever seen and still have no idea how to code for it, just brutal.
Later I went Z-80 and still have the Assembly Language Programming by Lance Leventhal, programming cards and books. That CPU is a great teacher, it was so much fun knowing "BC" stood for byte-counter...

My buddy sold his KIM-1 for $2,000 on eBay, someone has one for sale $9,995 right now. I don't see that much nostalgia for these.
Byte Magazine archives

 Same as my first one. But I don't know what your issue was, the 1802 is SUPER easy to code for. Most programs don;t even need a stack, not with 16 x 16 registers. Brutal? What's brutal about it? The instruction set is super simple, the first nibble for all register operations is the same, with the second nibble referring to which of the 16 registers. SUPER easy. After learning on that, trying to add an assembly routing to a BASIC program on a 6502 Apple II was a super pain. I fail to see, other than the cheap price, why the 6502 is so revered. THAT was a piece of junk, the price was low for a REASON.

I DO miss the good old days - it was easy to know ALL of the instruction set of those 8 bit CPUs. Hex keypads on Intel systems are odd - Intel's 8 bit instruction sets are aligned for Octal representation - and I really never did like octal. Hex I can do in my head, octal is just too weird. But today's processors - yeah, just try to know ALL the machine language opcodes. I last did assmebly with 8088's on original PCs, once things got much past that, the instruction sets just got way too complicated to not use a higher level language.

[Kleinstein]

How much one likes a processor is a lot influenced by what CPU one learns first. The is a tendency to like the one that one start with.

For the assembler names / parameters they also use a slightly different organization - about equally good, but the system you learn 2nd looks odd and confusing. It is similar with organizing 16 bit values - high or low byte first.

The 6502 is similar in speed to the Z80 at about 4 times the clock. AFAIK the 8080/8085 are a bit slower than the Z80. So performance wise  the 8085, 6800, 6502, Z80 were not that far apart. The 6800/6502 used memory mapped IO instead of a separate address space. This can make the HW a little more complicated, but allowed to have a combined RAM+IO chip (6532/6832) to build a relatively simple system. AFAIK the CMOS version of the 6502 came out relatively early too.

[coppice]

I don't know what your issue was, the 1802 is SUPER easy to code for. Most programs don;t even need a stack, not with 16 x 16 registers. Brutal? What's brutal about it? The instruction set is super simple, the first nibble for all register operations is the same, with the second nibble referring to which of the 16 registers. SUPER easy. After learning on that, trying to add an assembly routing to a BASIC program on a 6502 Apple II was a super pain. I fail to see, other than the cheap price, why the 6502 is so revered. THAT was a piece of junk, the price was low for a REASON.

The 1802 was definitely an easy to use platform for the kind of simple control programs people actually wanted to write for it. The 6502 is only revered because it was the first device many people used, and nostalgia is strong. Th first machine I used was a Honeywell SomethingOrOther. I believe the SomethingOrOther was one of their most popular products, but to me its only interesting quality was that I had some limited access to it to learn and write Fortran code.

[DrG]

/-/ The 6502 is only revered because it was the first device many people used, and nostalgia is strong. /-/

But you have to add in the REASON it was the first that so many used...the reason for the impact that it made....I guess that accessibility can breed reverence if you would not otherwise be able to get involved.

6502 price - US$25 https://www.eevblog.com/forum/vintage-computing/what-was-you-first-computer/msg2897904/#msg2897904 AFAIK, nothing close was cheaper.

For example, Z-80 (arguably MUCH more powerful and appearing after the 6502) US$200 (single quantities) https://www.americanradiohistory.com/Archive-Byte/70s/Byte-1976-08.pdf [p.38]

[coppice]

/-/ The 6502 is only revered because it was the first device many people used, and nostalgia is strong. /-/

But you have to add in the REASON it was the first that so many used...the reason for the impact that it made....I guess that accessibility can breed reverence if you would not otherwise be able to get involved.

6502 price - US$25 https://www.eevblog.com/forum/vintage-computing/what-was-you-first-computer/msg2897904/#msg2897904 AFAIK, nothing close was cheaper.

For example, Z-80 (arguably MUCH more powerful and appearing after the 6502) US$200 (single quantities) https://www.americanradiohistory.com/Archive-Byte/70s/Byte-1976-08.pdf [p.38]

The prices of one off parts relate to their volume price about as well as a politician's words relate to their actions. People used the 6502 because it was in the box they used. It was in the box because it was cheap, and the original sources - basically MOS Technology and Rockwell - were able to crank them out in adequate numbers (which, or course, relates to them being cheap). The Z80 wasn't all that expensive if you were serious about buying it, and Zilog did well at ramping production (later Zilog wrecked the Z8000 by being very slow at ramping production). The Z80 would never have got to second place in consumer systems, behind the 6502, if it hadn't been pretty cheap.