How was memory decoded in the 6052's heyday

[Dan Moos]

OK, I've got a single dip with 32k of RAM on it. But that wasn't available back on the early eighties, and the RAM was spread over many chips.

Did those chips work just like the modern ones, but just have less address lines? So instead of the simple decode we do to select the entirety of RAM on one chip, did they have a pile more glue logic to decode that same 32k spread over many chips?

Just curious. It would explain why old 8 bit computer's boards were so heavily populated.

[Dr Jefyll]

Dan Moos wrote:

more glue logic to decode that same 32k spread over many chips?

Yup. And, besides the extra glue, those RAM's generally required a lot more power than the modern equivalent. And, despite the meager capacity, those RAM's were expensive!

Some things about the good ol' days weren't so good!

[GARTHWILSON]

They used more cascaded logic, further cutting into the access time, and further holding the clock speed down in a day when memory was slow already. One IC I wish were available in the faster families (like 74AC__ and 74VHC__) is the '154. It's like a '138 but has four address (in addition to G1\ and G2\) inputs, and 16 outputs.

My first build was in 1985. Jameco had 8Kx8 SRAMs (6264), but they were $40 each! If I remember correctly, because of the price, I put off buying them until I was further along in the build, and during that time, the price came down to something like $8 each, because the Japanese were dumping! I grabbed two! (A dollar was more money back then than it is today.) I suppose that if I were doing it a few years earlier, I would have had to settle for a 2Kx8 or smaller. The C64 was already out, with 64K or RAM, but I believe it was DRAM, and still a lot of ICs.

I had a friend who assembled a COSMAC Elf computer kit (with the RCA 1802 processor) in the late 1970's, and I think he got an extra 1KB of RAM for it, which stood up on plug-in boards on the right side if I remember correctly, and gave off a lot of heat.

[BigDumbDinosaur]

GARTHWILSON wrote:

The C64 was already out, with 64K or RAM, but I believe it was DRAM, and still a lot of ICs.

The C-64 used 4416 DRAMs, which were also found in the C-128.

As Jeff said, not everything back in the good old days was all that good.

[BigDumbDinosaur]

Dr Jefyll wrote:

And, besides the extra glue, those RAM's generally required a lot more power than the modern equivalent.

Adding insult to injury, the early DRAMs were very susceptible to single-bit errors caused by background radiation. This susceptibility could cause a computer to crash for no apparent reason.

[GARTHWILSON]

BigDumbDinosaur wrote:

As Jeff said, not everything back in the good old days was all that good.

I think what was good was the excitement and optimism (although it was accompanied by uncertainty and mystery) about what was happening and what was going to happen in the near future of computing. I fail to find much exciting about today's RPi's and so on. I like what enso said in his post at viewtopic.php?p=35151#p35151, in the topic, "Getting young people into 6502."

[GaBuZoMeu]

Dr Jefyll wrote:

And, besides the extra glue, those RAM's generally required a lot more power than the modern equivalent.

I've noticed "2,34A" on the schematic of a 16KiB RAM card (32x 2114 + LSTTL bus driver & decoder logic). So roughly 7,5A rsp. 37W for 48KiB

[whartung]

GARTHWILSON wrote:

I fail to find much exciting about today's RPi's and so on.

The excitement of the RPi's and so on is that it's getting easier and easier for folks to take hard parts and wire them together with logic to make them do Interesting Things. The RPi's et al eliminate the bulk of the boiler plate necessary to get an LED to blink. Nobody wants to design, wire, and test memory decode logic for the same reason they don't want to write a TCP stack or a network driver.

I mean, decode logic is interesting but DRAM refresh is not? Which one is actually more challenging to pull off?

Folks want their sensors to sense, their stepper motors to step, and servos to serve. If they can do all that with a few wires, and not have to cycle count their interface stack to boot, it's a win.

Besides, if someone was going to do advance combinatorial or sequential logic outside of a CPU, they'd likely type it in to a program and load up a CPLD or FPGA.

[BigDumbDinosaur]

whartung wrote:

GARTHWILSON wrote:

I fail to find much exciting about today's RPi's and so on.

The excitement of the RPi's and so on is that it's getting easier and easier for folks to take hard parts and wire them together with logic to make them do Interesting Things.

I completely agree with Garth.

The "easier and easier" part is precisely what makes the Raspberry Pi, Arduino, etc., unexciting. Put into culinary terms (I got finished a while ago cooking and serving corned beef and cabbage to family members), working with one of those gadgets is about as exciting as opening and heating a can of soup. You certainly won't learn anything about what is involved in cooking a pot of homemade soup, and certainly will have little appreciation for what went into making that homemade soup flavorful and nourishing. In other words, one can of soup is just like another one and you really have little idea what went into that can. That also describes the Arduino and Pi.

Working with the "hard parts" at the chip level goes well beyond merely getting interesting things to happen. There is the immense satisfaction one can get when his contraption works and he knows that he created it, scratching his head to figure out the logic, soldering pieces to a board, and writing some code to get it to do something, rather than purchasing something that is already assembled and only needs some juice to run.

There is also a valuable learning adventure in there, especially when the design fails to work on the first try. You will never get that experience working with prefabbed hardware made with someone else's notion of how it should work.

It is precisely for these reasons that I scratch-designed POC V1 using only basic logic gates and I/O devices (and debugged it with nothing more than a logic probe). I certainly could afford to purchase a prefabbed toy ("canned soup") and then spend my time writing code to make it do things. I would not have learned nearly as much, however, and would have cheated myself out of the opportunity to do so.

To each his own, but don't forget this is 6502.org, and the Arduinos and Raspberry PIs of the world don't have a 6502 in them.

[Dan Moos]

I agree. Arduino and the like aren't that stimulating if you are looking for learning, and the satisfaction of building something complicated.

But as a tool, an Arduino sure is handy. I was able to get a serial terminal displaying my climbing addresses in my NOP test with very simple and short code. I could have hooked up a seven segment setup, but nothing to learn there, I already know how to. Do that. So I went ahead and used the Arduino.

But I agree, when people get Arduinos and Pis, and think that is electronics...

Here's the thing though. Kids aren't doing interesting stuff anymore. Arduino has sort of gotten them doing something at least. And I suspect that the ones with the knack will become bored, and move on to the real thing eventually. And who am I to tell the ones blinking LEDs with digitalWrite statements that that isn't a valid, albiet simple hobby.

[GARTHWILSON]

The RPi has very little general-purpose I/O, which is fine for those with little ambition. You don't need much to blink a light or have a light sensor and a solenoid. My workbench computer has over 80 bits of I/O plus fast A/D & D/A converters, anti-alias filter modules, speaker amplifier, eight different serial ports of different kinds, built-in keypad and LCD, and more. The next workbench computer will have a lot more.

Part of the lure of something like the RPi is that IDEs and so on are already there for you, Linux, the works; so you can supposedly get going quickly (if you can figure out how to use the supporting software black boxes). They have a lot of layers of underlying software; but although they do let you get down to the metal, it's not easy to do so. I find them daunting. They do have consumer-type things like USB and Bluetooth which are beyond the hobbyist to understand. If that's what I wanted, I guess I'd go for it. It's not though. People think that just because I'm into electronics and small computers, I'm also a PC guru and can answer all their PC questions or fix their PC problems; but actually, I know almost nothing about PCs, and I use my PC as an appliance. It does have slots however if I were interested in using them for projects.

[DerTrueForce]

I built Ittiara(my own 6502-based machine) because I wanted to try my hand at assembly programming, and I couldn't understand what the heck the Intel documents (for x86 asm) were trying to get across. That was probably not knowing where to start, though. A while later, I also figured out that Windows wasn't going to play along easily, anyway. Getting the DUART to work is next on my list, when I have the time. I'm a uni student, and I'm not as organized as I could be, so I get nearly no free time.

I've got an idea on the shelf for a pretty full-on hardware debugger using an Arduino. I don't think I need it yet, so it remains a design in my book.
About the only use I'd have for a RPi would be as the guts of a portable computer. And I haven't done that yet(not saying I won't).

[EugeneNine]

I tend to think of the Arduino as the C64 of today. Remember when we all bought out first 8 bit machines they turned on and ran some sort of basic? Most of us didn't start from scratch. The Arduino is similar you plug it in and it runs their own language (C based). I spent most of the 80's never opening my C64 or doing any hardware then finally in the early 90's I started interfacing to the user port and did the stereo SID. But still didn't build a system from the board up, most people don't. But if you get to that point you can download GCC, AVRDude, etc or go PIC and MPLABX and start from scratch as well. So I find them more similar than different.

[whartung]

BigDumbDinosaur wrote:

The "easier and easier" part is precisely what makes the Raspberry Pi, Arduino, etc., unexciting. Put into culinary terms (I got finished a while ago cooking and serving corned beef and cabbage to family members), working with one of those gadgets is about as exciting as opening and heating a can of soup. You certainly won't learn anything about what is involved in cooking a pot of homemade soup, and certainly will have little appreciation for what went into making that homemade soup flavorful and nourishing. In other words, one can of soup is just like another one and you really have little idea what went into that can. That also describes the Arduino and Pi.

Some folks are more interested in the dinner party than the food they serve. Witness the endless arrays of options for pre-made holiday meals. Someone else here, I think, mentioned they don't make pasta sauce anymore, and are content with a jar from the store.

Dan Moos wrote:

I agree. Arduino and the like aren't that stimulating if you are looking for learning, and the satisfaction of building something complicated.

Electronics are only a portion of the things folks are making with this stuff. The bevy of robots, drones, "animated clothing accessories", home weather stations, etc. etc. The overall project can be complicated, while using off the shelf, high level electronics.

GARTHWILSON wrote:

The RPi has very little general-purpose I/O, which is fine for those with little ambition.

And until the people have exposure to I/O at all, they may not know what missing ambition they have. As they grow, and learn, they may well bump in to the limitations of something like an RPi, and then have to figure out how to go beyond that.

I'm a expert and professional (yes, I consider them distinct) software developer. And I rejoice at end user programming tools. Whether it's Excel macros, or cut and pasting a bunch of scripts together along with Google Maps to make an interesting web page. This stuff is supposed to empower people. As the electronics get cheaper, more powerful, and easier to use, it opens up the domain to more and more people. People who may not have considered using electronics at all in their project, or felt their project was simply inaccessible to them and their level of expertise.

They look at these websites and look at projects, and find that simply wiring up stuff moves them downstream -- with almost complete ignorance of something like Ohm's Law or anything else. Eventually, they either stop, because they're satisfied with what they have, or they try to do something that doesn't work, and that's the gateway to potentially more learning. Only now they might have the basic tools necessary -- a VOM, a soldering iron, basic electronic tools.

Gradual and incremental education, starting with practical, rich, working components, rather than raw resistors, transistors and theory. This let folks become productive quickly. It favors productivity and progress over fundamentals and theory.

At the same, time, it's a valid point to suggest that such high level components may perhaps give people false hope, false success. They may get something to work, but they don't know actually know how or why it works. For many use cases (notably hobbies or other casual uses), that's enough. You're right, they don't know "electronics". But that may not have been their goal in the first place. Rather they wanted to work on something "electronic".

[BigEd]

I remember my grandfather talking about building cats-whisker radios, which essentially means finding a diode by probing. He might well not have seen the joy in getting a microprocessor system up and running - getting all of those thousands of transistors ready-made!

It's easy to lose perspective and fixate on what was leading edge when we first came to the subject. It's no coincidence that some of the older people here are failing to see the merits of Arduino and Pi systems, while it's the younger audience who are being brought into (something like) our world using them.

Those people who like to understand the universe from top to bottom will not be limited by the level of system that they first get started with. There will be chip designers making RAM cells, PLLs, and bandgap references who are not yet teenagers today. Someone has to understand transistor characteristics before any chips can be made.