Using 4000 series logic to drive....?

Those are obsolete. Use a micro

Reply to
gyansorova
Loading thread data ...

"To the man with only a hammer, everything looks like a nail."

Cheers

Phil Hobbs

Reply to
pcdhobbs

I can do what I need to do with ~4-5 chips that cost around 10 cent each Meanwhile the design using some logic is already done and works on the bench, aside from this lil interfacing issue (which I might have with a uP, too.)

Reply to
bitrex

A micro makes sense if what you need to do is very complex, there's no other way to do it. They're cheaper than any alternative.

A micro also makes sense if what you need is pretty simple, like say a

6-input OR then NAND-and-invert-some-of-the-outputs gate. Or a linear feedback shift register. An 8 pin micro will be cheaper than any comparable discrete logic option.

It's a tossup in the middle. What's easier and more cost effective if all you need is say a counter to drive an LED display? An at-least 16 pin uP that you need to program, or a binary counter IC and a 7 segment display driver - two cheapo chips you just wire together? It depends, I guess.

Reply to
bitrex

It's nuclear bomb-pumped anti-communist space lasers for me, or I'm not interested.

Reply to
bitrex

On Jul 31, 2017, bitrex wrote (in article ):

Well, for me the interesting bit is why that scheme didn?t work, although nobody doubted that LLNL could get those nuclear-powered X-Ray lasers to work:

All one had to do is fire a heavy iron hypervelocity missile right at one of these battle stations, and it was forced to detonate the nuclear bomb to activate the laser to destroy the incoming missile. But setting the driver bomb also destrored the battle station. Or, don?t shoot and let the hypervelocity heavy iron shot do the job.

Joe Gwinn

Reply to
Joseph Gwinn

Seems you are stuck in the 1970s. How will you program your micros? Same wa y as every other company that uses micros does the job! You can also use F PGAs or progammable logic - cheap as chips. Yes your solution is probably c heaper buy way less versatile if you want changes, and in any case, this is the 21st century, sure we can still use 741s, discrete logic and bipolar t ransistors, but why would you.

Reply to
gyansorova

I get tired of this sort of conversation. F^HMucking around with discrete logic chips is pure crap for nearly any design unless you only have poor embedded processor skills. If you think the tools for embedded processor work are big and clumsy you need to learn to use Forth tools. You can run the actual compiler on any but the tiniest of MCUs and be done with a simple design before lunch only needing a terminal emulator on a PC. I suppose you need a way to program the Forth system on the MCU but that is typically a USB serial port emulator module which costs $1 or so. Connect it between the your MCU card with a four pin header and the PC USB port and you can load the Forth system then load your application. It really doesn't get much easier.

It just seems like some BS to think making any sort of board is worthwhile for 100 units when you can buy MCU boards for $10 or even a lot less.

--

Rick C
Reply to
rickman

You need to get out more. Maybe in your world all you need is a MCU board, but that's far from true for the rest of us.

Nobody I know of still builds boards full of logic chips, but little bits o f MSI are often pretty useful. I've recently used MSI in frequency synthesi zers and lock-ins, for instance. When you need an accurate two-phase clock, a Johnson counter is good medicine. Also, as JL says, you don't have to pr ogram, document, and maintain a 74HC163.

Cheers

Phil Hobbs

Reply to
pcdhobbs

snipped-for-privacy@gmail.com wrote on 8/1/2017 7:52 AM:

Obviously somebody still uses the stuff or you wouldn't be able to buy it. But I don't remember using any for some 20 years. But then my designs typically have FPGAs. The way parts are designed now days, there just isn't much need for a single SSI/MSI device unless you count a single inverter or buffer. I do use voltage level converters since FPGAs often won't tolerate higher than 3.3 volt signals.

--

Rick C
Reply to
rickman

I design a lot with discrete logic and transistors. FPGAs and uPs just don't work for a lot of apps.

I don't use 741s, of course. There are far better opamps around.

--
John Larkin         Highland Technology, Inc 

lunatic fringe electronics
 Click to see the full signature
Reply to
John Larkin

[snip]

I'll bet that gyansorova is a member of that group I call "the copy and paste" crowd... incapable of original thought. ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
 Click to see the full signature
Reply to
Jim Thompson

Some people just don't have to (or get to) design challenging stuff: microvolts, picoseconds, gigabits per second, parts per million.

Life gets interesting below 1 ns. It gets downright difficult below

100 picoseconds.

Millivolt and microvolt, same idea.

--
John Larkin         Highland Technology, Inc 

lunatic fringe electronics
 Click to see the full signature
Reply to
John Larkin

I've essentially retired from FPGA design and embedded uP coding. Both have become very specialized and very complex, endless battles with opaque tools and kernals and libraries and test benches. Some kids enjoy all that, so let'em go for it.

Little bits of glue logic are handy. Buffers, schmitts, like that. Some Tiny CMOS gates and flops have 1 ns prop delay, and ECL/CML are low hundreds of ps. I can't get into and out of an FPGA in less than a few ns, and there's routing delay getting to the balls, and jitter and crosstalk, on a big chip. I can put a single gate wherever I want.

Besides, circuit design is fun, drawing like in kindergarten. FPGAs and uPs are all about typing.

--
John Larkin         Highland Technology, Inc 

lunatic fringe electronics
 Click to see the full signature
Reply to
John Larkin

Or enjoy the satisfaction of solving a problem that others have given up on.

...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
 Click to see the full signature
Reply to
Jim Thompson

Ahh, the heavy iron hypervelocity missile, of course! ;-)

Reply to
bitrex

The cost/benefit ratio of uPs goes down a lot when you're planning on doing a small-volume run. They're not as dirt cheap in onesies and twosies as they are in thousands, and usually I need a decent-sized board anyway to hold the jacks and connectors, anyway.

It's nice to be able to make changes but what happens if I discover a bug in my code after I ship it? It's too expensive to make every widget able to be user-flashed over USB or something, and what the f*ck am I gonna do, ship 500 users a new micro for them to socket in themselves?

A logic design of (in)sufficient complexity can at least be human-verified to be "correct". God only knows what that compiler is spitting out, many times.

Not saying this is true in the OP's case, I don't know, but "Just use a PIC" is often the go-to answer for people who have never had to actually use programmable hardware in a real product. The 2010s have a lot on the

1970s but it's still sometimes not just that simple.
Reply to
bitrex

And yeah, for the record I can write code for AVR/ARM and have worked on a couple real life projects before.

Reply to
bitrex

This shows how someone can completely miss the utility and ease of use of MCUs. John has only worked with large FPGAs and clumsy tools, in fact, he has worked with the hardest to use and most dysfunctional tools that support the xilinx zynq parts. He has never learned to do digital design with the same ease that many here do analog design. They use tools that are messy and get in the way accepting that is the only way to design digitally. It's simply not like that if you take some time to learn how to make it easy.

--

Rick C
Reply to
rickman

Much embedded code sucks donkey dick and I say this as someone who mostly writes code nowadays. It's great to be able to do something with non-programmable hardware when the task is feasible, as it encourages one to be non-sloppy, which is important on a design that one intends to front one's own cash to produce in small quantity for customers and can't really afford to write off bugs and duds.

Reply to
bitrex

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.