Using 4000 series logic to drive....?

I want a dishwasher with one big pushbutton, WASH. And one LED, DONE.

We bought a gas range for the cabin. I specifically didn't want a uP or a digital display panel or any of that nonsense. The uP-equipped gas ranges start at about $400. The non-electronic ones start at $2000.

It very much depends on the appliance.

I really like my dishwasher which has a knob that rotates as the cycle progresses. I can instantly see what it is doing and that it is progressing - and it has never, touchwood, failed in > a quarter of a century. OTOH I hate the "simpler" ones with an LED and a couple of buttons, because they "surprise" me since I cant see what they are/aren't doing.

OTOH, microwaves with knobs are a disaster. With a microwave I should be able to set 15s at 70% power, and it should do exactly that. Knobs fail to set the time accurately. But it is horrifying how many microwaves "improve" on that simplicity by not having simple time/power controls; some I (ant other people) can't even figure out what the ideograms mean :(

Even easier with an XMOS xCORE. The i/o ports contain counters so you can state "output X when the counter is Y". The i/o clocks (up to 250MHz/4ns) are independent of the core clocks (100MHz/10ns), and there are many independent cores.

If you pre-calculated the sine values, I would expect you could get to, say, 10MS/s per core without any difficulty. And you could still have other cores doing the boring guff like USB comms and front panel controls.

It has proved pleasantly simple to program deterministic programs in them, partly because the IDE guarantees loop times without executing the code.

My laptop is a bit weak in the speaker department. I had a very nice pair of Sony computer speakers someone freecycled to me which did the trick nicely. As an added benefit the speaker can sit right beside me and has volume knob I can turn to instantly set any volume level I want, nice when a web page comes up blasting. The one weakness of them was the audio plug which died on both. The inner conductor protrudes from the rest of the case and has the inner conductor of the coax soldered to it and is covered by rubber molding. So as the rubber flexes the wire is stressed and eventually breaks. The cord was a PITA anyway so I found a bluetooth module designed for auto use and could be powered by the internal 30 volt power supply for the speakers. A few hours of some very fiddly soldering and mounting and the Sony speakers were bluetooth enabled. Great! Best of all worlds... until after a year or so something gave up the ghost and the bluetooth won't connect anymore.

I've been looking for an off the shelf replacement for a couple of years and I can't find anything that has an actual volume knob and bluetooth! I saw another module for use in a car, maybe I should try it again.

When I saw Phil's post I thought, "Gee, who was that XMOS guy? Sounds like something he could do easily." But then an FPGA would do pretty well too.

Great, that's only 86 dB worse than the MSI part. Another triumph of modern technology. ;)

Not exactly an ATtiny though is it? It also doesn't cost 13 cents like an HC163.

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

Cheers

Phil Hobbs

I believe you were complaining about the complexity of the tools. Forth makes programming MCUs a *lot* simpler in my experience.

And some people think a crescent wrench is a hammer >:-} ...Jim Thompson

Hammers+nails can fasten things together as well as screws/screwdrivers. Shrug.

It's not the complexity of the tools exactly (they aren't particularly complex), it's the logistical drudgery of designing something with a MCU worth releasing, and most say hobbyist Arduino projects just aren't worth releasing.

I can write good C/C++ OK, but it takes time. If I decide to use Forth instead, well, are there implementations available for my platform? Who wrote those implementations? Have they been used in real products before or are they some hobbyist project on Github that may or may not be full of bugs? Great, now I have to look through the interpreter code and try to verify that it's correct, too. The design using some logic was finished in an afternoon.

Remember when D-Link released a WiFi enabled home-automation wall plug gizmo that had a straightforwardly exploitable stack overflow bug that allowed anyone on the network to gain root access and switch the load on and off as fast as it could go? Heh that was cool

Google IEFBR14

Particularly if you don't care about bugs.

Oh, there was Macintosh PortaAPL for the floating math (and matrix support). And, it has the same write-only qualtiy that you've grown to love.

Yes, I am cracked, but that's not relevant. The design documentation doesn't need to explain how each MSI chip works, but it has to explain how you are using them. Documentation of the HDL code would be no more complex. Come on, the code for an HC163 would be no more than x lines of code... rather than speculate, here's the code complete with boiler plate. Ten lines, the rest essentially documentation. I spent more time trying to read the TI data sheet than I did writing the code.

-- VHDL created by Arius, Inc copyright 2017

--------------------------------------------------------------------------------

-- Company: Arius, Inc

-- Engineer: Rick Collins

My god you make life difficult. Sure, if you have never used a tool before you need to vet it. That's a reason to never use another new tool?

You can do what you want of course. But your reasoning and justifications don't explain why anyone else should shy away from working with MCUs or FPGAs for the vast majority of design cases. MSI logic is a huge PITA to work with and isn't worth keeping fresh on in my opinion. I don't recall the last time I built a dead or live bug protoboard and I don't plan to start down that road unless it is absolutely required. I much prefer to work with wood if I'm going to get that hands on.

I barely use nails and much prefer through bolts to screws when building anything substantial. I just completed the woodwork on a kayak carrying rack for my truck. I still need to add a few tie downs and cover it with carpet. The only screws are 4 inchers into the cross members and the only fasteners remotely like nails are the staples I'll use to attach the carpet.

I got tired of tossing 65 lbs over my head to get the kayak on top of the cab.

Sorry if I insulted your baby, but it's still ugly. ;)

Cheers

Phil Hobbs

Yup. The everything-is-a-nail team appear to be assuming that every nontrivial design will have a computer in it. If that were the case (apart from the gross timing issues) their argument might have some force, but it isn't.

Cheers

Phil Hobbs

APL I'm more sympathetic with. I never got into it because it needed a spec ial keyboard, but it was so powerful (apart from the write-only aspect) tha t I might well have dived right in if I'd had one.

IIUC APL's power came from the combination of an expressive mathematical sy ntax and really clever use of lazy evaluation. You could write code that lo oked like it was inverting some ginormo matrix on a completely inadequate m achine, but APL would save your tuchis by computing only those elements you were actually using.

Cheers

Phil Hobbs