Meter with two consumer-side connections

[bernardgreen]

BAS

When you came back my initial reaction was good, I thought BAS and his valuable advice is back for the benefit of many. It was saddening to see your aggressive posts.

Please do not waste your useful knowledge by getting yourself banned

 

[SimonH2]

Here we go again.
How about you try a different tack. If someone writes something which you think is insulting/disparaging - just politely correct them, quoting what you disagree with (with enough context to see what it was about), and leave it for others to make their minds up.
Acting like a wallflower and demanding action just makes you look bad. If you just point out the errors, most people will be able to make their own mind up and if genuine then they'll add the person in the wrong to their "tossers" list.

Sadly Bernard is right, you are often aggressive. You tend to be aggressive (or at least write in a manner which many would perceive as aggressive) to "newbies". And you are very quick to pick up on typos and publicly show them up - again a mild form of aggression.
I'm sure it's mostly not deliberate - for some of us it can be hard to pick up on nuances of communication - but even where it's not deliberate, it's still there.

Your post just a little up the page is a case in point. Frankly, you get a lot of leeway from the Mods - many of us have wondered how you managed to stay so long without getting banned may years ago. Just accept that when you complain about someone else, your own actions will be taken into account - and stop DEMANDING special treatment.
If you can't accept it, then sorry - you can stay a bitter and twisted person, but it won't make any difference (well not for the better) to the outcome.

 

[ban-all-sheds]

The point is that a fixed-function digital IC doesn't have a "set of instructions", just as a PCB full of discrete components doesn't have a "set of instructions".

Surely it does.

Fixed or not, it still has a set of instructions which it carries out.

It all really comes down to the nature of the innocent-looking little bit of the AD71056 called "multiplier" in its block diagram. You are implying (I suppose actually saying!) that this multiplication of the digitised voltage and current signals is undertaken entirely by 'hard-wired' logic on the chip. If that is the case, such that the functionality of the chip is totally unchangeable, then I agree that even I would not call it a microprocessor!

It might be more likely called an ASIC or FPGA, I agree, but...

If that had been the case (which I now realise it probably isn't!), then I would probably have called the whole chip a 'microprocessor', even though the 'program' was not user-alterable.

Must admit I've never thought about it, but does a "microprocessor" have to have a non-fixed set of functions/instructions? The microinstructions are fixed, unless you re-write the microcode, so what if you wrote a drastically different one which locked down what the chip did?

What about embedded systems where the code is loaded from ROM and never changes?

If your chip has logic components, interconnects, memory, buses, clocking etc does it stop being a microprocessor if it's not re-programmable?

 

[JohnW2]

Surely it does. Fixed or not, it still has a set of instructions which it carries out.

I presume that what endecotp was saying is that if all one has on the chip is 'hard-wired' logic gates etc., there are no 'instructions' as such - any more than there are for anything built out of a set of physically interconnected components.

Must admit I've never thought about it, but does a "microprocessor" have to have a non-fixed set of functions/instructions?

To my mind, something is a "microprocessor" if it has the potential to function on the basis of more than one set of instructions, even if, in a particular application or instance, it has only one 'fixed' set of instructions, which cannot be changed once created.

What about embedded systems where the code is loaded from ROM and never changes?

If the instructions come from external source (hence could vary), then I would have no hesitation in calling the 'device being instructed' a "microprocessor". I don't think I would change my view if those instructions were 'permanently' stored (e.g. by fusible links) within the same chip - again I would call it a "microrocessor" if it could have been 'programmed' (permanently) differently.

If your chip has logic components, interconnects, memory, buses, clocking etc does it stop being a microprocessor if it's not re-programmable?

As above, if it could have been programmed differently, or if the instructions come from an external source, then I would say that it definitely remains a microprocessor, even if the programming is a "one bite of the cherry" exercise (e.g. with fusible links) that does not allow significantly for 're-programming'.

However, this is really all just semantics, so I can't get very excited about it.

Kind Regards, John

 

[ban-all-sheds]

I think you ought to stop saying things like "this is really all just about the meanings of words, so I can't get very excited about it".

 

[DetlefSchmitz]

I think that if someone offered for sale a microprocessor and it turned out to be a dedicated meter chip, I would be disappointed. I think that there is a point at which the flexibility to the end-user is what defines the name.
In the case of the meter chip in question, I don't think there is actually enough information in the data sheet to decide exactly how it works internally; as a dsp or as some fixed hardware generated by a Verilog/vhdl description. And as a user I don't need to care. I just need to know I won't be able to program it as a timer to time my alarm system or whatever.
So there are a least two different levels of product description: what the user can do with it, and how it works internally.

 

[JohnW2]

I think that if someone offered for sale a microprocessor and it turned out to be a dedicated meter chip, I would be disappointed. .... So there are a least two different levels of product description: what the user can do with it, and how it works internally.

Agreed - but you presumably wouldn't order the product without first discovering "what you could do with it" - any more than you would order something advertised as "an electric motor" without knowing whether it was a 'general purpose' one or one that had been constructed/configured/customers so as to be only suitable for one very specific dedicated purpose!

However, I think you have probably hit the nail on the head as to why there has been a small difference of opinion about terminology - i.e. those (at least) "two levels of description". I have been talking more in terms of the latter of those levels - if "how it works internally" is by having what is essentially a 'general purpose microprocessor', but together with an internal, unchangeable, set of
instructions which have rendered it, as created, suitable for only one task, then I would probably still call it a "microprocessor" (in the same way that a motor designed/customised so as to be only suitable for use in a Dyson vacuum cleaner is still "an electric motor"). Others, perhaps including you, think more in terms of the 'user programmability' of the device, and will not call it a "microprocessor" unless it is user-programmable.

Kind Regards, John

 

[ban-all-sheds]

I think that if someone offered for sale a microprocessor and it turned out to be a dedicated meter chip, I would be disappointed.

Hard to see how that could happen - even it it were a microprocessor you'd probably be looking at the specs for instruction set, clock speed, cache sizes, number of cores etc etc...

I just need to know I won't be able to program it as a timer to time my alarm system or whatever.

Or to have different rules than intended for recording consumption

 

[DetlefSchmitz]

(edit: reply to John)
Well, I was once accused of having 'matchbox philosophies'. I think this is one time when i can accuse someone else of the same!.

I would usually refer to a device by its part number or described functionality. if I was talking about how it worked then I would use terms relating to how things like that work, particularly it this was known from the data sheet description. If I was designing such a device, then I would use the most impressive description of the technology involved in order to get most sales. It all depends.

 

[DetlefSchmitz]

Hard to see how that could happen - even it it were a microprocessor you'd probably be looking at the specs for instruction set, clock speed, cache sizes, number of cores etc etc...

I think you are deliberately missing the point. Or maybe just didn't get it.

 

[bernardgreen]

I think that if someone offered for sale a microprocessor and it turned out to be a dedicated meter chip, I would be disappointed.

A micro-processor with on board program memory dedicated to a particular task could have its program held in re-programmable Read Only Memory ( ROM ). In which case erasing the ROM enable it to be re-programmed to perform a different task ( in different hardware in necessary ). But many dedicated micro-processors do not have re-programmable ROM anmd thus cannot be used for anything but the taskk they are dedicated to.

 

[DetlefSchmitz]

When we've got this thrashed out, I have a more interesting topic involving angels and pin heads.

 

[JohnW2]

(edit: reply to John) Well, I was once accused of having 'matchbox philosophies'. I think this is one time when i can accuse someone else of the same!.

The point seems to be that there is a whole spectrum of possibilities of 'programmatability' for a chip with 'processing' capabilities - from a set of instructions which is (unchangeably) incorporated at the time of chip manufacture (as with many 'dedicated' chips) to the situation is which the user is largely free to provide any set of instructions they wish, and to change those instructions as often as they want (as in a PC or laptop). Since the 'processor' is doing the same thing in all cases, the only difference being in the location and user-changeability of the 'instructions', I would be inclined to call them all "microprocessors".

Kind Regards, John

 

[JohnW2]

But many dedicated micro-processors do not have re-programmable ROM anmd thus cannot be used for anything but the taskk they are dedicated to.

Indeed, but a few of the people here are saying that they should then not be called "microprocessors", even though I (and perhaps you) probably would call them that.

Kind Regards, John