No-op microprocessors

As all microprocessor instructions finally boils down to 0/1 through output pins. This means that we can build a microprocessor that does nothing but just generates 0/1 through output pins, in a desired way. Functionalities can be built on the top of such microprocessors. Such processors would be very much simple (?) to build. Do we already have any thing like this?

Reply to
v4vijayakumar
Loading thread data ...

No they dont, they boil down to affecting a memory location. If that location is hardwired to a pin then the pin changes as well, but they could be hardwired to any other internal resource or just be a memory location.

Reply to
cbarn24050

The assumption is incorrect and hence the conclusions drawn are also

Reply to
Ian Bell

Sometimes, they also have input pins.

Rene

--
Ing.Buero R.Tschaggelar - http://www.ibrtses.com
& commercial newsgroups - http://www.talkto.net
Reply to
Rene Tschaggelar

Yes, there exist microprocessors with instruction sets devoted largely to digital I/O.

Perhaps the functionality you are desiring is more directly addressed with PLC's and/or ladder logic diagrams.

Most of the folks here are microprocessor dweebs and would rather write and debug code rather than do a very simple job of setting up a PLC or writing down the ladder logic. In fact the subjects of ladder logic and PLC programming are so simplistic that they often are not even done by engineers, and instead simply implemented by technicians.

And of course most PLC implementations today just use a regular microprocessor to implement the ladder logic or state machine, and also provide a pretty nifty user interface.

Tim.

Reply to
Tim Shoppa

. Allan Turing proposed a machine, now referred to as the "Turing Machine". It consists of:

1) A state machine 2) A tape reader that could only do 3 things: . Write a "1" or "0" at the current position on the tape Read the bit at the current position of the tape. Advance or retard the tape one bit position at a time. . Turing used this conceptual model in his work on computability. He proved that his machine could do anything that a genral purpose computer could do, given enough time, and a long enough tape. . Google "Turing". You'll get lots of hits. Regards, Jon
Reply to
Jon

All processors, memory and such all run on a 0/1 basis.

Luhan

Reply to
Luhan

Yes, there are many such processors. Many are even available with multiple processor cores on one die, for just a few cents!

For example, see the 7400, which has four fully independent cores! Parallel computing for pennies!

Reply to
cs_posting

On a sunny day (20 Dec 2006 06:30:17 -0800) it happened "Jon" wrote in :

This is correct, and as an extra, cryptology can be done on for example a 64 bit wide processor as 64 1 bit operations in parallel (indeed resulting in a huge speedup). Google for example 'ffdecsa'.

From the ffdecsa FAQ and technical doc:

FFdecsa is a fast implementation of the CSA decryption algorithm for MPEG TS packets.

Q: What does FF stands for? A: FFdecsa means "Fucking Fast decsa".

Q: Why would you use such a rude name? A: Because this code is f****ng fast, more than 800% the speed of the best implementation I'm able to find around at the moment.

TRICK NUMBER 2: parallel bitslice

--------------------------------- Implementing the algorithm as described in tricks 1 and 2 give us about

15% of the speed of a traditional implementation. This happens because we work on only one bit, even if our CPU is 32 bit wide. But *we can process 32 different packets at the same time*. This is called "bitslice" method. It can be done only if the program flow is not dependent of the data (if, while,...). Luckily this is true. Things like if(a){ b=c&d; } else{ b=e&f; } can be coded as (think of how hardware would implement this) b1=c&d; b2=e&f; b=b2^(a&(b1^b2)); and things like if(a){ b=c&d } can be transformed in the same way, as they may be written as if(a){ b=c&d } else{ b=b; } It could look wasteful, but it is not; and destroys data dependency.

Our codes takes the same time as before, but produces 32 results, so speed is now 480% the speed of a traditional implementation.

Reply to
Jan Panteltje

I remember when I first learned about digital.

Reply to
Charles Schuler

Yes and I remember a comment made to me by some wizened engineer in years gone by that "digital is just an analog signal with the gain set so high that it flips from rail to rail without much ado".

- mkaras

Reply to
mkaras

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.