New CPU Architecture !

New CPU Architecture

I was looking at old Microprocessors 6500, 68000, 6800 etc. and noticing that the instruction sets have only 3 types of operations: 1. data movement, 2. branch and 3. math - logical operations on registers so it would be only a JMP or BRANCH, a LOAD from Memory and an ADD (or AND etc.) Number to Memory or register, etc.

So given that these old CPUs had only 5,000 transistors, with the level of integration available today (billions of transistors) you could completely change the CPU architecture: do away with data movement operations, each memory location could have a CPU attached or something like that to do the logical operations, the branches could simply be signals to the memory locations, massively parallel CPU, completely new designs, very fast machines, etc.

If you consider that those old CPU systems had a 4k ROM for BASIC, then 4000 locations times 5000 transistors makes only 25 million transistors, it could be done or experimented, etc.

New CPU architectures, changing completely the way those CPUs were designed, make the memory the processor, make the logic diffused, try all possible combinations, etc.

CPUs in 50 years have only tried about 20 to 50 different combinations, we can now try thousands of new ideas (who cares if most don’t work, the possibilities are there, some can be breakthroughs!).

And imagine if we don’t even use logic or something, new ways of thinking and such. Just like pure blind forces created life and complexity exactly because those forces didn’t use any logic, anything could go, any combination, in as logic simply limits what can be while nature follows no rules it just keeps on making a mess until a brain pops out of the soup, so new CPU designs could be tried, billions of new combinations, with millions of transistors so many new connections and circuits and such…

APE

We can free ourselves from the Function Bounded Combinations of Matter into only some combinations not allowing us to explore the entire combinational space: like a car engine must only move a box from A to B and this bounds the design of the engine into only some types of contraptions not allowing us to try out all kinds of new contraptions: the engine could be full of valves, cylinders doing all kinds of things gears, all kinds of irrational mechanisms that are not going to lead to anything (imaginary functions ? wild functions, parallel functions, functions that are no longer point wise from A to B, but entire volumes into others ?) but are just exploring the combinational space of engine parts, coils, gears wires and so forth. Get rid of the bounded end functions and Matter can explore so many new entities, just as Matter did when it evolved the Brain from all kinds of random entities on the earth and a long chain of actions and reactions etc.

We can’t construct the origin of life because the random irrational combinations that Matter can assume are way too many for our puny cause and effect logic: blind chance, dumb luck is much richer and can produce much more than our bounded, simple logical mind.

So you can change the way Microprocessors are designed, for example you can have a one shot conversion of a Memory state into another, one million bytes directly connected to a new set of one million bytes, just one shot to change a million memory locations into another and you avoid all the sequential steps.

Or your CPU could have every point on it (or in the volume of the CPU) constantly branching into all possible directions, so not only does every memory location represent a CPU, a processor, but every point is always branching, trying all the combinations constantly and so forth, and all of the combinations in between, trillions of new CPU experiments and designs, so cool!

And a planet sized CPU where every point is a processor doing all kinds of things, if the earth keeps on producing a billion microcontrollers a year, in billions of years the earth will be a ball of Microprocessors all running wildly and randomly, a new evolution, and shove that planet in an ocean of electricity, in the Sun and see what happens.

On a side note, I noticed that these cheap 20 dollars Microcontrollers are becoming popular, Arduino or something like that, so now we can design new Personal Computers and try out all kinds of new ideas and no longer be bounded by Microsoft or Linux, but try all kinds of simpler and new Operating Systems, bring back to life Amiga and such, whatever.

Millions of new PC types, new Operating Systems, wild chips, a never ending stream of new experiments, new possibilities, and evolution will continue, breakthroughs will occur, we will be Free From the Past.

TARD

Other possibilities: clockless chips, serial to parallel converters, make a chip + program set into a parallel combinational circuit that executes in one shot, simply by signal delay across the complete combination of circuits equivalent to the program and so forth. After all a program in the 6502 (commodore 64) or 6800 could have a few thousand bytes of length, converting it into a parallel circuit could mean just 100 million transistors, no sequential steps, just a long chain of direct combinational signals, delay only by the serialization of the transit time through the transistors.

The slowest part of any system is the serialization delays, whether sequential of parallel…

And other possible machines inside the machines, so many registers and circuits could be connected with a 100 million transistors, or you could make them operate in linear mode, all analog signals and analog circuits, a hugely complex analog computer and so forth, and imagine all the combinations, all of the breakthroughs, you can do it man, just do it, first gear, its all right, second gear hold on time, third gear you’re out of sight…

TARD

CPU design hasn’t changed at all in 50 years, I was looking at old computers and CPUs, it is still exactly the same, just a jump instruction, a move and a logical - mathematical operation, etc. Very little has changed, funny how the most “innovative” field is in reality the most conservative, where the basics have been put in place decades ago with no real change (and even the higher level stuff, Android is just Unix made 40 years ago and so forth).

Only the integration has change and improved, just Silicon Science and lithography and such, but Moore’s law is running out, another 4 years and Moore’s law will be no more…

There is now room to challenge it all with billions of transistors available! so for example get rid of all move and logical operations and set them in a memory all calculated ahead of time: if you have 10 registers and another 10 registers and 10 operations (add, sub, xor) just avoid all moves between registers and put them all cross connected but simply equal to one long register having the contents of the first and the second register and all ten operations, you would get just a thousand registers with everything all done ahead of time, no need to move between registers:

10 registers time 10 registers times 10 operations gives you 1,000 memory slots:

1to 10 registers , then 11 to 20 registers then 1 to 10 operations…

register 1 | register 11 | and
register 2 | register 11 | and

and so forth for all of the combinations.

Then the branches and sequence would just be signals to access any of the 1000 combinations and maybe just a few moves to record back the result into register 1 and 11 and so forth. Or a recursive set of all the results giving you 1000 registers times 1000 reg. times 10 operations giving you 10 million registers (now composed of register 1, 11, register 2, 11 and the and of the previous and results), and then some moves and sequence - branch signals and so forth.

And all kinds of permutations on these ideas, there could be so many new kinds of experimental architectures, I can’t believe that in such a field we are still stuck with the same old instructions and CPU designs that we have been doing for 50 years, I can’t believe it, we need new bold minds and experiments, we need creative ideas, we need new designs, new architectures, just do it man, do it, break away from the past, do it man !!!

TARD