The Megaprocessor is a working CPU which is blown up in size to allow you to walk into it to watch how data is physically processed with your own eyes. There are 8,500 LED's in the core and another 2,048 for the memory which light up as data passes through the 15,300 transistors in the core and the memory's 27,000; though that total count includes the transistors which control the LEDs. The core's clock is a staggering 25kHz and there is 256 bytes of both RAM and ROM. The site actually provides you with the assembly language to write code for the processor if you are interested and you can visit the Centre for Computing History in Cambridge, England to see it in person. Drop by The Register for a quick look and for links to the project page for more details on the computer and build process, including a murderous vacuum cleaner.
"His ultimate goal other than the pure satisfaction of building the thing and getting it running, as El Reg reported in June this year, was to show the public how computers work by blowing the CPU up to a human-viewable scale."
Here is some more Tech News from around the web:
- Diamond Batteries That Last For Millennia @ Hack a Day
- Surface Studio torn down: Surprisingly upgradable storage @ Ars Technica
- 'DroneGun' Can Take Down Aircraft From Over 1.2 Miles Away @ Slashdot
- Neutralizing Intel’s Management Engine @ Hack a Day
- Snoopers' Charter: 'Draconian' IP Bill receives royal assent to become law @ The Inquirer
- Current switches insulator’s magnetic state @ Nanotechweb
- Storage newbie: You need COTS to really rock that NVMe baby @ The Register
- The Pokeball Power Bank @ Tech ARP
That is really neat!
That is really neat!
Very cool indeed!
Very cool indeed!
Not as cool if this is based
Not as cool if this is based on the Harvard architecture/modified Harvard architecture CPU designs that most modern microprocessor are based on. And remember a CPU’s architecture is not what made any single processor able to be fitted on a single monolithic die because that was mostly the process node fabrication technology/IP that made microprocessors on a single die possible. So this is an old fashion CPU made of many dies, or even individual transistors, with the individual dies having a rather limited number of individual circuits on the all the way back to the computers that where made up of individual Transistors or Tubes.
The Harvard architecture/modified Harvard architecture(1,2) CPU designs have their drawbacks mostly to do with security vulnerabilities concerning the stack overflow/underflow variety.
The stack machines(3) where of a whole different design that had the Stack Mostly as the focus of execution with every bit of code/data being directly accessed by stack hardware managed by the processor that had many stack pointers that more directly directly worked on the values pointed to by the data or code stack pointers on the stack machine. A list of features for the Burroughs stack machines follows:
“Unique features”
“All code automatically reentrant (fig 4.5 from the ACM Monograph shows in a nutshell why): programmers don’t have to do anything more to have any code in any language spread across processors than to use just the two shown simple primitives. This is perhaps the canonical but no means the only benefit of these major distinguishing features of this architecture: Partially data-driven tagged and descriptor-based design
Hardware was designed to support software requirements
Hardware designed to exclusively support high-level programming languages
No Assembly language or assembler; all system software written in an extended variety of ALGOL 60. However, ESPOL had statements for each of the syllables in the architecture.
Few programmer accessible registers
Simplified instruction set
Stack architecture (to support high-level algorithmic languages)
Support for high-level operating system (MCP, Master Control Program)
Support for asymmetric (master/slave) multiprocessing
Support for other languages such as COBOL
Powerful string manipulation
An attempt at a secure architecture prohibiting unauthorized access of data or disruptions to operations[NB 2]
Early error-detection supporting development and testing of software
First commercial implementation of virtual memory[NB 3]
Successors still exist in the Unisys ClearPath/MCP machines
Influenced many of today’s computing techniques”(4)
The old stack machines are more like the current VM virtual machines for example: Dalvik/ART/other based runtimes only with the stack machines all of this functionality was implemented in the stack machine’s hardware. So the Stack machine could run optimized in its hardware high level languages like ALGOL/etc more directly. What this artisan CPU project needs to tackle next is to implement some non Harvard designs like a Stack machine design to show some of the differences in actual computing architectures that have been used and some of the new ones not yet implemented in any widely used PC of today. There are some new designs like the old stack machine designs that have much more resilience to hacking.
(1)
“Harvard architecture/modified Harvard architecture CPU designs”
https://en.wikipedia.org/wiki/Harvard_architecture
(2)
“Modified Harvard architecture”
https://en.wikipedia.org/wiki/Modified_Harvard_architecture
(3)
“Stack machine”
https://en.wikipedia.org/wiki/Stack_machine
(4)
“Burroughs large systems”
https://en.wikipedia.org/wiki/Burroughs_large_systems
P.S. this Machine at the
P.S. this Machine at the Centre for Computing History in Cambridge, England is an actual Von Neumann(Very Smart, I’m a pea brain in comparison) architecture a predecessor of the Harvard based designs.
Read the part on “More Than Just Megahertz” linked to below.
“Modern Microprocessors
A 90-Minute Guide!”
http://www.lighterra.com/papers/modernmicroprocessors/
But can it run Crysis?
But can it run Crysis?
Very slowly it’s more like
Very slowly it’s more like watching the paint dry only much much much slower! Maybe not even one single frame per month sort of frame times!