The 4,000,000 core 1.2 nm Gen2PU. - Psych forums

Our partner

Jump to:

Thoughts from transgressing dimension
Here you can see some of my wild thoughts and you may find some good worldly ideas on here. I just love thinking and thought I should let my thinking be read.

highdimensionman: Consumer 2; Posts: 55; Joined: Tue Jul 08, 2014 5:38 pm; Blog: View Blog (1012)

Archives

November 2022

China should know they can change their fate and the worlds fate: Mon Nov 28, 2022 8:08 pm

September 2022
August 2022
July 2022
June 2022
May 2022
April 2022
March 2022
February 2022
January 2022
December 2021
November 2021
October 2021
September 2021
August 2021
July 2021
June 2021
May 2021
April 2021
January 2021
December 2020
November 2020
October 2020
September 2020
August 2020
July 2020
June 2020
May 2020
April 2020
February 2020
November 2019
October 2019
September 2019
May 2018
April 2018
March 2018
February 2018
December 2017
August 2017
May 2017
April 2017
March 2017
February 2017
December 2016
November 2016
October 2016
September 2015
August 2015
April 2015
March 2015
February 2015
January 2015
December 2014
November 2014
September 2014
August 2014

Search Blogs

Advanced search

The 4,000,000 core 1.2 nm Gen2PU.

by highdimensionman on Wed Apr 20, 2022 4:39 pm

Ok so the size of the bitwise data is a little bit bulkier not more than double though.
each core is CISC in nature it has a big set of little functions that process upto 6bits of data at a time.
most of these instruction's reconfigure in scratch the out come of a smaller set of base functions in 2 pipe stages the instruction is set then it is performed all in one cycle.
The bit codec it works with is a 4 bit codec that also accounts for 00 being a break point.
I call this break point coding.
So when doing a multiplication function shall we say it finds quick to hop to in bits preliminary information relative to say 24 bits and the 16 bit instruction aligns the 6 bit instructions on the extracted bits then new preliminary work is done and the final instructions, 2 cycles or with more pipe stages 1 cycle and to join up stages for bigger numbers could be done in parallel in a similar way.
Addition would be similar.
However there would be a good few extra speed up options for some math areas with known and not yet known transcendental numbers and their digit data which can be stored to some adequate digit precision
This CPU could perform 180TFlopps double precision real terms 150 Watts. So that's 1 Tflop to the watt and 1 Exa Flop to the mega watt. It'll probably be 2026-2028 until a 12-17 MWe 1 exa flop of double precision super computer arrives which will be more like 14 Exa flops of low precision and mixed precision at about 1.6nm node however this will likely also include 10,000's of optical accelerators and maybe some other AI accelerator chips taking the range closer to 100 Exaflops.

Diagnosis - Paranoid Schizophrenia although challenged.
Medication - Clopixol injection 4 weekly.
Personal diagnosis - I battle with the chemistry and biological effects of Clopixol every day trying to improve on the heath and range of my approach.
I try to improve my thinking. Hopefully some good ideas shine through.

0 Comments Viewed 2846 times

Who is online

Registered users: barry h, Bing [Bot], Google [Bot], Google Adsense [Bot], Majestic-12 [Bot], Yahoo [Bot]