mersenneforum.org  

Go Back   mersenneforum.org > Great Internet Mersenne Prime Search > Hardware

Reply
 
Thread Tools
Old 2016-06-21, 16:19   #1
tServo
 
tServo's Avatar
 
"Marv"
May 2009
near the Tannhäuser Gate

2·13·19 Posts
Default 1000 core chip that runs on AA batteries!

They claim it consumes 7 tenths of a watt executing 115 billion instructions/sec.
Interesting, not from a practical standpoint but to shows what's possible and maybe on the horizon. It sure would be the bomb, tho, to put several of these to work on our stuff.
https://www.ucdavis.edu/news/worlds-...processor-chip

It looks like they cut corners on the design all over the place ( cache, instruction set,
FP thruput ( single precision only, of course), but still ......
tServo is offline   Reply With Quote
Old 2016-06-21, 18:24   #2
science_man_88
 
science_man_88's Avatar
 
"Forget I exist"
Jul 2009
Dumbassville

203008 Posts
Default

okay that's about 6.086956 picojoules per instruction by my math so at full instruction speed in theory I get about 10.834782 joules used per second ( aka watts) but other than running multiple tests LL isn't parallelizable so it's only use might be TF or some other thing done in parallel.
science_man_88 is offline   Reply With Quote
Old 2016-06-21, 19:32   #3
airsquirrels
 
airsquirrels's Avatar
 
"David"
Jul 2015
Ohio

11·47 Posts
Default

FFT multiplication is (sort of) paralizable. Thus the good performance of clLucas/CUDALucas, etc.
airsquirrels is offline   Reply With Quote
Old 2016-06-21, 20:01   #4
xilman
Bamboozled!
 
xilman's Avatar
 
May 2003
Down not across

3×3,361 Posts
Default

Quote:
Originally Posted by science_man_88 View Post
okay that's about 6.086956 picojoules per instruction by my math so at full instruction speed in theory I get about 10.834782 joules used per second ( aka watts) but other than running multiple tests LL isn't parallelizable so it's only use might be TF or some other thing done in parallel.
Do you really believe that figure?

The raw data is 0.7W and 115e9 instructions per second. All we can take away from that is that the power draw lies between 0.75 and 0.65W, with the performance lying between 114.5e9 and 115.5e9 ips.

Thus all we can conclude is that the energy per instruction lies somewhere between 0.65/115.5e9 and 0.75/114.5 pJ, or between 5.6pJ and 6.6pJ.

IOW, there's a good chance that not even the first digit you quote is correct. Why, then, do you state nine digits? That amount of precision is total BS.
xilman is offline   Reply With Quote
Old 2016-06-21, 21:48   #5
science_man_88
 
science_man_88's Avatar
 
"Forget I exist"
Jul 2009
Dumbassville

26·131 Posts
Default

Quote:
Originally Posted by xilman View Post
Do you really believe that figure?

The raw data is 0.7W and 115e9 instructions per second. All we can take away from that is that the power draw lies between 0.75 and 0.65W, with the performance lying between 114.5e9 and 115.5e9 ips.

Thus all we can conclude is that the energy per instruction lies somewhere between 0.65/115.5e9 and 0.75/114.5 pJ, or between 5.6pJ and 6.6pJ.

IOW, there's a good chance that not even the first digit you quote is correct. Why, then, do you state nine digits? That amount of precision is total BS.
partly because I was typing it from a calculator. also I think you'll find that based on your own calculations, that 6 shows as the first digit in your range about 7 out of 11 times.
science_man_88 is offline   Reply With Quote
Old 2016-06-21, 21:56   #6
chalsall
If I May
 
chalsall's Avatar
 
"Chris Halsall"
Sep 2002
Barbados

2×3×5×7×43 Posts
Default

Quote:
Originally Posted by science_man_88 View Post
partly because I was typing it from a calculator. also I think you'll find that based on your own calculations, that 6 shows as the first digit in your range about 7 out of 11 times.
You are clearly too young, or perhaps simply too inexperienced, to understand experimental error.

Rather than giving six decimal points of meaningless precision, give a +/- on the measure.
chalsall is online now   Reply With Quote
Old 2016-06-21, 21:59   #7
science_man_88
 
science_man_88's Avatar
 
"Forget I exist"
Jul 2009
Dumbassville

26·131 Posts
Default

Quote:
Originally Posted by chalsall View Post
You are clearly too young, or perhaps simply too inexperienced, to understand experimental error.

Rather than giving six decimal points of meaningless precision, give a +/- on the measure.
I was just doing straight math from the given numbers. doing it from xilman's range of 5.6-6.5 pJ it gives 9.97 watts to 11.7 watts at full power rounded to 3 places.

Last fiddled with by science_man_88 on 2016-06-21 at 22:08
science_man_88 is offline   Reply With Quote
Old 2016-06-21, 22:01   #8
chalsall
If I May
 
chalsall's Avatar
 
"Chris Halsall"
Sep 2002
Barbados

215068 Posts
Default

Quote:
Originally Posted by science_man_88 View Post
I was just doing straight math from the given numbers.
And that was your mistake.
chalsall is online now   Reply With Quote
Old 2016-06-22, 02:04   #9
bgbeuning
 
Dec 2014

22·32·7 Posts
Default

Just think about the RAM bottleneck ...
bgbeuning is offline   Reply With Quote
Old 2016-06-22, 10:55   #10
science_man_88
 
science_man_88's Avatar
 
"Forget I exist"
Jul 2009
Dumbassville

203008 Posts
Default

Quote:
Originally Posted by bgbeuning View Post
Just think about the RAM bottleneck ...
Quote:
and they transfer data directly to each other rather than using a pooled memory area that can become a bottleneck for data.
so they don't rely on ram.
science_man_88 is offline   Reply With Quote
Old 2016-06-22, 16:55   #11
Antonio
 
Antonio's Avatar
 
"Antonio Key"
Sep 2011
UK

10238 Posts
Default

Quote:
Originally Posted by science_man_88 View Post
so they don't rely on ram.
So what magic do they use to transfer instructions and data into/out of the processors?

Last fiddled with by Antonio on 2016-06-22 at 16:55
Antonio is offline   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
short runs or long runs MattcAnderson Operazione Doppi Mersennes 3 2014-02-16 15:19
1000-Core Computer Rodrigo Hardware 15 2011-01-15 05:07
48-core chip - Intel hj47 Hardware 6 2009-12-20 13:59
Capicitor batteries. mfgoode Science & Technology 14 2006-07-03 15:48
Laptops, Batteries, and Prime95 Uncwilly Hardware 9 2004-12-22 10:56

All times are UTC. The time now is 04:49.

Mon Jul 13 04:49:50 UTC 2020 up 110 days, 2:22, 0 users, load averages: 2.13, 2.02, 1.96

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2020, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.