mersenneforum.org  

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

Reply
 
Thread Tools
Old 2016-02-07, 14:56   #12
Fred
 
Fred's Avatar
 
"Ron"
Jan 2016
Fitchburg, MA

11000012 Posts
Default

Quote:
Originally Posted by axn View Post
I'm sure this is a noob quesiton, but when they list "memory types" in the above linked table, such as "DDR4-1866/2133, DDR3L-1333/1600 @ 1.35V" for the 6500, it's confusing to me. Clearly folks are using DDR4 faster than 2133 with this cpu... so what do they mean by "DDR4-1866/2133"?
Fred is offline   Reply With Quote
Old 2016-02-07, 16:36   #13
Mark Rose
 
Mark Rose's Avatar
 
"/X\(‘-‘)/X\"
Jan 2013

22×733 Posts
Default

Those are the officially supported memory speeds. Anything faster is considered a memory overclock.
Mark Rose is offline   Reply With Quote
Old 2016-02-08, 07:48   #14
xtreme2k
 
xtreme2k's Avatar
 
Aug 2002

2·3·29 Posts
Default

I am actually surprised the difference between 6700K and others (6600/K/6500) are quite huge in applications such as rendering, encoding, sysmark and others. (Ignoring P95 for now)
Their costs difference is around $150, As a % of the overall system is only going to be about 10-15% (say a system for about $1K-1.5K), and yet you are taking a hit up to around 30% (up to around 40-45% in worst case) in performance.

http://anandtech.com/bench/product/1543?vs=1544
http://www.cpu-world.com/Compare/500..._i7-6700K.html

Last fiddled with by xtreme2k on 2016-02-08 at 07:57
xtreme2k is offline   Reply With Quote
Old 2016-02-08, 09:58   #15
LaurV
Romulan Interpreter
 
LaurV's Avatar
 
Jun 2011
Thailand

22×33×89 Posts
Default

Quote:
Originally Posted by xtreme2k View Post
Their costs difference is around $150, As a % of the overall system is only going to be about 10-15% (say a system for about $1K-1.5K), and yet you are taking a hit up to around 30% (up to around 40-45% in worst case) in performance.
Fixing the dollar bug for you...
edit: here much worse, due to it causes missing important text, because the % is treated as comment

Last fiddled with by LaurV on 2016-02-08 at 10:00
LaurV is offline   Reply With Quote
Old 2016-02-08, 12:43   #16
henryzz
Just call me Henry
 
henryzz's Avatar
 
"David"
Sep 2007
Cambridge (GMT/BST)

10110111110002 Posts
Default

Quote:
Originally Posted by xtreme2k View Post
I am actually surprised the difference between 6700K and others (6600/K/6500) are quite huge in applications such as rendering, encoding, sysmark and others. (Ignoring P95 for now)
Their costs difference is around $150, As a % of the overall system is only going to be about 10-15% (say a system for about $1K-1.5K), and yet you are taking a hit up to around 30% (up to around 40-45% in worst case) in performance.

http://anandtech.com/bench/product/1543?vs=1544
http://www.cpu-world.com/Compare/500..._i7-6700K.html
Does anyone have one? It would be interesting to see if that carries over to Prime95 at all.
henryzz is offline   Reply With Quote
Old 2016-02-08, 16:45   #17
Prime95
P90 years forever!
 
Prime95's Avatar
 
Aug 2002
Yeehaw, FL

2·53·71 Posts
Default

Quote:
Originally Posted by xtreme2k View Post
I am actually surprised the difference between 6700K and others (6600/K/6500) are quite huge in applications such as rendering, encoding, sysmark and others.
My estimate (based on Haswell data) is the 6700K would be about 7% faster than the 6600 paired with DDR4-3200.

Not enough faster to justify the extra upfront cost and electricity.
Prime95 is offline   Reply With Quote
Old 2016-02-08, 19:37   #18
VBCurtis
 
VBCurtis's Avatar
 
"Curtis"
Feb 2005
Riverside, CA

4,861 Posts
Default

Quote:
Originally Posted by xtreme2k View Post
I am actually surprised the difference between 6700K and others (6600/K/6500) are quite huge in applications such as rendering, encoding, sysmark and others. (Ignoring P95 for now)
Their costs difference is around $150, As a % of the overall system is only going to be about 10-15% (say a system for about $1K-1.5K), and yet you are taking a hit up to around 30% (up to around 40-45% in worst case) in performance.

http://anandtech.com/bench/product/1543?vs=1544
http://www.cpu-world.com/Compare/500..._i7-6700K.html
On most of the benchies at anandtech, the difference in scores is less than the difference in clockspeeds, as one would expect when a bench is even slightly memory-bottlenecked. Cherry-picking the biggest-difference scores only matters if you do those specific things with your machine. Do you? We do not.

"up to 30%" is not an average. try averaging all the scores for every benchie listed; eyeballing it gets me about 15% faster, in line with clockspeed increase. I haven't built a $1500 system since the 90s; even my 6-core i7 came in around $1000, so looks like more than a 15% increase in system price for most folks. If you're planning on a $500+ video card to reach $1500 system price, that's your business- but if you are, those games sure aren't going to run 15% faster either.
VBCurtis is offline   Reply With Quote
Reply



Similar Threads
Thread Thread Starter Forum Replies Last Post
(M48) NEW MERSENNE PRIME! LARGEST PRIME NUMBER DISCOVERED! dabaichi News 571 2020-10-26 11:02
disk died, prime work lost forever? where to put prime? on SSD or HDD? emily PrimeNet 3 2013-03-01 05:49
Goldbach's Conjecture: Hunting Counterexamples Christenson Math 19 2011-09-07 01:44
Speeding up Mersenne hunting m_f_h Math 8 2007-05-18 13:49
monster prime hunting (original in German) Jagd nach der Monsterprimzahl cyrix Lounge 6 2004-06-11 08:29

All times are UTC. The time now is 13:32.


Sat Jul 17 13:32:21 UTC 2021 up 50 days, 11:19, 1 user, load averages: 1.49, 1.53, 1.56

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, 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.