AOA Forums AOA Forums AOA Forums Folding For Team 45 AOA Files Home Front Page Become an AOA Subscriber! UserCP Calendar Memberlist FAQ Search Forum Home


Go Back   AOA Forums > Hardware > AMD Motherboards & CPUs

AMD Motherboards & CPUs Questions or comments on AMD products?


Reply
 
LinkBack Thread Tools Rate Thread
  #1 (permalink)  
Old 2nd February, 2004, 05:33 PM
ral ral is offline
Member
 
Join Date: March 2003
Location: Philippines
Posts: 351

Dual channel v. single channel @ 440FSB

Run a few more benches on my MSI K7N2 Delta-L:

Single channel: 2200MHz (10x220)

SuperPi 1M: 47s
Call of Dutry (Viperlair CPU Test Timedemo)
-167fps
-168.2 fps

Single channel: 2200MHz (11x200)

SPi 1M: 47s
COD: - 165.6fps
- 166.4fps

Dual channel: 2200MHz (10x220)

SPi 1M: 46s spi
COD: -170.8fps
-171.6fps

Dual channel: 10% increase in FSB = 3% performance increase.
Single channel: 10% increase in FSB = 1% performance increase.

I should probobly run more benchmarks. This time comparing 10.5x210 v. 220x10 on single channel mode. I dont suppose anyone with a single channel board would want to run further benchmarks? Dont want to move dimm around again.

Theories:

1) 400MHz does not fully utilize the 3.2GB available bandwidth. Most like a 404-410MHz FSB would yeild the most benefit with a single channel board.

2) Increasing FSB of the dual channel board beyong 440MHz would continue to yield increases in speed. Increasing FSB of the single channel board beyong 440MHz would not yield any increases in speed beyong the 1& increase.

Why bother. Lookign for optimal settings. Making overclocking an art

If might not be worth it to run fsb's much higher than 400MHz on a single channel board.

Hence on a KT600 or NForce2 400 @ 2300MHz it may be best to run something like 11x209.

On the otherhand on a NForce 2 Ultra 400 it may be best to run 10x230 or even 9.5x242.

There are also implications on what speeds you should aim for or what motherboard memory you should buy:

Previous dual channel benchmarks I ran seemed to show the the 400MHz FSB would start to bottleneck after 2200MHz-2300Mz.

2105MHz (10.5x200) - 161.2fps 15.3s
- 162fps 15.2s

2053MHz (9.5x216) - 164.2fps 15s
- 164.9fps 15s

2205MHz (11x200) - 164.5fps 15s
- 165.2fps 14.9s

2305MHz (11.5x200) - 166.9fps 14.8s
- 167.6fps 14.7s

2161MHz (10x216Mhz) - 167.7fps 14.7s
- 168.4fps 14.6s

2205MHz (10x220) - 170fps 14.5s
- 170.6fps 14.5s

If you overclock to try to get the best bang for the buck than this might be the best way to do so:

1. If you goal is to run 2200MHz (11x200 or 10.5x210) than a single channel NForce2 400 or KT600 paried with low cost DDR333/400 might be the best choice for money for performance.

2. If your goal is to run 2300-2500MHz than running higher FSB's on lower multipler would have greater benefit, so the higher cost of a dual channel board and top grade DDR400 or DDR433/466 memory might be worth it.

What do you guys think?
__________________
ASUS A7V600 (Bios Rev. 1.008)
AMD Athlon XP1800+ DLT3C JIUHB 0309 TPMW @ 2100MHz (10.5x200) @ 1.50v vcore
2x256MB Twinmoss PC2700 (CL2.5) @ 400MHz (CL2-3-3-7 2T) @ 2.55v vdimm
ALX-800 w/ 80mm HAC-V81 fan
MSI Star Force GeForce FX 5700 128TD 425MHz/555MHz @ 475MHz/650MHz
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #2 (permalink)  
Old 2nd February, 2004, 06:19 PM
Chief Systems Administrator
 
Join Date: September 2001
Location: Europe
Posts: 13,075

You didn't tell us what you did with the speed of the memory bus at all. Was it kept at a fixed speed, was it sync with the FSB or what?

Edit: 1% may well be within the margin of error (IE, not statistically valid)

Edit edit: How repeatable are your results?
__________________
Any views, thoughts and opinions are entirely my own. They don't necessarily represent those of my employer (BlackBerry).
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #3 (permalink)  
Old 2nd February, 2004, 06:22 PM
ral ral is offline
Member
 
Join Date: March 2003
Location: Philippines
Posts: 351

Oh sorry. Memory running 1:1, timings all at 2-3-3-11. Only did two runs of each COD test and one run of SuperPi.
__________________
ASUS A7V600 (Bios Rev. 1.008)
AMD Athlon XP1800+ DLT3C JIUHB 0309 TPMW @ 2100MHz (10.5x200) @ 1.50v vcore
2x256MB Twinmoss PC2700 (CL2.5) @ 400MHz (CL2-3-3-7 2T) @ 2.55v vdimm
ALX-800 w/ 80mm HAC-V81 fan
MSI Star Force GeForce FX 5700 128TD 425MHz/555MHz @ 475MHz/650MHz
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #4 (permalink)  
Old 2nd February, 2004, 06:24 PM
Chief Systems Administrator
 
Join Date: September 2001
Location: Europe
Posts: 13,075

Point 1) isn't valid then, as you bring up the FSB when running the RAM in sync, the memory bus speeds up too. Hence, at 220MHz clock, the memory can deliver 3.5Gb theoretical peak.
__________________
Any views, thoughts and opinions are entirely my own. They don't necessarily represent those of my employer (BlackBerry).
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #5 (permalink)  
Old 2nd February, 2004, 06:26 PM
ral ral is offline
Member
 
Join Date: March 2003
Location: Philippines
Posts: 351

Even on single channel mode?
__________________
ASUS A7V600 (Bios Rev. 1.008)
AMD Athlon XP1800+ DLT3C JIUHB 0309 TPMW @ 2100MHz (10.5x200) @ 1.50v vcore
2x256MB Twinmoss PC2700 (CL2.5) @ 400MHz (CL2-3-3-7 2T) @ 2.55v vdimm
ALX-800 w/ 80mm HAC-V81 fan
MSI Star Force GeForce FX 5700 128TD 425MHz/555MHz @ 475MHz/650MHz
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #6 (permalink)  
Old 2nd February, 2004, 06:26 PM
Chief Systems Administrator
 
Join Date: September 2001
Location: Europe
Posts: 13,075

Quote:
Originally Posted by ral
Oh sorry. Memory running 1:1, timings all at 2-3-3-11.
Would 2-3-3-9 be a better choice than 2-3-3-11, assuming that both the memory controller and RAM can turn the bus around fast enough?
__________________
Any views, thoughts and opinions are entirely my own. They don't necessarily represent those of my employer (BlackBerry).
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #7 (permalink)  
Old 2nd February, 2004, 06:29 PM
Chief Systems Administrator
 
Join Date: September 2001
Location: Europe
Posts: 13,075

Quote:
Originally Posted by ral
Even on single channel mode?
Single channel @200MHz clock provides 3.2GB/sec theoretical peak.
Single channel @220MHz clock provides 3.5GB/sec theoretical peak.
Assuming everything else remains constant (which is sometimes harder than it sounds!)

Dual channel @200MHz provides 6.4GB/sec theoretical peak (Ahem)
Dual channel @220MHz provides 7GB/sec theoretical peak...
Of course, DC is great until you hit the bottleneck that is the FSB...
__________________
Any views, thoughts and opinions are entirely my own. They don't necessarily represent those of my employer (BlackBerry).
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #8 (permalink)  
Old 2nd February, 2004, 06:34 PM
ral ral is offline
Member
 
Join Date: March 2003
Location: Philippines
Posts: 351

Have not tested 2-3-3-9.

I was under the wrong impression that a single channel board could only provide 3.2GB of bandwidth, even atat above 400MHz FSB.

So much for the theory.

Well you learn something new everyday... I guess I better stop benching and finnish playing Cal of Duty.

Well, it was fun for the time that I tohught I had a good idea
__________________
ASUS A7V600 (Bios Rev. 1.008)
AMD Athlon XP1800+ DLT3C JIUHB 0309 TPMW @ 2100MHz (10.5x200) @ 1.50v vcore
2x256MB Twinmoss PC2700 (CL2.5) @ 400MHz (CL2-3-3-7 2T) @ 2.55v vdimm
ALX-800 w/ 80mm HAC-V81 fan
MSI Star Force GeForce FX 5700 128TD 425MHz/555MHz @ 475MHz/650MHz
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #9 (permalink)  
Old 2nd February, 2004, 06:35 PM
ral ral is offline
Member
 
Join Date: March 2003
Location: Philippines
Posts: 351

By the way, thanks Áedán
__________________
ASUS A7V600 (Bios Rev. 1.008)
AMD Athlon XP1800+ DLT3C JIUHB 0309 TPMW @ 2100MHz (10.5x200) @ 1.50v vcore
2x256MB Twinmoss PC2700 (CL2.5) @ 400MHz (CL2-3-3-7 2T) @ 2.55v vdimm
ALX-800 w/ 80mm HAC-V81 fan
MSI Star Force GeForce FX 5700 128TD 425MHz/555MHz @ 475MHz/650MHz
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #10 (permalink)  
Old 2nd February, 2004, 06:39 PM
Chief Systems Administrator
 
Join Date: September 2001
Location: Europe
Posts: 13,075

The 3.2GB/sec figure is only at 200MHz. Bring that speed up or down and the bandwidth changes too
__________________
Any views, thoughts and opinions are entirely my own. They don't necessarily represent those of my employer (BlackBerry).
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #11 (permalink)  
Old 2nd February, 2004, 06:44 PM
ral ral is offline
Member
 
Join Date: March 2003
Location: Philippines
Posts: 351

Well, I guess I did find out one thing, my board is better off running dual channel than single
__________________
ASUS A7V600 (Bios Rev. 1.008)
AMD Athlon XP1800+ DLT3C JIUHB 0309 TPMW @ 2100MHz (10.5x200) @ 1.50v vcore
2x256MB Twinmoss PC2700 (CL2.5) @ 400MHz (CL2-3-3-7 2T) @ 2.55v vdimm
ALX-800 w/ 80mm HAC-V81 fan
MSI Star Force GeForce FX 5700 128TD 425MHz/555MHz @ 475MHz/650MHz
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #12 (permalink)  
Old 2nd February, 2004, 11:11 PM
Daniel ~'s Avatar
Chief BBS Administrator
 
Join Date: September 2001
Location: Seattle Wa.
Posts: 45,606

That alone is worth it Ral! You also refined your parameters a bit. I for one would be interested to see your new results. I'm sure Aidan would be glad to help you set up a reliable test...Which we would then publish front page...that's if your still interested.

Setting up a test and getting reliable results can be quite challenging, but it's rewarding to know a thing first hand because you tested it personally.{ or at lest know a bit about a thing":O}
__________________
"Though all men live in ignorance before mystery,
they need not live in darkness...
Justice is foundation and Mercy ETERNAL
."
DKE

"All that we do is touched by Ocean
Yet we remain on the shore of what we know."
Richard Wilbur

[img]/forum/attachments/random-nonsense/16515-sigs-dan_drag.jpg[/img]
Subscribers! Ask Pitch about a Custom Sig Graphic

Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #13 (permalink)  
Old 3rd February, 2004, 04:47 PM
ral ral is offline
Member
 
Join Date: March 2003
Location: Philippines
Posts: 351

Thanks Daniel ~. Am always interested

If Áedán is game, so am I. Am may not be able to do a "good" test this time, but I hav to start somewhere
__________________
ASUS A7V600 (Bios Rev. 1.008)
AMD Athlon XP1800+ DLT3C JIUHB 0309 TPMW @ 2100MHz (10.5x200) @ 1.50v vcore
2x256MB Twinmoss PC2700 (CL2.5) @ 400MHz (CL2-3-3-7 2T) @ 2.55v vdimm
ALX-800 w/ 80mm HAC-V81 fan
MSI Star Force GeForce FX 5700 128TD 425MHz/555MHz @ 475MHz/650MHz
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #14 (permalink)  
Old 4th February, 2004, 07:07 AM
ral ral is offline
Member
 
Join Date: March 2003
Location: Philippines
Posts: 351

Am a bit confused though...

Why do Intel 848 single channel boards perform below the level of their 865/875 dual channel counterparts?
__________________
ASUS A7V600 (Bios Rev. 1.008)
AMD Athlon XP1800+ DLT3C JIUHB 0309 TPMW @ 2100MHz (10.5x200) @ 1.50v vcore
2x256MB Twinmoss PC2700 (CL2.5) @ 400MHz (CL2-3-3-7 2T) @ 2.55v vdimm
ALX-800 w/ 80mm HAC-V81 fan
MSI Star Force GeForce FX 5700 128TD 425MHz/555MHz @ 475MHz/650MHz
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #15 (permalink)  
Old 4th February, 2004, 12:35 PM
Chief Systems Administrator
 
Join Date: September 2001
Location: Europe
Posts: 13,075

Possibly because it's Intel's 'value' offering? Whats the performance on the 865/875 when running single channel?
__________________
Any views, thoughts and opinions are entirely my own. They don't necessarily represent those of my employer (BlackBerry).
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #16 (permalink)  
Old 4th February, 2004, 03:43 PM
Staz's Avatar
Member/Contributer
 
Join Date: June 2002
Location: Lafayette LA
Posts: 3,241
Send a message via Yahoo to Staz

P4 can handle more memory bandwidth than an XP so there isn't the same bottleneck when you go to dual channle.
__________________
How come whenever I have a 50/50 chance I'm wrong 80% of the time?

What goes in a computer? Click me to find out.
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #17 (permalink)  
Old 4th February, 2004, 04:14 PM
ral ral is offline
Member
 
Join Date: March 2003
Location: Philippines
Posts: 351

Yes, I think I am startin to understand. Read an article at Extremetech:

http://www.extremetech.com/article2/...1163519,00.asp

I don't understand it all, but I guess it has to do with the chips memory controller.
__________________
ASUS A7V600 (Bios Rev. 1.008)
AMD Athlon XP1800+ DLT3C JIUHB 0309 TPMW @ 2100MHz (10.5x200) @ 1.50v vcore
2x256MB Twinmoss PC2700 (CL2.5) @ 400MHz (CL2-3-3-7 2T) @ 2.55v vdimm
ALX-800 w/ 80mm HAC-V81 fan
MSI Star Force GeForce FX 5700 128TD 425MHz/555MHz @ 475MHz/650MHz
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
Reply



Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools
Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is Off
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
Difference between single and dual channel RAM p01n7e AMD Motherboards & CPUs 2 16th August, 2006 10:55 PM
Single or Dual channel komis EPoX MotherBoards 5 13th September, 2003 09:24 PM
nForce2 Single Channel DDR chipset overthrows Dual Channel DDR? ravenlot AMD Motherboards & CPUs 4 13th July, 2003 09:33 PM
8RDA3+ FSB200+ dual VS single channel miko EPoX MotherBoards 18 3rd July, 2003 03:29 PM
Dual Channel or Single Memory? Redrum EPoX MotherBoards 5 21st June, 2003 09:23 AM


All times are GMT +1. The time now is 12:59 PM.


Copyright ©2001 - 2010, AOA Forums
Don't Click Here Don't Click Here Either

Search Engine Friendly URLs by vBSEO 3.3.0