Go Back   Hardware Canucks > HARDWARE > CPU's and Motherboards

    
Reply
 
LinkBack Thread Tools Display Modes
  #1 (permalink)  
Old March 20, 2008, 07:09 AM
CMetaphor's Avatar
Quadfather
 
Join Date: May 2007
Location: Montreal, Canada
Posts: 4,991

My System Specs

Default AMD AM2 6000+ vs. Mobile 4000+ - A benchmark comparison

Hey all, I was just wondering just how much a newer architecture and dual core and faster core speeds, etc really affect overall framerates in certain games. Therefore, this weekend I will be directly comparing my socket 754 AMD Mobile 4000+ @ 2.6ghz against my current system (In my signature). For the purpose of this test i will be using one of my 8800gts 320mbs in the 754 system and one in the AM2 system. I'll also be removing all but 1gig of ram from my AM2 system to make the comparison more fair.
Side by side system specifications:
MoDT HTPC system (754)--------------------Primary system (AM2)
CPU: AMD Athlon Mobile 4000+ @ 2.6Ghz-----AMD 6000+ @ 3.0Ghz
RAM: 1g OCZ Plat. EL (2-3-2-5)-------------OCZ Plat. Rev.2 (4-4-4-15)
Video Card: eVGA 8800GTS 320mb-----------eVGA 8800GTS 320mb
Mobo: Asus K8N-VM-------------------------Asus Crosshair
PCIx16: 1x Ver. 1.0--------------------------2x Ver. 1.0 (Will be using slot #1)

For the benchmarks I'm not totally certain which system i'll use. I was thinking of comparing the two systems using the Half Life 2 : Lost Coast video stress test at the same graphics settings for each then comparing the framerates. Any other suggestions? No, I don't have Crysis, lol.
__________________
"Backed by common sense and physics!" -Squeetard
Opteron Server for Sale! http://www.hardwarecanucks.com/forum...ade-ideas.html

Last edited by CMetaphor; March 20, 2008 at 07:20 AM.
Reply With Quote
  #2 (permalink)  
Old March 20, 2008, 09:09 AM
MVP
 
Join Date: Feb 2007
Location: Calgary
Posts: 479
Default

By far, the biggest difference is going to come from having the extra core in the AM2 system. After that, most of the difference will be from running dual channel memory vs. single channel and from the increased clockspeed. After that, the differing cache on the processors will make some difference.

DDR1 vs. DDR2 and the differing HTT speeds each make very little difference in overall performance.
Reply With Quote
  #3 (permalink)  
Old March 20, 2008, 09:25 AM
CMetaphor's Avatar
Quadfather
 
Join Date: May 2007
Location: Montreal, Canada
Posts: 4,991

My System Specs

Default

I agree about the dual-channel vs. single channel memory but not about the extra cores. Most games I play do not take full or even partial advantage of the extra core. I monitor my gaming using my G15 performance display and I've rarely seen the second core above 20-30% usage, so I don't think that will be a big factor in the performance comparison.

On a side note, I've decided that I'll test with HL2: LC at many different resolutions. If my guess is right, a new video card, even in an old system like the 754, will be able to handle resolutions up towards 1280x1024 or so without that much a difference compared to the AM2 system. If thats true then Newer systems are more for people with big screens (greater than 22'' wide), whereas an old system with some decent graphics horsepower will manage just fine for the rest of us .
__________________
"Backed by common sense and physics!" -Squeetard
Opteron Server for Sale! http://www.hardwarecanucks.com/forum...ade-ideas.html
Reply With Quote
  #4 (permalink)  
Old March 20, 2008, 09:32 AM
Thund3rball's Avatar
Hall Of Fame
 
Join Date: Nov 2007
Location: Vancouver
Posts: 3,525

My System Specs

Default

You can download the Crysis SP demo and use the GPU benchmark in there too. I just tested my 8800GT SC vs my 8800GTS 640MB (G80) using both the Crysis benchmark and the HL: LC demo.

8800GT
HL:LC - 79 FPS
Crysis - 26FPS

8800GTS
HL:LC - 79FPS
CRysis - 17FPS

As you can see the Crysis benchmark has a big difference in performance whereas HL:LC is no difference? Not sure why, maybe has something to do with improved SM3 on the GT and HL not utilizing SM3 like Crysis?
__________________
"this is not troll. this is real deal!" - hohohee

Desktop:
See the drop down!

Laptop: Dell XPS M1530 \\ T7500 2.2GHz \\ 3GB 667 \\ 6 Cell \\ 8600M GT \\ 200GB 7.2K \\ WSXGA+ \\ Bluetooth/Wireless \\ MS Blue Track Mini \\ Vista Ult. (RED)


HEATWARE
Reply With Quote
  #5 (permalink)  
Old March 20, 2008, 09:35 AM
CMetaphor's Avatar
Quadfather
 
Join Date: May 2007
Location: Montreal, Canada
Posts: 4,991

My System Specs

Default

link for the download? sounds like a good idea to bench it too then... My poor poor HTPC won't know what hit it, lol.
__________________
"Backed by common sense and physics!" -Squeetard
Opteron Server for Sale! http://www.hardwarecanucks.com/forum...ade-ideas.html
Reply With Quote
  #6 (permalink)  
Old March 20, 2008, 09:37 AM
Thund3rball's Avatar
Hall Of Fame
 
Join Date: Nov 2007
Location: Vancouver
Posts: 3,525

My System Specs

Default

Demo Versions: Crysis Single Player Demo - Demo Movie Patch Download Section - GamersHell.com

Crysis SP Demo Download - FileShack - Download PC Video game patches, videos, betas and trailers for PC, PlayStation, Xbox 360 and Wii

Another great game to bench CPU performance with is Supreme Commander

Demo Versions: Supreme Commander Demo - Demo Movie Patch Download Section - GamersHell.com

FilePlanet: Supreme Commander Demo
__________________
"this is not troll. this is real deal!" - hohohee

Desktop:
See the drop down!

Laptop: Dell XPS M1530 \\ T7500 2.2GHz \\ 3GB 667 \\ 6 Cell \\ 8600M GT \\ 200GB 7.2K \\ WSXGA+ \\ Bluetooth/Wireless \\ MS Blue Track Mini \\ Vista Ult. (RED)


HEATWARE
Reply With Quote
  #7 (permalink)  
Old March 20, 2008, 10:40 AM
CMetaphor's Avatar
Quadfather
 
Join Date: May 2007
Location: Montreal, Canada
Posts: 4,991

My System Specs

Default

Sounds good, 3 games to benchmark with. Im still not sure how much punishment ill inflict upon my 754 htpc... maybe medium settings crysis will be torture enough :P
__________________
"Backed by common sense and physics!" -Squeetard
Opteron Server for Sale! http://www.hardwarecanucks.com/forum...ade-ideas.html
Reply With Quote
  #8 (permalink)  
Old March 20, 2008, 10:50 AM
Top Prospect
 
Join Date: May 2007
Location: Alliston, ON
Posts: 75
Default

Sounds interesting. I'm still single core and will be going dual or quad in about 6 months (money permiting I admit I would have switched by now). For gameing I have found that having enough ram and a good GPU have done me well in this age of duals and quads. I like to think that if I had a crappy video card but a quad core with 4gb of ddr2 ram I wouldn't be able to play the games I can currently.
Reply With Quote
  #9 (permalink)  
Old March 20, 2008, 11:25 AM
CMetaphor's Avatar
Quadfather
 
Join Date: May 2007
Location: Montreal, Canada
Posts: 4,991

My System Specs

Default

Whats your system matty? Maybe we could get you to contribute some numbers too.
__________________
"Backed by common sense and physics!" -Squeetard
Opteron Server for Sale! http://www.hardwarecanucks.com/forum...ade-ideas.html
Reply With Quote
  #10 (permalink)  
Old March 24, 2008, 07:24 AM
CMetaphor's Avatar
Quadfather
 
Join Date: May 2007
Location: Montreal, Canada
Posts: 4,991

My System Specs

Default

Didn't have as much time as I would've liked this weekend to do tests but at least I got some preliminary numbers from the Half Life 2 : Lost Coast video stress test:

My results were fairly interesting... not to mention surprising since HL2 isn't a very old game at all:
1680x1050----------------- HTPC 754----------AM2 System----------------
Avg. FPS: Lowest Settings:--104fps------------132fps
Avg. FPS: Medium Settings:--102fps------------130fps
Avg. FPS: High settings:-----101fps------------131fps
Avg. FPS: High settings
(2x AA, 2x AF, HDR Bloom):--100fps------------133fps
Avg. FPS: High settings
(4x AA, 4x AF, HDR Full):-----97fps-------------126fps
Avg. FPS: High settings
(8x AA, 16x AF, HDR Full):----95fps-------------128fps
Avg. FPS: Really High Settings
(8x AA, 16x AF, HDR Full)
(Color correction on)
(V.Sync enabled)------------58fps--------------54fps

is all i can say.... Especially with the last test.... how on earth can a budget HTPC with an 8800GTS beat the system in my Signature? Also note that I FORGOT to remove the extra ram in my main system, so it was 1gig of ram vs. 4 gigs... I double checked those last results a few times and it was always the same... I'm not sure why the results came out this way exactly, but it certainly is interesting. It seems that HL2 is far more graphics dependant that I previously thought, So for those planning on only playing this game or a similar one- An expensive system is Not Nessesary! I will be running these tests again with my soon-to-arrive ATI 3850 256mb since theyre supposed to perform similar to the 8800gts 320mb. More results coming tomorrow!

PS: Please note that the same tests, when run using the Turion ML-37 instead of the Mobile 4000+ showed a noticable drop in fps, probably due to cpu bottlenecking. So as long the cpu isn't a bottle neck.... nothing else seems to matter much, not even 4x the amount of ram!
__________________
"Backed by common sense and physics!" -Squeetard
Opteron Server for Sale! http://www.hardwarecanucks.com/forum...ade-ideas.html
Reply With Quote
Reply


Thread Tools
Display Modes