AMD Processors
Decrease font size
Increase font size
Topic Title: X2 5200+ Performance
Topic Summary: vs. 3000+
Created On: 06/25/2007 05:34 AM
Status: Read Only
Linear : Threading : Single : Branch
Search Topic Search Topic
Topic Tools Topic Tools
View similar topics View similar topics
View topic in raw text format. Print this topic.
 06/25/2007 05:34 AM
User is offline View Users Profile Print this message

Author Icon
znort
Junior Member

Posts: 4
Joined: 06/25/2007

Hello. I have been using an Athon 64 3000+ based pc for some years, recently I upgraded to an X2 5200+, with 800 MHz RAM, thus I am able to make some comparison.
I am running my own applications, which essentially are optimization processes, requiring lot of floating point calculations as well as access to wide and complicated data structures.
I have been a little bit disappointed to measure only a 1.33 to 1.42 increase in speed (which means nearly 25 to 30% less time) on these computational intensive applications, as one would expect about a 1.7 increase, by published AMD Overall Performance benchmarks.

On the positive side there is that running 2 applications in parallel doesn't degrade performance, so we can really count on the power of 2 CPU's, but the speed of each CPU is somewhat lower than expected.

I thus wonder if this is also your experience.

A great thing instead is the Cool'n'Quiet feature, it doesn't compromise performance, but reduces significantly power consumption and noise.
 06/25/2007 09:10 AM
User is offline View Users Profile Print this message

Author Icon
zir_blazer
Code Warrior

Posts: 2840
Joined: 03/19/2004

Your results are almost what could be expected. The benchmarks that you sayed are probabily from a Multithreaded application, but as your application seems to be Single Threaded, it uses only one Core, so though you got an A64X2 the other Core wouldn't do anything for it (But you can run two applications at the same time very well as you sayed). However, the performance for your application seems to scale almost linearly with Frequency, because if your old A64 was the Socket 754 version that runned at 2 GHz, an extra 600 MHz (30% higher) plus the DDR-II improved by that much your performance.
 06/25/2007 10:53 AM
User is offline View Users Profile Print this message

Author Icon
znort
Junior Member

Posts: 4
Joined: 06/25/2007

Thanks zir_blazer.
What you say makes sense, however the benchmarks used standard sw, which should be mostly single-threaded, e.g. see: http://www.amd.com/us-en/asset...ance_Comp_Q2-2006.pdf.
Both single core and double core processors are compared and we have a +125% for a 3800+ and a +134% for an X2 3800+, thus the benchmarks took little advantage from the availability of two cores and should account mostly for the brute force of each core.
For these reasons I was expecting more computational power from each core. As we know, this is not determined only by clock frequency, but also by architectural design of the processor, which I thought it was improved from the 3000+ design, but which is perhaps the same instead, this thus explains my results, as you said, but at this point is not clear to me how AMD could report a +163.5% performance for an X2 5000+, which implies an estimated +170% performance for an X2 5200+.
 06/25/2007 11:07 AM
User is offline View Users Profile Print this message

Author Icon
Charlie22911
Voodoo Programmer

Posts: 3278
Joined: 04/09/2004

do a google for cinebench and download it, compare results with the 3000+ and the 5600+. that will show you the difference .

-------------------------
Desktop:
Phenom II x6 1055T @ 4GHz | 4x2GB Patriot DDR3 1600 @ 2000 | 3x AMD Radeon HD6970 Crossfire

Laptop:
Core i7 2960xm @ 4.2Ghz | 4x4GB Kingston DDR3 1866 | 2x GTX 580m SLI OC 725/1450/1500
Statistics
112018 users are registered to the AMD Processors forum.
There are currently 0 users logged in.

FuseTalk Hosting Executive Plan v3.2 - © 1999-2014 FuseTalk Inc. All rights reserved.



Contact AMD Terms and Conditions ©2007 Advanced Micro Devices, Inc. Privacy Trademark information