Author: William Henning
Editor: Howard Ha
Publish Date: Monday, June 4th, 2007
Originally Published on Neoseeker (http://www.neoseeker.com)
Article Link: http://www.neoseeker.com/Articles/Hardware/Reviews/65nm_am2_x2_5000/
Copyright Neo Era Media, Inc. - please do not redistribute or use for commercial purposes.
AMD quietly started shipping 65nm Athlon X2 parts some time ago; and recently they sent us a 65nm X2 5000+ to take a look at.
In the past, we reviewed the 90nm X2 5000+ ... so now that we have its 65nm counterpart in hand, we will set up the exact same system and see what differences - if any - there are in performance and overclocking elbow room between the two chips.
Now before we start, you may be wondering why AMD is doing such a "soft" launch of the new 65nm parts.
It's quite simple really.
AMD's current Athlon 64 X2 processors are simply not as fast as Intel's Core 2 Duo offerings, thus it makes little sense to trumpet the introduction of newly shrunk devices when they will not outperform the competition.
While the AM2's may not have the highest 'bang' - they definitely do have a good 'bang for the buck', with Athlon 64 X2 3600+ processors being widely available for as little as $59 OEM / $69 boxed, the E4300 at $114 boxed is still $45 more expensive.
By switching the production of Athlon X2's to 65nm, AMD gets many more die's per 300mm wafer, thus reducing its costs; which allows them to sell processors at significantly lower prices than Intel - thus maintaining market share with a lower performance product until their new Barcelona products are ready.
Mind you, Intel is aware of this, and will soon introduce the E2100 series of Core 2 Duo's for the low end.
In order to keep the testing as fair as possible, we will use the following test platforms:
ASUS' M2N32-SLI motherboard is an excellent performer, and allows us to try very high FSB speeds.
Naturally, we also used a high-end 975X-based motherboard for our LGA775 benchmarking. In this case it was the ASUS P5B-E Premium -- a favourite of ours here.
Software used during testing consisted of the following:
Please note that we are showing overclocked results in all the charts - we are not holding you in suspense until the end of the article. :-) We are also presenting you with a FULL set of overclocked benchmarks - our test sytems were all stable at the settings shown with air cooling. Please note that chart labels incorporate a lot of information about the test configuration. The first line shows the socket type and the model of the processor. Since all the processors shown are dual-core devices, we did not specify that on the charts.The second line shows the "FSB/HT clock rate" x "CPU multiplier" followed by the effective DDR memory speed. Please note that all DDR2 tests were run at 4-4-4-12 timings unless otherwise specified.
At stock speeds, the 65nm AMD Athlon X2 5000+ slightly outperforms its 90nm predecessor, and it also outperforms the Core 2 Duo E4300, E6300 and E6400!
For content creation, at stock speeds we again have the 65nm 5000+ outperforming the Core 2 Duo's its going up against!
It looks like for every day office use, the Athlon 64 X2 5000+ is an excellent choice.
Unfortunately we switched what version of Sandra we used for benchmarking a few months ago; however the results are still quite comparable. The Core 2's beat the Athlon X2 5000+ for Sandra CPU scores.
At stock speeds, the AM2 X2 5000+ only beats the E4300; however the chart shows you what a difference overclocking can make to memory performance.
We can see that the read performance of the X2 5000+ at stock speeds is only slighly better than the E4300, but scales nicely when overclocked.
The AMD Athlon 64 X2 5000+ easily trounces the Core 2 Duo's at stock speeds for write performance - and scales nicely when overclocked.
The on-chip memory controller means that the AMD X2 5000+ simply destroys the Core 2 Duo's when it comes to latency.
At stock speeds, the 5000+ does not fare well compared to the Core 2's for bandwidth.
At stock speeds, the X2 5000+ beats the E4300, however it needs to be seriously overclocked to beat the E6300 and E6400 at their stock speeds.
As we've only recently started running the multi-threaded version of LAME, we only have a couple of results for you.
The AMD X2 5000+ managed to do very well at TMPGEnc at stock speeds, beating the E4300 and the E6300!
The AMD X2 5000+ did well at CineBench; at stock speeds it beat both the E4300 and E6300, and was only slightly bested by the E6400.
There is a definite pattern emerging... again, the X2 5000+ beat both the E4300 and the E6300 at stock speeds!
Call of Duty
The X2 5000+ beat the E6300 for call of duty, however its performance was lackluster compared to the E6400.
Again, the X2 5000+ beats the E4300 and E6300 at stock speeds!
Well, the AMD X2 5000+ at stock speeds beat the E4300 - but was otherwise trounced by the E6300 and E6400.
Surprisingly the X2 5000+ slighly beat the E6300 and also beat the E4300... and thoroughly trounced the older 90nm 5000+
In an interesting turn of events, at stock speeds the 65nm AMD X2 5000+ beat its older brother, the E4300, E6300 and E6400 for Jedi Knight!
The X2 5000+ beat the E4300 and E6300 for Unreal Tournament 2004.
Overclocking the 65nm AM2 X2 5000+ was an interesting experience.
I managed to get it running with FSB speeds as high as 345MHz - but only with significantly reduced multipliers. In order to get the best possible balanced score, I spent a lot of time trying different multiplier and FSB combinations, and frankly my experiments were made needlessly frustrating by the unpredictable ways it would choose the memory speed dividers.
If only the BIOS would have let me set the memory ratios directly, I am certain I could have achieved even better scores than I did.
Regardless, 3.192GHz for a socket AM2 processor, with air cooling, is no mean feat, and is definitely a decent overclock - we've just been spoiled by the insanely great overclocking headroom available on Core 2 Duo processors, where even 1.8GHz processors can regularly hit 3.4GHz with air cooling.
How did I get to 3.192GHz?
Please do not redistribute or use this article in whole, or in part, for commercial purposes.