Athlon 64 Revision E: Unofficial DDR500 Support
by Anand Lal Shimpi on July 11, 2005 12:05 AM EST- Posted in
- CPUs
Final Words
Based on the tests that we’ve seen here today, AMD’s reluctance to move to higher bandwidth DDR2 offerings makes a lot more sense. The plain fact of the matter is that at the current clock speeds at which the Athlon 64 and X2 line are running, most desktop applications see virtually no benefit from higher bandwidth memory. It is possible that server usage models may show a greater performance boost, but it is highly unlikely for a mission critical server to be equipped with anything that isn’t an officially supported standard - especially memory.While some have been critical of AMD’s unwillingness to embrace DDR2 when Intel did, it would appear that the quest for more bandwidth simply wasn’t in AMD’s best interests. These Athlon 64 and X2 cores that we have here today are far better suited for use with low latency and lower priced DDR400 than anything that offers higher bandwidth.
Down the road, as CPU speeds and the sheer number of cores goes up, then higher bandwidth memories will definitely make much more sense. But for now, for the majority of the population, these new memory dividers won’t do much for you.
The performance improvements themselves aren’t tangible, but if you are trying to squeeze every last ounce of performance out of your system, then these new memory dividers offer you one more avenue to do so. If you have memory that can run at higher than DDR400 speeds without any reduction in latency, then by all means, explore the new dividers; just don’t expect them to change your life.
The one exception to the rule seems to be heavy multitasking scenarios. As we saw from our simple DVDShrink + Doom 3 test, when you run two very memory bandwidth dependent applications on a dual core processor at the same time, the benefits of these faster memory speeds really starts to show itself. We measured a 6.5% increase in performance in the aforementioned test, but next to no performance improvement in other lighter multitasking scenarios. As we continue to develop our multitasking benchmark suites, we will now start looking at how added memory bandwidth, made possible through these new dividers, changes the performance picture.
Facebook
Twitter
RSS
37 Comments
View All Comments
ElJefe - Monday, July 11, 2005 - link
Well, I could say this would have been a great test if only one thing: to show the effect of TIMINGS on this. I know the OCZ had sickest tight timings, but im talking about the difference of say, using 4 single gig memeory modules and then messing around with the timings, showing which is best. I know at 1 gig, most of those l33t ram flashy types of dual channel matches drop off and kinda go into hum-drum land.If i had an X2 system it would have 2 gigs on two chips or 4 gigs on 4 chips. I couldnt see a dual proc system built for simply faster gaming, it's meant to be a cookie monster of processes.
KristopherKubicki - Monday, July 11, 2005 - link
bupkus: DDR2 comes with the M2 socket next year. At least, thats according to the roadmap.Kristopher
fishbits - Monday, July 11, 2005 - link
"The ceil() function is a pretty basic mathematical function that returns the smallest integer value greater than its argument..."Think that should be "greater than or equal to." ceil(10.0)=10. Not that it exactly matters much in this context. Anyhoo...
I'm glad AMD is further ahead of supporting mem speeds than needed, as opposed to playing catch-up. Additional options and capability for their future items will pay off down the road if not now.
brownba - Monday, July 11, 2005 - link
#1"keeps getting better and better."
huh? did you look at the benchmarks?
what is better?
Anand just showed us to save our money, we don't the very fastest ram, so I guess that is better.
Wesley Fink - Monday, July 11, 2005 - link
#2 - The OCZ PC3500 GX is specially binned BH5. Anand had planned to use VX memory but it was out of stock as this article was being developed. OCZ binned some BH5 at 3.3V for 2-2-2 operation at DDR500 to meet the requirements of this review and turned it around in a few hours. The point is that OCZ PC3500 GX will NOT all operate at 2-2-2 at DDR500. The OCZ VX and Mushkin Redline, on the other hand, are both rated 2-2-2 at DDR500 at 3.5V. The VX/Redline are based on Winbond CH5 blanks.Wesley Fink - Monday, July 11, 2005 - link
#20 - There is a new official DFI BIOS in the works that is to post soon. However, as I mentioned in the nF4 Ultra roundup there are over 60 BIOS revisions avaialble for the DFI nF4 boards. Oskar Wu of DFI has posted many of the BIOS' at www.dfistreet.com, the DFI Forum website, or in the Forums at www.xtremesystems.org.yacoub - Monday, July 11, 2005 - link
[In keeping with our recently growing practice,] "We chose three CPUs to investigate the impacts of these new memory dividers: [The three you are least likely to own]: the Athlon 64 X2 4800+ (2.4GHz/1MB L2), the Athlon 64 X2 4200+ (2.2GHz/512KB L2) and the Athlon 64 FX-57 (2.8GHz/1MB L2)."hehehe
ksherman - Monday, July 11, 2005 - link
I know this is going to sound like a dumb question, so here it is. Where did you get the BIOS update for the DFI board? I have that board, and their website only lists an update throung March, nothing as recent as July...Stinger22 - Monday, July 11, 2005 - link
One small correction..There is also an Athlon 64 3000+ that is a Socket 939 and is Revision E.
Hacp - Monday, July 11, 2005 - link
#13, I know of one other mobo that supports 3.3 volts and many other ram sticks can get 2-2-2-5 at ddr500.