Athlon 64 Memory: Rewriting the Rules
by Wesley Fink on October 1, 2004 12:45 AM EST- Posted in
- Memory
Final Words
With the latest memory chips from Samsung, Micron, and Hynix, the real excitement in DDR memory performance has shifted to the Athlon 64 platform. In our first full-scale roundup of memory on the Dual-Channel Socket 939, we found several DDR400 memories that reached a stable overclock of DDR600 and several more that came very close to this memory speed. It is clear that DDR memory manufacturers are paying close attention to the Athlon 64 platform, since we are seeing familiar memory reaching further on Athlon 64 than what we saw on our Intel 478 platform.A 50% overclock of memory is nothing short of incredible, but that is exactly what we are talking about with most of these memories. In comparing memory at the same CPU speed with different memory bus speeds, we also see that, in general, the improvements in memory performance are real. The latest memories are still fast enough at higher memory speeds to outperform DDR400 2-2-2 performance.
The very top of the memory tests are the most revealing results here. Despite the fact that several memories reached DDR600, highest performance was at the fastest speed the memory could achieve with a 1T Command Rate. This varied from DDR546 to DDR590, and it was at those 1T memory speeds that the best memory performance was consistently achieved.
All six memories tested here performed very well in our Athlon 64 tests. They all outstripped our expectations when we first set up the Athlon 64 test bed. However, a couple of memories do stand out. The OCZ 3200 Platinum Rev. 2 was fastest at both DDR400 and it also achieved the highest 1T speed that we found in our tests. Since these are the same chips used in four other tested memories, we can only suspect that OCZ is doing something unique in their SPD programming. We would suggest that the PCB might also be responsible except for the fact that a couple of other TCCD memories are using what appears to be the same PCB. As we saw in our 2-2-2 roundup, the Crucial Ballistix also stood out for the incredibly fast timings that the Micron-based Ballistix achieved in the DDR433 to DDR533 speed range. The Ballistix was faster through much of the tested range, and the OCZ was best at the top and bottom.
This should not take anything away from the excellent and consistent performance of the G. Skill TCCD and Geil 3200 Ultra X. Both exhibited a very wide range of memory speeds and they were both competitive at every speed. The PQI 3200 Turbo was generally a bit slower, probably the result of tweaking for the Intel platform. We suspect that this would quickly change if PQI updated their SPD programming for better Athlon 64 performance. Still, there is nothing to really complain about in the performance curve of the PQI memory.
It is a bit early in Athlon 64 testing to select an Editor's Choice, but the OCZ 3200 Platinum Rev. 2 and Crucial Ballistix are former Gold Editor's choices. These two memories were also the standouts in a group of standouts in these tests. The Hynix-based OCZ 3700 Gold Rev. 3 takes a different approach to memory performance, but it is still an excellent choice, if it can be bought for a lower price than the Samsung TCCD and Micron-chip memories.
We learned that the Athlon 64 quest for a 1T Command Rate is worth the search, but you also will need the tools to allow the best overclock with memory on the Athlon 64. As surprising as it will seem to some, that should include the highest quality power supply that you can find at 500 watts or more. We found that replacing a well-respected 465 watt PS with a 520 watt PowerStream allowed even higher memory overclocks. This was true with both the power-hungry nVidia 6800 Ultra as well as the more mainstream ATI 9800 PRO. If top memory overclocks on the Athlon 64 is your goal, don't skimp on the power supply. Putting the best PS that you can find in your system will pay off in higher memory overclocks with greater stability.
The memories tested here were a cross-section of the best current memory that you can buy. They used Samsung TCCD, Micron G die, and Hynix DT-D5 memory chips. All of the memories based on these current chips performed incredibly well on the Athlon 64 Socket 939. The Samsung TCCD, in particular, seems to be much better on Athlon 64 with recent chips than what was seen in early testing of TCCD. With Athlon 64 performance this good, we can only wonder how long it will be until AMD makes DDR500 or DDR533 a standard DDR memory speed on the Athlon 64 on-chip memory controller.
47 Comments
View All Comments
saechaka - Friday, October 1, 2004 - link
i can't seem to find a legit place to buy that ocz 3200 rev. 2. any suggestionsAvalon - Friday, October 1, 2004 - link
Excellent article. It's good to know what different memories can do on the Athlon 64 platform.ramclocker - Friday, October 1, 2004 - link
14The psu is probably around 20A on the 12V...I know from my testing 20A doesn't cut it anymore on a high end gaming/benching rig....you also have to remember that at high speed the ram will be drawing high levels of current also and the board will draw higher current due to heat etc.
I found the article an excellent read due to the fact it finally proved to me with reasonably tight timings running high fsb over 2-2-2 at 200 is the way to go...running 2.4gig for all tests Wes was the wise move here...great work.
Blappo - Friday, October 1, 2004 - link
The computer would probably wouldn't use more than 250W. I understand that you don't want to mention the make and model. The nVidia 6800 Ultra draws most of its power through the 12V connection to the PSU, where the ATI 9800 Pro draws its power from the AGP slot. What is the max current rating on the 12V rail for the 465W PSU that you were using? I agree that a high quality PSU is needed (although not necessarily high max rating).Wesley Fink - Friday, October 1, 2004 - link
#12 - The Asus A8V is reviewed at http://www.anandtech.com/mb/showdoc.aspx?i=2128 and compared to other 939 boards. In memory testing we use a standard test bed to minimize variables.#11 - The 90nm Athlon 64 tests should appear next week. We have just received 90nm 3500+ and 3000+ processors. AMD did not do a media launch on these processors, so we had to find them on the open market
#9 and #11 - A Value RAM roundup is in the works, but it has been moved out a while because of a large number of new launches this month.
Deuce - Friday, October 1, 2004 - link
It sure would be nice with tests also conducted on the Asus A8V. I'm still deciding between the two.PrinceGaz - Friday, October 1, 2004 - link
Just to follow up that comment, I suppose the DDR533/2.4GHz results are actually the most useful out of them all when it comes to comparing those particular modules. All of them were fastest (at 2.4GHz) at that speed, except for the OCZ PC3200 Plat Rev.2 which was marginally faster at 8x300 for DDR600.Anyway, I'm looking forward to the reviews of the desktop 90nm A64 processors, and especially finding out how well each of them overclocks.
And also the Value Memory review you promised a few weeks ago :)
PrinceGaz - Friday, October 1, 2004 - link
Although all the (admittedly premium quality)memory could reach very high speeds, that didn't have much impact on performance.Taking the highest clocking brand as an example, the OCZ PC3200 Platinum Rev.2, from the DDR400 2-2-2 speed to the DDR534 2.5-4-3 speed which was the best result still at a CPU speed of 2.4GHz, the results were
Quake 3: 516.3 -> 525.8
Super PI: 80 -> 79 (lower is better)
Wolf: 110.8 -> 112.7
So running the memory at DDR534 instead of DDR400 provided less than 2% increase in performance. This is to be expected when you compare the real-world performance of S754 and S939. The only thing that is important is that the memory can do 1T command-rate to the maximum overclock of your A64 at default multiplier.
I think the results on the highest memory performance page are probably misleading to some readers. It shows the Crucial Ballistix coming in at 536.5fps on Q3. Looking at the results I see that was at 9x278 for a CPU speed of 2.5GHz. Your CPU was able to reach over 2.6GHz so the performance in real world tests would have been somewhat higher with a 10x multiplier. Sandra results are irrelevant to most people.
It would be better if you included an additional test in addition to Highest Memory Speed, and Highest Memory Performance. This would be Highest CPU Speed where the CPU is maxxed out, and the memory run at whatever multiplier gives best performance on real-world tests (ignoring Sandra). I suspect the results would be a *lot* closer.
AtaStrumf - Friday, October 1, 2004 - link
Now about some value RAM tests? These modules are just too expensive for most of us.Jalf - Friday, October 1, 2004 - link
Or maybe the "average" user would rather blow $200 on 1GB memory ;)