AMD Performance Test Configuration

The Corsair TwinX1024-4400C25 were first tested on the new Athlon 64 memory test bed. You can find more information on the new memory test setup in Athlon 64 Memory: Rewriting the Rules.

The A64 test bed includes components that have been proven in Socket 939 Athlon 64 benchmarking, such as the Gold Editors Choice MSI K8N Neo2, the completely unlocked Socket 939 FX53, and the OCZ Power Stream 520 Power Supply. Since the Athlon 64 tests represent a new series of DDR testing, we have chosen the current generation nVidia 6800 Ultra video card for benchmarking. We have found the 6800 Ultra to be a particularly good match to nVidia nForce3 Ultra motherboards.

All other basic test conditions attempted to mirror those used in our earlier Intel memory reviews. However, test results are not directly comparable to tests performed on the Intel test bed.

 AMD nForce3 Ultra Performance Test Configuration
Processor(s): AMD FX53 Athlon 64
(2.4GHz, Socket 939, Dual Channel, 1000HT)
RAM: Corsair TwinX1024-4400C25 (DS) 2X512MB
G. Skill TCCD (DS) 2X512MB
PQI 3200 Turbo (DS) 2X512MB
Crucial Ballistix (DS) 2X512MB
Geil PC3200 Ultra X (DS) 2X512MB
OCZ PC3200 Platinum Rev 2 (DS) 2X512MB
OCZ PC3700 Gold Rev 3 (DS) 2X512MB
Hard Drives Seagate 120GB PATA (IDE) 7200RPM 8MB Cache
PCI/AGP Speed Fixed at 33/66
Bus Master Drivers: nVidia nForce Platform Driver 4.24 (5-10-2004)
Video Card(s): nVidia 6800 Ultra 256MB, 256MB aperture, 1024x768x32
Video Drivers: nVidia Forceware 61.77
Power Supply: OCZ Power Stream 520W
Operating System(s): Windows XP Professional SP1
Motherboards: MSI K8N Neo2

We have found the fastest performance on AMD Athlon 64 chipsets (nForce3, VIA K8T800 PRO) to be achieved at Cycle Time or tRAS of 10. Athlon 64 platform benchmarks were therefore run with the tRAS timing of 10 for all A64 benchmarks.

Test Settings

The FX53 is completely unlocked, which allows a different approach to memory testing that truly measures performance differences in memory speed alone. All tests were run with CPU speed as close to the specified 2.4GHz of the FX53 as possible, with CPU speed/Memory Speed increased at lower multipliers to achieve 2.4Ghz. This approach allows the true measurement of the impact of higher memory speed and timings on performance, since CPU speed is fixed, removing CPU speed as a factor in memory performance.

The following settings were tested with the Corsair DDR550 on the Athlon 64 test bed:
  1. 12x200/DDR400 - the highest stock memory speed supported on K8T800-Pro/nF3-4/SiS755-FX motherboards.
  2. 11x218/DDR436 - a ratio near the standard DDR433 speed
  3. 10x240/DDR480 - a ratio near the standard rating of DDR466
  4. 9x267/DDR533 - a standard memory speed used in testing other high-speed memory
  5. Highest Memory Speed - the highest memory speed that we could achieve regardless of the multiplier. This setting was generally achieved at a 2T command rate and performance is often poorer than slower memory timings at a 1T Command Rate.
  6. Highest Performance - the highest memory performance settings that we could achieve. This setting is normally the highest stable speed using a 1T Command Rate.
Command Rate is not normally a factor in Intel 478 tests, but it is a major concern in Athlon 64 performance. A Command Rate of 1T is considerably faster on Athlon 64 than a 2T Command Rate. For this reason, we had added the Command Rate to the timings and voltage reported for each memory speed setting.

We ran our standard suite of memory performance benchmarks: Quake 3, Return to Castle Wolfenstein-Enemy Territory-Radar, Super Pi 2M, and Sandra 2004 Standard and UnBuffered. Since the results for Athlon 64 tests are new, we are now including Sandra Buffered (Standard) test results and Sandra UnBuffered test results. RTCW Enemy Territory has also been added as a standard memory benchmark.

Corsair PC4400 Intel Performance Test Configuration
Comments Locked

23 Comments

View All Comments

  • kmmatney - Tuesday, January 4, 2005 - link

    The review did have some mid-range memory in the tests. Their value was noted in the last sentence of the review, but it should be highlighted more.

    According to the review, the PQI turbo 3200 performed almost as well, but is almost half the price! The money saved can be spent on a cpu or video upgrade.

    A quick glance at NewEgg shows PQI turbo 4000 memory priced at $95.50, which I think is a good deal.


  • PrinceGaz - Tuesday, January 4, 2005 - link

    #6- I've been wondering about the value/mid-range round-up too. Looking at the AMD 533/2.4GHz results page, the largest difference between the best and worst memory types in the real-world applications is just over 3%, and these are applications that were chosen because they are more affected by memory-bandwidth than usual. At the end of every review of premium memory it should say "it doesn't really matter which premium module you get for an AMD system so buy the cheapest as it'll make no real difference in performance, and the money would be better spent on a higher-rated CPU or better cooling". Maybe the worry is that if they test some budget/mid-range memory, they'll find you'd be better off getting that instead of the premium modules because it makes so little difference, and that would annoy the companies that send them premium samples :p

    Intel platforms are more affected by memory bandwidth, but they are moving to DDR2 which makes DDR tests increasingly irrelevant for them.
  • HardwareD00d - Tuesday, January 4, 2005 - link

    Wes - Thank you for the clarification.

    I'll trade you my 2 sticks of ShikaXRam for your Corsair sample. ;)
  • Wesley Fink - Tuesday, January 4, 2005 - link

    #9 - No conspiracy intended. Our AMD memory test bed was not even set up when we tested the Shikatronics memory in June. We have stated many times that Hynix B, the chipsets used in the Shikatronics, does not generally perform as well on the Athlon 64 platform as it does on Intel. TCCD usually does better on AMD than Intel. Also the timings are 3-4-4 on the Shika which are slower than TCCD at the same memory speed.

    Just to make sure we weren't blowing smoke, I popped the Shikatronics 550 into the AMD test bed. It did it's specified DDR550 on the A64 at 3-4-4-10 2.85V, which is excellent for Hynix B on AMD. However, I could not make it to even DDR560 as the memory topped out at 554 on the AMD platform.

    We are testing new memory on both AMD and Intel, but we have not gone back to past Hynix B dimms for an AMD update, nor do we plan to.
  • HardwareD00d - Tuesday, January 4, 2005 - link

    I've got the Shikatronics PC4400 DC kit for my Athlon64, and I'd like to see that compared against the Corsair modules. It IS on the Intel side, and is the clear winner. Seems strange that you wouldn't compare the "priorly fastest" memory (per a previous review) against the new Corsair modules. This omission seems a little fishy to me.
  • erinlegault - Tuesday, January 4, 2005 - link

    Maybe other PC4400 rated memory from other companies should be compared, especially the OCZ PC4200 Pt Series ram. All the ram used was PC3200, except for the PC3700 OCZ, and overclocked from the rated 400MHz to near 600MHz. The Corsair PC4400 memory was only overclocked from the rated 550MHz to 636MHz. I would personally like to see if any other PC4000 or higher rated memory could do any better.
  • Wesley Fink - Tuesday, January 4, 2005 - link

    #2 - We used 5X HT at stock speeds, 4X HT for 218 and 240, and 3X HT for 267 up. All other settings are in the reviews.

    #4 - With the huge number of memories reviewed at AnandTech, we feel comfortable in talking about relative positioning of tested memory. With 28 memories compared in performance on the Intel charts and seven different DDR400 2-2-2 memories in the AMD charts, there is certainly comparative info to draw conclusions. The Corsair 4400C25 proved to reach the highest FSB, but it was not the best choice in the DDR400 to DDR450 range. We state that clearly in our conclusion.
  • eetnoyer - Tuesday, January 4, 2005 - link

    Wow, another elitist memory review. As if TCCD weren't reviewed to death already. By the way, still waiting for that value memory round-up. Any idea when you will be able to get around to reviewing memory products that are useful for the majority of your users?
  • Marlowe - Tuesday, January 4, 2005 - link

    Really interesting IMO.. But I would love if you could review the A-Data Vitesta PC4800 ram also. They are based on TCCD and have maby Brainpower PCB. On my P4C800 my 2x512 kit can do 2-2-2-5 at 220MHz and tops out around 275-280 MHz on 2,5-3-3-5, both on 2,85V. Well that's what I could do with my 3,0E and watercooling anyways :P Also they're quite affordable in comparison.
  • arswihart - Tuesday, January 4, 2005 - link

    anandtech reviews are more and more praising a product as ultimate, best, etc... all based on in some cases, insufficient testing to say such things. granted all review sites do that to some respect, its just the conclusions pages are getting kind of narrow viewed as if they have a whole picture and are 100% sure of there recommendations, while often they haven't tested enough competing products or taken all of their recommended product's potential drawbacks or limitations into consideration. still, its a really good review site

Log in

Don't have an account? Sign up now