Searching for the Memory Holy Grail - Part 2
by Wesley Fink on August 26, 2003 11:11 PM EST- Posted in
- Memory
Does Memory Speed really matter in the Real World?
There have been skeptics in the computer industry who have boldly stated that Dual-Channel 266 memory will give performance as fast as you can get — in the real world — with a Canterwood or Springdale motherboard. Another popular variation of this is that memory over DDR400 (PC3200) will make no difference in real-world performance of an 875/865 system. To determine whether these statements held any validity, we decided to first take a look at the influence of memory speed alone on our chosen benchmarks.
We wanted to isolate performance variation as much as possible to Memory. The very design of the Intel 875/865 chipsets makes this difficult, since we have very limited settings for memory ratio. We finally decided to test stock 800FSB settings with different memory speeds, and to also test the highest standard CPU FSB setting we could run at different memory timings. To keep variables at a minimum, we looked at our benchmark results for memory that would perform at both 533 and 400 at the same timings. We settled on two 512 MB DS OCZ 4000 DIMMs. We ran the memory at stock 2.4GHz at 2.5-3-4-6-1 and at a high overclock of 1066FSB at the same 2.5-3-4-6-1 timings. Since we required 2.75V for stable operation at 1066FSB at these timings, we decided for consistency to set the vDimm to 2.75V for all tests. At each setting, we varied only the memory speed at the available 1:1, 5:4, and 3:2 ratios.
Double-Sided Memory
| Standard 800FSB (2.4Ghz) Performance at Varied Memory Speeds — 2 x 512 MB DS DIMMs |
||||||
| Memory DDR Speed | Quake3 fps | UT2003 Flyby fps | UT2003 Botmatch fps | Sandra UNBuffered | Sandra Standard Buffered | Super PI 2M places (time in sec) |
| 266 MHz | 300.30 | 188.66 | 65.55 | INT 1854 FLT 1814 |
INT 3759 FLT 3787 |
144 |
| 320 MHz | 313.70 | 193.26 | 67.57 | INT 2138 FLT 2123 |
INT 4254 FLT 4256 |
138 |
| 400 MHz | 328.07 | 198.27 | 69.16 | INT 2594 FLT 2640 |
INT 4700 FLT 4724 |
132 |
| 1066FSB (3.2GHz) Performance at Varied Memory
Speeds — 2 x 512 MB DS DIMMs |
||||||
| Memory DDR Speed | Quake3 fps | UT2003 Flyby fps | UT2003 Botmatch fps | Sandra UNBuffered | Sandra Standard Buffered | Super PI 2M places (time in sec) |
| 355 MHz | 382.67 | 235.82 | 85.13 | INT 2415 FLT 2394 |
INT 5043 FLT 5039 |
107 |
| 426 MHz | 403.56 | 239.96 | 87.82 | INT 2924 FLT 2875 |
INT 5711 FLT 5688 |
104 |
| 533 MHz | 424.5 | 249.24 | 91.53 | INT 3532 FLT 3542 |
INT 6308 FLT 6252 |
100 |
Since there is much confusion about when and whether Intel’s PAT is activated, and its effect on performance, each setup was checked with CPU-Z 1.18C. Under the “Memory”’ tab, with this version of CPU-Z, there is a box for “Performance Mode”, which will indicate “enabled” or “disabled”.
In all configurations, except one, with the ASUS P4C800-E with 800FSB or higher selected, both Synchronous and Asynchronous, CPU-Z indicated Performance Mode “enabled”. We will talk more about the exception later.
As you can clearly see from the tables above, gaming performance continues to improve as memory gets faster on the Intel 875 motherboard. As we move from an 800FSB CPU running 266 memory to the CPU running the same speed with DDR400 memory, we see Quake3 frame rate increase a bit over 9%, while UNBuffered Sandra increases about 40% in memory bandwidth. At 1066 constant CPU speed with memory increasing from 355 to 533, we see a larger increase of 11% in Q3 frame rate, while Sandra UNBuffered increases about 46%. The increase in Quake3 frame rate from 800/266 to 1066/533 — which admittedly includes a large boost in CPU speed — is significant, at 41%.
UT2003 also shows a similar pattern of increases over the same range, with increases over the whole range of 32%-39%. Sandra UNBuffered measurement of memory bandwidth shows a 90% increase over the same range from 800/266 to 1066/533. As we would expect, both UT2003 and Quake3 appear to respond more to a CPU speed increase than a memory increase, but alone, the increases in game benches from memory speed increases are real and significant.
Our pure number-crunching benchmark, Super PI, is shown to be more sensitive to memory speed than you might expect. We saw increases of 7% to 9% in the “memory only” increases at a fixed CPU speed. These increases are only a little smaller than those found in our Quake3 tests.
It is conclusive from these benchmarks that Memory Speed does matter in real-world performance on Intel 875/865 motherboards. Game benchmarks, Super PI, and Sandra Memory Tests all benefit from increases in memory speed. This was true up to the highest memory speed that we tested — DDR533. The differences, when we looked at just the effect of memory alone, varied from 9% to 11% at a given CPU speed, with the limited memory ratios Intel has provided us on the 875/865 chipsets. You will have to decide if the increases in performance from using faster memory are worth the cost of that speedier memory. For some, these increases will matter a great deal, while for others, they will not be worth the cost.
Facebook
Twitter
RSS
77 Comments
View All Comments
dshodson - Friday, September 5, 2003 - link
So what is the best memory to put in my new dell server coming in a week which has the 875 chip. I heard u cant really overclock this system although i havent received mine yet.Anonymous User - Tuesday, September 2, 2003 - link
Sorry about the typos ( its 11:36 PM here )Thanks
OCZGUY
Anonymous User - Tuesday, September 2, 2003 - link
Uberclocker ,If your having a problem with a OCZ part , email me directly , Most problems are related it simple setup issues , and your problem is verry likely fixed in a few simple steps , or with a bios update
My email is oczguy@ocztechnology.com
I will not like answer your email until tomorow ( tuesday ) as today is a holiday and I am not in the office today
Thanks
OCZGUY
Anonymous User - Monday, September 1, 2003 - link
Hello, I have followd your recommendation and bought an OCZ 4000 in Germany, this weekend. Unfortunately it has not performed to its specification. Can you tell me if there is a special part number I should look for to get their special 'reviewers golden sample' ?UberClocker
Anonymous User - Saturday, August 30, 2003 - link
I think if you have a cpu that will do 250 FSB and you purchase memory based on this article and you get within 98% of the results posted in this article, you owe Wesley $10.00. If 20% of the people with favorable results donated, there would be no more need to advertise. Now for the people who did not purchase ram based on this article and are bitching about it being bias, what are you bitching for?. This article has not cost you a cent.Thanks Wesley, you saved me a $100 on ram. I almost got caught up in all the hype about 4000 and 4200.
Fossil
Anonymous User - Saturday, August 30, 2003 - link
This article goes part way to answering the question of what memory configuration is fastest but I was disappointed that there was no variation in ratio.I would like to see a review that tests agressive timings against high frequency to see what produces the best results.
Also a test of all P4C chips running @ 3.6Ghz to see whether there's any benefit in buying the fastest chip.
With most motherboards easily hitting 1Ghz are we likely to be seeing a P4D which supports a 1066Mhz FSB?
Anonymous User - Friday, August 29, 2003 - link
I think the way to do these reviews is to tell the manufacturer you'd like to include them in a review, purchase the memory from a store you trust, and then get re-imbursed by the manufacturer. That way, there'd be no selected modules and you would know what you were going to put into the review (4 ss vs 2 ds).So, when do we get that part 2 of the 865/875 mobo roundup ?
retrospooty - Friday, August 29, 2003 - link
Wesley,I have run these tests myself on an IC7G at 200 , 230 250 and 280 FSB and found that 5:4 2-2-2 always beats 1:1 at 2.5-4-4, even 2.5 4-3 ... However SOME motherboards arent as efficient at 5:4 with some types of ram as others... If you try the same test on several different mobo's I think you will find that 5:4 2-2-2 is quite a bit faster in non-synthetic mem test benchmarks.
Oldfart # 64, you are right about the reviewers and using the synthetic tests to boost sales. good point.
Anonymous User - Friday, August 29, 2003 - link
If your not going to overclock, get low latency pc 3200 or 3500, the PC4000 wont help you at alleven IF you ARE going to overclock, the PC4000 will not likely get you better performance.
Anonymous User - Friday, August 29, 2003 - link
#67 I'm not an expert by any means but here is how I understand this discussion:If you are planning to use one of the faster processors such as the 3.0C you won't be able to achieve a big overclock because the multiplier for that processor is 15. If you want to use a higher speed processor such as the 2.8C or the 3.0c, you really dont need to spend the money on faster memory such as the PC4000 because you probably won't be able to overclock to a point that you will be able to reach a front side bus speed of 250mhz which is the lowest rated speed (x2=ddr500) of the memory module. So as the article has explained,you would be better off with
slower ram that has faster timings. However, if you have a slower processor such as the 2.4C or the 2.6C, you should be able to achieve a higher overclock because these processors have lower multipliers. In this case you would need faster ram such as the PC4000, because with these processors many have been able to set their front side bus up to and beyond 250mhz. In most scenarios with fast processors/slower ram vs slower processors/faster ram, IMHO, the performance difference may be fairly equal. What is really happening is a cost/benefit consideration. As always be careful!!! Overclocking can damage your system...you do it at your own risk!