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
Anonymous User - Wednesday, August 27, 2003 - link
#37Point well taken, But remember depending which bencmarks you use ,The diference you see could be less than you expect for example look at buffered vs unbuffered sandra scores.
Half the motherboards I have played with wont run 5/4 above 280 anyway.
Anonymous User - Wednesday, August 27, 2003 - link
Let the ignorance run wiiiilllldddd!!!!For sponsored links, I also see it at the bottom of tomshardware and the bottom of extremetech.com. I figure a company is selling those spots on multiple review sites so whoever wants a spot can grab it and have a text link on those sites.
I see big banners for corsair and googlegear on the front page of anand, which I assume would cost at least 10x more than a text link, so why didn't corsair win? They obviously pay more for advertising! Gee Kingston has full color banners too. Why didn't they win? Maybe because their modules didn't test as high, oooh, what a thought, the ram that performs the best wins, i can't believe it!!!! /sarcasm
Anonymous User - Wednesday, August 27, 2003 - link
What this (and most other) article fails to mention is that you get better performance a high FSB with a 5:4 ram divider at low latency...In other words,
275FSB at 5:4 2-2-2-5 is WAY faster than
275FSB at 1:1 2.5-4-4-7
If you have ram taht can run at 2-2-2, test it for yourself.
In short, last years low latency PC3200 and 3500 2-2-2-6 ram is faster and cheaper than todays PC4000 with rediculously horrid timings like 3-4-4-8
Radelon - Wednesday, August 27, 2003 - link
The simple fact in the matter is anybody can take the same sets of ram from all different brands and will see approximately the same results. In all my tests, OCZ is the leader, sometimes less than others but fact is, it's still on top. I've done these tests on 4 different canterwood/springdale motherboards and OCZ has always come out the best for me. "Don't knock it until you try it" That seems to be the statement of the year. The people that do knock it before they try it, are only hurting themselves and the others they influence.Anonymous User - Wednesday, August 27, 2003 - link
I've been in the hardware website business before. People don't know that they will get more hardware from a company if the give them their props. The posters who bitch about OCZ have a valid claim.You have to look at the whole picture and not just what they are doing today. I'm sorry to say that OCZ, even if they have "good" products still wouldn't be a choice by me or even recommending it.
I've heard some bad rumors with OCZ and other websites it's not even funny. 3DGameMan I heard used to give raging reviews because of getting more hardware. Overclockers something used to be in the same ballpark.
It's hard to judge reviews these days. To recommend a brand over another brand just because of the results you received is flat out ignorant at best. So many variables play into account. Corsair and Mushkin will always get my money. Even if it's more money. Their products have been around for awhile and have proven to be noteworthy.
GeIL is another company that raises an eyebrow. The owner of GeIL is the owner of an online store. All he does is buy and overclock memory and then sells it at a premium. It helps to do research on these companies.
Anonymous User - Wednesday, August 27, 2003 - link
What about Mushkin? They didn't even get entered into the test, and it used to be that Anand was always touting them, right?Hmmmmmmmmm. Anand should explain what happened to wipe Mushkin from his list totally...
AgaBooga - Wednesday, August 27, 2003 - link
I found the link, here it is:http://forums.anandtech.com/messageview.cfm?catid=...
Look down the page for Anand's post, kind of long, but it explains it! :)
I hope someone appreciates that link, hehe, it actually took about 15 minutes to find, not that long, but its not the most fun thing to do, but I had to since its for AT...
Anonymous User - Wednesday, August 27, 2003 - link
Sadly, it is clear that people still don't trust OCZ, and will go as far as to say that a positive review automatically means the web site in question (Anandtech in this case) has sold out. What's illogical about this argument is that Mushkin, Corsair, and Crucial ALL advertise on Anandtech as well, and have been advertising on Anandtech for MUCH longer than OCZ. Hopefully anyone who has read all these comments now realizes that their argument is completely invalidated by this fact. Not only that, but the writer of this particular Anandtech article even says that he has no affiliation with any of the ads that get displayed in a review, and another Anandtech editor says that OCZ isn’t even a direct advertiser. If people would learn the facts we would haven’t so many ignorant comments such as #11, #15, #16/#17, and #23.Secondly, you'd have to be blind not to see that OCZ memory is clearly one of the best solutions out there today. Anandtech is NOT the only web site that has found OCZ memory to be of the highest quality. Please search Google if you are not aware of this fact. Any review in the last 9 months will prove my point.
AgaBooga - Wednesday, August 27, 2003 - link
Hey guys, go through and look through some old posts in the archive section that Anand Lal Shimpi has talked in. One of them discussed advertising. It may be outdated, but atleast its something to look at, and if this is mentioned already, sorry, I haven't read through all the responses yet.Anyway, Anand clearly stated that the advertising portion is done by another group of people and that they do not talk directly with him or any of the article writers... hope that helps, I'll try searching around for the link...
pastorjay - Wednesday, August 27, 2003 - link
I am appalled that anyone would think that Wesley would do anything to compromise the integrity of his reviews. I have read many, many reviews at many other sites, and they have come to a similiar conclusion. OCZ has got several good products on their hands at the moment. THey are doing a spectacular job of producing quality products NOW. THey also happen to have the best Customer support in the industry, whether it would be Ryan or Sean or Bo... whoever I have dealt with, they have all been a terrific help in solving problems, and making sure I am happy. Now, I am no OCZ fanboy. I will use what i feel is the best on the market at the time... and to me... OCZ is it right now.