DDR2 v DDR3Post Date: 2008-09-25 |
Post Reply
|
| Author | |
Lawdog
Newbie
Joined: 18 Aug 2008 Online Status: Offline Posts: 64 |
Quote Reply
Topic: DDR2 v DDR3Posted: 25 Sep 2008 at 3:21am |
|
Ok so I found myself messing with the configurator again and I know that all I hear is that DDR3 is not worth the money right now as you get more bang for your buck with DDR2.... If that's the case how did this happen
Am I missing something? It is quite late.
If I'm not missing anything why wouldn't you go DDR3?
|
|
![]() |
|
DST4ME
DS ELITE
Joined: 14 Apr 2008 Online Status: Offline Posts: 36758 |
Quote Reply
Posted: 25 Sep 2008 at 3:41am |
|
not till you hit ddr 1600 and above, does the ddr3 show its difference with ddr2.
but if the prices are the same then ya go with ddr3, why not. |
|
![]() |
|
widdlecat
DS Veteran
Joined: 11 Mar 2008 Online Status: Offline Posts: 840 |
Quote Reply
Posted: 25 Sep 2008 at 3:46am |
|
Keep in mind that the type of memory will affect your choices of mobo. Very few mobos use both ddr2 and dd3. Those that do will favor ddr2 in operation usually.
|
|
|
|
![]() |
|
gamerk2
Groupie
Joined: 28 May 2008 Online Status: Offline Posts: 198 |
Quote Reply
Posted: 25 Sep 2008 at 8:52am |
|
Agreed; I see little diffrence using DDR3 1333 as opposed to DDR2 800. There is a slight increase with DDR3, but not worth it in most cases.
Here's how to calculate the "best" match of RAM for your system (assuming you want 1:1 linked with the FSB):
Divide the listed FSB (1333 or 1600) by 4
Multiply by two (Dual Channel RAM, hence the double)
So:
FSB 1333 = (1333 / 4 = 333.25 * 2 = 366.5 = DDR2 667)
FSB 1600 = (1600 / 4 = 400 * 2 = DDR2 800)
That being said, can someone run a quick test if you have access to DDR3 RAM. I'm wondering if running a 2:1 FSB ratio (like DDR3 1333 instead of the 1:1 DDR2 667) would be faster than using another ratio with faster DDR3 (such as DDR3 1600). Basically, I want to see, on a 1333 MHz FSB, weather DDR3 1333 is faster than DDR3 1600, due to a better link ratio.
I might run some tests myself if I can get my FSB up to 1600 (im running 1500 now, kinda in between). If so, ill let you know how the diffrent configs match up. I DO know that DDR2 667 will beat DDR2 800 on a FSB of 1333 in most instances (although timings do affect those as well, but i dont want to get into THAT discussion...)
|
|
![]() |
|
Tyler Lowe
Newbie
Joined: 14 May 2008 Online Status: Offline Posts: 0 |
Quote Reply
Posted: 25 Sep 2008 at 11:13am |
|
Just make sure you're not running tests that specifically look for RAM bandwidth. I ran a series of benchmarks months ago for someone on these boards and showed something like a 2% difference in performance between DDR2 667 and DDR2 1000, and nearly all of that difference came between DDR2 667 and DDR2 800. You can cherrypick a game that will benefit from additional RAM bandwidth to "prove" how much better DDR3 is, but give me enough time with the sort of testing methodology I see on some sites and I could "prove" anything.
Also, the Rampage motherboard is not a standard P5 deluxe. If you're looking for a reason the prices are the same, price those two motherboards out.
For a true apples to apples comparison, you should be looking at the two Rampage boards. The price of DDR3 continues to drop weekly so the difference in price directly related to the memory is going to continue to shrink. Given the increased latency of DDR3, however, you will not see real gaibn over the faster DDR2 RAM available until you hit DDR3 1600Mhz. Here again, how much gain is application dependent. Most of the time, we are looking at incrimental gains in *max* framerates in games which do nothing for *minimum* framerates which is what really determines the quality of that experience.
Naturally, if you want an SLI capable system, the comparison is between a 780i and a 790i, and the price difference looks more like the two Rampage boards.
|
|
![]() |
|
Kliebor2
Senior Member
Joined: 22 May 2008 Online Status: Offline Posts: 659 |
Quote Reply
Posted: 25 Sep 2008 at 11:34am |
|
My best friend has a system similar to mine, a 790i, same processor as I have, similar graphics cards (9800GTX+ where I have 9800GTX) and when we play our framerates are virtually identical in Warhammer and Bioshock, the two games we tend to play most.
He gets a little gain on me in bioshock, and I seem to cut that gain down to almost nothing in Warhammer which may be that Warhammer is more network latency/GPU bound perhaps than Bioshock is, not sure, but at best he only beats me by 2 or 3 FPS at best. So real world, I have seen little difference between our PCs... other than his is a MainGear and obviously there fore less worthy of mention than mine. Dave ![]() |
|
|
Digital Storm 950Si - Q9450 Quad Core @ 3 Ghz
Dual PNY OC2 GTX 560Ti 8 Gigabytes DDR2/800 2 Western Digital 500 GB SATAII 7200 RPM HD |
|
![]() |
|
Post Reply
|
| Forum Jump | Forum Permissions ![]() You cannot post new topics in this forum You cannot reply to topics in this forum You cannot delete your posts in this forum You cannot edit your posts in this forum You cannot create polls in this forum You can vote in polls in this forum |