As usual almost no difference with increased bandwidth:
At 2400Mhz same timings:
DDR400 though DDR533 are virtually identical, less than 5% within one another on the largest game, wolf.
It is'nt until you bump processor speeds (which I never liked this testing method for memory) do the numbers begin to change signifigantly with added bandwidth. Due to bandwitdh? nope, due to running 2.72Ghz A64 instead of 2400Mhz.
Anyway I think these sticks are a waste of $ for the most part. Primarly due to the fact you must push signifigant volts though them and fan them and maybe destory them if not well learned. Noise and only DFI can provide this voltage eliminates lots of people. One is much better off IMO with low volts TCCD or some crucial 8T running 2-2-2 ~200Mhz at low volts all day without the noise of additional ram cooling.
VX/REDLINE/UTT is nice and everything, but I don't get why people will buy it since BH and CH based models are much cheaper and perform very similarly with a bit less voltage, and can also run stock at 2.6v-- which makes them incredibly flexible. best ram ever.
The point of that thread is to show that in *real-world* situations, the measurable effect of 2T over 1T seems to be no more than 3%. Also keep in mind that at 2T, you can OC the RAM higher and tighten your other timings, and thus you can OC well with 4 sticks or more than 1G of RAM.
Nice review....I couldn't help but notice that in most of the benchies that mattered, hi-po ddr400 2-2-2 was at the top of the charts. Especially OCZ Plat Rev 2 and Crucial Ballistix....kinda makes me wonder if DFI mobo with high voltage ram is just a big waste of time and money.
Which thread? As far as all AT benches have shown, 2T cripples performance on Athlon 64 testbeds...
Anyway, great review, but I gotta wonder just what we're going to be doing to cool our cases in a few years--active cooling for the RAM, northbridge, CPU, and video card (the latter two of which account for 150+W)... Why do I get the feeling that I'm going to turn on my computer one day, and there'll just be a tiny mushroom cloud where the case used to be?
Any hopes of some 2T command rate testing/overclocking potential-ing?
A thread has shown 2T makes very little difference to performance, but may help overclocking a bit, so mabye we could see if this stuff can be pushed even further?
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
Facebook
Twitter
RSS
41 Comments
View All Comments
jmke - Monday, May 16, 2005 - link
I agree with you completely Zebo; is an increase of 2-5FPS in an already OLD game like Wolfenstein worth $300 (memory + ddr booster/new motherboard)?I don't think so
Zebo - Monday, May 16, 2005 - link
My above comments excludes cyber-athletes participating in cyber olympics. Mainly geared twards average enthusiast at AT.Zebo - Monday, May 16, 2005 - link
As usual almost no difference with increased bandwidth:At 2400Mhz same timings:
DDR400 though DDR533 are virtually identical, less than 5% within one another on the largest game, wolf.
It is'nt until you bump processor speeds (which I never liked this testing method for memory) do the numbers begin to change signifigantly with added bandwidth. Due to bandwitdh? nope, due to running 2.72Ghz A64 instead of 2400Mhz.
Anyway I think these sticks are a waste of $ for the most part. Primarly due to the fact you must push signifigant volts though them and fan them and maybe destory them if not well learned. Noise and only DFI can provide this voltage eliminates lots of people. One is much better off IMO with low volts TCCD or some crucial 8T running 2-2-2 ~200Mhz at low volts all day without the noise of additional ram cooling.
thumbs down to extreme modules.
cryptonomicon - Monday, May 16, 2005 - link
VX/REDLINE/UTT is nice and everything, but I don't get why people will buy it since BH and CH based models are much cheaper and perform very similarly with a bit less voltage, and can also run stock at 2.6v-- which makes them incredibly flexible. best ram ever.bersl2 - Monday, May 16, 2005 - link
#5: http://forums.anandtech.com/messageview.aspx?catid...The point of that thread is to show that in *real-world* situations, the measurable effect of 2T over 1T seems to be no more than 3%. Also keep in mind that at 2T, you can OC the RAM higher and tighten your other timings, and thus you can OC well with 4 sticks or more than 1G of RAM.
flatblastard - Monday, May 16, 2005 - link
Nice review....I couldn't help but notice that in most of the benchies that mattered, hi-po ddr400 2-2-2 was at the top of the charts. Especially OCZ Plat Rev 2 and Crucial Ballistix....kinda makes me wonder if DFI mobo with high voltage ram is just a big waste of time and money.Shinei - Monday, May 16, 2005 - link
Which thread? As far as all AT benches have shown, 2T cripples performance on Athlon 64 testbeds...Anyway, great review, but I gotta wonder just what we're going to be doing to cool our cases in a few years--active cooling for the RAM, northbridge, CPU, and video card (the latter two of which account for 150+W)... Why do I get the feeling that I'm going to turn on my computer one day, and there'll just be a tiny mushroom cloud where the case used to be?
Lonyo - Monday, May 16, 2005 - link
Any hopes of some 2T command rate testing/overclocking potential-ing?A thread has shown 2T makes very little difference to performance, but may help overclocking a bit, so mabye we could see if this stuff can be pushed even further?
JustAnAverageGuy - Monday, May 16, 2005 - link
P1: "OCZ st arted it all with their VX series memory."P2: Redline is available as DDR433 (PC3500) p arts.
For these reasons, Mushkin Redline memory was only tested on the DFI LANP arty nF4 SLI-DR Athlon 64 Socket 939 test bed.
P3: The Mushkin Redline XP4000 was tested with the DFI LANP arty nF4 SLI-DR Athlon 64 Socket 939 motherboard.
We have found the 6800 Ultra to be a p articularly good performance match to NVIDIA motherboards
Don't worry though, all those macros are spelling error free ;)
classy - Monday, May 16, 2005 - link
Great performance, but the price is really high. Especially considering the fact the OCZ 3200 Gold is not far behind and is almost $100 cheaper.