Corsair Releases Tri-Channel Memory Kits, Including DDR3-1866

Rob Williams

Editor-in-Chief
Staff member
Moderator
From our front-page news:
Yesterday, I posted about OCZ becoming the first memory vendor out the door with a tri-channel kit for Intel's upcoming X58 platform, and it didn't take too long for another to follow. Corsair today announces three new kits also, with two variants each (3GB & 6GB). One thing the Corsair release includes that the OCZ one lacked though, is pricing information.

The lowest kit that Corsair will offer is DDR3-1333MHz (CL9), and will retail for $120 (3GB) and $230 (6GB). Moving up we have what will likely become the most popular kit among enthusiasts, the DDR3-1600 (CL8) will retail for $175 (3GB) and $300 (6GB). Lastly, Corsair wants to make sure they're among the few to offer insanely fast kits, and the DDR3-1866 (CL9) kit deserves that title. At $250 (3GB) and $475 (6GB), it's not going to be for everyone.

Among the new kits, the top two include revised Dominator heatspreaders, while the top-end kit includes a revised flow fan, which appears slightly wider and replaces three small fans with two larger ones. This is to increase airflow while reducing noise... two good things. Corsair's DDR3-1333 part is available now (which means it should creep up in e-tailers soon), while the others will likely follow in the coming weeks.

corsair_tri_channel_dominator_102108.jpg

Each memory speed will feature kits of three 1 Gigabyte or 2 Gigabyte modules for a total of 3 Gigabyte or 6 Gigabyte kits. The 1600MHz and 1866MHz DDR3 memory kits will also feature the industry’s highest performance memory technology, Corsair’s patented Dual-Path Heat Exchange (DHX), for maximum performance and reliability. With six total products available, Corsair is launching the most comprehensive product line available for the Intel Core i7 processor.

 

THUMPer

Coastermaker
The granularity is 64MB, in other words, the smallest size on one interleave would be 192MB, although it's unlikely that someone would end up with such a configuration, as the smallest DDR3 modules are 512MB. Again, this is good news for people planning to upgrade, as it's possible to run with unevenly configured memory, but according to Intel you get the best performance if you keep the same amount of memory in the same channel, although this doesn't mean the same size modules in each channel.

For example, if you own a 2x2GB DDR3 kit today, you can complement it with a 2x1GB kit and split it into 2GB per channel for the best overall performance. This means that triple-channel memory kits might not be as popular as first expected and it could possibly upset a few memory manufacturers that hoped to sell a lot of triple-channel DDR3 memory kits.

http://www.fudzilla.com/index.php?option=com_content&task=view&id=10039&Itemid=1
 

Rob Williams

Editor-in-Chief
Staff member
Moderator
For example, if you own a 2x2GB DDR3 kit today, you can complement it with a 2x1GB kit and split it into 2GB per channel for the best overall performance.

http://www.fudzilla.com/index.php?option=com_content&task=view&id=10039&Itemid=1

That's the first I've heard of that, but if I have time, I'll test it out. When a board has six DIMM slots, each set of two are in their own channel, so that could make sense.

I still think, if you own a 2x2GB right now, the smart thing would be to just pick up another single stick if possible. There's also the problem of whether or not that RAM will work in the new motherboards, because Nehalem is going to be fussy with SPDs and voltages. We'll see I guess.
 

THUMPer

Coastermaker
I keep hearing that 1.65v on the cpu is the limit. But the memory voltage should be able to go up there. For instance, my DDR3 kit is 1.9v.
 

Kougar

Techgage Staff
Staff member
I keep hearing that 1.65v on the cpu is the limit. But the memory voltage should be able to go up there. For instance, my DDR3 kit is 1.9v.

1.65v is referring to the memory voltage, not the CPU Vcore. Although 1.65vCore is probably also extremely high for a 45nm chip ;)

Best article I've found on the issue, I'll expect they will have more devoted to the topic at some point: http://www.anandtech.com/memory/showdoc.aspx?i=3426
 

Rob Williams

Editor-in-Chief
Staff member
Moderator
I am not sure just how severe the possibility of disaster is if you go above 1.65v, but even the thought of killing a CPU with overvoltage is keeping me away from it, for now. It's such a low limit, it sucks. Most of the best overclocks floating around have involved using a much higher than 1.65v VDIMM. I guess I'm just not that ballsy.
 

THUMPer

Coastermaker
1.65v is referring to the memory voltage, not the CPU Vcore. Although 1.65vCore is probably also extremely high for a 45nm chip ;)

Best article I've found on the issue, I'll expect they will have more devoted to the topic at some point: http://www.anandtech.com/memory/showdoc.aspx?i=3426

Well it WAS both...

"It’s not all rosy with Nehalem. We've learned that in spite of its 4GHz+ overclocking potential on air, that there are some limitations. You simply cannot get more than 1.65V to the chip which might be a limiting factor for high-end overclocking."

Its gets better, as the CPU and memory voltage have to be synchronous which implies that not all DDR3 memory will run with Nehalem. At this time you can forget all DDR3 2000 modules with voltages over 1.65V, as they simply won’t work on current Nehalem motherboards.
http://www.fudzilla.com/index.php?option=com_content&task=view&id=9072&Itemid=35

LMAO
THEN this
"Core i7 has no Voltage limitations "
http://www.fudzilla.com/index.php?option=com_content&task=view&id=10037&Itemid=35

Now I know FUD. It is what it is. But he has been right a few times. I like the anand article too.
 

Kougar

Techgage Staff
Staff member
The only reason FUD is right is because they will publish three articles saying yes, no, and maybe so. :D

Vcore is going to be board dependand, but I've seen screenies with as much as 1.7-1.8v piped into Core i7 chips...
 

Rob Williams

Editor-in-Chief
Staff member
Moderator
Vcore is going to be board dependand, but I've seen screenies with as much as 1.7-1.8v piped into Core i7 chips...

That seems rather insane to me. That would be high for current-gen chips... Core i7 requires less voltage overall and seems to become much hotter with more voltage than current chips. I could not imagine going above 1.55V on Core i7, much less 1.7V+.
 
Top