Tuesday, October 21st 2008

Corsair Releases Tri-Channel DDR3 Memory for X58 Platform

Corsair has announced the availability of two tri-channel DDR3 memory kits for the upcoming Core i7-supportive X58 platform. Corsair claims to have extensively tested these kits for stable operation on various X58 chipset-based motherboards. The kits are value-oriented, and come under the XMS3 series.

There's a 3GB (3x 1GB, model: TR3X3G1333C9) kit, and a 6 GB (3x 2GB, model: TR3X6G1333C9) kit. Both kits operate at 1333 MHz, with timings of 9-9-9-24. The modules use simplistic aluminum heatspreaders, with no fancy DHX cooling mechanisms. This could be attributed to the fact that they are value oriented, and that they could be operating at low voltages of 1.60V, and therefore, small thermal footprint. The 3 GB kit is priced at US $167, while the 6 GB kit is priced at$ 316.

Source: TechConnect Magazine

22 Commentson Corsair Releases Tri-Channel DDR3 Memory for X58 Platform

#1
Well if all these tri-channel memory kits are being aimed at X58, lets just hope they get the voltage requirements right so you dont fry your nice new CPU just to get your ram upto speed :shadedshu
Posted on
#2
alexp999 said:
Well if all these tri-channel memory kits are being aimed at X58, lets just hope they get the voltage requirements right so you dont fry your nice new CPU just to get your ram upto speed :shadedshu
Like I already mentioned, Corsair tested these on various X58 boards.
Posted on
#3
btarunr said:
Like I already mentioned, Corsair tested these on various X58 boards.
High RAM voltage could be a long term issue for Nehalem and hat's hardly something they can test right now. They had no time...
Posted on
#4
DarkMatter said:
High RAM voltage could be a long term issue for Nehalem and hat's hardly something they can test right now. They had no time...
Nobody can test for a potential long-term issue in such short time. Each of these modules operate at 1.6V, for the know, but they have been tested stable on X58.
Posted on
#5
btarunr said:
Nobody can test for a potential long-term issue in such short time. Each of these modules operate at 1.6V, for the know, but they have been tested stable on X58.
I would say that Intel can and so is they only admit RAM voltages up to 1.65V. If these are indeed 1.6V, that is AFAIK well inside Intel's recommendations and that responds to Alex's concerns much better than your previous post. I'm not going to talk for him, but I was 100% sure his comment were regarding other DDR3 modules that operate at 1.8V and above, which BTW are the mayority.
Posted on
#6
I'm not concerned with his comments on the broader market scene, all I'm concerned with, is that this kit was tested for X58, and my telling him that this kit works with X58.
Posted on
#7
I think you don't understand our point of view. Any DDR3 module can work on X58 motherboards, but those above 1.65V (if intel is to be believed) can severely damage Nehalem CPUs over the time. So for me, any RAM manufacturer saying they tested on various X58 mobos means nothing, taking into account they had no time to test if their modules will damage the CPU over the time. They only test stability, and I have yet to see overvolting causing stability issues (not related to heat increase derived from that overvolting). Usually the more voltage the more stable the system is (as long as you keep it cool), but much more prone to electromigration and severe damage over the time. You get it now?

http://www.anandtech.com/memory/showdoc.aspx?i=3426

They are not sure if it is a problem only related to RAM voltage, but apparently they did witness CPUs dying only after few days of operation with 2.0V Vdimm. So if 1.65V is safe and anything above 2.0V will damage the CPU in days, would you use 1.7-1.8V RAM, because they tested it and worked properly in tests that couldn't last more than a few weeks? IMO concerns about the voltage at which the DDR3 modules work are very well founded.
Posted on
#8
1. I get it.
2. To test if a 1.6v kit damages the CPU over time requires..well, some time.
3. All they can test, is stability, and they do that perhaps with memtest86 for a said amount of time.
4. If a 1.6v kit damages a CPU, blame Intel, because it falls well within its 1.65V threshold.
5. If an overvolted kit damages a CPU, blame yourself, not Corsair or Intel.
Posted on
#9
dumm 316$for rams and with 1333 Posted on #11 Doesn't matter, Corsair is not a small company, I trust them. All of my rams are Corsairs. and I have 12 models from them, included, single, and dual-kit, range from DDR1 -> DDR2. And I didn't have any problems with any of them. There is no need to doubt Corsair. Posted on #12 Wow, didnt mean to start an arguement :o . Was just saying that I hope memory MF's are able to give us some good performing modules at low voltage. only time will tell. Posted on #13 Man my problem was that instead of just telling us they were 1.6V, and acept our concerns with higher VDimm modules, you kept saying they had been tested and worked as if our initial concern about VDimm was not legitimate. It is very possible they could have tested 1.8V modules and worked PERFECTLY for 2 MONTHS, BUT that doesn't mean they would not damage the cpu OVER 3-6 MONTHS. Even you mentioned there's no way for them to test if the modules will damage the CPU in long term, but as long as they volt the modules below 1.65V there shouldn't be any problem. INtel DID test that and decided 1.65V was the max. Look better on my posts, because once you mentioned modules were 1.6V I already said they were good and I only continued replying to the fact that you replied to our VDimm concerns. Posted on #14 DarkMatter said: Man my problem was that instead of just telling us they were 1.6V, and acept our concerns with higher VDimm modules, you kept saying they had been tested and worked as if our initial concern about VDimm was not legitimate. It is very possible they could have tested 1.8V modules and worked PERFECTLY for 2 MONTHS, BUT that doesn't mean they would not damage the cpu OVER 3-6 MONTHS. Even you mentioned there's no way for them to test if the modules will damage the CPU in long term, but as long as they volt the modules below 1.65V there shouldn't be any problem. INtel DID test that and decided 1.65V was the max. Look better on my posts, because once you mentioned modules were 1.6V I already said they were good and I only continued replying to the fact that you replied to our VDimm concerns. This is so obvious that I didn't have to ASK. This module is absolutely under 1.65v. They don't need to test it to see if it will damage the CPU or not, they only need to test stability of the memory, follow Intel's warning and use modules under 1.65v. Posted on #15 Just out of interest what is the JEDEC voltage for DDR3? Posted on #16 kid41212003 said: This is so obvious that I didn't have to ASK. This module is absolutely under 1.65v. They don't need to test it to see if it will damage the CPU or not, they only need to test stability of the memory, follow Intel's warning and use modules under 1.65v. And because Corsair already have many types of DDR3 1333, so there is no reason for Corsair to put Corsair Releases Tri-Channel DDR3 Memory for X58 Platform", If it just the same like others modules. Everyone knows really WELL that Nehalem requires the ram to run under 1.65v to not damage the CPU. Even YOU know this, then why would Corsair not? Or you think they ignore Intel's warning? BS. :slap: http://forums.techpowerup.com/showthread.php?t=72926 A-Data Releases Tri-Channel Memory Kits for Intel X58 Platform A-DATA Technology Co., Ltd., a worldwide leader in high performance memory products, announced today that it is expanding its overclocking DRAM dual channel line with tri-channel kits at 1600 MHz and 1333 MHz. Recent next generation Intel Core i7 platform using the latest DDR3 technologies has proven that i7’s internal controller gives a notable performance increase while using DDR3 tri-channel modules. A-DATA is now ready for production of DDR3-1600+ and DDR-1333+ tri-channel kits. These new modules will be available in 3GB and 6GB configurations and are expected to be available this quarter. These latest announcements will obviously impress our highly demanding enthusiast customers, as well as push A-DATA even farther into the lead as world memory solutions manufacturer. Please find below further information about specifications and features: Specification: * DDR3-1600+ & 1333+ 3GB/6GB tri-channel module kits are tested at 1600MHz/1333MHz * Test in tri-channel configuration at latency settings 8-8-8-24 2T for DDR3-1600+ and 7-7-7-20 2T for DDR3-1333+ at 1.65V-1.75V * SPD (Serial Presence Detect) programmed at JEDEC standard for basic system booting Features * All memory chips are verified using overclocking criteria for better quality * High quality 6 layers PCB (Printed Circuit Board) * Optimized for 64-bit OS (operating system)* * Backward compatible with DDR3-1333, 1066 * Lifetime limited warranty As you can see the title does not grant anything and you should not take anything as granted from an announcement. Or do you think A-Data didn't test their modules?? Ask always, as we did, and we were just expecting the response: "Don't worry, they are 1.6V". EDIT: Oh BTW. Because the tone of your post and because I had no chance to dodge your (:slap:) here you have: :slap::slap::slap:. God knows you deserve them well. Posted on #17 We need to calm down. All of us, and get this thread back on track. A high temperament is always encouraged. It's just that sometimes temperament feeds its evil cousin, temper. Thanks for the valuable posts. Back to topic. :toast: Posted on #18 Also Corsair as a company has not given me (or many others for that matter) any reason to doubt them. There are many ram companies out there that I take with a grain of salt but Corsair has proven themselves to both be a reliable company and one worth working with when there is a problem. alexp999, I didn't see a problem with your comment. Good to see some companies getting ready for the x58 release, even if the prices suck. Posted on #19 Wow, hopefully cooler heads will prevail,…. Look, the way I see it, the system integrator is responsible for knowing the specs / requirements and is therefore responsible for acting accordingly. Its very simple, if it violates the tolerances (spec) don’t use it. Also note that while we would all like to see lower voltage DDR3 1333RAM due to the recommended 1.65v DIMM Vcc upper limit for Intel Core i7, not all DDR3 RAM is predestine for X58 boards. There are a number of LGA775 boards that can probably handle 1.7v and higher just fine. Also, finding DDR3 1333 RAM with 1.5v spec and or at or below 1.65 spec isn’t as hard as some people seem to make it out to be. Such in spec 1.5v DDR3 RAM doesn’t even have to be particularly expensive. You can buy 3GB (3x1GB) DDR3 1333 1.5v for under$100 USD.

I don’t really intend to buy any DDR3 1333 RAM with a spec above 1.5v but I wouldn’t be surprised if some of the RAM with higher operational voltage ran fine at lower voltage. Some of the higher voltage RAM may SPD to 1.5v for what its worth.

Even if that’s not the case it doesn’t matter. This isn’t necessarily an issue of buyer beware its an issue of builder / buyer know your stuff and that always applies,….always!
Posted on
#20
Dissapointing, but i guess it is value orientated. Hopefully they will bring out dominator ones soon with much better latencies!!
Posted on
#21
Darkrealms said:
Also Corsair as a company has not given me (or many others for that matter) any reason to doubt them. There are many ram companies out there that I take with a grain of salt but Corsair has proven themselves to both be a reliable company and one worth working with when there is a problem.
alexp999, I didn't see a problem with your comment.

Good to see some companies getting ready for the x58 release, even if the prices suck.
I neither distrust this RAM, much less Corsair as a company, LOL, just take a look at my specs. My only problem was to know if these modules operated below 1.65v and with the way he responded to our inplicit question about thhat: continously mentioning they were tested RAMs, when that didn't respond to our concerns AT ALL. In previous platforms -> tested RAM, a big OK. In X58 -> tested RAMs + 1.6V OH YEAH; no voltage specified = not for me thank you very much, not yet, not until I know it's safe. We ALL know how they are tested and the testing methods CAN'T ensure that violating Intel's specs would not damage the CPU over the time, basically because they HAD NO TIME.

NOW the market is full of DDR3 with voltages well above Intels specs for X58, including newly released ones (announced as X58 tri-channel versions) that are above specs (A-DATA 1.75V) or just in the threshold (OCZ 1.65V). Being those the precedents I find it legitimate to have concerns about that issue, being Corsair or not. I have had many Corsair RAMs and they are of best quality and performance, but they DID use to have a little bit more voltage than other ones in order to achieve better timings/speeds. This of course, does not mean they would do it now, being that now lower voltage is a must, but doesn't mean they SURELY had to be of lower voltage, when other BIG PLAYERS are releasing modules with voltages higher than the requirements (mentioned above). A-DATA and OCZ surely tested their RAM too, but they are above the specs, AND just as Corsair they had no time to test the possible damage that high voltage could cause over the time. And YET they are selling those modules as tested RAM aimed for X58, just as Corsair here. The only way to ensure that, right now, is by following Intel's recomendations, which Corsair did BTW, and that's the difference, the one me and alex (I think) were hoping to see here, not if they were tested RAMS o not. ALL brands sell tested RAMs, but that doesn't mean those modules are free of troubles. It's neither a matter of trusting one company or not, they just can't reproduce the kind of stress that would demostrate if a high voltage could cause any damage in say 6 months, when they only had 2. Some manufacturers (according to Anand) have seen that 2.0V can destroy the CPU in days though. If that's not any indication for you I don't know what would be. So where's the limit? Anything below 2v is safe? 1.9v? 1.8v? I'd say NO. If 2.0 = days, 1.9v = weeks, 1.8v = months. As of now only 1.65v is safe.

And honestly, I don't care how much you trust one company, if it is to the point to blindly buy their tested RAM that according to another company is above te safe threshold. And obviously I'm not talking about Corsair, but many people trust OCZ and A-DATA and I wouldn't be safe running Nehalem with those TESTED modules from them. That's why I asked (implicitly) about the voltage of Corsair's modules. Period.

\end of rant.
Posted on
#22
How much the world changes in hours. Let's put an end to this comedy. Yes, DM is right, my thinking his comments were with malice was fabricated , but out of pure misunderstanding on my part, for which I apologize. Every time a "kit for X58" comes up, the first question that should come to mind is its module voltage, and not if it is stable/tested. I will draw constructive criticism out of DM's argument, and make sure to back up statements such as "....operate at low voltages", and "tested stable for X58" with "operates at < insert voltage >" in the future.

And since things only got worse after my polite request to calm down, please don't mind removal of some comments.
Posted on