Upgrading time

Hi,

I am very curious to see what these new Platinum Dominators are all about but not sure it's worth upgrading. Right now my system is an Asus Z68 Deluxe 2600k rig with Corsair cas 7-8-7-20 xms3 modules (8gb=4x2gb). Like to go 2133mhz but not sure my CPU is up for it with just a Zalman CNPS9900 Max cooler. Also not sure if 1.5v kit or 1.65v kit is way to go? And why do the single 8gb kit (2x4gb) have tighter timings at 9-10-11-27 vs the 16gb kit having 9-10-11-30 @ 4x4gb? Can't I just buy two separate 8gb kits to have tighter timings?

Thanks.

: Also you guys recommend 2x8gb or 4x4gb considering stress on imc?

I can vouch for your CPU cooler! I use it exclusively on all my machines. And that RAM I can vouch for as well. :):

Definately the 1.50v kit. I've done extensive testing with those kits, & can't praise them enough. I actually was able to get them to run at 2200 with the exact same timings @ 1.507v's verified by DMM. Even mixed two matched kits which is really a no-no & not recommended, but it did work without a problem. My results may not be typical, just so it is clear on that.

It all depends on your CPU's IMC really. As far as 16GB 4X4 or 2X8, I like the 2X8 for what you seem to be aware already of, it is often more likely to have less IMC issues.

If it is allowed to link this, here is a user review I wrote;

http://www.hardwaregurus.net/index.php?topic=152127.msg1126548#msg1126548

Hi,

Thanks for replying. The kit I have right now is actually two kits of the same ram. I just wonder if Corsair laxed the timings slightly on the 4x4gb kit vs the 2x4gb kit and wonder if they are the same ram? Because I'd like to double my pleasure if it would be slightly tighter when buying two of the same kits?

Thanks.

Hmmmm.......possibly the same chips, but not sure. The IC's could be checked. If they are the same, since a 4 stick kit, they probably programmed them to be optimized.

As mentioned, the mixing of kits is not really recommended or guaranteed as to the results. I will stick my neck out & say I have mixed 4 of the 2X4GB kits without a problem in a couple X79's, but here again, results of this practice are never guaranteed. It is a testimony of the how good & consistant this RAM really is though. :):

Well this isn't really mixing kits but doubling them. As I understand duel channel architecture is suppose to be able to handle it as long as each channel is the same. Any Corsair guys care to give me the green light?

Thanks

Well this isn't really mixing kits but doubling them. As I understand duel channel architecture is suppose to be able to handle it as long as each channel is the same. Any Corsair guys care to give me the green light?

Corsairs official stance on this is that it is still considered mixing/combining memory and can not guarantee correct operation or rated speeds.

There is no way to tell how any two random kits will behave with each other even if they are the same exact part number, version number and so on. And why they take the time to test and match kits.

You are more than welcome to try what you want, but if you run into any issues (which does happen a lot) , you do so at your own risk. And you would have no recourse except to RMA a kit in hopes you get another that plays well with the other. Then that puts you right back where you started as far as mixing memory is concerned with no guarantees of correct operation. With two unmatched kits,DDR3 1333mhz would be suggested.

For complete compatibility and guaranteed rated speeds all your memory must come from one single tested package of the capacity you want. IE One kit per MB.

If you want to wait for RamGuy to respond to this you are more than welcome too, but the answer will be the same.

Just to add to this dual channel is a function of the MB and not the memory. So if you used a quad channel kit with your Mb it would still run in dual channel with your MB. Or if you used just one module it would run in single channel.

Don't get me wrong, people do this all the time. It was a practice that started along time ago when you could throw just about anything in your machine and it would just work. But when they moved the memory controller from the MB to the CPU the memory controllers became pretty sensitive to unmatched memory. One slight difference between the two kits can cause throw the whole system off.

Thanks for your words of wisdom. Guess ill play it safe. 3 points in last setting won't make a big difference anyhow. But the ram I have right now is actually two of same kits doubled and only bsod on me once out of 500 sessions. Goes to show the quality of Corsair modules. I swear they're my favorite!

Thanks

Thanks for your words of wisdom. Guess ill play it safe. 3 points in last setting won't make a big difference anyhow. But the ram I have right now is actually two of same kits doubled and only bsod on me once out of 500 sessions. Goes to show the quality of Corsair modules. I swear they're my favorite!

 

Thanks

You are welcome, & hope your questions were answered satisfactorily. What was the BSOD btw? Stop: error code 050, 124, 101?

Hi,

I don't remember which bsod I had. Could have been 124 or 101. Like I said was very random and single occurrence out of 100s of windows sessions and not overclocked at all. Anyhow I ordered the 16gb 4x4gb matched pair 2133mhz kit. Didn't want to take the chance mixing sets. Wonder if there will be any performance going from cas 7 1600...

Thanks.

With a BSOD of 124, it is possible the BIOS is setting too low a CPU V. once in awhile when under a certain condition, stress, or load.

To keep the BSOD screen up to read it, uncheck 'Automatically Restart'.

http://i1249.photobucket.com/albums/hh504/ClipperSnipper/AutoStart.jpg

Wonder if there will be any performance going from cas 7 1600...

Recommend to do some benchmarks at the current settings & then compare the results with the benchmarks of the new RAM.

Hi,

I got my new modules. They say 9-11-10-30 on the sticker and CPU-Z shows the XMP 2133 profile as 9-12-11-31. Slightly off. No big deal but its not accurate?

Thanks.

PS: Its the 16gb 4x4gb 1.5 volt kit.

http://i228.photobucket.com/albums/ee28/davidm671/CPTimings.jpg

When I first looked at your post and thought maybe you were getting tested and SPD latencies mixed.

But even SPD or 1333mhz is off too. Advertized is 9-9-9-24, and CPUz is reading 9-9-9-25...thats weird!

http://www.corsair.com/en/dominator-platinum-with-corsair-link-connector-1-5v-16-gb-dual-channel-ddr3-memory-kit-cmd16gx3m4a2133c9.html

I wonder which is wrong? Was it just a misprint on the website, misprint on the label, Or if there was a revision that didn't get changed on the website. Either way, it's off!

When I first looked at your post and thought maybe you were getting tested and SPD latencies mixed.

 

But even SPD or 1333mhz is off too. Advertized is 9-9-9-24, and CPUz is reading 9-9-9-25...thats weird!

 

http://www.corsair.com/en/dominator-platinum-with-corsair-link-connector-1-5v-16-gb-dual-channel-ddr3-memory-kit-cmd16gx3m4a2133c9.html

I wonder which is wrong? Was it just a misprint on the website, misprint on the label, Or if there was a revision that didn't get changed on the website. Either way, it's off!

I think CPU-Z got it wrong. Will have to try a newer version. Aida64 gave the correct timings. Since then I've set the XMP profile and so far so good. Thanks.

I think CPU-Z got it wrong.

I tend to believe that as well. The reason is when I go into my UEFI/BIOS, my Memory-Z reads the profile correctly for the kit. I have never seen that CPU-Z anomoly either until just recently. It also does that odd reporting for another kit, but I can't remember which one. I should keep better notes, lol.