• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Question on memory clockspeeds

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

Dravenspur

Member
Joined
Dec 13, 2015
I had a question about memory and CPU's. I bought the ASUS Maximus VIII Gene for my new motherboard, which says it supports DDR4 memory at 2133 up to 3600 (OC). I'm looking at the Intel i5 6600K and it says it supports DDR4 memory at 2133 or 1866 directly from Intel's site. So does this mean I can't get memory at, say, DDR4 3000 or higher even though my motherboard supports it? I haven't built a computer in a long time and I'm confused. Or is it not worth paying more to go higher than 2133 because my processor would throttle down the memory to 2133 even if it was higher than that?
 

Mr.Scott

Beamed Me Up!
Joined
Jun 9, 2013
Quick answer, Intel will not tell you to overclock. ;)
You buy the fastest you can afford and the board will support.
The processor will run it just fine.
 

trents

Senior Member
Joined
Dec 27, 2008
Most people are using DDR4 3000 or faster because the performance hit inherent (at this point in time, at least) in the bigger latencies of the DDR4 platform are offset by the higher frequency.
 
OP
D

Dravenspur

Member
Joined
Dec 13, 2015
Most people are using DDR4 3000 or faster because the performance hit inherent (at this point in time, at least) in the bigger latencies of the DDR4 platform are offset by the higher frequency.

But the 3000 is fine for my setup? Will it be throttled by the processor? That's my question...what matters more, the fact that the motherboard will take speeds up to 3600 or that the CPU says it only deals with DDR4 at 2133? Will my 3000 speed memory actually be 3000 speed?
 

wingman99

Member
Joined
Dec 10, 2003
You will have to overclock the memory to 3200MHz with the motherboard.

Product and Performance Information
1Warning: Altering PC clock or memory frequency and/or voltage may (i) reduce system stability and use life of the system, memory and processor; (ii) cause the processor and other system components to fail; (iii) cause reductions in system performance; (iv) cause additional heat or other damage; and (v) affect system data integrity. Intel assumes no responsibility that the memory, included if used with altered clock frequencies and/or voltages, will be fit for any particular purpose. Check with memory manufacturer for warranty and additional details. For more information, visit http://www.intel.com/content/www/us/en/gaming/overclocking-intel-processors.html. http://www.intel.com/content/www/us/en/gaming/extreme-memory-profile-xmp.html
 
Last edited:

wingman99

Member
Joined
Dec 10, 2003
Well I contacted Intel Tech/RMA in California last week and she said the number one problem their having with defective CPU's is memory overclocking she gets 4 calls at least a day with that damage, she told me it could take weeks or six months.

From what I see when overclocking memory like 3200Mhz with XMP on a ASUS and Gigabyte board raises the CPU Vccio and Vccsa a lot. XMP profile, Vccio original 0.960v changed to 1.165v
Vccsa (System Agent) original 1.060v changed to 1.275v
 
Last edited:

EarthDog

Gulper Nozzle Co-Owner
Joined
Dec 15, 2008
Location
Buckeyes!
.2V isn't remotely needed...it is typical for changes like this to occur. I highly doubt that will kill anything though...and it can also be lowered back to stock in most cases(3200Mhz on down).

The board I am using, Maximus VIII Extreme, is .97/1.07 then with XMP is 1.16 and 1.14 for SA and IO respectively. Same CPU since very shortly after launch day running 3000-3200MHz and those voltages.
 

wingman99

Member
Joined
Dec 10, 2003
Well I'm just the messenger from talking to Inlet about increasing memory speed, It is a big problem she said for Intel causing defective CPU's, then she went on to explain why increasing the memory speed increases memory controller speed and they do fail a lot.

EathDog I'm sure you will do fine you never have any problems since 2008 here.
 

EarthDog

Gulper Nozzle Co-Owner
Joined
Dec 15, 2008
Location
Buckeyes!
If it was as big a problem as you are making it out to be, one would think Intel to mention it to the AIBs causing it with the increased voltage.

I was simply giving an example of a working chip twice as long as your cue card reading CSR at Intel said it would die. But, I have been pretty lucky throughout. ;)
 

wingman99

Member
Joined
Dec 10, 2003
Well Inlet knows how to void warranties very well if you don't purchase the Intel tuning plan.;) Give Intel a call and see what they have to say don't believe me.
 

LutaWicasa

, Immutable, Administrator
Joined
Dec 22, 2000
Location
Huntsville, AL
Certainly not to defend Intel (or most any other mega-corp) but....it is we who void warranties by by running hardware beyond their stated specifications ;)
This used to be understood and mostly accepted by the ocing community.
I automatically assume the warranty is basically void on most hardware I buy and play with :D
 

trents

Senior Member
Joined
Dec 27, 2008
It seems to reasonable to me that going for a memory frequency that is somewhat higher than the rated IMC frequency of 2133 is probably safe but to go a lot higher is asking for trouble. The IMC is probably the most fragile component of the CPU. Besides we all know that the performance gain of running high speed RAM doesn't amount to a hill of beans. So what's the point?
 
Last edited:

EarthDog

Gulper Nozzle Co-Owner
Joined
Dec 15, 2008
Location
Buckeyes!
XMP isn't overclocking the memory. If it says it's supposed to run at X speed with Y timings, that's stock. We could all really care less that it, like every other IC, was cut from 2133/2400 pedigree. It's how the binning process works in memory. But it's not overclocking the ram sticks. It says that speed right on the packaging.

I just dont buy that what all boards do, increase voltage with xmp profile, is a problem that will rear it's head often, particularly within 6 months as the intel CSR said. Asrock does this too...so does MSI. Again, if it was really as big a problem as she said it was... we would have heard about skylake cpus dieing left and right... think about it.

Sorry, I simply don't consider a queue card reading tier 1 CSR a reliable source...particularly never seeing a problem like this across a couple of forums.
 

LutaWicasa

, Immutable, Administrator
Joined
Dec 22, 2000
Location
Huntsville, AL
Of course it's not a rampant problem. What it is ...another way to squeeze a few more bucks per purchase, a guarantor of a single RMA and a clear identification of ocers to further curb future RMA's ;)