lostamerican13 Posted November 17, 2007 Share Posted November 17, 2007 I'm not to pleased my ram went dead on me but I guess when you actually USE the hardware in your computer, <= Hardcore Gamer, hardware tends to die a bit sooner than those just using a computer to check e-mail and what have you. So I put in an RMA request and waiting on that to go through. My question is, if anyone can answer it, I still have the original packaging for the ram and that's where their housed at for now. Should I send them off in a box while still in the original packaging or wrap them up in anti-statics? Link to comment Share on other sites More sharing options...
DerekT Posted November 17, 2007 Share Posted November 17, 2007 Can you please tell us what you mean by your DRAM went dead? Both sticks do not even boot the system up? Even singly? Link to comment Share on other sites More sharing options...
lostamerican13 Posted November 17, 2007 Author Share Posted November 17, 2007 One stick is definately dead for sure, doesn't even post to the BIOS when used, the other works but shows errors at about....95% through test 5 on memtest. The first stick I tried individually in every slot, no go. The second stick I also tried in every slot but still got errors at 95% or so through test 5 on memtest. So i guess one dead is but the other is not dead..but close enough. Edit: I'm a Tech at CompUSA and used a store unit to verify my findings. The store computer came up with the same errors I had on my computer using the Corsair memory and running memtest. When I used the P*Y memory we have in our store unit on my computer, my computer boots up fine with the ram in any of the slots. Link to comment Share on other sites More sharing options...
DerekT Posted November 17, 2007 Share Posted November 17, 2007 You will have no problem with RMA as Corsair does not fiddle with RMA's. Staff here work business days and hours so look to a response on Monday. Keep in mind that when both sticks error out, this usually is indicative of external issues residing in the system and regarding damage to the passive DRAM modules. You may wish to trouble-shoot your system as well. If the problem was and is external to the DRAM (and this is very likely) then you do not want to repeat the issue with the same results. Good luck Link to comment Share on other sites More sharing options...
lostamerican13 Posted November 17, 2007 Author Share Posted November 17, 2007 Well the only overclocking the ram ever got was the SLi configuration...I believe I have good cooling but maybe I'm wrong... 200mm fan on top, 120mm in the back, 120mm on the side, 2x 120mm in the front...I never mess with the stuff in my case and I have my CPU zalman fan pulling air from the front of my case, over the Corsair memory heatsinks and out the back... Could you go into further detail as to what kind of external source would cause this damage? Lets hypothise that I get an RMA and two new sticks of memory, is there a way to purchase the fan that'd go on top of them? Edit: And I agree with you, I don't want this issue to happen again as I'm planning on buying a second set of the same memory from work today! =D I believe in Corsair as the answer to any user's memory needs and are excellent memory modules that I recommend to anyone looking to get top of the line memory. So please, any suggestions are more than welcome on preventing this issue again. Link to comment Share on other sites More sharing options...
DerekT Posted November 17, 2007 Share Posted November 17, 2007 I have personally found that although some are driving high voltages through the Microns there are more who are finding them damaged with time. The prevailing view "was" 2.4v but that has now been seen as too extreme for long term use and 2.2 - 2.3v is seen as far better. I have an 80mm fan on the 8800GTX and it is pointed directly on the DRAM. Yes it's a bit ghetto but it works very well. Use the touch method. Run Memtest and touch your DRAM after 1 full pass. What does it feel like? You will have your answer. As well, higher voltages, even when the hardware is kept cool can be trouble. There is a condition called "Threshold Voltage" which is the amount of power necessary to turn on and switch the transistors gates. There is a formula for the switching power: Power ~ ½ CV ² Af This formula states that dissipated power is linear with the effective capacitance, activity and frequency. Power increases quadratically with the increase of the DRAM's voltage. There is a characteristic of the condition, that being a spread of the spectrum which allows for electronic transference between strobes. In audio, this is called "Crosstalk" or "Harmonic Resonance". When this characteristic is in play along with an increase of voltage, the threshold voltage since it is necessarily higher moves beyond the strict limits and can create errors. This creation will continue on a system that is not turned off and is left running all the time. This will increase error rates and in time WILL damage DRAM. So, if you are one who uses your system 24/7, then you need to keep your DRAM voltages lower. Or, you need to turn your system off and allow the buildup to dissipate in the inactive state. How long is just a guess and since I am not certain of this and since I have damaged my share of Microns, I now do not go past 2.25V with active cooling. I have had no issues since then. Keep in mind, that if you move from 2 X 1024 to 4 X 1024 of 8500C5, then you will need to overclock to get that speed back. You will need to drop the DRAM to 800Mhz and then overclock to get back to ~1066Mhz and the higher end of upclocking the DRAM will not be found. For example, I have 2 X 1024 of D9GHM 8500C5s running at 1200Mhz with 5-5-5-12 with 2.25V. If I move to 4 X 1024 I will be lucky to make 1066 and I will have to change the ratio as I will need to overclock the DRAM. The onboard memory controller reacts far differently with four DRAM banks filled than it does with two DRAM banks filled. Link to comment Share on other sites More sharing options...
lostamerican13 Posted November 18, 2007 Author Share Posted November 18, 2007 ...Wow...I wasn't expecting such an epic response but, right on! Yea I understand what you're talking about...thing is I wouldn't expect the SLi configuration setting that uses the EPP to overvoltages the modules if that's what happened...I spent a pretty penny so I wouldn't have to overclock myself, Lol. Oh well, lesson learned here is self overclock the memory, don't use profiles. I just really try and stay away from self overclocking as it tends to void warranties in the process... Where can I get one of those active cooling fan units that sit over the dominator memory? Anyone? :confused: (answered) Edit: If this is the case, does that mean my warranty was voided by the SLi configuration having possibly set the voltage higher than 2.1v? RamGuy please tell me this isn't true....... Link to comment Share on other sites More sharing options...
Corsair Employees RAM GUY Posted November 19, 2007 Corsair Employees Share Posted November 19, 2007 Let's get them replaced, please use the On Line RMA Request Form and we will be happy to replace them. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.