RAM Issues on X58 Boards

I had this issue once awhile back on my Asus P6T Deluxe v1 board with 3x 2GB Corsair XMS3 DIMMs. Reseated and rearranged DIMMs and never saw it again.

Kyle, did the Asus rep ever manage to get over to your place and see your proof of the issue?

Sounds to me like we're dealing with some narrow tolerances and if any DIMM is a little weak, in a slot it doesnt like, your timings are a little off, or voltage is too high/low you can have some frustrating problems.

Here's to hoping that Intel, Mobo mfgs, and RAM mfgs can improve things before we see Socket 2011 with X68 and quad channel RAM.
 
Well, my the system in my sig is doing the same thing.

i7 860 @ 3.8 | EVGA P55 FTW | G.Skill 8GB DDR3-1333 | Win 7 Pro x64 |

I've got it clocked down to stock for now since I was messing with all of the BIOS settings. Windows would report 8GB, 3.99GB usable. The BIOS would just report 4088MB even though there was 8GB installed. .. I install memory in DIMMs 2 and 4 and it boots but when I install DIMMs in slots 1 and 3 it does not boot. I install DIMMs in all slots and it boots fine. I changed the voltage on the RAM to 1.51 to 1.54 and it seems to have fixed it. I flashed the BIOS to A72 because there was some issues that were fixed due to XMP crap...

I have now had to fix this 3 times now. The last time was the most frustrating since it took almost 3 hours to get it to work.

Edit:

I have these GSkill sets
 
Last edited:
heelix said:
Got an Asus Rampage II Gene, i7-950, and 3x4G G.Skill RAM (DDR3-1333, 1.5v), PC&P 750W ps. All 12G are showing up.

Mainboard reports BIOS to be 0501, so will update that shortly. Everything is stock while I do the first round of sanity testing.
Click to expand...

Added another 3x4G of RAM Mushkin (DDR3-1333, 1.5v) to my box and all 24 shows up.

However, buddy of mine just added 3x2G more to his gigabyte board, and sees 10 of 12G, so looks like I'm going to join this party after all. Having him reseat/shuffle the RAM and check the BIOS first.

Gigabyte X58A-UD5, 6x OCZ gold PC3-16000

A reseating/shuffle of the RAM looks like it did the trick.
 
Last edited:
Add another one here. Rage III Formula dropped the C2 DIMM slot one day. After I had added a 6Gb SATA III HDD. Nothing would bring it back, not even pulling the New HDD. It was showing all 24GB of matched Patriot RAM for a week prior. Then poof, only 20
 
Asus Maximus Formula III - open box and brand new from newegg. Only 2GB usable out of 4GB. No problems in single channel, only in dual channel.
 
I got the same issues has you.

Motherboard: Sabertooth X58
CPU: i7 950
Memory: Corsair Vengance 3x4gb (12Gb)

Only 8gb detected, i tried each stick of ram and only one doesn't boot, however, i tried the same stick in another board and it worked.

No success:
Reseating CPU
Changing voltage

My big interrogation is, what's wrong, RAM or the Motherboard?
 
Here we go again. I have got EVGA P55 200 SLI from newegg and it is the same f... problem. Single channel works, dual gets detected (sometimes) but only 2GB are usable. I came to a conclusion that a CPU's memory controller if f... up. Another dud from Intel after I bought Sandy Bridge. I am going to RMA that bloody CPU back to Intel and literally demand a new one. I tested three different boards, three sets of RAM (work great in dual channel in my SB rig). I feel bad about returning another great board to Newegg saying it is faulty.

Also, please refer to this thread.... 30 pages of problems with RAM.

http://vip.asus.com/forum/view.aspx...rd_id=1&model=P6X58D Premium&page=1&count=301
 
MSOlhao said:
I got the same issues has you.

Motherboard: Sabertooth X58
CPU: i7 950
Memory: Corsair Vengance 3x4gb (12Gb)

Only 8gb detected, i tried each stick of ram and only one doesn't boot, however, i tried the same stick in another board and it worked.

No success:
Reseating CPU
Changing voltage

My big interrogation is, what's wrong, RAM or the Motherboard?
Click to expand...

Please try single channel and I assure you it will work great. Dual channel is the problem.
 
I have got EVGA P55 FTW 200 SLI and still the same issues. I called Intel and sent back my brand new i5-7600. I will keep you updated when I get it back next week.

Cheers.
 
I have an ASUS P6T Deluxe V2 with an i7 920 D0. My C0 never did, and has not yet done anything like this in either the P6T Deluxe V2 or my eVGA X58-758A

Went from a 3x2gb OCZ Gold[XMP] kit to a G.Skill F3-12800CL7T-6GBPI. After running for about 2 weeks without a hitch a DIMM dropped.

Reseated G.Skill kit: Failed to report 6gb
Swapped back to OCZ Gold kit: Reported 6gb (ran this for a couple days)
Swapped back to G.Skill kit: Reported 6gb (has been running fine for a month now)
 
I had issues back in Jan 2010 when I built my i7 system. OCZ Gold 1600mhz 8 8 8 on an EVGA 3X SLI (e758). OCZ Was of no help and EVGA kept insisting it was OCZ.. Eventually gave up and bought Mushkin redlines, I had no issues for nearly a year. Upgraded to RIIIE because I needed something better for multiple video cards AND sound card, etc.

The northbridge cooler on that e758 was stupidly loud and the mosfet also loaded at nearly 105c.. Had to ziptie a delta fan to it to get it to stable temps.
 
So update. I upgraded to a 950. I yanked the patriot to put in the wife's new (my old 920) and she booted to 6 GB no prob. I left the GSkill in mine. Now I was reporting only 2GB of the 6. I swapped between the two again, the Patriot showed 6GB in mine and 4 in hers. Sounds like defective RAM to me. I have seen a lot of these memory issues related to GSkill. Poor manufacturing?
 
hello Devil22,
I have tested my motherboard X58A-UD5 rev 1.0 with latest bios F7e and I did not experience this problem.
Here is my system configuration:
Motherboard: X58A-Ud5 rev 1.0
Bios version: F7b and F7e.
CPU: Intel I7 980X
Memory#1: Kingmax DDR3 1333 FSB 2 GB x 6 pcs.
Memory#2: Corsair DDR3 1600 FSB 2 GB x 3 pcs.
OS: Window 7 64 bits.

Test done:
1. Using bios default value, check windows under default.--> No problem. Detect the memory 6GB at 3 pcs. memory in triple channel and 24 GB at 6 pcs. memory in triple channel.
2. OC my memory to 1600 FSB of Corsair.
a. At bios setup/MIT/ set XMP to profile 1.
b. Set performance enhance from Turbo to standard.
--> Check system from window. No problem. 24GB is detected.
Please try to check Gigabyte website of latest beta bios F7e.
http://www.gigabyte.com/products/product-page.aspx?pid=3304#bios
 
K i read more or less every post in this forum thread and only one or two people out their seems to know what their talking about.

and no one that i can see that posted did what they should have to testing their ram and that is

"A" Use one stick of ram and get your self in to the bios.
"B" Disabled auto everything on the ram topic.
"C" Forget SPD manual setup only.
"D" Ddr3 is 1.5v and if you have ram heat sinks give it a little something more like .05% more volts or something just don't over heat the ram.
"E" If your ram is 1666mgz run it slower "example 1333mgz"
"F" Never mind SPD&auto\plug and play anything if your memory is 100% all the same brand and make~version and the memory timings are CL8-8-8-24 then run it at 14-14-14-34
"G" Don't over clock anything in your system when testing memory default settings only and or safe mode settings all the way around for the computer.
"H" long story short manual setup only and run the memory at -50% of rated speed all the way around.
"I" And ya maybe update bios to newest version.Or at lest check out whats new in this part of the world.
----------------------------------
If after doing all of the above it still doesn't work then you have a right to complain because if you haven't then you simple don't know what your doing in my book.
You know their was a time when their was no SPD auto anything and you all ways had to setup the memory up manual stile.

Seems to me too many people out their are dependent on SPD~auto mode and the min that doesn't work anymore they don't know what their doing.
banghead.gif

Have you ever stopped to think that maybe memory company "X" is bull shiting about what memory speed\timeing\voltage their ram can deal with ?
And remember the more ram dimms\sticks you use the more you have to slow the ram down to stabilized the system as far as the memory controller~ram is concerned.

Remember what this is people ? No SPD~auto anything here .
72PinSIMM.gif

30pin.jpg

And don't forgot to run memory diagnostics after everything said and done to make shore everything's ok and check your temps "duh" on your ram if you over volt {under full load}.
 
Last edited:
Wow no one got anything to say now or what lol .Guess everyone testing their junk or something :D
Ok part 2 then:

More info for you all.

So to take a wild guess ............... mobo manufactures are trying to win benchmarks by setting their default memory settings to tight\fast and at the same time memory company's are selling bogus ram with memory timing that are for the most part lies and or are trying to over volt their ram so as to try and hold unrealistic memory timings.

And i keep hearing story's about this topic right left and center about memory problems including but not limited to computer shop owners .
Funny too seeing as i have three computers here all with mismatched ram in em and duel channel memory setups and i just got my self more ram for my main computer and on my way out the door the shop owner was like "it wont work!" .
Funny it douse and i all way setup the ram manual stile seeing as i don't care about auto detect anything .Imho its retarded and doesn't know what its doing half the time.
It's not for nothing i use to call plug and play {plug and pray}
As well one of the computers in question has one of the first chip-sets ever made to support dual channel memory setups.:p

To quote someone from a different forum seeing as i can't put it better then buddy.

The problem with the DDR3 ram and the X58 platform that you are seeing has to do with the Secondary timings or Sub timings as they are called that the boards detect based on the profiles loaded into the ram and the compatibility profiles the boards have listed in the bios.

This is why you see many board manufacturers such as Asus update bios just for a memory compatibility change. The IC's on the ram decide what timings are detected by the board and also the latency for many timings boards don't show.

These IC's also will decide what latency the sticks can handle for the primary timings and the secondary. The problem is the current bidding market with the short manufacturing cycles are causing Manufacturers to build less at a time and order parts more often. This is pushing component prices up in the bidding going on especially for IC's that are performing better than others.

The Ram companies however are trying everything they can to keep prices down do to the economy. This means they are often changing IC's constantly especially for the lower end IC's where price is the overall deciding factor. These IC's are normally your 1333 sticks at CL9. First off find a brand from the QVL that the vendor is testing. This means likely they are getting ram compatibility updates from the manufacturer and the different speeds should not make a difference much in compatibility even if the particular model is not on the supported list.

Second thing is look for DDR3 1600 sticks with CL6 timings if they are 2GB sticks and CL9 for 4GB sticks. Likely these IC's are the better crop since they are able to handle better timings. The last thing when buying 2 kits is look at the S/N. Often the S/n tells you on the sticks if they were made in the same or close production run. This likely means they have the same IC's in both kits and will run together.

The problem most people have is with this last one. If you have 2 sets of DDr3 1600 with 2 different IC's on each set, that often causes the problem you mention. One way to fix it though is disable the XMP profile and manually set all of the primary timings, trc, trfc, and command value.

That will often fix this problem. If not 1 of the sets has to be replaced with one with a similar IC or a bios update has to be made to update the compatibility for the board with those modules. Like Harm has said in the past, if you are going to buy 6, try and find a Hex kit where you know the sticks are from the same production run. That is just a little background and why this problem is so prevalent when you check forums.
Click to expand...

Last but not lest software can only test for memory problems up to a point.
Better then that is to use a dedicated hardware memory tester but only "some" computer shops have this kind of hardware.

----------------------------------------------------------------------
Software based PC Memory Testers.
Memtest86+ 4.20
Memtest 1.0
DocMemory PC RAM Diagnostic
------------------------------------------------------------------------

ramcheck_lx_ddr3_ddr2.jpg
ramcheck_lx_ddr1.jpg


Ps real world speed of ddr3 1666mgz is 200mgz and in some cases single channel ram is faster then triple channel ram setups because it can react faster and theirs less overhead to synchronize the ram with the memory controller.

Meaning a single ddr1sdram {"Type one" Double Data Rate Synchronous dynamic random access memory} 266mgz memory module at Column Address Strobe (CAS) latency {1} can react piles faster then ddr3 1666mgz in triple channel setups using ddr3 1666mgz memory at Column Address Strobe (CAS) latency {10} but has less theoretical max memory bandwidth.
burnout.gif


Further reading
[1] [2] [3]
 
Last edited:
I think we're all familiar with how to test ram, war zone. Also, you seem to be preaching to us as if your method is somehow scientifically based, test the ram at 50% speed for instance, I doubt is based on any real science but more just a random figure you pulled from wherever. Also what's with the pointless pictures (that you are stealing bandwidth on)?
 
Yes, I remember 30 and 72 pin SIMMs. Yes, I know how to set up RAM using the BIOS. No, Your condescension is not helpful.
 
Seems like someone is a legend in his own mind. BTW, I like setting up my RAM manually, so assuming that everyone is dependent on SPD is absurd. And not many people have access to a hardware mem tester. If you do, then great! Pat yourself on the back.
 
I have Asus P6T Deluxe with most recent BIOS. I originally bought this is January 2009 with a OCZ XMP 6GB set. I recently bought another 6GB XMP set from OCZ. I was running previous 1333MHz memory at 1442MHz with 8-8-8-20 timings. New memory is 1600MHz I've tried running at 1333MHz and 1442MHz with 8-8-8-20 timings and sometimes I get 8GB others I get the correct 12GB. Typically when I get 12GB system is unstable and will hard lock after a period of time, could be 1 minute or 4 hours, random really. I recently switched the memory around so that the new 1600MHz stuff is in A channel and the older stuff is in B channel and I've been running almost six days now with no issues. This doesn't mean its resolved as I expect it to happen again. I'm pretty disappointed to be honest.
 
N1GHTRA1N said:
I have Asus P6T Deluxe with most recent BIOS. I originally bought this is January 2009 with a OCZ XMP 6GB set. I recently bought another 6GB XMP set from OCZ. I was running previous 1333MHz memory at 1442MHz with 8-8-8-20 timings. New memory is 1600MHz I've tried running at 1333MHz and 1442MHz with 8-8-8-20 timings and sometimes I get 8GB others I get the correct 12GB. Typically when I get 12GB system is unstable and will hard lock after a period of time, could be 1 minute or 4 hours, random really. I recently switched the memory around so that the new 1600MHz stuff is in A channel and the older stuff is in B channel and I've been running almost six days now with no issues. This doesn't mean its resolved as I expect it to happen again. I'm pretty disappointed to be honest.
Click to expand...

trying to run ram million miles per hour but not overclocking the cpu with its IMC you will get anything but problems
 
Spooony said:
trying to run ram million miles per hour but not overclocking the cpu with its IMC you will get anything but problems
Click to expand...

I'm sorry. Not sure I follow you 100%. I do overclock my CPU. 190 BCLK works pretty well on air with some stability problems I run at 180 instead since it is 100% stable, minus the issue sometimes with RAM amount not showing correctly. I run that standard 20x multiple, 21x works pretty good but again some heat and stability issues. I've been running at 3.6Ghz for about 18 months with no problems.

I just noticed in my bios with my manual memory settings it tells me that only 800 or 1066 memory is supported. I want to confirm I'm actually running my memory at the right speed BIOS shows 1442 MHz.

Here are a couple CPU-Z shots and HWMonitor at idle. Feel free to offer suggestions.

cpu1q.jpg


cpu2u.jpg


tempsn.jpg
 
N1GHTRA1N said:
I'm sorry. Not sure I follow you 100%. I do overclock my CPU. 190 BCLK works pretty well on air with some stability problems I run at 180 instead since it is 100% stable, minus the issue sometimes with RAM amount not showing correctly. I run that standard 20x multiple, 21x works pretty good but again some heat and stability issues. I've been running at 3.6Ghz for about 18 months with no problems.

I just noticed in my bios with my manual memory settings it tells me that only 800 or 1066 memory is supported. I want to confirm I'm actually running my memory at the right speed BIOS shows 1442 MHz.

Here are a couple CPU-Z shots and HWMonitor at idle. Feel free to offer suggestions.

cpu1q.jpg


cpu2u.jpg


tempsn.jpg
Click to expand...

sorry there mate thought you tried the ram first. To be honest OCZ ram I stay away from. Its nothing but problems on the x58. They even label they're stuff wrong. Saw a set labeled I5 and I7 supported 1.65v ram!
I5 tells you to f off with ram rated 1.65v in it lol
 
Spooony said:
sorry there mate thought you tried the ram first. To be honest OCZ ram I stay away from. Its nothing but problems on the x58. They even label they're stuff wrong. Saw a set labeled I5 and I7 supported 1.65v ram!
I5 tells you to f off with ram rated 1.65v in it lol
Click to expand...

Yeah. I may never buy OCZ RAM again. I had a couple bad double channel kits from CORSAIR a few years ago so got OCZ in Jan 09 when I built my i7 920 rig. I haven't had any issues with it other than this x58 issue which seems to effect pretty much everything. I'd probably go with Corsair for my main rig in the future upgrade. I just bought 4GB of G.Skill for a HTPC build, I've used it it other builds without issue.
 
N1GHTRA1N said:
Yeah. I may never buy OCZ RAM again. I had a couple bad double channel kits from CORSAIR a few years ago so got OCZ in Jan 09 when I built my i7 920 rig. I haven't had any issues with it other than this x58 issue which seems to effect pretty much everything. I'd probably go with Corsair for my main rig in the future upgrade. I just bought 4GB of G.Skill for a HTPC build, I've used it it other builds without issue.
Click to expand...

OCZ are responsible for a lot of issues on the x58 platform. Dunno is it due to their labeling or what. I've even see at tomshardware their part numbers being all mixed up and a set of 1.65v part number came out to 1.5v part number.
Corsair is primo stuff just read this lol

http://www.hwbot.org/forum/showthread.php?t=21576
 
Just bought a new ASUS Rampage III Formula that would only recognize 4GB of the 6GB installed (3 x 2GB). On the ASUS drivers disc, there's a button that brings up your mobo info in your browser. Last night it indicated that the middle red slot wasn't recognizing the RAM. It shows DIMMS numbered 0, 1, 2, 3, 4, & 5. Only they don't exactly match how the manual labels the slots. 2 is the first red slot, 4 is the second red, and 0 is the third red. This is from left to right. So only the RAM in first and third slots was being recognized.

So, bad RAM slot on the board? Nope. First I ruled out a possible bad RAM stick by loading up my defaults in BIOS and booting three times, each time with a different single stick in the first red slot. Booted up all three times.

Next, I ruled out the possible bad slot by booting with a single stick in the middle red slot. Again, booted right up with the 2GB stick registered.

Then, instead of booting with a single stick in the third slot (I knew that one was working), I added a second stick in that slot and booted with no issues. 4GB registered.

One slot left. I added my last stick into the first red slot and booted with all three sticks. BAM! All 6GB registered. I then loaded up my OC profile and rebooted a final time. Again, all 6GB recognized.

While I haven't the first clue as to why the RAM issue occurs or if it's specific to X58 boards, I can only come to the conclusion that those affected should simply add one stick at a time and boot after each module is in place. I hope this helps some of you folks out!
 
Interesting. Just replaced an i7 930 with a 990x and 12gb is now only 8 usable. Some interesting notes to this...

Classified 4 way sli (762) has been running for 12 mnths at 4.5ghz (215 bclk) with utterly no issues (and all 12gb's of it's Trident ram). Plucked it and installed the 990x (bios flash/cmos reset too). Fucked around with it no end (including trying known perfect dominator GT's) and couldn't revive 12gb. <Incidently, it was posting as only 8gb - so forget windows!>

Put the 930 back... all is fine. Put the 990x in again, sadness. The 930 is back in again and its fine currently. The 930 has never once in 12 months done this - the 990x simply refuses to see all 12...

The only variables here are cpu, bios and socket. My gut feeling is the socket/mount.

I think an important point to make and keep clear in peoples minds is that there are likeley numerous reasons for the 'missing' ram problem. I have no doubt that many of these are bad ram, poorly seated ram, defective sockets, crook bios settings etc. It's a fools errand I think to persue just one cause as the basis for this problem overall.

I also want to add that I overtighten sockets. I can tell you that this 930 has been more than firmly pressed into that EVGA board for some time. (the socket looks perfect, I wouldn't bet that its not deformed in some way though) I just gotta wonder how the statistics on this problem would decrease if everyone used 'clip on' Intel stock coolers..!

cheers
 
I have found that using the QVL list leads to few if any problems. At least with ASUS. My last 5 motherboards were all ASUS and all had RAM from the QVL and I have never had a memory problem that wasnt the RAM stick being bad.
 
Necro-bump!

I've decided that I'm finally fed up with my EVGA X58 LE board and its RAM issues. Thing is, NO ONE in this thread or from what I can find on Google has the problem I do.

When I had 6GB of RAM (3x2GB), my BIOS and Windows would only detect 4GB after a cold boot. Shut down completely and start back up, all 6GB are detected in the BIOS and Windows.

Upgraded to 12GB of the exact same kit and the same thing happened, but this time only 8GB are detected after the cold boot. Shutting it down and starting back up and all 12GB are available.

This issue is 100% repeatable when I start the machine when its been OFF for awhile. I've actually gotten in the habit of booting it up and going right into the BIOS, then shutting it down and then starting it back up again. Retarded.

Been through 3 BIOS updates and this "issue" persists.

Weird.
 
criccio said:
Necro-bump!

"""I've decided that I'm finally fed up with my EVGA X58 LE board and its RAM issues. Thing is, NO ONE in this thread or from what I can find on Google has the problem I do.

When I had 6GB of RAM (3x2GB), my BIOS and Windows would only detect 4GB after a cold boot. Shut down completely and start back up, all 6GB are detected in the BIOS and Windows.

Upgraded to 12GB of the exact same kit and the same thing happened, but this time only 8GB are detected after the cold boot. Shutting it down and starting back up and all 12GB are available.

This issue is 100% repeatable when I start the machine when its been OFF for awhile. I've actually gotten in the habit of booting it up and going right into the BIOS, then shutting it down and then starting it back up again. Retarded.

Been through 3 BIOS updates and this "issue" persists.

Weird.
Click to expand...

I feel your pain now -- im running :
*x58 usb3 rev2 MOBO, 6gb corsair dominators (@2000mhz,3x2gig DIMMS) in tri-channel with the i7 core at 3.06ghz.

after a while my PC started falling into the DRAM LOST LOOP , which means it restarts after detecting a drop in either the voltage (clocking error?) and fails to link the *third* ram slot meaning a loss of the entire DIMM..
the PC was running fine, i did not go fiddle with the OCing again after i got it to where i wanted it, and if it does go wonky i have dualbios backup for ressetting to safe workable settings.

what i have done so far -- swapped DIMMS in and out to check for physical damage on MOBO or DIMMS.testing each dimm independantly, and then in seets of 2, and then in 3, in different positions.
the memtest is fine for ach dimm in ech slot, cept for EVERY dimm placed in channel C or channel 3. always 4gigram. i have adjusted the OC in a variety of ways following threads from Tomshardware, but to no avail.
I get my clocks/cycles/speeds almost identical on each dimm (actually 100% identical after an emotional war) also no effect.
i get the PC to boot fully into windows every 4th or so attempt by randomly hoppnig and chopping and changing to a variety of saved BIOS configs, even after flashing CMOS the problem persists. after about 1 hour everytime the machine freezes/ reboots

I have formatted the machine, cleaned the DIMMS, cleaned the MOBO, thermal repasting, tride the ram on diff machines..... etc.

im new to Hardforum, sum1 save me plz. alexknox (MCSE,MCSD etc etc)

can post image of OC stats and voltage tables...
 
i am aware that the trichannel sequence follows as such -- > CHANNEL A has slot 2 first then slot 1, CHANNEL B has slot 4 then 3, and CHANNEL C has slot 6 then 5 :
hence, my 3 x2gig DIMMS will be in slots 1 - 3 - 5 as per the x58-usb3 MOBO. My drive is a 1tb seagate internal.

it does pick up 6gb of ram, but only 4 usable, ive adjusted the voltage to 1.65 to compensate (after checking that without change from my side. Gigabyte boards.. useless..

has anyone come up with ANY sort of solution ?
 
Hey looks like this super old thread got bumped last Sunday. Thought I'd sign up to share the issues I'm having.

I upgraded my system years back with an ASUS Rampage II Gene. Because it was so long ago I don't quite remember if my ram worked correctly with this system. I think I ran it at some point both with a 2 DIMM and 1 DIMM config, but I can't for the life of me remember if the problem I'm seeing now was present all those years back.

I've been away at uni for 4 years and before going I bought a laptop - so the PC I had never really got any use.

Then recently I got my first job and it would be useful for me to have my own PC at work. We're in an interesting time at the moment where dual channel setups are "affordable" and quad channel setups from Intel are all 2011-3 socket and right now are super expensive. Additionally DDR4 is here and again, it's new and expensive.

AMD are releasing Zen in early 2017 (hopefully) and so it makes no sense for me financially to buy new hardware now. I'm just waiting to start this job and my cash flow is zero.

However, this does mean that I managed to find 6x DDR3 4GB sticks, rated 1866 MHZ for less than £100!

ANYWAY: Getting to the main part of this saga... I put all 6 DIMMS in and the system "had several attempts at booting before the right beep sounds were made". - Not very descriptive but you know what I mean.

I found that it was showing 16 GB out of 24 GB. So I pulled all the RAM out and tried one stick per "primary" slot of the triple channel slots. What I mean here is that I put only 1 DIMM in at a time, and tried it individually in each of the 3 "main" slots where you put memory for a triple channel setup... Does that makes sense?

The result was for the "primary slot" closest to the CPU (not the absolute closest slot but the next one along - the primary one for the closest pair) the system wouldn't boot. Tried "MEM-OK"'ing it - nothing.

I vaguely remember experiencing some problems like this years ago, but I think I put it down to not having fully populated all slots, because I knew these systems will only work with RAM in certain slot configurations.

Anyway I'm convinced that unlike everyone else here, I have a system with a dead slot... It seems like most people here can at least get their slots to work in a single DIMM config, whereas I cannot...

Any ideas? It's now 2016... Can I RMA this board to ASUS? I'm assuming not, as it must be about 6 years old now...
 
hyper-nova said:
Hey looks like this super old thread got bumped last Sunday. Thought I'd sign up to share the issues I'm having.

I upgraded my system years back with an ASUS Rampage II Gene. Because it was so long ago I don't quite remember if my ram worked correctly with this system. I think I ran it at some point both with a 2 DIMM and 1 DIMM config, but I can't for the life of me remember if the problem I'm seeing now was present all those years back.

I've been away at uni for 4 years and before going I bought a laptop - so the PC I had never really got any use.

Then recently I got my first job and it would be useful for me to have my own PC at work. We're in an interesting time at the moment where dual channel setups are "affordable" and quad channel setups from Intel are all 2011-3 socket and right now are super expensive. Additionally DDR4 is here and again, it's new and expensive.

AMD are releasing Zen in early 2017 (hopefully) and so it makes no sense for me financially to buy new hardware now. I'm just waiting to start this job and my cash flow is zero.

However, this does mean that I managed to find 6x DDR3 4GB sticks, rated 1866 MHZ for less than £100!

ANYWAY: Getting to the main part of this saga... I put all 6 DIMMS in and the system "had several attempts at booting before the right beep sounds were made". - Not very descriptive but you know what I mean.

I found that it was showing 16 GB out of 24 GB. So I pulled all the RAM out and tried one stick per "primary" slot of the triple channel slots. What I mean here is that I put only 1 DIMM in at a time, and tried it individually in each of the 3 "main" slots where you put memory for a triple channel setup... Does that makes sense?

The result was for the "primary slot" closest to the CPU (not the absolute closest slot but the next one along - the primary one for the closest pair) the system wouldn't boot. Tried "MEM-OK"'ing it - nothing.

I vaguely remember experiencing some problems like this years ago, but I think I put it down to not having fully populated all slots, because I knew these systems will only work with RAM in certain slot configurations.

Anyway I'm convinced that unlike everyone else here, I have a system with a dead slot... It seems like most people here can at least get their slots to work in a single DIMM config, whereas I cannot...

Any ideas? It's now 2016... Can I RMA this board to ASUS? I'm assuming not, as it must be about 6 years old now...
Click to expand...


Same here. I'm reviving this old system I have and just faced the same RAM counting problem. Thing is, I still remember how to fix it. At least for me in this Gigagbyte GA-EX58-UD3R. I do use 3 XMS3 Corsair 2GB sticks (totals 6GB) in this triple channel X58 chipset and that's where the problem lies. Let's get some things straight:

1) This issue is a BIOS problem related to the triple channel setup. This has been addressed and corrected for many X58 mobos. So please, update your BIOS first and try again.

2) In the early days there was a solution that worked most of the time. Mostly, when hardware changes were made, or when trying to OC, that's when the problem arise. I read this whole thread and didn't find the obvious fix: RESET YOUR BIOS. After that, enter the bios and load defaults. Shut down (not reset) your PC and turn it back on again.

3) Another solution involves setting the BIOS option to verify DMI pool data. It slows down startup but might help recognizing correctly your RAM.

4) If nothing else works. Try setting RAM timings and voltage manually in the BIOS.

If nothing of this helps, you might be facing a hardware problem. Nevertheless, I bet 2/3 of the problems reported over here could have been solved with a simple BIOS reset, shutdown, load defaults, shutdown.

My two cents...
 
86 5.0L said:
So many noobies necro posting =\
Click to expand...

Sorry for bringing this thread back to life again. I might be a noob over here but I surely know most of the online forums TOS allow this for a reason. I read the rules and they say it's ok as long it continues the discussion at hand and it doesn't go off topic =/

I could start a brand new thread about this same issue but I think It makes sense to continue this one.
 
It's been a long time but if I recall correctly, one of the possible cause of the X58 memory issues is that the 1366 socket latch wasn't strong enough and the pressure is distributed unevenly which in turn caused the CPU to "lose memory slots/channels" due to bad contact. Intel remedied the issue by using the dual latch mechanism on socket 2011 and forth. If possible, you may want to reinstall the processor and make sure the contacts between the pins and pads are all good.
 
MarcoFSG said:
Sorry for bringing this thread back to life again. I might be a noob over here but I surely know most of the online forums TOS allow this for a reason. I read the rules and they say it's ok as long it continues the discussion at hand and it doesn't go off topic =/

I could start a brand new thread about this same issue but I think It makes sense to continue this one.
Click to expand...

No worries, it just threw me off haha

Anyway, I know the memory controller in these chips are very picky. They're only rated for 3 dimms @ default specs of 1066(IIRC) 6 dimms @ faster speeds needs a higher voltage for the controller, and over time it will burn it out
 
You must log in or register to reply here.
Back
Top