CMOS button batteries

[Clunkclick]

4 hours ago, StevesTruck said:

RTFM and make sure your memory is installed correctly (check the pairs don't need to be in 1+3, 2+4 rather than 1+2, 3+4).

 

Download and boot off a Linux live boot dvd/memory stick (hirens is old, but good for this), when the boot loader comes up, you'll get an option to do a memory check. 

Memory is currently  installed 1+2 (Single-sided DIMMS), 3+4(Double-sided DIMMS) but is reported in Windows as 1+3, 2 +4.

 

After further starts-up I did this morning, its pointing towards a memory issue. I had previously noticed a problem in Windows, where certain actions required, unusually, a double-click in order to get them started as opposed to the normal single click and this seems to be mirrored in start-up (Hitting the warm-boot, after the boot has stalled, starts the system first-time) - is there a known problem in the design of the memory controller interface on the high-density single-sided DIMMS, perhaps in the bank switching circuitry ?

 

N

Edited 16 May, 2017 by Clunkclick

[RainbowFire]

10 minutes ago, Clunkclick said:

*******s !

If consumer electronics, especially home computing, is so reliable and infallible, why are their pages and pages of reports on the internet on people not being able to get systems to start, systems failing-over etc - a good 50% of which don't get resolved. Is it possible, that white male middle class professionals aren't the oracles they make out ?

 

N

 

By, now, in your advancing years, I would have thought you should have realised that, in consumer electronics, as with many other things, you get what you pay for. You paid ASUS money, and got an ASUS product........not sure why you have a problem with that really.

 

You'll find that may of those people that "have an issue" have tried to "adjust" or "upgrade" their kit on their own. So many people think they know better than a manufacturer, whether it's Skoda, Microsoft, ASUS, VW etc. For cars this has resulted in people thinking a £100 spoiler from Halfords will make their 1.0 Corsa stick to the road better. For computers there's a whole raft of people who think we still live in the '70s and that computers are for hobbyists and that, armed with no more than a soldering iron and a couple of uncalibrated meters, they'll be able to diagnose and fix anything, along with all those people who believe that there is a piece of software that will make Windows go faster, or that there's a Nigerian prince wanting to give them millions of dollars. Sadly, all of these people will never admit their failings, or what they did to break the item in the first place, and will never report back when what they broke has been fixed. There are pages and pages of tech reports on the internet on people "thinking they knew best" and ended up paying a tech to fix. All tech support people have stories about what users did to stuff up their own kit.

 

 

"Is it possible, that white male middle class professionals aren't the oracles they make out ?" given that, time and time again, you ignore advice that's given to you by several people on here, that's something that you'll never find out, isn't it.

[Clunkclick]

I thought Asus were pretty good - I've seen worse constructed stuff poked into major manufacturers kit

 

This board's design is over 10 years old, the board was bought 8 years ago. No problems until this issue -obviously of my own making through the memory upgrade.

The board has run all Microsofts Os's  from XP up to and including W10 and does so at final output speed to the user that would rival lower end I7's(My mid-high end I5 HP laptop is slower).

 

All the memory is Corsair, reportedly another good brand.

 

I'd like to see the MTBF comparison charts between Asus + Corsair and the generic components that are stuffed into most the high-street branded stuff made in China.

 

Agreed, the problem is rare and obtruse and I won't be able to tell  whether the problem is in one of the sets of Memory DIMMs or in one of the set of memory banks until I have all the banks populated with the exactly the same type of memory.

 

Postscript

 

Postman has just pitched up with new memory DIMMS.

 

Standby - I'll need a couple of hours to do a proper powerdown.

 

N

 

 

Edited 16 May, 2017 by Clunkclick

[RainbowFire]

Having spent an hour resolving some vandalism issues, it occured to me.......

 

Why is it that you have such a hard time with the battery, and thus thinking that the battery has an issue straight out of the packet, but you believe that the "less than a year old" £140 Fluke is completely infallable?

 

[Clunkclick]

Looks like the problem is solved, although I won't say definitely, until the system has had a proper overnight power shutdown.

 

E-Bay supplier delivered  Corsair XMS3 CMX4GX3M2A1600 vers 8.16, which is populated with memory chips only on one side (Confirmed Single-sided by CPU-ID software under windows) - not as advertised in E-bay  picture (Version 2.3 Double-sided).  Again, the outside packaging and labelling on the DIMMS made no reference to whether it was single-sided or double-side DIMM.

 

OK, the DIMM I was expecting was Double-Sided and it turned out to be single-sided, so I just changed tack and made all of the installed memory single sided and luckily the Mboard was apparently able to deal with an all- single sided installation. So, it would appear that certain motherboards don't like mixed installation of single and double-sided DIMMS.

 

So, simply removed the original 2 x Double-Sided (Vers 2.3) in Memory banks 3 & 4  and substituted the version 4.2 single-sided (The previous upgrade) and stuck the vers 8.16 in Banks 1+2, removed the CMOS battery, shorted out the CMOS memory on the Mboard pins, reset all the latter and restarted. All OK.

 

Whilst I was at it, I tested the installed CMOS battery (Panasonic purchased this week) - 3.2v (Fluke 325) and 2.96v (Ansmann) - (Commentators will note that this is the nearest to official spec) whilst there is a small difference in the readings from the various meters, when the measured source changes, the readings all change in the same direction and don't Fluke given a calibration certificate which is good for at least the length of time equivalent to the warranty ?

 

Be interesting to see whether the 8.16vers DIMMS are compatible with the 2.3 vers if they are not that could well  point to damage to 2.3 vers DIMMS

 

Nick

 

 

 

Edited 16 May, 2017 by Clunkclick

[Clunkclick]

What on earth is this doing here:-

 

 

N

[Clunkclick]

Just for the sake of completeness, I took out the ver 4.2 single-sided DIMMS in Banks 3+4 and re-installed the ver 2.3 Double sided DIMMS, leaving the single-side ver 8.16 DIMMS in Banks 1+2. Popped the CMOS battery, cleared the CMOs using the Mboard pins, re-instated same and pushed the power-on.

 

Thing wouldn't boot at all, even with coaxing - red "MEM-OK" light illuminated and despite pushing the "MEM-OK" button and later the system warm boot button, wouldn't respond at all - had to chop the power in the end. So, it would appear that the vers 8.16 + vers 2.3 combo is even more dysfunctional than the 4.2  + 2.3 combo in this motherboard - remembering that when installed on their own into Banks 1+2, with banks 3+4 empty, vers 2.3 DIMMS worked fine.

 

The only  discernable distinguishing   characteristics between a double-sided Corsair XMS 3 DIMM and its equivalent single-sided DIMM are that, on this model, that , on close inspection,the chip legs are visible on both sides of the DIMM on the double-siders and the labelling on the double-side DIMMS shows "XMS" in white and "3" in blue with white surround, whereas the single-siders labels show "XMS3" all in blue with white surround and DDR3 printed vertically in white against a blue background.

 

Nick

Edited 16 May, 2017 by Clunkclick

[Clunkclick]

Corsairs official memory guide states with reference to AMD processors  that Corsair XMS supports the Athlon II processor, 4 populated memory banks, all three combinations of single and dual ranked memory and has a maximum memory frequency in this configuration of 800Mhz - the CPU-ID program is reporting that is exactly the speed that all 4 memory modules are running at.

 

https://corsair.secure.force.com/knowledgebase/apex/KnowledgeArticle?id=kA1400000008UVI&q=XMS+memory+compatibility&l=en_US&c=Technical_Support%3ADDR3&fs=Search&pn=1

 

Voltage is reported as being 1.65v at this frequency, again within spec, so hopefully, no damage is being done.

 

Still a mystery why it won't run single and dual ranked together.

 

I know the microprocessor lost one leg at installation(Outside rank, rhs), but that didn't seem to effect it to date.

 

N.

 

 

 

 

[gadgetman]

5 hours ago, Clunkclick said:

Well, the Fluke 325, is only a year old, has only been used half a dozen times since then, and I would therefore hazard a guess that its still "On station" as far as agreeing with the factory calibration standards - i.e. good enough to be used as an index for home use . . . . blue touch paper lit.

 

Nick

Look on the back of the meter. 

 

Out of the box the error margin is up to +/-10%

 

But anyway you clearly know more than me and have ample time to waste trying to work it out that it's not the battery. 

 

 

 

 

[Mort]

What's the model of the MoBo?

[Clunkclick]

23 hours ago, gadgetman said:

Look on the back of the meter. 

 

Out of the box the error margin is up to +/-10%

 

But anyway you clearly know more than me and have ample time to waste trying to work it out that it's not the battery. 

 

 

 

 

Look

 

I'm not saying two swallows make a summer, but,. . . . . .  20 years old, bought second hand from E-bay mailed From to US to UK, then after 2 years use sent to Fluke for calibration, comes back with clean bill of health, except for one item - frequency. And, if it does need "Calibration" then this seems like a home job (For the hobbyist/emthusiast):-

 

 

Different kettle of fish for professionals as they will need a certificate from a third party for Professional indemnity cover.

 

Edit #1: And if the button batteries are good down to 2v (As the manufacturer spec I quoted above suggested) against a newly supplied voltage reference of somewhere between 3-3.4V then +/- 10% is irrelevant.  But manufacturers positions on the same kit can be different - see post below.

 

I had four button batteries as replacements. Only one was substantially above 3v and since using that battery in the Mboard, CMOS values have been  retained over powerdowns. Using any of the other 3 batteries, with voltages below 3v meant that some CMOS values were lost. But, its probably worth testing them again as the I suspect that the mixed memory combination I was using may have been causing a leakage voltage from the CMOS. Now the memory problem is sorted perhaps a different result will be observed with the < 3V button batteries.

 

Postscript

 

I suppose the real difficulty that the home bod would have, unlike Fluke, is getting an electrical source with more stability and less variation than the margin markers of the multimeter.

 

N.

Edited 17 May, 2017 by Clunkclick

[Clunkclick]

3 hours ago, Mort said:

What's the model of the MoBo?

Asus M4A88TD-M EVO - can't recall the version number, but the manual is E5341, First edition, version1, March 2010 - comes with 2 X USB 3 ports as a mini add-on PCI board. Can be run with anything that's AM3/AM3+ socket compatible, including Phenom II 1095T which runs the memory at 1033Mhz (Believe another Briskodian had this set-up and thought it good). Mines running on an Athlon II 630  with 8GB Corsair  DDR3 XMS3 1600 (Now all single-sided) and is fine - manual says its limited to 3GB if you are running 32 bit Windows and earlier AM3 socket processors are limited to DDR3 1033Mhz.

 

Later versions had the USB3 connectors surface mounted on the MBoard:-

Asus M4A88TD-M/USB3 &

Asus M4A88TD-M EVO/USB3.

 

Odd, cause the  Asus manual says it only supports Double-Sided Corsair XMS DIMMS  of the type listed above, whereas the Corsair on line table (Posted above) says the board is OK with all single-sided, double-sided or a mix. Mixed no longer works for my system (It used to), Double-Sided DIMMS up to 4GB in Banks 1+2 or Banks 1+3 does work and Single-Sided up to 4GB works in Banks 1+2 and 1+3 and 8GBs with Banks 1-4 populated - the latter being the current configuration.

 

It s still working this morning after an overnight power down . There was a slight stall on boot when it came to initialising USB, this was overcome by disabling legacy USB support in the BIOS CMOS.

 

Nick

[StevesTruck]

On 2017-5-16 at 16:35, Clunkclick said:

 

 

I know the microprocessor lost one leg at installation(Outside rank, rhs), but that didn't seem to effect it to date.

 

N.

 

 

 

 

 

Seriously? 

 

I'd be suspecting that way before I worry about the CMOS battery voltage to within 0.1v.

[Mort]

Did you try the original pairs in 1+3 (both blue) and 2+4 (both black) like they're meant to be?

[VWD]

On 16/05/2017 at 07:09, gadgetman said:

But none of your meters are calibrated. 

 

There's a reason professionals get their meters calibrated, and you've proved it. 

 

Seriously, stop wasting time on this. The battery is fine. 

There is another reason, that professionals/firms get their meters/test kit calibrated- Accreditation, and if working as a contractor for a large girm/Gov't organisation, part of the contract is that all test kit is covered by individual calibration certs, even down to readings etc having a space for the individual instrument/tool number to be recorded. In one industry it's common practice ,at start of work to crimp test pieces of a test joint ,and label them with the date /site etc.

I used to take test kit for regular calibration ,and I've yet to be told by the calibration house( or see it noted on the cal cert)that one of the Flukes was out of cal, and that's meters in constant use in all weathers in rough conditions.

Possibly Nick's problem is caused by dodgy memory connectors ?

Common method of cleaning older connectors of this type was with an eraser, but I'd wait for confirmation that this will not cause a static problem.

 

As for meter calibration - I've three meters, two Flukes of different qualities, and one old Argos type one. Years ago, out of curiosity , I tried all of mine against a meter fresh from calibration. None were outside of industry tolerances.

Edited 18 May, 2017 by VWD

[Clunkclick]

15 hours ago, StevesTruck said:

 

Seriously? 

 

I'd be suspecting that way before I worry about the CMOS battery voltage to within 0.1v.

Its been like that since 2012 , witrhout substantial deleterious effect.- one pin out of four which  were bent whilst the processor was out of the socket, broke off when being re-straightened and was replaced by the leg of a Rexel Bambi staple. The only noticeable effect is a one-pixel wide vertical line which appears on the rhs of the screen  whilst the BIOS software has control of the boot. It disaapears as soon as control of the boot is passed to Windows.

 

CMOS battery voltage became the main suspect in my mind because 1). The button battery that I removed was  completely dead, no voltage no nothing, 2) The machine had only been used 2-3 times a month over the last year - therefore the majority of the time the CMOS was being powered by the button battery  and 3). During the same period I started switching off all electrical systems which ran standy mode (To save power) - in standby mode the CMOS is drawing its power from the mains rather than the button battery. QED the demands on the button battery increased substantially over the last year.

 

It well be that i'm going up entirely the wrong avenue and that the apparent solutions found i.e. making all installed memory uniform rather than mixed, just happen to produce effects which stabilise the system only time will tell.

 

I want to avoid the cost of a replacement  processor (That would probably be a  used Phenom II @ £50-£80) so that I'm not encouraged to start taking the existing processor out of the Mboard socket just to find out the condition of the Bambi staple leg ! Once lucky, . . . twice a miracle.

 

Anyway, all OK. Booted first time from overnight power down without any errors. A few more days like this and I'll start  incrementally re-installing the other boards and drives. Hoorah !

 

N

Edited 18 May, 2017 by Clunkclick

[Clunkclick]

13 minutes ago, VWD said:

There is another reason, that professionals/firms get their meters/test kit calibrated- Accreditation, and if working as a contractor for a large girm/Gov't organisation, part of the contract is that all test kit is covered by individual calibration certs, even down to readings etc having a space for the individual instrument/tool number to be recorded. In one industry it's common practice ,at start of work to crimp test pieces of a test joint ,and label them with the date /site etc.

I used to take test kit for regular calibration ,and I've yet to be told by the calibration house( or see it noted on the cal cert)that one of the Flukes was out of cal, and that's meters in constant use in all weathers in rough conditions.

Possibly Nick's problem is caused by dodgy memory connectors ?

Common method of cleaning older connectors of this type was with an eraser, but I'd wait for confirmation that this will not cause a static problem.

 

As for meter calibration - I've three meters, two Flukes of different qualities, and one old Argos type one. Years ago, out of curiosity , I tried all of mine against a meter fresh from calibration. None were outside of industry tolerances.

Exactement. Life as we know it is so Contract protocol driven, that common sense empirical evaluation is precluded at risk of liability and is  buried somewhere in a corner.

 

N.

[Clunkclick]

7 hours ago, Mort said:

Did you try the original pairs in 1+3 (both blue) and 2+4 (both black) like they're meant to be?

According to the Mboard manual 1+3 is only really  for overclocking - I only had 4GB of the Double-sided original memory 2 x 2GB DIMM - but it worked fine  as 1 +2 and was tested 1 + 3, as  fine.

 

The original memory is apparently treated as twin banks (With twin controllers) whilst the single sided memory is treated as one bank. What has just occurred to me is that the BIOS CMOS settings offer a choice between unganged and ganged mode  for system memory(The default is unganged). I 've always been running the system memory as unganged, even when I only had 4GB of double-sided DIMMs installed. I don't know whether the effects of that are just limited to performance of memory of whether  setting it inappropriately for the type of memory installed would cause the boot stalls.

 

The other thing is that the memory controllers are  located in the Microprocessor which we know has a damaged pin . . . hmm ? Beyond my limited expertise.

 

When, I've got more time, I may play around with the ganged and unganged settings to see what they do.

 

N

Edited 18 May, 2017 by Clunkclick

[Mort]

23 minutes ago, Clunkclick said:

According to the Mboard manual 1+3 is only really  for overclocking - I only had 4GB of the Double-sided original memory 2 x 2GB DIMM - but it worked fine  as 1 +2 and was tested 1 + 3, as  fine.

 

According to the manual it supports "A: One module inserted in to any slot as a single-channel memory configuration. B: Supports one pair of modules inserted into either the blue slots or the black slots as one pair of dual-channel memory configuration. C: Supports two pairs of modules inserted into both the blue slots and the black slots as two pairs of dual-channel memory configuration."

 

So, again, did you try the original (mixed) pairs in the correct config?

 

 

[Clunkclick]

3 hours ago, Mort said:

According to the manual it supports "A: One module inserted in to any slot as a single-channel memory configuration. B: Supports one pair of modules inserted into either the blue slots or the black slots as one pair of dual-channel memory configuration. C: Supports two pairs of modules inserted into both the blue slots and the black slots as two pairs of dual-channel memory configuration."

 

So, again, did you try the original (mixed) pairs in the correct config?

 

 

Just tried it.

You're right  !

 

It will boot without problems with the original Double-ranked DIMMS (Vers 2)  in mboard memory slots A1 + B1 (Blue) and the single-ranked DIMMS (Vers 4.3) in A2 +B2(Black). Only difference compared with the all single ranked set-up is that its a tiny bit slower.

.

Hands-up, that's  the original set-up by me in 2012 of 1 x2GB Double-ranked (Vers 2) in A1 + A2  with the same logic carried on for last year's upgrade in slots B1+B2 of the single sided (Vers 4.3).

 

I presume the reason I did that was because of the prior information on page 1-11 of the manual - the table which states:-

Channel A - DIMM A1 + A2

Channel B - DIMM B1 + B2

 

. . . and because the original 2012 set-up (2 x 2GB in A1 +A2) worked with no problems for 5 years !

 

Postscript

 

I also note the entries on the DDR3-1600 O.C. Table 1-13  under DIMM socket support A, B, C.

 

Postscript #2

 

I've also had the chance to examine the original XMS (Vers 2) memory more closely and found out that they are not double-sided. In fact they are single-sided with the same number of chips (8) as the other later supplied DIMMS (Vers 4 & 8) - so I take it that each chip is 256Mb in size (8 x 256 = 2GB).

 

However, the vers 2 DIMMS are reported by the CPU-ID program as Double-ranked and vers 4 & 8 as single-ranked, so, does that mean that the vers 2 DIMMS are addressed as two banks of 128  bit memory whilst the vers 4 & 8 DIMMS are addressed as a single bank of 256  bit memory ?

 

VMT

 

Cheers !

 

 

Edited 18 May, 2017 by Clunkclick

[VWD]

On 17/05/2017 at 15:55, Clunkclick said:

Asus M4A88TD-M EVO - can't recall the version number, but the manual is E5341, First edition, version1, March 2010 - comes with 2 X USB 3 ports as a mini add-on PCI board. Can be run with anything that's AM3/AM3+ socket compatible, including Phenom II 1095T which runs the memory at 1033Mhz (Believe another Briskodian had this set-up and thought it good). Mines running on an Athlon II 630  with 8GB Corsair  DDR3 XMS3 1600 (Now all single-sided) and is fine - manual says its limited to 3GB if you are running 32 bit Windows and earlier AM3 socket processors are limited to DDR3 1033Mhz.

 

Later versions had the USB3 connectors surface mounted on the MBoard:-

Asus M4A88TD-M/USB3 &

Asus M4A88TD-M EVO/USB3.

 

Odd, cause the  Asus manual says it only supports Double-Sided Corsair XMS DIMMS  of the type listed above, whereas the Corsair on line table (Posted above) says the board is OK with all single-sided, double-sided or a mix. Mixed no longer works for my system (It used to), Double-Sided DIMMS up to 4GB in Banks 1+2 or Banks 1+3 does work and Single-Sided up to 4GB works in Banks 1+2 and 1+3 and 8GBs with Banks 1-4 populated - the latter being the current configuration.

 

It s still working this morning after an overnight power down . There was a slight stall on boot when it came to initialising USB, this was overcome by disabling legacy USB support in the BIOS CMOS.

 

Nick

Nick- I've got an Asus MB,and after looking into  problems seeing an USB booted stick not booting, I've discovered that after a Windows UPDATE booting into w7 creates problems wit the secure boot facility in Asus Bios. Simply put , from what I can gather, it's a security device to stop malware auto running on Asus MB. However, again what from what I’ve found ,there is a fix for this update. But to get secure boot off, you have to go into bios at startup- "DEL "key, then Advanced and fing boot -change "FAST BOOT" TO DISABLED. Then look at "Security".This then brings up the "SECURE BOOT " option. Which MAY be the cause of the problem.

[Clunkclick]

You must have a different BIOS - deactivating Fast Boot just outputs the extended BIOS screen start-up messages. Can't find a "Security" option.

 

My machine, now fitted with a re-chargeable Li-on button CMOS battery, rated @ 3.6v (!), and a mixed array of double rank and single rank DIMMS (Doubles in A-1, B-1 and singles A-2, B-2) and operating from the onboard Radeon graphics chip (MSI radeon graphics board not re-installed) is now as happy as and is starting normally every time.

 

Looks like the fault was caused by the incorrect  positioning of pairs of DIMMS during the memory upgrade from 4GB to 8GB - the machine appears not to tolerate mixed rank DIMMs when paired A1+A2, B1+B2.

 

There were also other contributory faults:-

 

Definite was:-

1). A depleted CMOS battery which may have been  corrupting system values on start-up and losing CMOS values during power-off periods. 

 

2).The MSI graphics card interfering with CMOS values (You recall it appeared to) - I have not re-installed it yet.

 

Possible:-

 

1) Malware in the CMOS - prior and whilst this was happening I had been getting regular calls from the sub-continental "Hello Sir, there is something wrong with your computer" artists. These calls have now stopped - presumably they are now fully employed righting the BA computer system !

 

Anyway, if there was anything dodgy in the CMOS, the battery has been in and out so many times over the last 3 weeks, that any peculiar  values will have been flushed.

 

Nick

 

 

Edited 30 May, 2017 by Clunkclick

[jjc]

I have seen this issue many times and is due to the PSU not being man enough to provid the power requirements. The issue is that PSUs are listed as their "total" power and not the amount each voltage rail provides. so one PSU may have lots of spare power on a specific rail at the expense of another rail. the way to test this is try with just two of the memory modules in and do numerous tests is all OK test again with the other two modules is still OK test with all four if it then fails it will probabley be the PSU.

 

John

[Clunkclick]

John - noted.

Be a shame if it was the PSU, as its only 18 months old.

 

I swapped - out the pair of double rank DIMMS yesterday, for single rank, so the set-up is all single rank now and, again, it appears to be fine.

 

As it seems stable now, with memory pairs planted in A1+B1 and A2+B2 (No matter whether they are single or double rank) and as the motherboard voltages reported in CMOS are within limits, I'm inclined to leave the memory alone, hope it will keep going- if only to prevent further wear and tear on the DIMM connectors, ultimately with a view to upgrading to AMD4 socket MB and Ryzen processor, when the prices come down.

 

I will be re-installing the MSI radeon graphics board. That is a big power draw, the board has its own 6 point power connector - voltages will drop slightly, even if the PSU is OK. If the PSU, or one of its rails is faulty, a problem will probably show up when I re-install that, especially when I run the furry bouncing ball stress test.

 

Also,there is another persistent problem, which may or may not be related. When closing down W10 on this machine, I'm getting a "Cannot write to this memory as its 'Read Only'. This problem persists, no matter which of the three pairs of DIMM memory I install in the 2 bank pairs on the MB. So, perhaps there's a failing component on the MB. Something to do with the "No-Execute bit" facility on x64 processors ?:-

 

https://en.wikipedia.org/wiki/X86-64

 

Postscript

 

The other thing I've noticed is that the AM3(938) processor is reported by the CPU-ID program as being from the Propus family of Athlon II 630 4 core processors (Family F, model 5, stepping 2, Ext. family 10, Ext.model 5, revision Bl-C2), with 4 x level 2 cache of 512Kb, 16 way, whereas the following document records the Propus as a 3 core Athlon II, with 3 x level 2 cache of 512KB and with the fourth core disabled:-

 

http://www.cpu-world.com/info/AMD/Unlocking_cores_and_L3.html

 

I'm just wondering whether the final seller (OC UK) or their supplier tweaked the processor as it was sold as 4 - core in conjunction with the overclocking Asus M4A88 board and that this may be causing the issues.

 

Nick

Edited 31 May, 2017 by Clunkclick

[gadgetman]

Nothing to do with the damaged CPU pin you mentioned earlier 

 

Before you waste any more time I'd be looking at that as the likely cause of your issues.