Overclocking 3770k on Msi z77 Mpower

I would have been surprised if it did turn out to be a RAM issue. I can't remember the last time I had to RMA memory but I suppose it has to happen to someone at some point!

I still haven't entirely figured out why my chip doesn't seem to like it at anything more than 4.5, but I do have a hunch. When I have my chip at say 4.6ghz it boots into windows fine and everything, but what actually happens is that the various cpu testing apps actually "stop responding and are forced to quit." but the computer itself never actually freezes or crashes or anything like that, plus the temperatures are all fine. When I run cinebench sometimes it stops responding and sometimes it has an error message.
Yea in my experience that is a classic sign of a very tiny undervoltage. You've got enough volts to load windows but not quite enough to do the testing and the volts are dropping just a little bit too low under load to make it all work but you still have enough for windows to recover as the volts increase again. Adding 1 step up on the LLC control would be the easiest remedy I'd imagine.

Overall though I think it's the old silicone lottery. Once you try for more than 4.3GHz you need to start getting lucky. My first 3570K would boot up to 4.9 but I couldn't get stable at 4.7 for anything. Lady luck is a bitch.
 
I would have been surprised if it did turn out to be a RAM issue. I can't remember the last time I had to RMA memory but I suppose it has to happen to someone at some point!


Yea in my experience that is a classic sign of a very tiny undervoltage. You've got enough volts to load windows but not quite enough to do the testing and the volts are dropping just a little bit too low under load to make it all work but you still have enough for windows to recover as the volts increase again. Adding 1 step up on the LLC control would be the easiest remedy I'd imagine.

Overall though I think it's the old silicone lottery. Once you try for more than 4.3GHz you need to start getting lucky. My first 3570K would boot up to 4.9 but I couldn't get stable at 4.7 for anything. Lady luck is a bitch.





for this kit i would have to agree since it is failing prime now, but the sleep wake issue is definitely another issues though because i've tried a kit of kingston stuff that is jedec 1066 i think with an xmp of 1600 and the same issue it wont wake from sleep when its set to xmp mode for that stuff

the gskill ripjaws stuff i tried works running prime for hours but it will not wake from sleep either so i still dont know what that is


the problem is mine will do 4.4ghz at 1.23V, but it won't even boot into windows at 4.5ghz until 1.35V and at that point the temperatures are way too high anyway even though it is stable at those settings , 4.6 have no idea what voltage would be required for that but it's not looking doable in a reasonable way.

i have tried the various vdroop settings when i was doing this i had it at 100% so minimal droop but still no luck unfortunately


i am able to do 4.4 ghz turbo no problem though which i am happy with



not sure what i am going to do about the ram i dont really want top send it back i also want to get another kit of 8 to make it 16gb
 
my 8gb kit of viper extreme stuff died completely today it is failing prime and giving me bluescreens, so i am going to return that stuff and replace it with a 16gb kit of viper 3 which is newer stuff anyway and lower voltage so i will report back how that stuff performs.

right now the 16 gb kit of gskill ripjaws 2133mhz 1.6V is in there and it runs fine at 2133mhz i just havent been putting it to sleep

i was planning on getting another kit of 8gb xtreme to add to the 8gb i had already but since it's not working properly anyway i figured i just return it and get a 16gb kit of the newer lower voltage viper 3 stuff instead.


i will have an update shortly on my build log and report the results here as well.
 
just a quick update -- i've replaced the patriot viper xtreme 1.65v stuff with the newer intel extreme masters 1.5v stuff that is designed for ivy bridge. this stuff is rock solid at 1.5V and it even has slightly faster copy/read/write speeds than the gskill kit i used temporarily.



still won't wake from sleep though but i've tried 5 kits now and none of them have been able to wake from sleep at XMP settings so i know it's not the ram but the motherboard or cpu at fault.

not a huge deal since i rarely sleep the machine anyway.

8315544926_e5431c586d_h.jpg
 
I would have been surprised if it did turn out to be a RAM issue. I can't remember the last time I had to RMA memory but I suppose it has to happen to someone at some point!


Yea in my experience that is a classic sign of a very tiny undervoltage. You've got enough volts to load windows but not quite enough to do the testing and the volts are dropping just a little bit too low under load to make it all work but you still have enough for windows to recover as the volts increase again. Adding 1 step up on the LLC control would be the easiest remedy I'd imagine.

Overall though I think it's the old silicone lottery. Once you try for more than 4.3GHz you need to start getting lucky. My first 3570K would boot up to 4.9 but I couldn't get stable at 4.7 for anything. Lady luck is a bitch.

Ohh I see.

If you can pass memtest86+/prime small/32M then it's not actually memory issue. It migh have something to do with those S1/S3 power options, that determine the way the computer "sleeps" and "wakes". I doubt that messing with timings and voltages will solve that kind of problems.

It's BIOS/chipset (acpi) drivers/OS/crashing drivers that causes those troubles, most of the times...



hey guys

i believe ive found out the problem with this current setup. i have no idea why i didn't notice this soon, but for some reason i did not.

the voltages for the mpower board are way off, the 12V is only at 9V and in addition to that all of the voltages have massive spikes.


this is a screenshot of the voltages for my laster motherboard, the sabertooth x58, everything looks pretty normal to me so that indicates that the power supply is working properly.

8132470379_3658121148_o.png



Now have a look a tthe voltages with the current mpower motherboard


as you can see the current value for the 12V is only 9V, but it has recorded a spike of 22.4V, 10.7V for the 5V, and 2V from the CPU vcore.

8323153145_613bc12190_o.png





I am going to go on monday in to the store and see about getting the board replaced, if i can't get it done in store i guess i will have to send it to msi.
 
hey guys

i believe ive found out the problem with this current setup. i have no idea why i didn't notice this soon, but for some reason i did not.

the voltages for the mpower board are way off, the 12V is only at 9V and in addition to that all of the voltages have massive spikes.


this is a screenshot of the voltages for my laster motherboard, the sabertooth x58, everything looks pretty normal to me so that indicates that the power supply is working properly.

8132470379_3658121148_o.png



Now have a look a tthe voltages with the current mpower motherboard


as you can see the current value for the 12V is only 9V, but it has recorded a spike of 22.4V, 10.7V for the 5V, and 2V from the CPU vcore.

8323153145_613bc12190_o.png





I am going to go on monday in to the store and see about getting the board replaced, if i can't get it done in store i guess i will have to send it to msi.
hope it will all get sorted!
 
Don't mean to thread hijack, but this was so similar I thought posting here would be better than making a new one; has anyone managed to figure out what the deal with Z77 XMP and Sleep Mode is?

I'm using Kingston HyperX Predator 2400MHz on the same board, and mine will also not boot from Sleep Mode; no input works, so I'm eventually forced to hit the power button, which starts the computer briefly, then restarts it, at which point I get a fail message about overclocking settings. When I boot it again from cold, the memory is still running at the 2400MHz XMP settings though, so it doesn't reset them.

My voltages are also a bit off; although I'm not seeing the huge spikes that the OP is, my Z77's 12v runs at 9.102v, and the 5v at 5.98 to 6.02v.
 
Don't mean to thread hijack, but this was so similar I thought posting here would be better than making a new one; has anyone managed to figure out what the deal with Z77 XMP and Sleep Mode is?

I'm using Kingston HyperX Predator 2400MHz on the same board, and mine will also not boot from Sleep Mode; no input works, so I'm eventually forced to hit the power button, which starts the computer briefly, then restarts it, at which point I get a fail message about overclocking settings. When I boot it again from cold, the memory is still running at the 2400MHz XMP settings though, so it doesn't reset them.

My voltages are also a bit off; although I'm not seeing the huge spikes that the OP is, my Z77's 12v runs at 9.102v, and the 5v at 5.98 to 6.02v.




As far as the voltages I found out it was just an issue with the reading from CPUID HWMonitor and that other MSI boards have had the same issue.

if you check in control center, AIDA64 or the bios those all report the same voltages which are the correct ones


i dont have a fix the to sleep issue unfortunately i just run it at XMP and dont sleep it, i dont have any issues when it restarts though.

here is a screenshot of my bios, control center and aida64 voltages is, i suspect of you check your voltages in one of these places you will see the same. my dad says that it would not even run if it actually was 9V or so.




8325425326_35f35e7247_o.jpg


8326489077_9e8c867af2_o.png


8326509509_ac282285cc_o.png
 
Hi mate.

Seens i have the same board, i will try to see what will go on when i put my pc in sleep mode. The only thing is that we can help, and maybe get the problem done and out ;)
 
Hi mate.

Seens i have the same board, i will try to see what will go on when i put my pc in sleep mode. The only thing is that we can help, and maybe get the problem done and out ;)

from the various kits of memory i've tested none of them have had a problem with sleep at 1866mhz, it's only at 2133 mhz + is when it has trouble waking from sleep, so if you're signature is accurate and you have 1966 memory then you shouldn't have trouble with sleep with that memory
 
from the various kits of memory i've tested none of them have had a problem with sleep at 1866mhz, it's only at 2133 mhz + is when it has trouble waking from sleep, so if you're signature is accurate and you have 1966 memory then you shouldn't have trouble with sleep with that memory

I also have a 2133 kit here and a 2400 kit. But that is to some mates builds i will do, when he get the last parts he miss. So will try and see if it works, but after what i seen the 1.65v have that problem, so will see whit the 2400 kit and the 2133 kit is 1.5 and do hope that work.
 
Weird behaviour this. Not waking from sleep if at 2133 ( pretty high frequency ). Have you encoutered someone else experiencing this?
 
Don't mean to thread hijack, but this was so similar I thought posting here would be better than making a new one; has anyone managed to figure out what the deal with Z77 XMP and Sleep Mode is?

I'm using Kingston HyperX Predator 2400MHz on the same board, and mine will also not boot from Sleep Mode; no input works, so I'm eventually forced to hit the power button, which starts the computer briefly, then restarts it, at which point I get a fail message about overclocking settings. When I boot it again from cold, the memory is still running at the 2400MHz XMP settings though, so it doesn't reset them.

My voltages are also a bit off; although I'm not seeing the huge spikes that the OP is, my Z77's 12v runs at 9.102v, and the 5v at 5.98 to 6.02v.

Weird behaviour this. Not waking from sleep if at 2133 ( pretty high frequency ). Have you encoutered someone else experiencing this?

yes the guy above with 2400 mhz hyper x predator is having the same issue. i haven't herd too many cases of it but i have tried 5 different memory kits so i know for sure it's not the memory and so it has to be either the CPU or the motherboard.

I also have a 2133 kit here and a 2400 kit. But that is to some mates builds i will do, when he get the last parts he miss. So will try and see if it works, but after what i seen the 1.65v have that problem, so will see whit the 2400 kit and the 2133 kit is 1.5 and do hope that work.



i thought at first it may be a 1.65v issue as well, but the patriot extreme masters stuff i have now is 1.5V and it does the same thing. it's making me think it might be the CPU because it also doesn't overclock well.



i also know that 2400mhz kits run fine one other ivy bridge CPUs and they are often 1.65V so i dont think it is a voltage issue either.
 
You were spot on about CPUID being the issue on the voltage detection, AIDA64 shows my 12v running at 12.02v. I'm guessing the sleep issue is just down to MSI's motherboard, hopefully it's a driver issue rather than something to do with the hardware and it will be fixed over time(perhaps we should all fire an email to MSI?).
 
You were spot on about CPUID being the issue on the voltage detection, AIDA64 shows my 12v running at 12.02v. I'm guessing the sleep issue is just down to MSI's motherboard, hopefully it's a driver issue rather than something to do with the hardware and it will be fixed over time(perhaps we should all fire an email to MSI?).



hi there, yes other people have been reporting the same with hw monitor recording wrong voltages for some msi boards.

i actually just submitted a tech support request for the issue with msi just a few minutes ago, figured it was worth a try. there was a massively detailed sheet of info that needed to be filled out so i hope that will prove fruitful and maybe we can get a real answer to this. it's interesting to me because when i google for it i dont really see much at all.



oh a side note how are you liking the hyper x predator aside from the sleep issue? i've been really eyeing those and liking the look of them more and more recently and i can definitely see myself using some 2400mhz sticks in the future. i need to figure out a cooler solution though if i want to have memory that is any taller than the patriot stuff i have now..
 
I was having issues with the HWmonitor as well. For some strange reaon, it wasn't reading my +12V right ( I have an MSI board ) and I installed "Control Center" ( Special Utility for this motherboard ) and it works. But what puzzles me is that I have Kingston Hyper X Blu memory sated at 1.65 V and in the Control Center it's reported as being 1.60V. ( I left it on "Auto" in the Bios ) Now, my system is completly stable and all in case you are wondering. I don't know if it's a reading problem or the bios has selected this value since it's on "Auto". The closest value to it's rated value when I try setting it manually is 1.67, but I am gonna leave it on Auto since it's stable. BTW I heard people OC'ing this type of memory up to 2200MHz. What do you guys think?

( Sorry for the off topic post )
 
LLC on?
Need more volts mate. Lots more. Follow TTL's way of oc'ing and you should be ok. I hope you have some good cooling to keep her cool.
Maybe you just lucked out on the silicon lottery as well.

Where would one find this guide?
 
i thought at first it may be a 1.65v issue as well, but the patriot extreme masters stuff i have now is 1.5V and it does the same thing. it's making me think it might be the CPU because it also doesn't overclock well.

i also know that 2400mhz kits run fine one other ivy bridge CPUs and they are often 1.65V so i dont think it is a voltage issue either.

After some test this night, i found that the 2400 kit work well on my CPU and have no problem whit sleep mode. But then i testet it on a i5-3570k and that have the problem whit sleep mode on a Asus mobo, so after what i think. It is a CPU problem.

The 2133 kit worked out well on the i5 and my i7. Just needed to test the i5 in my build, and it do OC better then my i7. Can hit 4.8 whit 1.27 volt on it. Almost the same volt to my i7 whit 4.5 ghz. I do use 1.26 volt on that..
 
You were spot on about CPUID being the issue on the voltage detection, AIDA64 shows my 12v running at 12.02v. I'm guessing the sleep issue is just down to MSI's motherboard, hopefully it's a driver issue rather than something to do with the hardware and it will be fixed over time(perhaps we should all fire an email to MSI?).

After some test this night, i found that the 2400 kit work well on my CPU and have no problem whit sleep mode. But then i testet it on a i5-3570k and that have the problem whit sleep mode on a Asus mobo, so after what i think. It is a CPU problem.

The 2133 kit worked out well on the i5 and my i7. Just needed to test the i5 in my build, and it do OC better then my i7. Can hit 4.8 whit 1.27 volt on it. Almost the same volt to my i7 whit 4.5 ghz. I do use 1.26 volt on that..



The whole thing is quite strange. I guess it depends on the combination of hardware you have.


I sent MSI an email and they sent me a newer bios to try that is newer that the current release version, but it did not seem to do anything different.


http://forum.overclock3d.net/showpost.php?p=592711&postcount=255
 
Damn, that's a shame, hopefully they can figure out what's going on and fix it. In the meantime, I tried using the second XMP profile at 2133, and still it hates Sleep Mode; can anyone link to a beginner guide to overclocking RAM? I want to try downclocking to 1866, but I'll have to do it manually(there are only 2400, 2133, and 1333 available by default in the BIOS) and I've never gone anywhere near RAM OC'ing before.
 
Back
Top