MOVED: BIOS 3.4 - destroyed overclock and wierd CPU prob

This topic has been moved to Overclockers & Modding Corner.
BIOS 3.4 - destroyed overclock and wierd CPU prob

Happy to see you have problem cause only few mention this.
I have it too with Venice 3500+. Forced multiplier 5, forced min. voltage and I get a cool 5x200. To fix it I must use HTT of 5, then rest acts normal. Anything but 5 and it defaults to this minimum setup. Yes, C&Q is turned off. May be if we set up C&Q in OS that would help but not going to happen.
Solution for me is to use bios 3.2 which dont have either RAID bug or 219 bug - which actually is a 22x bug for me. Ive done all P95 tests at 11x246 for min. 12 hours. Im not happy with solution or 3.2 though and hope new beta also fix this problem not so many talk about. Think Ive seen 5-6 crying about it in public. With HTT 5 you wouldnt notice, guess most use that.
If you test new beta bios and it does fix this please post so I know 3.5 is worth waiting for. Beta is not 100% compatible with our mobo I think - not sure you should flash unless you got bios savior, some have done it but Im not going to.
Link to beta https://forum-en.msi.com/index.php?topic=82493.0 but be careful...

Similar Messages

BIOS 3.4 - destroyed overclock and wierd CPU prob

Well i have just upgraded to BIOS 3.4 for the Platinum SLI and it has:
1: ruined my overclock, was stable at 2.8 and now at around 2.4, amd trying to find out what setting is causing aggro but much higher than this and it seems to just hjang after saving BIOSsettings - am up to 220 FSB but much higher kills it, was at 270!!!!!
2: CPUz etc. now detect my multi as 5x, even though it is set to 11x in BIOS and Cool 'n' Quiet isnt on?!?
this is on my 3700+ San Diego
Any help???
thanks
nana

Happy to see you have problem cause only few mention this.
I have it too with Venice 3500+. Forced multiplier 5, forced min. voltage and I get a cool 5x200. To fix it I must use HTT of 5, then rest acts normal. Anything but 5 and it defaults to this minimum setup. Yes, C&Q is turned off. May be if we set up C&Q in OS that would help but not going to happen.
Solution for me is to use bios 3.2 which dont have either RAID bug or 219 bug - which actually is a 22x bug for me. Ive done all P95 tests at 11x246 for min. 12 hours. Im not happy with solution or 3.2 though and hope new beta also fix this problem not so many talk about. Think Ive seen 5-6 crying about it in public. With HTT 5 you wouldnt notice, guess most use that.
If you test new beta bios and it does fix this please post so I know 3.5 is worth waiting for. Beta is not 100% compatible with our mobo I think - not sure you should flash unless you got bios savior, some have done it but Im not going to.
Link to beta https://forum-en.msi.com/index.php?topic=82493.0 but be careful...

MOVED: [P67A-C45 B3]About overclocking and the frequency

This topic has been moved to Overclockers & Undervolting & Modding Corner.
https://forum-en.msi.com/index.php?topic=154190.0

As HU16E replied, all seems correct. Voltages are right in the ball park. Personally I would advise to stick to a 100 BCLK and rather increase the multi to 44. There are enough articles out there warning against increasing the BCLK to 104 or beyond.

MOVED: K8N Neo2 question about overclocking and sata ports 1 & 2

This topic has been moved to Overclockers & Modding Corner.
https://forum-en.msi.com/index.php?topic=130785.0

Quote from: SteveH647 on 24-September-09, 04:39:34
But is it the HDD itself that fails at the higher frequency? Or is it the HDD controller on the motherboard that fails? What about DVD drives? And what if the HDD is SATA 300 instead of SATA 150?
Quote
But is it the HDD itself that fails at the higher frequency?
Yes.
Quote
Or is it the HDD controller on the motherboard that fails?
No
Quote
What about DVD drives?
Can't tell you, haven't maked such experiments.
Quote
And what if the HDD is SATA 300 instead of SATA 150?
It will not change anything.

General Slowness and Wierd CPU Usage

Writing on behalf of my uncle. He has a dual quad-core 2.8GHz Mac Pro with 2GBs of RAM (I verified this). And yet his system is slow and very, very crash-prone. Daily tasks such as emailing, using iWork, and Firefox or the cats pressing keys on his keyboard freeze his system.
I called him up and had him look at Activity Monitor. Even with "All Processes" selected, the top users of his CPU come in at 2%. Meanwhile, at the bottom, the value of "% User" is fluctuating from 30 to 60. (It's not RAM usage; he had ~1.5 gigs free).
Any ideas why this is happening or what can be done to speed up his system? Thanks.

the value of "% User" is fluctuating from 30 to 60.
That is actually pretty high and would say that you are hitting 3-5 cores at 100% and that there should have been a process (or group of processes at the top, viewing by cpu usage) eating up A LOT of cycles.
I think somewhere along the line, either the original OS build itself was bad, which happens, or bad hard drive, or bad RAM even. Or installed something wacky.
The OPs title and symptoms sure sound like MA gone wild, and sadly, the old "reinstall to cure" in such a case - people think it isn't needed or that is "... so Windows-like" and this time, take notes, exactly what steps they took and what they did that triggered the problem, maybe next time have "Missing Manual" in front and take notes and be more careful.
But Migration Assistant wasn't involved, and there was no G4/G5 involved. But I would think someone writing a program would recognize safe vs harmful and avoid killing a system the way it can, and which I've walked people through, often grudgingly and dragging to have to "do it all over again" and a fresh install of everything (clone your system before you make changes folks!).
For each burn project you use to need 24-40GB of free unfragmented space for best results. I guess that could double, and then you are talking very low disk space. But that was not their problem (low space).
People thing leaving a system on overnight to run scripts. Okay, don't do it for three months and see what happens or not. Test the theory.
My personal worst example of system meltdown (I saw a meltdown in 10.5.1 era too caused by pre-mature syncservices gone wild and the system refusing to respond, and did a fresh install finally as I installed 10.5.2, but got by using Disk Warrior which found a number of errors - errors that got there I know not how or why.
Yesterday I was moving drives to a new case, resized some partitions, it looked okay, ran First Aid, started copying files... and the system just froze. This is a $200 FirmTek dual-drive case.
Reboot after hitting power. Zilch. Nada. Had to power off the drives I was working on. Nada. Popped one of my 'spare' boot drives into Mac Pro and booted off that. Checked existing drives. Rebooted with the external drives and those checked out, but reformatted anyway just to be safe.
Disk Utility changed in Leopard. Creating partitions I sometimes have to quit and relaunch Disk Utility. That the resizing/creation of partitions for me still looks like it isn't right. And that a reboot may have been needed.
Disk Utility even said it "lost communication with a drive" and to relaunch Disk Utility. I think I saw that once back in 10.5.0 or 10.5.1 and was why I felt it was looking more like Final Candidate than ready for prime time. But I thought 10.5.4 was prime time.
I've zeroed one drive. Have run Speedtools Integrity and Media test, and has the option to map out bad blocks automatically, but I don't think those will find anything, I just thing OS X has gremlins.
My worst fear was the sound, lack of, when drives go silent when they are always running (and is a RAID which don't spin-down) or mechanical.
Second fear: heat from the external drive case. Summer. Drive reported 49ºC.
Third: the entire system stopped responding. Could not force quit and even Activity Monitor was beachballing. I had just quit OmniWeb and it wouldn't. And in the Finder it was doing a copy of disk images to backup which I do a lot, and later redid 300GB copy without issue.
Fourth: a new ergonomic keyboard caused the system and Finder to freeze and lockup???
Anyway, I really try to figure out why things happen. This is harder than the old mainframe that didn't have this many variables to worry about!!

Bios 3.5 and raised cpu temps

i just updated my bios to version 3.5 and my cpu temps went from 32c idle to 52 idle .now ive returned to bios 3.4 and my temps have gone back to 32c idle ..
has anyone else had this strange problem..i know the 3.5 bios updates the cpu microcode and maybe this is the problem or is it that the new bios is giving the right readings and the 3.4 bios giving the wrong readings.!!!!!!any help from the moderators is welcome on this subject

Quote
Originally posted by Del UK
Additional:
QTEC PSU and errors?????? Be there done that and brought Enermax or Antec PSUs..............
Just a HINT
HINT Q-Tec PSU.......
Be careful..... you may lose ALL
Del
i disagree Del.
you may lose all....
it would more be YOU WILL LOSE ALL

MOVED: Overclocking and RAID, do they mix???

This topic has been moved to Overclockers & Modding Corner.
Overclocking and RAID, do they mix???

What board do you have? It makes a difference in this case. For example, the Via K8T800 based boards do not have PCI/AGP locks, so raising the FSB overclocks all the other components as well, including the HDD controller, and it has been observed that for these boards, RAID arrays will become unstable at FSB settings above about 225 MHz, at most, so overclocking and RAID do not really gotogether on these boards. The nforce3 based boards, on the other hand, should have PCI/AGP locks, though I think (but I'm not sure) that you have to explicitly enable them in BIOS, and I've seen some threads questioning whether or not they actually work, so again the board might have problems with high FSB settings, although I'm pretty sure an nforce3 board would have a working lock and thus be able to overclock with a RAID array defined. Finally, if you have a nforce4 based board, it should definately have a functional lock on it, and it should handle the RAID just fine when overclocking.

MOVED: P6N SLI-FI overclock and suspend to ram

This topic has been moved to Overclockers & Modding Corner.
https://forum-en.msi.com/index.php?topic=108460.0

To overclock my cpu from 8*266 to 8*333 i just change the cpu voltage from 0.0000 to 0.10000, with cpu voltage to 0.000 my computer don't start
To understand why when i resume to suspend to ram my computer freeze i try this.
No overclocking but just cpu voltage 0.100000
I start windows and read the vcore with cpu-z ( 1.325 + 0.1 V )
suspend to ram
resume
read cpu-z ( 1.325 + 0.0 V )

MOVED: P67A-G45 Windows 8 Overclock Speedstep Issues

This topic has been moved to Overclocking, Undervolting.
https://forum-en.msi.com/index.php?topic=162736.0

Quote from: celltech on 24-October-12, 21:13:49
So my bad...there is an Intel Turbo Mode setting under the OC tab. It was enabled and I tried to disable it. No change...
So then I started turning off C1E and C-State stuff. That only made it worse. According to CPU-Z the CPU was getting full voltage all the time. At least with C1E/C-States enabled the voltages will back off with no load on the CPU.
The Win 8 Task Manager is very strange. It is showing the frequency bouncing around 3.3-4.2 while CPU-Z shows it hard locked at 4.2 I am not sure what the Task Manager is using for its calculations. But even the Task Manager hard locked at 4.2 with everything disabled.
I have the same board (P67A-G45) and I am having the same problem. When I set the multiplier in the BIOS, the CPU will bounce between 3.3GHz and whatever I set the multiplier to. I am running the V1.18 BIOS. With Windows 7, I could set the multiplier to whatever I wanted, enable C-States, and the CPU would throttle down to 16x without any issue or any interaction from me.
You can use ThrottleStop to check the multiplier more accurately then CPU-Z. Also in ThrottleStop, you can click "Set Multiplier" -> "Power Saver" -> "Turn On" and the CPU will run like a champ at 16x multiplier, so its definitely possible. This 16x get reported in CPU-Z and in the Windows Task Manager as well. Once you disable Power Saver, its back to 33x + multiplier land.
I hope a solution to this will be found, seems really silly to install a new OS and have CPU multiplier issues...

MOVED: h67ma-e35 won't overclock

This topic has been moved to Overclocking, Undervolting.
https://forum-en.msi.com/index.php?topic=163374.0

Quote from: Bernhard on 16-November-12, 18:39:21
Oh no, they are not, as you can OC the GPU and the BCLK if you so wish.
First peruse the facts before making wild comments that are untruthful. http://www.anandtech.com/show/4830/intels-ivy-bridge-architecture-exposed
It does help to gather technical facts before purchasing HW that one expects to support the intended outcome.
No H67 board out there supports "K" edition CPU multiplier overclocking. Sorry mate. Cry on Intel's shoulder.
Never buy a Yaris and then tell the manufacturer that he forgot to give you the Lexus luxuries.
OK, just gotten the box of the mainboard, let me qoute from the box:
OC Genie is an auto-overclocking technology for any type of user. Just enable the OC Genie function and the CPU memory, and iGPU will be automatically increased for top performance!
Benefits:
Auto detect and overclock CPU, memory, and iGPU in 1 second
Unlock CPU Ratio and overclock to 4.2GHz
Unleashed 36% iGPU increase
Tell me how this is not a lie, overclock to 4.2 GHz...

Overclocking AND Undervolting

...I wonder, has anybody tried out this combination? I just did (because the other day I stumbled across an article talking about how well the Athlon64 handles undervolting), and so far the results are pretty good...when I was overclocking my FSB to 215 MHz, I ran into some stability issues and boosted my vcore to 1.525 volts, but ultimately the instability traced back to the fact that my memory timings were just too aggressive for the higher speed, though I left my vcore at the increased setting anyways, until I found that article. So today, I went back to my BIOS and specified a vcore of just 1.4 volts, and the system runs stably, even at the overclocked speed. Furthermore, the thermal properties of the CPU are *much* improved:
Athlon64 3000+ @ 215 MHz FSB, 1.525V vcore:
idle temp = 4 to 7 degrees Celsius above ambient case temp
full load temp = 48 to 50 degrees
Athlon64 3000+ @ 215 MHz FSB, 1.4V vcore
idle temp = 1 to 3 degrees Celsius above ambient case temp
full load temp = 43 degrees
...other than that, my system stability so far seems to be exactly the same as it was at the higher voltage. Benchmark scores remain identical to what they were before the undervolting...the only change is that the CPU now runs *much* cooler. I'd always thought that overclocking required higher CPU voltages in order to maintain stability, but I guess not always. Has anybody else experimented with undervolting an overclocked CPU, or am I the first? Is there any reason for me to rush back to my BIOS and restore the previous voltage setting?

I stumbled upon a little utility called ClockGen while reading a Tom's hardware article, and among its many features is the ability to set the vcore to values much lower than those present in the BIOS (it also allows multiplier manipulation, though my system crashed when I tried to raise mine...lowering it may be possible though). I set my vcore to 1.30 V (still overclocked to 2.15GHz), and was stable for a few minutes, then got a spontaneous reboot...but anyways, SpeedFan did accurately show that my vcore voltage dropped from 1.4V to 1.3V before the reboot, and it's a pretty neat little utility if you're not too scared of making your system crash, so maybe some of you with non-overclocked CPU's can try going to 1.3V (or lower) and see how that goes. The link is:
http://www.cpuid.com/download/CG-ICS950403.zip
NOTE (and this is important): the above version of the utility is ONLY for people running the K8T Neo mainboard (i.e. MS-6702)...I don't know what would happen if you tried it with a different board, and I don't advise trying to find out.

MOVED: Incompatibility between nVidia based motherboard and ATI based graphics?

This topic has been moved to Off-Topic Technical.
https://forum-en.msi.com/index.php?topic=125552.0

Sounds to me you run short on IRQ's .
Make sure to "Load Optimized Defaults" in mobo BIOS so set IOAPIC=Enabled and MPS version = 1.4 . Then reinstall OS.
If not this is done it's a no go.

MOVED: Slow boot with P6N SLI and Vista

This topic has been moved to Vista problems.
https://forum-en.msi.com/index.php?topic=108306.0

Quote from: HenryW on 12-August-07, 08:15:55
Hey Jack did you read the whole thing?
Obviously not good enough. I had a look at the official BIOS update changelogs and saw that v1.2 included fixes that have to do with S3.
Quote
thought S1 was for sleep and S3 for hibernate.
Not exactly. Just to make sure we are all talking about the same thing:
Most common Important ACPI Standby-States:
- S1
This is the most shallow sleep state. Power to CPU and RAM are maintained, fans keep spinning.
- S3 (Suspend to RAM)
Power to RAM is maintained.
- S4 (Suspend to disk, Hibernate)
All content of main memory is saved on the hard drive, preserving the state of the operating system, all applications, open documents etc. After everything was saved to disk, the system will turn off.
If I am not mistaken, all you can do in BIOS is choose between S1 and S3. In Vista you should be able to choose beetween Standby/suspend (S1 or S3, depending on BIOS settings) and Hibernate (S4).
So again, which of the three states are you referring to?

MOVED: MSI X79A-GD45 8D overclocking 3930k question about Power Saving Features

This topic has been moved to Overclocking, Undervolting.
https://forum-en.msi.com/index.php?topic=163826.0

Quote from: Froggy Gremlin on 29-November-12, 19:15:02
VDroop set to 100% is all you are going to get. 1.344v's should get you about 44-46X on the cores on 'Auto'. The voltage goes down from there with the lower VDroop settings. With CPU V. set to 'Auto', you get power saving, soon as the CPU V. is taken out of 'Auto', you lose it. No, the CPU speed may lower, but the voltage going to it will not.
VDroop is not a true offset feature like can be found on other brands of boards. https://forum-en.msi.com/index.php?topic=162779.0
Most folks around the industry recognize 1.350v's as the recommended limit for socket 2011 CPU's, so 1.344v's is fairly close to that anyway.
So when i set the voltage and the cpu multiplier i loose power saving features right?
Thanks for the reply

Overclocking and to D.O.T or not to D.O.T?

First off, please forgive my very newbish question here. While I have built a couple of systems, this is my first journey into the world of overclocking.
A bit of background:
The rig I have built in my sig, with no overclocking, has been performing very well and rock solid with gaming and general use for 8 months now. I have just started to venture into the world of video editing and converting my videos into a format that can be streamed. That is when I hit the performace wall, and realized for this kind of CPU hungry use I needed to either upgrade my build after a pitifully short 8 months, or look into overclocking my system to squeeze every ounce of performance I can get from it. Not having wads of cash readily available, I decided to get my feet wet in overclocking.
My first foray into overclocking has been to just stick my toes into the water and let the system overclock for me using the D.O.T options in the BIOS. Doing some experimenting, I have found that I can get some fairly good gains in performance ~+10% in benchmarks by simply bumping the D.O.T level to 'colonel'. The 'general' setting is unstable, giving me the BSOD whenever the overclocking kicks in.
Here is some hopefully relevent info on my setup. This is the CPU-Z display when D.O.T has kicked in.
Temp wise I go from 37C (idle) to 47C (Overclocked Load)
And here are my benchmarks when using D.O.T: (3DMark05)
Main Test Results
3DMark Score 5321 3DMarks
CPU Score 5156 CPUMarks
Okay, now to the meat of my question.
How much CPU headroom do I have on my hardware, above and beyond what I am getting with the D.O.T settings I am using? I have reached the limit as far as using D.O.T('general' setting being unstable), but could I get significantly better gains if I manually overclocked?
Or.. would I be better off staying ignorant of the mysteries of manual overclocking, and just save my money to one day buy an Athalon X2?
Thanks!

first of all, if you are using DOT, you have no control on the OC. the system is overclocked using some defaults settings programed into the cell chip. that means that it doesn't take into consideration the hardware you are using, and its limitations. this might result in instability.
second, the max OC with DOT is 11% if i remember correctly. that is a puny OC attempt, especially if you are running a A64 3000+ like i do. i can get to almost 50% OC for the CPU using different methods than DOT, instead of 11%.

MOVED: BIOS 3.4 - destroyed overclock and wierd CPU prob

Similar Messages

Maybe you are looking for