That may have been it in your case, but the fact is that overclocked processors throw their standard TDP out the window. At 3.8GHz your CPU was no longer just pulling 125 watts. So a rating of 140watts is essentially meaningless if your CPU goes past that. I'm not saying yours did as your explanation for your issue seems pretty cut and dry. I don't know the exact wattage a given CPU would be at a specific clock speed but the fact is that some cases, it's not a simple heat sink issue but rather VRMs that either aren't up to specification or are being pushed past specification by the end users. If your board is only rated for 140watts and you overclock any processor and it ends up pulling 150 watts, your VRMs are being abused harder than they were designed to be.
![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
MSI Motherboard caught fire! What really happen?
- Thread starter ins-
- Start date
That may have been it in your case, but the fact is that overclocked processors throw their standard TDP out the window. At 3.8GHz your CPU was no longer just pulling 125 watts. So a rating of 140watts is essentially meaningless if your CPU goes past that. I'm not saying yours did as your explanation for your issue seems pretty cut and dry. I don't know the exact wattage a given CPU would be at a specific clock speed but the fact is that some cases, it's not a simple heat sink issue but rather VRMs that either aren't up to specification or are being pushed past specification by the end users. If your board is only rated for 140watts and you overclock any processor and it ends up pulling 150 watts, your VRMs are being abused harder than they were designed to be.
Old Hippie
Supreme [H]ardness
- Joined
- Oct 31, 2005
- Messages
- 6,013
I wouldn't think so.
They could be held liable if the conquences of this flaw were the cause of a lawsuit.
You seem to be blaming the heat damage on sloppy construction but I'm wondering if lower spec'd components aren't to blame.
It's probably a combination of both.
Motherboard and CPU replaced and installed. Wonder should I try to clock it to 3.8ghz and stress test it for 1 hr? then try 4.0ghz... Since its still under warranty.
What you guys think? Of course I will play safe and supervise it. Maybe even record a video? LOL!
What you guys think? Of course I will play safe and supervise it. Maybe even record a video? LOL!
While you're under warranty, I would OC the hell out of it and stress test it 24/7. You would rather expose the weak points now than later.
Edit: you should run folding @ home on that bad boy for Team #33. That'll be a good test.
Edit: you should run folding @ home on that bad boy for Team #33. That'll be a good test.
MassiveOverkill
[H]ard|Gawd
- Joined
- Jul 8, 2005
- Messages
- 1,748
My X6 at idle:
at load @ 3.8 Ghz:
That's 232 watts my CPU is using, 107 over the 125 watt spec. It takes alot more voltage to go from 3.8 to 4.0. You're asking for another blowout if you try for 4.0, which is why I don't even bother.
at load @ 3.8 Ghz:
That's 232 watts my CPU is using, 107 over the 125 watt spec. It takes alot more voltage to go from 3.8 to 4.0. You're asking for another blowout if you try for 4.0, which is why I don't even bother.
Well oc is never guaranteed. It does suck when they blow but sometimes stuff just isn't designed to take certain loads.
sirmonkey1985
[H]ard|DCer of the Month - July 2010
- Joined
- Sep 13, 2008
- Messages
- 22,414
id say its more like 200w tops. you also have to take into consideration the efficiency of your psu at say 137w compared to 370w. it may be more efficient at the lower vs the higher. also have to consider how efficient the VRM's are. the hotter they run the less efficient they are. but most 140w boards are designed to support up to 200w with a heatsink. most 125w boards are good up to about 160-175w.
MSI's just using cheap ass components, especially on the 890GX and lower boards.
Nothing is garantee, but considering this is their top teir board and they state this board is as followed: (Don't quote me, this is directly from their website.)
"The Best Choice for Ultimate Performance, Stability and Easy Overclocking"
"Military Class Components Enhance Stability and Board Longevity"
"MSI's Exclusive OC Genie Intelligent Overclocking Function Offers a New Experience for Easy Overclocking"
"Unlock CPU Core Unleashes CPU Performance Potential"
"these boards targets advanced users who demand ultimate performance and beyond with overclocking."
"The MSI 890FXA-GD70 and AMD 800 series mainboards features server-grade Military Class components. Certified by the Department of Defense, these components ensure that you are using the highest grade components to power your board. Hi-c CAP (highly-conductive polymer capacitor), Solid CAP (solid capacitor), and Icy Choke (low-temperature inductors) fully enhance product lifespan, heat-resistance, and stability. The specific characteristics of Hi-c CAP are extreme heat-resistance, high stability, and high performance, offering 8X the lifespan of ordinary solid state capacitors. Icy Choke inductors are the result of the application of the latest materials and structural designs to deliver high heat resistance, low power consumption, and low noise. Compared to ordinary capacitors, Solid CAP offers a lifespan that can extend your board up to 10 years, not only providing admirable mainboard longevity, but also offering flawless performance and reduction crashes or lagging due to overheated hardware. These materials power defense systems and they are going to power your gaming system as well. "
Now, does this not promise to be the best it can be especially for overclocking? Especially, nothing drastic of an overclock? Yes, I feel ripped off on the first board, well see how this one holds. Who knows, maybe I just had a lemon?
Sometimes the marketing department and the engineering department don't work closely enough together, or a product wasn't thought out well enough. But what can you do, hey at least they are still doing rmas. Sucks though.
Matthew Kane
Supreme [H]ardness
- Joined
- Dec 1, 2007
- Messages
- 4,233
Holy fuck.....reflex ftw....
MassiveOverkill
[H]ard|Gawd
- Joined
- Jul 8, 2005
- Messages
- 1,748
You do have a good point about efficiency, but the opposite of what you're thinking It's about 82% efficient at idle, and 90% efficient at load (power supply only).
http://www.jonnyguru.com/modules.php?name=NDReviews&op=Story&reid=183
BecauseScience
[H]ard|Gawd
- Joined
- Oct 9, 2005
- Messages
- 1,047
There are many reports of that on MSI 880GX boards as well, including some very recent ones.
I don't mean to nitpick but UPS's aren't very accurate when they report loads, especially when the load is a psu with power factor correction. Also, due to their intended use they tend to overestimate the load / underestimate the amount of runtime. Better to have you shut down with a couple of minutes left than have you run short.
Just thougth I'd chime in on this...
Just upgraded my CPU last week and had the same issues as everyone else. Was running a 550 BE unlocked at 3.630ghz @ 1.53v on my board thats in the sig. no issues prime for 24hrs stable ran fine since last Jan... Upgrade to a 1090T O/C to 4.009ghz @ 1.44v and the board blows up!! Was running prime for about 40min and had the side cover off... wife was standing there watching and it just shut down... looked inside everything was fine... waited about 30-45 secs and hit the pwr button and BLAME!! sparks come shooting out scared the hell out of my wife... starts yelling at me about burning the house down! After shuting it down I took a look inside to see the one of the mosfets is burnt/discolored but the board seems to work fine as long as I dont prime with it. So its seems more or less a power phase design issues with this X6's when O/Ced. Now I seen they were getting warm when I first stared testing this chip so I placed some sinks on the mosfets but it seems to not help when O/Cing x6's on a 4+1 board. So I guess I will be upgrading to a 8+2 board to fix this.
On a side note I did see that Asus makes a 16+2 phase design for Intel chips... wheres the love??
Just upgraded my CPU last week and had the same issues as everyone else. Was running a 550 BE unlocked at 3.630ghz @ 1.53v on my board thats in the sig. no issues prime for 24hrs stable ran fine since last Jan... Upgrade to a 1090T O/C to 4.009ghz @ 1.44v and the board blows up!! Was running prime for about 40min and had the side cover off... wife was standing there watching and it just shut down... looked inside everything was fine... waited about 30-45 secs and hit the pwr button and BLAME!! sparks come shooting out scared the hell out of my wife... starts yelling at me about burning the house down! After shuting it down I took a look inside to see the one of the mosfets is burnt/discolored but the board seems to work fine as long as I dont prime with it. So its seems more or less a power phase design issues with this X6's when O/Ced. Now I seen they were getting warm when I first stared testing this chip so I placed some sinks on the mosfets but it seems to not help when O/Cing x6's on a 4+1 board. So I guess I will be upgrading to a 8+2 board to fix this.
On a side note I did see that Asus makes a 16+2 phase design for Intel chips... wheres the love??