So I've had the Intel DX58SO2 desktop motherboard and Intel Core i7-940 CPU for over 2 years now. I had been using it as a Hyper-V Server 2008 R2, and using another system to RDP in to. Recently, I decided to convert it to a gaming system. So I upgraded my graphics card and started interfacing with the system directly.
I started playing games like StarCraft 2, Diablo 3, Team Fortress 2, and so on. Not very graphically intensive games. I started noticing that after a certain amount (what seemed to be random) amount of time, my system would freeze up. My first instinct was that my system was overheating. So I opened up my case and felt around to see if I could find out what was overheating. What I found out was that my GPU was often VERY hot to the touch. My first instinct after that was I thought there was something wrong with my GPU (EVGA GeForce GTX560).
So I borrowed a friend's GPU, an AMD ATI Radeon card (I can't recall the model, but it was something released about a year ago). I also used the old GPU from my still operating, old system (EVGA 8800 GTS). Neither of these cards have a history of overheating. In every case, the GPU seemed very hot to the touch. So I determined there's nothing wrong with the GPU, but something else.
To further prove that the GPU is not to blame for my system freezing, I've noticed my system freeze when I'm simply watching YouTube videos. Yes, I know that browsers these days do take advantage of GPU processing power, especially for playing video, but I mean YouTube seriously should not be causing my system to freeze.
My next step for troubleshooting was my CPU. I had been using the stock heatsink (which came with the CPU), with the stock thermal paste (that is pre-applied at the factory). I thought maybe my CPU was generating too much heat now that I'm using the system for gaming. So I purchased a Therlamtake Frio Advanced which has an enormous heatsink and 130mm fans on both sides. I actually had to remove the fan on the far side (inner-most) because it would have been physically interfering with my RAM. So I thought that wasn't a big deal because it has an enormous fan and heatsink, no problem. Still my system froze up.
Then I thought maybe my case didn't have proper cooling because I was using a pretty small case and only had a single fan in the back of the case. So I purchased and installed the Cooler Master Sniper (previously was using the Cooler Master Elite 330). This new case has manual auxiliary fan control on the case, has a 200x30mm fan on the front (directly in front of the HDD bays), on the side panel, and on the top of the case, in addition to a 120x25mm in the rear of the case for exhaust from the CPU. Once again, system is still freezing up.
At this point, I'm extremely frustrated about just throwing money at this problem to fix this freezing problem and obviously none of it is making a difference. On top of that, I'm not even overclocking any of my components. So I started monitoring my system temperature with software (I wish I had done this first). I noticed that none of my components were really that excessively hot, with the exception of 3 different items, called SMIOVT4, SMIOVT5, and SMIOVT6. All 3 seem to register the same temperature no matter what, so they're tied to each other in some way. At idle, with my just running my operating system without doing anything computationally or graphically demanding (just using a web browser and Excel, these 3 different components read 94 degrees Celsius. That is very alarming to me as I don't believe any component inside a computer system should ever be that hot.
What are the components SMIOVT4, SMIOVT5, and SMIOVT6? I can't seem to find any worth references to these components on the Internet.
Are there any other system software monitoring and diagnostics tools you'd recommend that I use to figure this out? Preferably a motherboard diagnostics tool.
At this moment, I'm leaning towards pointing at my motherboard as the culprit that it has a defect that is causing this, but the other potential culprit I suppose could be my CPU having a defect. Although in my almost 6 years working as a systems analyst, I've never come across a bad CPU, but have had plenty of malfunctioning motherboards.
Another thing I forgot to mention is that when I am watching videos on YouTube or playing games I mentioned previously, I noticed these SMIOVT components reaching 98 degrees Celcius.
The SIO device on the DX58SO2 board is NOT used for thermal monitoring or fan speed control. The SIO's SMIOVT monitors are of absolutely no consequence; on this particular board, the BIOS doesn't initialize them, most are not connected to anything and none of them have been calibrated. Thermal monitoring and fan speed control decision-making is implemented on this board using a standalone ASIC that is compliant with Intel's Heceta 6P requirements specification. Your board will have either an Andigilog aSC7621 or (more-likely) an ON Semiconductor/Analog Devices ADT7490. While both of these devices have their own unique set of capabilities and device-specific register set, for the "standard" set of capabilities that we define in the specification, there is a (more or less) common register subset and the devices are pin-compatible. These devices are accessed across the System Management Bus (SMBus).
Intel provides a Windows-based tool, called Intel® Desktop Utilities, that allows you to monitor the various temperature, voltage and fan speed sensors included in these ASICs. There are also plenty of third-party tools that can provide this support as well. Here's a non-exhaustive, randomly-ordered list:
Note that Intel does not validate the level of compatibility of any of these tools.
Now, all that said, there are plenty of reasons for system hangs and the thermal issues don't start until way, way, way far down the list. I am not saying that you are wrong; I am just saying that you are going to need to do a lot more to verify it. I am being skeptical because you haven't said anything that would conclusively point me towards a thermal issue ("hot to the touch" means absolutely nothing; there are always going to be components that are hot to the touch, regardless of the load being placed on the system). As well, your system description is pretty anemic; you need to provide more details. If you have any concerns regarding what you say in a public forum, you can send it to me via private message.
If you need pointers on what to include, here are some ideas:
1. What other components are plugged into your system? Describe all add-in cards, all internal peripherals (HDDs, SSDs, ODDs, etc.), etc. Do you use RAID or similar capabilities?
2. What changes have you made in the BIOS Configuration? Have you verified that you still see the issues with the BIOS configured to defaults? Yea, there are going to be cases where you absolutely have to have some configuration changes in place for a system to work. If so, document these settings! What is the O/S configuration? What software is (always) running?
3. What fans are being utilized? How many inlet fans are there and what sizes are they? How many outlet fans are there, what are their sizes and where are they located (top? rear?).
You mentioned a new heatsink-fan unit for the CPU - a heatpipe tower. I hope you picked one that provides cooling for 130W. I also hope that you are using one that (also) provides support for pushing air down onto the motherboard and providing cooling for the voltage regulation circuitry and the memory. There are many solutions out there that do a great job of cooling the processor but leave the voltage regulation components to slowly fry. There are even some that, in addition to not providing airflow themselves, block much of the airflow generated by other sources....
As a first step, I suggest that you install Intel® Desktop Utilities. It will generate onscreen alerts for over-temperature situations and will track them until removed. I also recommend that you try version 3.2.6. This version has a lot of robustness and capability improvements over previous versions. While it is not specifically tested on DX58SO2 any longer, it still retains support for it and should work reliability (the DP55KG board also utilizes these ASICs and I have tested it on this). Version 3.2.6 was released to manufacturing just this past Friday. It may take a day or two for it to be made available in download center. You can use the previous version (3.2.4) in the meantime, but I recommend that you switch to 3.2.6 as soon as it is available to you. When you are installing, you will likely see a warning that the application is not being tested on this board; simply continue the install - and don't forget to reboot the system when it says to...
Thank you very much for that very informative and prompt reply! Below, you'll find the information that you've requested:
Intel Desktop Utilities v3.2.4 (I can't seem to find the newer version that you refer to at the moment). All temperatures that it reports do not produce any alarms. I've been using the software for a couple weeks now, and have been using SpeedFan for a couple of months now, and just installed HWMonitor earlier this morning.
Windows 8 Enterprise x64
Hyper-V feature installed
I use IE10, Google Chrome, Adobe Reader XI, Adobe Flash, Microsoft Silverlight, HP Software for my all-in-one HP printer, Java 7 Update 21, Microsoft Office Professional Plus 2013, Microsoft SkyDrive
Games I play are StarCraft II, Diablo III, Team Fortress 2
Board - Intel DX58SO2 on firmware SOX5820J.86A.0879.2011.0824.0021
Processor - Intel Core i7 940 @2.93 GHz
Memory - Patriot Viper II Sector 7, 6x DDR3 DIMMs being used (each DIMM has 4 GB @1067 MHz) = 24 GB
Case - Cooler Master Storm Sniper (1 front 200mm inlet fan, 1 side 200mm inlet fan, 1 top 200mm outlet fan, 1 back 120mm outlet fan)
HDD - 1x Western Digital WDC WD2002FAEX-007BA0, 1x WDC1001FALS-00J7B0, 1x WDC WD1002FAEX-0023A0
GPU - EVGA GeForce GTX 560 SC (Have also tried using an AMD ATI Radeon card, and an EVGA 8800 GTS)
Sound Card - Sound Blaster X-Fi Xtreme Audio using the PCIe x1 slot (Have also tried using a Sound Blaster Xtreme Gamer using the PCI slot)
Power Supply - OCZ OCZZ850M (Modular, 850W)
CPU Cooler - Thermaltake Frio Advanced CLP0596 (Compatible with Socket LGA 1366, Supports 230W, also tried using stock heat sink, fan, and thermal paste)
I'm not using RAID or any ODDs or SSDs
Installed on a Basic Disk, will soon be installing on a GPT disk and configuring BIOS to use UEFI
Configuration -> Onboard Devices -> Audio = Disable, default is Enable
Configuration -> Onboard Devices -> 1394 = Disable, default is Enable
Configuration -> SATA Drives -> Hard Disk Pre-Delay = , default is 
Configuration -> Video -> Primary Video Adoptor = Ext PCIe Graphics (PEG), default is Auto
Security -> Intel VT for Directed I/O (VT-d) -> FLR Capability = Enable, default is Disable
Power -> System Power Options -> Wake on LAN from S5 = Power On, default is Stay Off
Power -> System Power Options -> PCIe ASPM L0s = Enable, default is Disable
Power -> System Power Options -> PCIe ASPM L1 = Enable, default is Disable
Boot -> Boot Menu Type = Advanced, default is Normal
Boot -> Boot Device Priority = (can't recall the order), default is
Boot -> Boot to Optical Devices = Disable, default is Enable
Boot -> Boot to Network = Disable, default is Enable
Boot -> Boot to USB Devices First = Enable, default is Disable
Boot -> Boot Display Options -> On Screen Options -> POST Function Hotkeys Displayed -> Display F7 to Update BIOS = Enable, default is Disable
Boot -> Boot Display Options -> On Screen Options -> POST Function Hotkeys Displayed -> Display F10 to Enter Boot Menu = Enable, default is Disable
Boot -> Boot Display Options -> On Screen Options -> POST Function Hotkeys Displayed -> Display F12 for Network Boot = Enable, default is Disable
I'll get back to you if I notice these problems using default BIOS configuration
Also, the front, side and top fans are controlled with an auxiliary dial on the front of the case. When I do have the fans running at 100% (Pretty noisy), I do seem to be able to get farther along than not running the fans at all. However, my system does eventually end up freezing at some point even with all of those fans at 100%. I even turn my GPU fan up more than what is slated for its position in its heat/fan curve. I use EVGA's Precision X utility to control the GPU fan.
Note: Rear fan is always running and is not controlled via this auxiliary dial.
I didn't realize they made CPU fans that provide cooling for the motherboard and memory. Mine looks like the fan points directly out to the outlet fan for the rear of the case.
Note: the stock heat sink/fan just pushed air directly upwards from the CPU, so in my opinion, my new heat sink does a much better job of the stock heat sink. Also like I mentioned before, I don't overclock any of my components, so I'm not sure how I would need to provide cooling to my memory or motherboard. Also, my memory have heat sinks mounted on the top of each of the sticks.
I'll also try to remember to take a picture of the inside once I've got the case opened again to show you in case you think the CPU cooler is blocking airflow (Although once again I was having this same problem with the stock cooler that was less than half the height of this new cooler).
First of all, as I said, we RTM'ed the new version of IDU just this past Friday. It will take a day or two for things to trickle down and the files ot actually appear on download center. You are ok with 3.2.4 for now; watch for 3.2.6 to be posted and use it once it is...
Second, the stock heatsink-fan unit's fan pushes air down, through the heatsink and then out in all directions along the surface of the motherboard. This is the optimal airflow pattern. I have seen tower heatpipe solutions that have their cooling fins aligned so that they deflect some of the air down onto the motherboard. One of the places on the motherboard that we monitor is one of the major VR hotspots (we put a thermal diode as close to there as we can (barring routing issues)). If you are seeing the temperature for this sensor staying below 80, it's fine (though 70 would be better).
Seeming to run a little longer to lock up with the fans cranked up is not much of a proof of it being an overheating problem. You've got to watch the temperatures of everything that you can and watch for specific zones reaching levels that are concerning. If the temperatures are not rising to concerning levels, it's time to look at other possibilities. In a very significant number of the situations that I have analyzed, it has been the audio solution that ended up being the culprit. Most of the time it was the audio driver (why these are so often responsible for hangs, resets and BSODs, I am not sure). Now that I think about it, however, I remember a situation in which heat from the video card was affecting the audio card beside it. Hhmmm, you said you disabled the onboard audio and used an add-in card? Can you move this card any further away from the video card? If not, maybe you should pull this card for a while and use onboard audio and see if this positively affects the system's stability.
Another thing - the zone around the PCI connectors often get the least airflow. The front inlet fan is the one that best affects this zone. It's worth having this fan be the one running faster. Another way to loook at it is overall acoustics. The CPU and GPU fans are typically the worst. In both cases, however, providing them with cooler air to work with means they don't have to work as hard (spin as fast) to provide the same cooling benefit. Bottom line, speeding up the inlet fans especially (but speeding up the outlet fans helps as well) will often result in a reduction in overall acoustics. Think about it...
This board has reached its End of Life and it marked to be as End of Interactive Support, so Intel may not longer provide interactive support for it. More information on this is available here:
However, I can recommended you update to the latest BIOS available 0888, since according to the information provided, you have BIOS version 0876 installed on your board.
Sylvia, thank you for informing me that interactive support on this board will be ending on 6/28/13.
Per your recommendation, I have installed the BIOS update that you identified.
Scott, thank you for your input on motherboard cooling. Please remember I was experiencing this same problem with the stock cooler, and that the ONLY reason why I am now using a 3rd party cooler is because I was trying to get more cooling for my system, even though I am not overclocking.
I also did as you recommended, removed my sound card. I played StarCraft 2 for over 2 hours without any issue. This also was my CPU, rear, and GPU fans at the software-controlled speeds, and leaving the front, side, and top fans at the minimum speed according to the dial on the case. I think you may have found the culprit. I shall get back to you after I've performed more testing.
If the sound card is indeed the culprit, what would you recommend I do for sound? I run Windows 8 x64, and there are no compatible drivers for the Realtek integrated audio. Needless to say, the integrated audio does not support 5.1 surround sound. Do you suppose that SoundBlaster is at fault for not making more heat resistant components? Or have you seen this also with other manufacturers?
How would you suppose I resolve this issue if the sound card is to blame? I currently have my GPU in the primary PCIe x16 slot, and I had the sound card in the next available PCIe x1 slot. Do you think I should move the sound card to the primary PCIe x16 slot and move the GPU to the secondary PCIe x16 slot? I seem to remember reading something stating to install your GPU in the first available PCIe x16 slot, that is closes to the CPU.
There are a couple of things that you can do. The first would be to try and verify that the issue is indeed the sound card locking up the system due to overheating (as opposed to it perhaps being some sort of driver-related issue). The easiest and most obvious thing is to ensure that you have the absolute latest version of the driver for the card. If you are already on the latest, you might want to talk to the manufacturer to see what they have to suggest (and whether they can investigate this issue further for you). Dealing with Creative Labs is, in some respects, a boon since they are both the silicon and the board manufacturer...
From a thermal standpoint, what you need to do is provide a thermal break between the graphics card and the audio card. The most obvious thing to do is to move the cards as far away from each other as is possible. Keep the GFX card in the X16 slot (it needs the lanes!) and put the Audio card in the farthest away PCIe slot that you can (it doesn't have to be a X1). This may be a complete solution all by itself. If it isn't, next thing to do is ensure that a fan is blowing cool air into the volume between the two cards (or is exhausting heat from the air in this area). If you can't do it using one of the fans already present (just running it faster), you can fairly cheaply add one. Some cases have a spot to mount an extra fan to the side panel that will blow cool air onto the cards. If yours doesn't, you can always get one of the fan or blower units that mounts in a card slot position. I checked NewEgg, Amazon and Buy.com; there are solutions as low as $6US (though I don't know how bas their acoustics are going to be)...
I am sure that I am going to get inundated with emails as my reward for answering this...
I am asoftware architect in Intel's Desktop Boards organization. In addition to owning the development for most of our value-add software applications (Intel® Integrator Toolkit, Intel® Desktop Utilities, etc.), I also own most of our hardware monitoring and fan speed control hardware and software (within both the BIOS and runtime environments). I wrote the requirements spec for the Heceta ASIC that is used on this board. I was also the architect and lead developer for the Intel® Quiet System Technology feature in the 965 and 3, 4 and 5 Series Chipsets (and which makes its return in our 8 Series boards). Bottom line, I live and breathe thermal and acoustic issues all day long.
Most of the time that I spend assisting folks on the Communities site is my own. With the exception of the Intel Customer Support folks, most of the Intel folks on the site are here voluntarily, simply because we care about our products and our customers...
Thanks for the reply! I understood that you had some understanding of these issues, little did I realize the true nature of your comprehension! Its fantastic to see input from the developers, engineers, team leaders etc in a forum like this. It is a shame that it is on your own time. It makes me, and I'm sure everyone, feel more connected to the team behind these products if you guys lurk and provide answers that demonstrate a true understanding.
Just wanted to chime in and say excellent work, well done. I sure do have a few questions regarding IDU and XTU, but I'll bite my tongue