1 2 3 4 5 Previous Next 71 Replies Latest reply on Sep 21, 2011 6:57 PM by bunny Go to original post
      • 15. Re: Calling all DG45ID users with Intel Desktop Utilities 3.1.3
        ryangibson77

        ...So, I know this is somewhat unrelated, but I don't want to start a  new thread...I just noticed that my CPU fan NEVER spools up. Since I couldn't get IDU loaded, I  downloaded SpeedFan and ran a simulation tool that maxes out all four cores on my Q8300 processor. Speedfan tells me that my core temp is exceeding 70°C, but the fan speed never automatically changes. I can MANUALLY change it using speed fan, but no auto.

         

        I then went back into the BIOS and changed my processor zone fan response to "Aggressive," thinking that the last BIOS update simply made the processor fan less sensitive to changes in core temp, since a lot of folks have been complaining about their fan speed changing a lot...but nothing...it just sits there at about 1040 RPM, no matter the temperature...unless I adjust it manually...

         

        Any ideas?

        • 16. Re: Calling all DG45ID users with Intel Desktop Utilities 3.1.3
          Flying_Kiwi

          When you say the fan never speeds up, does it rev up at boot for a few seconds? The one on my DG45ID does this but (just as with yours) I never hear a peep out of it otherwise (as long as CPU C-States is disabled so the DG45IDs well known fan speed irregularity bug  - with certain CPUs at least - doesn't appear). I wouldn't worry about the fact it's staying quiet when the CPU temp is showing in the software as being 70 degrees - sometimes these temps aren't accurate (just look at the temps shown in IDU on the DG454ID immediately after a cold boot).

          • 17. Re: Calling all DG45ID users with Intel Desktop Utilities 3.1.3
            ryangibson77

            Yes, specifically, as long as I have the "detect fans at startup" enabled, it does the fan "surge." Otherwise not.

             

            I WISH I could run IDU to see what the temps are after a cold boot...currently waiting on some bugs to be fixed. I wish Intel would roll back IDU to the previous revision, which apparently worked just fine...but they don't offer ANY of the older versions, unfortunately...

             

            In any case, I am concerned that if those measured temperatures aren't accurate, then what is being used to ramp up my CPU fan when it gets hot? Other than at bootup, I never hear the thing spool up, even when I max out all four cores for several minutes, and the "measured" temperature exceeds 70°C! Does Intel's warantee on the processor cover damage by their own motherboards that don't cool the processor correctly??

            • 18. Re: Calling all DG45ID users with Intel Desktop Utilities 3.1.3
              Flying_Kiwi

              R. Gibson wrote:

               

               

              I WISH I could run IDU to see what the temps are after a cold boot...currently waiting on some bugs to be fixed. I wish Intel would roll back IDU to the previous revision, which apparently worked just fine...but they don't offer ANY of the older versions, unfortunately...

              Doh! sorry I forgot they don't keep the older version of IDU available online (not even for a short while ) after a successor arrives (and that not everyone has kept a copy to reinstall). I didn't mean to rub it in!

               

              If you have an all Intel affair (mobo, CPU and CPU fan) I can't see why it shouldn't be covered by their warranty as long as it's all professionally assembled according to standard trade practices - after all, who else could they blame for any failures? The only sticking point might be if any of it's OEM (rather than retail boxed) components. Of course having IDU quickly back up and running with your PC is going to be the best way to ease your concerns - maybe a progress update from someone at Intel in the know could be forthcoming?

              • 19. Re: Calling all DG45ID users with Intel Desktop Utilities 3.1.3
                yf38

                Hi,

                Of course my DG45iD / E8400 stepping E0 has (testing only one, I have two) the same problem as all others with  IDU 3.1.3.

                windows 7 x64 SP1 all updates.

                Fortunatly I  kept the previous release.

                 

                However, you can go to this site:

                http://www.gtopala.com/siw-download.html

                and at the bottom of the page, download the file siw.exe at the line:

                "SIW Without Installer (English-Only)"

                 

                starting this program, (no installation) and going to the hardware menu, select the line "sensors" and you will see the temperatures and fan speeds.

                I compared with IDU running aside (not recommended, but just to check) and the temperature and fans speed are the same for both utilities.

                More of that, SIW records and shows the minimum and maximum values aside the current ones.

                 

                This program helped me also to track the infamous spurious fan speed surges discussed in another thread...

                • 20. Re: Calling all DG45ID users with Intel Desktop Utilities 3.1.3
                  Flying_Kiwi

                  Just a note to all affected, I've received a brief reply to my idufeedback email from the relevant Intel manager about an hour ago and he advised that they hope to have a 'hot fix' released 'by next week' (quite an appropriate term when you consider what IDU monitors) I'll let him know this also affects other 4 series IDU compatible motherboards (for the person who posted in this thread with other than a DG45ID) so they can test the fix accordingly. I'll also ask for Intel to consider leaving the previous version up for a short while (maybe a month) after new versions are introduced, just in case there are bugs that have slipped through so people can revert until a fixed new version comes along.

                  • 21. Re: Calling all DG45ID users with Intel Desktop Utilities 3.1.3
                    sueron
                    • Thank you, funny me I thought I could fix this because I installed it wrong.Imagine my supprise when the first hit is the same issue  I am pulling out my hair over.By the way i am no code engineer so you know this was a total waste of time fix HA HA, Glad I keep all  driversfor moments like this,just thought the last one would be on download section,But when it runs IT RUNS


                    • 22. Re: Calling all DG45ID users with Intel Desktop Utilities 3.1.3
                      spearson

                      Flying Kiwi is correct, we have indeed root-caused the bug (in my code, as it turns out ) and a fix is currently in validation. We hope to have this available sometime next week (our processes require specific defect regression followed by at least 3 days of stability testing, with no issues seen, across the various boards and O/Ss supported)...

                      • 23. Re: Calling all DG45ID users with Intel Desktop Utilities 3.1.3
                        sueron

                        Now I have discovered for some unexplainable reason I cannot remove or install the older (working) version of desktop utities. IDU 3.1.2.022 Just have to wait foe the updated version when it arrives at least all else is working fine. I did notice the fan problem as well but hot or not it keeps on ticking

                        • 24. Re: Calling all DG45ID users with Intel Desktop Utilities 3.1.3
                          yf38

                          You cannot remove IDU because it is running.

                          You can avoid running an installed IDU by unchecking it from the list of services launched at startup.

                          start/ administrative tools / system configuration / services (maybe not exactly the names...)

                          Another working way is to stop the two or three processes related to IDU with the task manager.

                          • 25. Re: Calling all DG45ID users with Intel Desktop Utilities 3.1.3
                            Flying_Kiwi

                            spearson wrote:

                             

                            Flying Kiwi is correct, we have indeed root-caused the bug (in my code, as it turns out ) and a fix is currently in validation.

                            Hi spearson,

                             

                            It's now over a week so could we please have an update on this matter? I see that the previous version has been brought back and is now listed as the latest version suitable for the DG45ID. I see that the new version 3.1.4.031 is listed as compatible with series 4,5 and 6 boards but the specific few listed 4 series boards do not include the DG45ID.

                             

                            Being the wreckless daredevil that I am , I've tried this latest version with my DG45ID (which has just had a fresh install of Win 7 SP1 x64) and it seems to have little difference to the previous version except it is a little slower at showing the various screens after clicking the options (or perhaps that's just because all my previous use with this PC was under x86 versions of Win 7). I see that the only other real 'issues' I have with this version (as well as the previous version now listed as suitable for the DG45ID) are:

                            1. The Maximum System Memory (shown in System Information) is only listed as being 8GB (even though with recent BIOS versions - such as the one I'm using - it's actually 16 GB).
                            2. The MCH Vcc graph shows a voltage which is consistently 1.116 ~ 1.125 V (I've never seen it any higher under any circumstances) yet the guage pointer is right at the lower limit of the green range (in stark contrast to the rest which look safely near the middle).IDU Screenshot.jpg
                            • 26. Re: Calling all DG45ID users with Intel Desktop Utilities 3.1.3
                              ryangibson77

                              I have also noticed that IDU does not recognize (or at lease indicate) what my various mobo temp sensors or fans are. I am at work, and don't recall exactly what it says, but like the northbridge temp sensor just reads something like "misc/unknown temp" but does actually seem to read a value. Also, it doesn't recognize which chassis fan is which...it just says "unknown/other fan." Neither are huge issues, especially considering that this latest version at least doesn't crash like the previous version...but it would be nice if it showed the correct sensor names...

                               

                              Also...and I think I've read this somewhere before, and may be more of a BIOS issue...the Northbridge temp sensor does not seem to display the ACTUAL temperature, but rather some minimum temperature that is considerably higher than the actual temperature...and if I recall correctly, will only start indicating the actual temp if it EXCEEDS that preset minimum temp...which is fairly annoying...

                              • 27. Re: Calling all DG45ID users with Intel Desktop Utilities 3.1.3
                                Curious592

                                The thermal margin approach has prompted multiple comments and seems to defy primary purposes of software; elimination of read and write errors.  This has prompted me to consider the relationship of fabrication process to the variation of limits for boards in addition to fans, sinks and layout.  It would seem that a thermometer that automatically adjusts for air pressure would be similarly confusing except where variation makes interpretation too unreliable (calculating a difference).  There may be a similar reason for this approach but I never saw one given.  At least this way a service tech can interpret reports more easily.

                                 

                                The sensors should have meaningful names since the locations are associated with discrete functions.  I seldom need be concerned but some of the newer boards would be better supported.

                                 

                                I have had a problem with detection upon removal of a fan.  I now realize that the BIOS can probably be reset instead of disabling fan detection.  Since the only system I have that I has a problem with 3.1.3 lacks an onboard PS fan connector (newest PS) I wonder if that is another complication for the utility.  Even in older systems I don't think power management ever controlled the PS fan speed so it shouldn't matter.

                                • 28. Re: Calling all DG45ID users with Intel Desktop Utilities 3.1.3
                                  Flying_Kiwi

                                  R. Gibson wrote:

                                   

                                  I have also noticed that IDU does not recognize (or at lease indicate) what my various mobo temp sensors or fans are.

                                  IDU must be installed as an Administrator and if any other hardware monitoring software is installed, that may conflict with IDU. If it's your DG45ID you're referring to then as it works in this regard fine with mine (under Win 7 SP1 x64), it must be something specific to your setup.

                                   

                                  If you have a look in IDUs help, there's a section titled Fan Changes which provides instructions on how to redetect fans - are you saying this doesn't work (after the reboot) on your PC?

                                   

                                  As for your final paragraph, yes I've booted this PC from cold before and as soon as Windows was up, went into IDU and found one of the temperatures at 60 degrees or so - that's to hot to have warmed up so much in the minute or so from power on. I don't recall which temp it is (perhaps MCH?) but I can live with that if its down to a hardware limitation - IDU can't work miracles.

                                  • 29. Re: Calling all DG45ID users with Intel Desktop Utilities 3.1.3
                                    spearson

                                    I will try to respond to most of the questions/complaints...

                                     

                                    First, Flying_Kiwi...

                                     

                                    1. The boards that are listed as being supported are those that are included in our validation matrices. We simply cannot claim that Intel(R) Desktop Utilities (IDU) works with boards that we no longer validate it with. Boards are removed from our validation matrices once they have past their end-of-life date (i.e. we're no longer building them). This is the case for the DG45ID; it past its end-of-life date recently and thus was removed from the validation matrices. Now, if you ignore the G41 boards (which have never been supported by IDU), (a) not all 4 Series boards have past their end-of-life date (so we're still actively verifying support for them) and (b) all of the 4 Series boards are supported using the same software stack, so boards like the DG45ID should still, in theory, work with the new release (but you're not going to get it automatically downloading to your system).

                                     

                                    2. The BIOS tells us how much memory can be installed. If you are seeing an issue with the number being incorrect, this is a BIOS issue. I will report this to the sustaining team for the BIOS...

                                     

                                    3. The thresholds that are used to establish the green (normal), yellow (non-critical) and red (critial) sections of the gauges are, in this particular board's case, obtained from Intel(R) Quiet System Technology (QST). It appears that the QST configuration has the normal voltage range skewed a little too high. Since, in Intel(R) Desktop Boards, the BIOS owns the QST configuration, this is also a BIOS issue. I will report this to the sustaining team for this BIOS...

                                     

                                    4. In the 4 Series MCHs, there was indeed a limit to the range of temperatures that its sensor could expose. If I remember correctly, one version of the MCH had a lower limit of 66 degrees and the other had a lower limit of 59 degrees. Regardless, I wouldn't worry about it; the Tcontrol temperatures for these devices are way up in the vicinity of 100 degrees...

                                     

                                    Next, Mr. Gibson...

                                     

                                    1. The QST algorithms were specifically designed to improve system acoustics and psycho-acoustics (a fancy name for the concept of eliminating patterned noise that can become so annoying). Part of how it accomplishes this is by using a PID (Proportional-Integral-Derivative) algorithm. This is the same algorithm that most thermostats use to ensure that they don't turn your furnace on and off too often. There's lots of information on the web that you can read regarding PID algorithms, so I won't waste space discussing their theory here, but I will discuss impact. PID algorithms add time as a variable in the decision-making process. If temperatures are rising fairly quickly, the algorithm anticipates that the temperature will soon get too hot and will begin accelerating the fans earlier, in order to prevent any significant overshoot. If the temperature is rising very slowly, however, it will not begin accelerating the fans until it approaches its thermal limit. Either way, it will not let the temperature stray too far above the thermal limit. Now, the thermal limit is based upon the processor's Tcontrol temperature. By definition, this thermal limit is one that the processor can sustain for its entire warranted lifetime. That is, it could be this hot all of the time and you wouldn't need to be worried about it failing (or voiding your warranty). Bottom line, for your situation, the processor is just not getting hot enough to warrant the algorithms responding...

                                     

                                    Setting the response to "Aggressive" will not make a difference in the results that you see; this would only make a difference if the temperature is rising fairly rapidly. If you really, really want the processor to stay cooler (I realize that some people are more uncomfortable with higher temperatures) - and presuming that you are willing to put up with the louder fans that would result - the answer is to lower the temperature limit to some level below the processor's Tcontrol level. Yea, I know, we offer no BIOS support (like what I designed for the 6 Series board's BIOSs) nor any tools that you can use to make this change yourself. The tools I developed for QST were included in the chipset firmware integration package for board designers and thus were placed under NDA. I have been trying to isolate enough time to put one together that can be freely given out, but they are keeping me way too busy architecting support for next generation boards. If you are a software developer, there is an SDK that I published last year that you could use to develop such a tool; that's all I can offer at this point (I am still interested in developing this tool; its all a matter of time)...

                                     

                                    2. You should not be seeing IDU saying "Unknown/Other" for fans and you should *never* be seeing "Unknown/Other" for a temperature (since the board design predicates what they are). If you are, this is an indication that the data in the registry is incomplete or corrupted. This could occur if you have an issue (such as my 3.1.3.030 bug) that terminates the initialization process early. With the exception of 3.1.3.030 (which needs to be replaced, obviously), uninstalling and then reinstalling the application should fix this situation. If it doesn't, send me a message and we can try to figure out why. I am willing to bet, however, that you saw this issue mostly with versions before 3.1.2.022 (we fixed a number of issues in that release that resulted in similar symptoms). With 3.1.2.022, you might see this if you are using a RAID array on a 6 Series board. This was due to a bug in the 10.0 and 10.1 versions of Intel(R) Rapid Storage Technology (RST). We included a fix in 3.1.3.030 that addressed this from a stability standpoint, but at the cost of not being able to support HDD temperature monitoring (the forthcoming RST 10.5 release will address this issue)...

                                     

                                    ...Scott