Where Are The Gigabyte GA-N680SLI-DQ6 Motherboards?

[justinburton]

Blazer7,

I guess you didn't read the inquirer article: "We learned that the affected CPUs are the Core 2 Duo E4000/E6000, Core 2 Quad Q6600, Core 2 Xtreme QX6800, QX6700 and QX6800"

Also I believe the patch fixes the microcode. But I am not even sure what microcode is. Maybe really small code at the CPU level?

 

[fartucci]

Hi, I am new to this forum, and to Gigabyte motherboards. I just received a 680i-DQ6 Rev. 2 to replace my EVGA 680i A1 board. The reason for the replacement is because it is too difficult to cool the chipsets on the EVGA board. I have water cooling on the NB, but I have had trouble installing a water block on the SB, and the tubing becomes a mess since I am also water cooling the CPU and a pair of graphics cards. So, I am not using my SB water block, but instead am using the noisy EVGA SB fan, which hurts my ears and gives me headaches.

I am currently running my system at 1333 FSB, 3.33 CPU, with a QX6700. I have not had any problems OCing my FSB with the EVGA board, although I did RMA my original AR for an A1 because I am using a quad core.

I am wondering, does anyone know whether or not Gigabyte has fixed the FSB/quad core problem with the Rev. 2 board? I would like to run the Gigabyte board at the same speed as my EVGA board. Not a huge overclock, just 1333 FSB and 3.33 CPU. I am skeptical that problems with overclocking the FSB can be fixed with a bios update. It required a hardware fix with the EVGA boards. I am very much hoping that Gigabyte addressed this issue with the Rev. 2 board.

Also, I am wondering how hot the NB and SB get. I have water blocks for both, but I want to just use the passive cooling that comes with the board -- this is the main reason I am replacing the EVGA board. My NB right now runs at 31C with a Danger Den water block. The SB is much hotter, I think around 60, since the EVGA fan does not work all that well. I know that I cannot match my present NB temps using passive cooling, but I hope my chipset stays in the 50s. Can anyone report how hot their NB and SB get (with just the passive cooling, no fans attached)? By the way, I'm using a Mountain Mods U2-UFO case, so I'm getting pretty good airflow through the case.

Thanks.

 

[Blazer7]

Mr Justinburton

I have indeed read the inquirer article and it was me that mentioned it in the 1st place (see my 1st post). Q6600 is basically 2 E6600 put into the same package, and this is what puzzles me. Q6600 is affected but E6600 is not.

I'm still trying to figure out what this alleged bug is but found nothing more so far.

 

[justinburton]

Once again Blazer7...""We learned that the affected CPUs are the Core 2 Duo E4000/E6000, Core 2 Quad Q6600, Core 2 Xtreme QX6800, QX6700 and QX6800"

I read somewhere that all Core2Duo processors are affected. This also means all Core2Quad processors up to date are affected too, by definition.

There is no such thing as an E6000, they just mean anything that starts with E6XXX.

 

[TheBeagle]

Good Evening Mr. Blazer & Colleagues.

I'm not at all familiar with the workings of the Face Wizard feature so I will not even attempt to comment on that (a task better left for those who possess greater knowledge - of whom there are likely many). However, I can briefly comment on the very limited legitimate information currently being bandied about regarding the new series of 1333 FSB procs from Intel, and especially the E6850. If you look quite a way back in this Thread, you will see some notations by Gary indicating his preliminary experiences with a sample version of the E6850. From those comments, and other private discussions I have had with him, it would appear that the latest round of BIOS upgrades have fairly well resolved a few issues attendant to the new 1333 FSB DUAL-CORE procs. Notice I did NOT refer to the 1333 FSB QUAD-CORE procs!

The issues with the 1333 FSB Quad-Core procs are much more complicated than just the fact that there appears to be two (2) Dual-Cores dies placed on one 775 chip body. Some articles, like the ones referenced above, have begun to appear, in which some discussion about various matters associated with these high speed procs are now being discussed, and for the most part, most of those issues are being successfully addressed. Simply stated, we don't yet know if the Rev 1 boards are going to effectively operate a Quad-Core 1333 FSB proc, let alone allow it to be reasonably subject to OC. However, the "fish or cut bait" point is now on the horizon, and the various component manufacturers (board makers, chipset makers and Intel) are going to have to tell us the final truth on these matters rather shortly.

Quite frankly, the lack of definitive information of Quad-Core 1333 FSB performance on our Rev 1 boards, is a bit disconcerting at this point in time. However, if it doesn't work, I suspect those corporations are likely to be visited by swarm of lawyers with class-action lawsuit papers in their sweaty palms; the likes of which will make a swarm of hungry locusts look tame at harvest time! In such a dreadful scenario, it will at least be some consolation to watch each of the entities point the finger of blame at the others, kinda like a feeding frenzy when sharks smell blood.

However, I digress. Back to the E6850. All indications are that the E6850 proc may just be the outstanding CPU for this board, since it might get close to 4.0 GHz on good air cooling. After all, the E6850 is really just a mature E6600, clocked at a high speed. Indications are that there is likely to be quite a bit of headroom available on the E6850, in combination with an equally mature BIOS and very good memory. As for my immediate choice, although I would have possibly sprung for a Quad-Core 1333 FSB proc, I'm not going to do that until I'm reasonably sure it's not going to give me a bad case of heartburn brought on by the frustration of learning that all those entities screwed up, and that proc won't work effectively.

Therefore, I find that the E6850 is a quite reasonable alternative, and since there aren't any significant apps or games that will be able to tap into that Quad-Core architecture in the near future, a somewhat lesser, but nevertheless powerful new proc will do nicely. Enjoy. TheBeagle

PS. Thanks for the generous comments about the genuine hospitality on this Thread and the self-policing by the contributors and moderators to keep the flamers at bay. We all try to do our part, and that has seemed to be a bit of the "magic" on this Thread.

PSS. Hello Mr. Justin. Seeing as how Mr. Blazer is a newly acquired colleague, might we give him a wee bit of slack on the tone of our responsive rhetoric? Just a suggestion.

 

[justinburton]

My apologies Blazer7.

 

[sskmercer]

Originally posted by: TheBeagle

The issues with the 1333 FSB Quad-Core procs are much more complicated than just the fact that there appears to be two (2) Dual-Cores dies placed on one 775 chip body. Some articles, like the ones referenced above, have begun to appear, in which some discussion about various matters associated with these high speed procs are now being discussed, and for the most part, most of those issues are being successfully addressed. Simply stated, we don't yet know if the Rev 1 boards are going to effectively operate a Quad-Core 1333 FSB proc, let alone allow it to be reasonably subject to OC. However, the "fish or cut bait" point is now on the horizon, and the various component manufacturers (board makers, chipset makers and Intel) are going to have to tell us the final truth on these matters rather shortly.

Quite frankly, the lack of definitive information of Quad-Core 1333 FSB performance on our Rev 1 boards, is a bit disconcerting at this point in time. However, if it doesn't work, I suspect those corporations are likely to be visited by swarm of lawyers with class-action lawsuit papers in their sweaty palms; the likes of which will make a swarm of hungry locusts look tame at harvest time! In such a dreadful scenario, it will at least be some consolation to watch each of the entities point the finger of blame at the others, kinda like a feeding frenzy when sharks smell blood.

There is plenty of info about the lack of Quad Core O/C problems around the place, my biggest gripe is the fact that Gigabyte wont own up to it, say like eVGA did. They realised ther was a problem and fixed it, then did what i suspect alot of poeple wouldn't expect and exchanged everyones boards.

The unfortunate thing for us Rev1 owners and possibly Rev2 owners is the problem is at a hardware level, as there is a few sites already that give detailed instructions on how to Modify your board with the req' transistors needed to feed the Quads with the juice they need.

There will possibly be some interesting times ahead regarding the Lawsuit side of things, though i have a fealing that it's going to be more confined to the USofA unfortunately for those of us that dont live there. This of course is no guarentee, but here's to hoping atleast something will be done about it in some way (Glad i keep all my reciepts.).

@Beagle that chip is definately looking the goods from reports so far. What do you mean by the highligted/underlined bit.

Also i would love to know what this Microcode update actually does. I understand it's something to do with the underlying CPU marchitecture, but would be interesting to know what it's about. The MS link describes improved stabilty, but i've never had a problem in that regards.

Cheers

 

[Blazer7]

Mr Justin,

No need to apologize. It was really my bad as despite my years of experience I completely missed the E6xxx point. Thank you very much for taking the time to reply to me a 2nd time.

------------------------------------------------------------------------------------------------------------

Mr Beagle,

Thank you also for sharing with us your insight regarding the E6850 cpu.

------------------------------------------------------------------------------------------------------------

If anyone has any clue as to how face wizard interacts with the various BIOS files, any help will be appreciated.

------------------------------------------------------------------------------------------------------------

***** Here is something new regarding our ?bug?

http://www.rage3d.com/board/sh....php?threadid=33889730

 

[TheBeagle]

Hello Mr. SSK.

If you are asking me to explain the meaning and usage of that phrase, I'll give it a go as follows. The phrase "fish or cut bait" is a colloquial saying that is intended to convey the idea that someone can't sit around in a boat, doing nothing, on an indefinite basis. I believe that it comes from a time when men went "down to the sea in boats," and actually made a living by means of manually casting lines into the sea for fish.

In such times, each man was expected to do his part, and so while onboard, he either had to fish, or cut bait for the others to fish - but doing nothing was not an option. Over the years, the phrase has evolved, at least in the USA, to have a wider meaning, and can be used in many different circumstances, including a circumstance in which a person or entity who has heretofore been doing nothing about their responsibilities concerning a particular matter, will be forced to do so in a very short time. One source has explained the phrase and its origins it this way:

This phrase is of US origin. To cut bait means to stop fishing. It appears to have been introduced to the public consciousness, and may well have been coined by, US Judge Levi Hubbell. It came up in 1853, in a legal dispute over land ownership between US Attorney General Caleb Cushing and a William Hungerford. Cushing was displeased with Hubbell's conduct of the case and threatened to have him impeached. Hubbell's response was: "Judge Cushing has commenced a suit in the United States Court. Judge Cushing must either fish or cut bait."

And used in the manner in which I stated, it likewise means that this issue about Quad-Core performance in our boards is about to become acute, since we are in the advent of the imminent release of the 1333 FSB Quad-Core procs, and that someone will buy one, and it may not perform "as advertised." Then, and if that non-performance occurs, according to the meaning of the phrase, Gigabyte will be finally forced to acknowledge that their product, which they advertised as being capable of Quad-Core 1333 FSB performance, doesn't deliver. Thus the onslaught of lawyers with class-action lawsuits.

BTW, there are a number of ways in which folks who live in other places around the globe might be able to be parties in such lawsuits, and be able to have their rights protected under American laws in such lawsuits. Gigabyte has a major corporate presence in the USA, and therefore can have a lawsuit brought against it in the USA, by anyone (citizen and non-citizen alike) who feels they have be defrauded by Gigabye's conduct. In class-action lawsuits brought in the USA, the successful parties' lawyers get paid by the other party (Gigabyte), and thus those lawyers will be very pleased to include anyone and everyone whose rights have been violated, since at least part of the lawyers' professional fees ($) will be based on the number of parties in the successful class of plaintiffs being represented by them - the more, the merrier!

Sorry if my use of that phrase was unclear. And I really didn't mean to get into a short diatribe about this whole thing again. However, I am more than a bit displeased and disappointed in Gigabyte's silence about all this. That's either a display of non-loyalty to its customers, or raw corporate arrogance, probably some of both! Best regards. TheBeagle

 

[cdfire]

Well guys, I can't say for sure yet, but I did borrow a Samsung 906BW monitor from a friend and well, so far everything seems to be working good. Will keep using this monitor untill the weekend and if still no issues, then I will purchase a new monitor. (Man This Samsung Is Bright)

Thanks again for the help. BTW has anyone installed that Core2 updated posted. Just wondering if I should do so or not. Don't want to be messing anything up now that things are working again.

 

[dlbetz]

I installed the Microsoft update I didn't notice anything bad or good. The artical did say it was rare to effect someone. but better safe then sorry. Just my opinion.

 

[Fallen Kell]

Originally posted by: justinburton
Blazer7,

I guess you didn't read the inquirer article: "We learned that the affected CPUs are the Core 2 Duo E4000/E6000, Core 2 Quad Q6600, Core 2 Xtreme QX6800, QX6700 and QX6800"

Also I believe the patch fixes the microcode. But I am not even sure what microcode is. Maybe really small code at the CPU level?

A microcode update is an update the the actual CPU instructions and how they operate. Things like ADDSUBPD (Add Subtract Packaged Double), or HADDPS (Horizontal Add Packaged Single). These are the CPU instructions themselves that all programs are converted (compiled), into which inform the CPU what tasks to perform. A microcode update is an update the the process that the CPU performs when it is told to perform one of these instructions.

If you still don't understand what I am saying, continue reading.

A brief overview of how computers work. The CPU is now a mix of hardware and software which understands how to perform a fixed set of instructions. These instructions are the "CPU Instruction Set". For an Intel Core 2 processor, these are include the x86 instruction set, the EM64T instruction set, SSE, SSE2, SSE3, SSSE3, and others. The motherboard BIOS directly interfaces with all the hardware on the motherboard like the different system buses and CPU. This BIOS has a set of defined functions for checking the hardware exists and is working properly as well as knowing how to boot and load an operating system onto the system from different devices and modes that is has access to (i.e. boot off floppy, CD/DVD, USB, network, IDE, SATA, etc., etc.). The operating system has a kernel which know how to interface with the hardware architecture that it was designed to run on (i.e. 32 bit x86, 64 bit, SPARC, SPARC64, etc., etc.). The operating system kernel knows uses only the basics of the know instruction set to be most compatible with different CPUs. But it also can then load drivers for access to more complex functionality. The programs then run on top of the operating system kernel and have access to the computer hardware through the operating system kernel. Most applications are programmed in a higher order computer language like C, C++, C#, VB, JAVA, etc., etc., etc... Most of these languages are then compiled using a compiler into the machine code for the specific CPU architecture it is designed to run on. The compiler is told what CPU you want to compile the program for, and it knows of the different instruction sets that CPU is aware of and then converts the higher level language in to a fairly optimized machine level code (much like translating something from Spanish to English). The CPU only knows how to work with machine code, not the higher level language we prefer to program in nowadays. With the translation having been made, the code can be run by the CPU.

Now on top of all this, modern day CPUs being a mix of hardware and software, have many higher level instructions which are really a grouping of other instructions. These are to help make the people who write the compilers have easier jobs by taking a group of things that are commonly done and make them a single instruction to the CPU instead of having to send the full combination. This is where the microcode updates come into play. The microcode update changes how the CPU executes these grouped instructions. However, the CPUs can not be permanently updated since you can not write to the section in the CPU where these instructions are stored. However, we can override the instructions by placing the override information into a special section in the CPU, however this is not permanent memory, and will be lost when the CPU loses power. A BIOS update is the ideal way to apply a microcode update as it will send the update to the CPU each and every time the system gets turned on. However, you can also send a microcode update from the OS kernel. But, the problem with this method is that it is a different patch that would be needed to be written for each and every operating systems that may exist. For instance, Windows 2000, 2003, XP, XP Pro, XP Business, OS X, Sun Solaris 9 x86, Solaris 10 x86, Red Hat Enterprise Linux 3, RHEL 4, etc., etc., etc. In other words, hundreds of different operating systems exist which can run on the x86 platform which this CPU is a member. Which means it is best to have the motherboard manufactures issue a BIOS which contains the microcode update.

 

[fartucci]

Originally posted by: fartucci
Hi, I am new to this forum, and to Gigabyte motherboards. I just received a 680i-DQ6 Rev. 2 to replace my EVGA 680i A1 board. The reason for the replacement is because it is too difficult to cool the chipsets on the EVGA board. I have water cooling on the NB, but I have had trouble installing a water block on the SB, and the tubing becomes a mess since I am also water cooling the CPU and a pair of graphics cards. So, I am not using my SB water block, but instead am using the noisy EVGA SB fan, which hurts my ears and gives me headaches.

I am currently running my system at 1333 FSB, 3.33 CPU, with a QX6700. I have not had any problems OCing my FSB with the EVGA board, although I did RMA my original AR for an A1 because I am using a quad core.

I am wondering, does anyone know whether or not Gigabyte has fixed the FSB/quad core problem with the Rev. 2 board? I would like to run the Gigabyte board at the same speed as my EVGA board. Not a huge overclock, just 1333 FSB and 3.33 CPU. I am skeptical that problems with overclocking the FSB can be fixed with a bios update. It required a hardware fix with the EVGA boards. I am very much hoping that Gigabyte addressed this issue with the Rev. 2 board.

Also, I am wondering how hot the NB and SB get. I have water blocks for both, but I want to just use the passive cooling that comes with the board -- this is the main reason I am replacing the EVGA board. My NB right now runs at 31C with a Danger Den water block. The SB is much hotter, I think around 60, since the EVGA fan does not work all that well. I know that I cannot match my present NB temps using passive cooling, but I hope my chipset stays in the 50s. Can anyone report how hot their NB and SB get (with just the passive cooling, no fans attached)? By the way, I'm using a Mountain Mods U2-UFO case, so I'm getting pretty good airflow through the case.

Thanks.

I am hoping that someone has some thoughts on the following three questions:

1) Did Gigabyte fix the FSB/quad core overclocking issue with the Rev. 2 board?

2) What are the temps for the NB and SB using just the passive cooling that comes with the board?

3) If the FSB/quad core overclocking issue has not been fixed, are there any difficulties with bumping up the CPU clock in a quad core by increasing the CPU multiplier?

I am deciding whether or not I will keep this board. If it will not perform as well as my EVGA board, then I will just keep the EVGA board and try to install the SB water block again. If I can run the Gigabyte board at 1333 FSB, 3.33 CPU, and still have acceptable NB and SB temps, then I will install the Gigabyte board.

Thanks

 

[cdfire]

Hey Fartucci and welcome to the forums. I don't think many here can answer you questions for most of us have the REV 1 board. I can tell you the north bridge gets warm on the Rev 1 and the south bridge is hotter then, well lets just say hotter then it is way down south where the guy with horns lives. But like I said thats REV 1. Sorry I can't help you with rev 2 info, but maybe someone else can help you out.

BTW I installed that update patch thingy. Don't notice much differance but so far it did not cause any issues that I know of.

 

[Blazer7]

Same here cdfire. Patch applied and everything seems to be working fine. One thing though, the post on rage3d is dated April 29 with a mention that Intel had released an update for the problem around April 10. This means that Intel knew of it before then. The patch in microsoft's site is dated June 11. That's a 2 month gap (at least). Nice... :shocked:

 

[darren1394]

1) Did Gigabyte fix the FSB/quad core overclocking issue with the Rev. 2 board?

some here have rev2's however, without quadcores... I'm wondering if someone posted pics of rev1 vs rev2. we might be able to see if the quadcore volt mod was added.

 

[Blazer7]

Mr cdfire,

I know that this post comes may-be too late but you may be interested to know that you are not the only one that had this Hz experience with your monitor. I had a similar experience with mine also (LG L1980Q).

My monitor can do 60 when Digital connection is used (DVI) & 75 when analog is used. At some time XP reported that I was doing 75 when digital which was quite odd. As the monitor?s specs are for 60 I was more than sure that something was wrong. To fix that I had to remove nvidia?s drivers & also the monitor from the device list. After reinstalling the monitor?s drivers and nvidia?s latest everything worked fine. My understanding is that, may-be, at some point I switched from analog to digital and the 75 value remained in the system. This could have been due to a driver or OS issue. After all, this kind of things are not unheard-of.

In order to be sure that the monitor & VGA were completely removed from the system I used HP?s System diagnostics utility as this seems to be able to remove keys from the registry that may have been left there even after a device has been ?removed?. This utility is also able to remove registry keys that are left in the system by previous devices that have been replaced by new ones (like cpus, VGAs, monitors etc). You can find this utility easily in HP?s site.

As for the 75Hz many believe that this is a must. This is primarily because it is common knowledge that the human eye ?operates? at ~72, so when doing more everything appears stable (flicker-free).

When talking about CRT monitor?s I couldn?t agree more. You can test it yourself. Find a CRT (if possible ) and set it below 72. You?ll most certainly see some difference. But this is hardly the case with modern TFT?s. Many of the TFT?s out there, don?t do more than 60 when in native resolution. And this is almost the rule when dealing with notebooks, but still, they don't suffer from this kind of problems.

Like Smitty I believe that there is nothing wrong with your hardware. In any case if you are not happy with your monitor, then listen to the Beagle?s advise. If I was to change my monitor now, and Hz was a factor, I would most certainly go for something that can do no less than 75 at the monitor's native res while connected to DVI.

I hope this helps.

--------------------------------------------
GA-N680SLI-DQ6 rev 1- BIOS F5b
C2D E6600@stock
4x 1GB DDR2-1066@4-3-3-5-2T
ASUS EN8800GTX - 768MB - 158.22
LG 1980Q
WinXP SP2

 

[Blazer7]

Inquirer article - Intel explains the Core 2 CPU errata

http://www.theinquirer.net/default.aspx?article=40627

---------------------------------------------------------------------------------------------------------------------

Linus Torvalds post on Core 2 Errata

http://www.realworldtech.com/f...hreadid=80534&roomid=2

 

[bigbillyt99]

Okay, I managed to resolve the issues I was having with the NVIDIA RAID controller and the associated RAID 5 array.

Unfortunatley, I did 3 things at once (only two of which could have been the actual fix).

First, a quick backstory for all this:

1. I have a nearly identical machine sitting next to this one. The only differrences are:
a.) Intel D975XBX2 motherboard
b.) E6600 C2D proc as opposed to the Quad Core
c.) 4gb Kingston RAM instead of 8gb of the Patriot memory

Things that are the same:
a.) Same Seagate SATA2 7200.10 drives in RAID 5 array on an onboard RAID controller
b.) Windows Server 2003 x64 OS
c.) Power supply
d.) Running large VMWare environments that put a pretty heavy IO load on the sytem (although in this case successfully)

So, in my mind at least, this pretty much rules out the specific model hard drives as the culprit. Just because they are not specifically designed for RAID use does not mean they are entirely unsuitable.

Also, I tried running the 4 drives as RAID 5 using just software RAID in the OS after disabling the RAID functionality on the NVIDIA SATA controller and experienced similar problems (this time as event ID 11 in the system event log). This made me think the NVIDIA SATA controller was behaving flaky in a more general sense than just the RAID issues might indicate.

So, I was at the point where I was ready to rip out that motherboard and put it on Ebay. Unfortunately, I had tested the memory in the Intel board already just to see if it would work and it did not. This is probably the thing that compelled me to go through a few more hoops before giving up.

Anyway, I applied the microcode update from Microsoft for the processor AND updated to the newer F5b BIOS provided by our most wonderful freind Gary. I followed the procedure outlined earlier in this thread exactly.
1. Jumpered the Clear CMOS pins for a good 5 mins.+
2. Rebooted and let the BIOS go to default settings
3. Rebooted and went into Q-Flash and applied the F5b BIOS to both Main and backup BIOS
4. Rebooted and applied desired settings to BIOS

BAMM!!!! I rebooted once more, hit F10 added all 4 disks to RAID 5 array ( I had removed data and partitions beforehand). Logged into Windows and gave the array a drive letter in disk manager, along with a quick format this time, as I have become impatient now with all this. Copied over 4 VMWare instances of W2k3 x64 with a DC and Exchange 2007 and fired 'em up!

No errors!! Not a spec of red in my event log. I then ran load generator for about an hour against the Exchange mail store and no problems!!!

I feel pretty confident this is fixed now. But just in case I have 8gbs of Kingston on the way for the Intel machine anyway.

Only a couple of interesting sidenotes / questions still lingering for me.
1. The whole system, particularly the processors seem to run much cooler now. I doubt just the new fan did this as it does not seem to move that much more air. Is it possible this is a result of just the new BIOS?
2. The processor now reports in PC Wizard 2007 and CPU-Z that it is running at 1599 - 1600mhz until a load is applied then it speedsteps right to 2400mhz. It was not doing that before these updates. Is the a result of the "microcode" update, from Intel's description of what changed that seems unlikely. It is not impossible that I just did not notice this earlier, but I doubt it. So, is this a result of the different BIOS?
3. Lastly, is it possible, or likely, the "microcode" update for the proc is in the new BIOS update as well? Might I have applied it twice?

Either way, I am a happy camper. Another big thanks to Gary for the update.

Oh, by the way, Gigabyte support responded to my support inquiry with a few additional questions. I guess I need to let them know this is fixed. They need to post the F5b BIOS up on their site.

System also now runs WAY cooler overall!

Temps (according to PC Wizard):
Before updates: CPU 111-116F Cores 122-132F Mainboard 108-111
After New Fan and New BIOS: CPU 108-111F Cores 116-128F Mainboard: 85-92F
I also did a touch test on the north and southbridge heatsinks upon rebooting both before and after the changes and there does seem to be huge difference. The first time they were very warm to the touch, now, not so much.

NEXT: A little OC action! I will let you all know how that goes. I am a straight up noob when it comes to overclocking so this will be interesting.

One other tip: I have the Intel based system in a Coolermaster Stacker 810 and the Gigabyte system in a CM Stacker T01. The Coolermaster stacker 810 case is a significantly better design over the T01, at least as far as the thermal properties go. The two 120mm fans really keep the inside of the case cool , and I doubt I will ever need a second power supply anyway. So, I may move the considerably warmer quad core system into the other case at some point. If I should muster an energy surplus at some time in the future.

 

[cdfire]

<div class="FTQUOTE"><begin quote>Originally posted by: Blazer7
Mr cdfire,

I know that this post comes may-be too late but you may be interested to know that you are not the only one that had this Hz experience with your monitor. I had a similar experience with mine also (LG L1980Q).

My monitor can do 60 when Digital connection is used (DVI) & 75 when analog is used. At some time XP reported that I was doing 75 when digital which was quite odd. As the monitor?s specs are for 60 I was more than sure that something was wrong. To fix that I had to remove nvidia?s drivers & also the monitor from the device list. After reinstalling the monitor?s drivers and nvidia?s latest everything worked fine. My understanding is that, may-be, at some point I switched from analog to digital and the 75 value remained in the system. This could have been due to a driver or OS issue. After all, this kind of things are not unheard-of.

In order to be sure that the monitor & VGA were completely removed from the system I used HP?s System diagnostics utility as this seems to be able to remove keys from the registry that may have been left there even after a device has been ?removed?. This utility is also able to remove registry keys that are left in the system by previous devices that have been replaced by new ones (like cpus, VGAs, monitors etc). You can find this utility easily in HP?s site.

As for the 75Hz many believe that this is a must. This is primarily because it is common knowledge that the human eye ?operates? at ~72, so when doing more everything appears stable (flicker-free).

When talking about CRT monitor?s I couldn?t agree more. You can test it yourself. Find a CRT (if possible ) and set it below 72. You?ll most certainly see some difference. But this is hardly the case with modern TFT?s. Many of the TFT?s out there, don?t do more than 60 when in native resolution. And this is almost the rule when dealing with notebooks, but still, they don't suffer from this kind of problems.

Like Smitty I believe that there is nothing wrong with your hardware. In any case if you are not happy with your monitor, then listen to the Beagle?s advise. If I was to change my monitor now, and Hz was a factor, I would most certainly go for something that can do no less than 75 at the monitor's native res while connected to DVI.

I hope this helps.

--------------------------------------------
GA-N680SLI-DQ6 rev 1- BIOS F5b
C2D E6600@stock
4x 1GB DDR2-1066@4-3-3-5-2T
ASUS EN8800GTX - 768MB - 158.22
LG 1980Q
WinXP SP2
</end quote></div>

I must have misread the View Sonic Specs, but I could have sworn it said default 60 yet it supports 75. Like I said I ran it at 75 for months, not only on this pc but on my old pc.

In any event, on this Samsung 906BW, with the latest Monitor Drivers, it defaults to 75Hz right off the bat even though Samsung recommends 60, it does state it supports 75, in fact unlike the View Sonic where 75 was hidden, the Samsung shows 3 refresh rates to chose from without un-checking the hide non supported box in display properties. Every thing runs great again. I have not had one issue (Knock On Wood) since installing the Samsung monitor, so I am pretty sure that View Sonic just could not handle it any more or like you said maybe a clean install of drivers was needed, but I did format several times with the same issue.

I know what you mean about CRT and refresh rate. The thing is I can also tell a big difference between 60 and 75 on a LCD. For one, things are brighter and colors in games are much better. Also in games, it seems to me at 75 things are much clearer and smoother. As for desktop and web, I really don't notice much difference between 60 and 75. It may be a personal taste type thing, but 75 just seem to work better for me.

Anyways, thanks for the reply. I love these forums and the good people here. It?s unlike any I have visited before. Good to be here and sharing and helping with info with everyone.

 

[Blazer7]

Mr bigbillyt99,

I'm also using cpu-z, especially when fine-tuning my RAM or whenever I'm applying a new BIOS and have to re-apply my "default" settings. I've also noticed the 1600-2400MHz idle/load variation in cpu-z readings you describe. I have recently applied Microsoft's microcode patch as well but this 1600-2400 thing started before that, just after I applied BIOS ver F5b. I'm sure that cpu-z didn't show anything else but 2400MHz with any BIOS ver before F5b and I have used every BIOS I could get my hands on after the initial F2.

I've also noticed my cpu temp dropping by 1-2 degrees without having changed anything else in my system. C2D E6600@stock + Arctic Silver5 + Thermalright SI-128 + Thermaltake A2018 120x120x25 Smart Case Fan. I'm now doing cpu 22/23C - system 45C with environment temp @30C. This is nothing dramatic but unless it is a direct result of the cpu working at a lower frequency when idle, I can't explain it.

May-be this BIOS fixed something that allows the cpu to run, or run longer @1600 when idle, thus the temp drop.

Anyone else notice anything similar ? Can anyone post a list as to what has been changed with BIOS F5b ?

 

[cdfire]

CPU-Z bounces between 2399 and 2401. Looking at core temp, my temps are down to around 23 to 24, normally there at about 27 to 28 idle. (Before F5B 27 to 28)

 

[justinburton]

If your theory of the magical temperature drop is true, then if you uninstall the patch and the temps should climb back up. Same with the BIOS, you can check if the temps go up by reverting back the the old bios.

Tonight I will monitor my temps before and after the patch and see if the rumors are true.

 

[eklock2000]

At Blazer and BigBillyT99...

The CPU throttling is due to an EIST setting in the BIOS, I think it's under Advanced - CPU EIST Function, but someone else may know exactly which one it is. I have not noticed any performance decrease by having it on, and the CPU throttles right up to 2400 when any load is applied. I am not overclocking tho, and most on this thread that are seem to have EIST disabled.

Peace,

EK2K

 

[cdfire]

I uninstalled the patch just to see and that did not change temps at all, I do have the A/C cranked up, but thats normal for me. As for the BIOS, I am going to leave it at F5b. Been a couple weeks since I have got to enjoy my pc and with my luck, I am better leaving that alone for now, but I am guessing if there is a temp change, it's BIOS related, not the patch.