Printable View
Don't expect all E6600's to do 3.6GHz stable at all - any voltage.
My week 23 wouldn't do it. My week 27 does.
What about 3.4GHz? What does it take to get there? I think 3.4GHz is a great OC - an even 1000MHz OC.
In fact, I have since backed mine down to 3.4GHz as my 24/7.
Now I just need to figure out how to cool and OC my new 8800GTX.
My week 26 does OC at 3.6 GHz @ 1.4v after c/o BIOS 1506.:toast:
It is suggested that this BETA BIOS 1602 includes:Quote:
Originally Posted by Psitech
-Fix CPU Temp (previous versions showed too low temperatures)
-Update Kentsfield Quad Core uCode
However, some users determined and reported that the Fix CPU Temp actually has NOT been fixed the temp monitoring problem!
I'n not in for this beta-bios-version!
You're talking OS level again. The BIOS does not know what OS you are running and cannot dynamically determine that. I could be running Ubuntu Linux for all it knows. The point is the OS did not determine what this problem was nor did it cause it. The problem was never at the OS level, which you keep insisting is the problem.Quote:
Originally Posted by JokerCPoC
i am struggling for 3.3ghz. i mean, sure its 900mhz and all (IF i can stabilize it) but i mean come on, i didnt buy a P5W DH and Team memory to get the same overclock as those on E6300 + ASrock boards, etc... yet again, premium hardware + bad luck = bad luck...Quote:
Originally Posted by Brahmzy
Hi everyone, sorry for my english it’s not very good. I don't know if this is really the place for this post but the title of the thread says Problems + Fixes... so I think it maybe right here, if not, tell me please.
It’s been one month since I’ve built my new computer and I have a problem which I couldn’t have fixed yet.
My systems specs are:
CPU: Intel Core2Duo e6600
MotherBoard: Asus P5W DH Deluxe
RAM: 2x1GB DDR2 TeamXtreem 800Mhz (4-4-4-10) – Before I had 2x1GB DDR2 GEIL 800Mhz (5-5-5-15)
Primary HDD: Western Digital Raptor 74GB
Data HDD: 2 x Western Digital 320GB in Raid 0
Power Supply: Hiper HPU-4M670-SU (Version 1) – Before I had Tacens Valeo Pro 600W
Graphic Card: Ati Sapphire X1900XT 512MB
The problem:
When I turn on my computer it makes 4 beeps (1 large and 3 short), and I get no POST screen (no signal on the monitor) but the fans are working. If I let it run a while in this mode and then restart it the computer boots well and then I can play, work, etc, with no problems. (Sometimes it takes me up to 3 or 4 restarts to boot it well).
Regarding to AMI bios beep codes: 1 large and 3 short mean: Memory test failure A fault has been detected in memory above 64KB.
I’ve tried inserting just one stick of memory in every slot of the motherboard (I’ve tried with all sticks I have). I thought it was a memory problem so I bought a new pair of Team Xtreem memory (before I had GEIL memory) and the problem persists.
I’ve also tried updating bios (first I had 1201, then updated to 1407 and yesterday I’ve updated to 1506). But nothing of this solved the problem.
I thought it could be a PSU problem so I have also tried changing the PSU (before I had Tacens Valeo Pro 600W) now I have Hiper HPU 4M670 which is supposed to be compatible with my graphic card
My computer passed satisfactory all tests I have run on it (BurnInTest Pro, Memtest, 3DMark06 and PCMark05).
The symptoms:
The matter is that all I have described before just happens when the computer has been some time turned off, I mean when it is working, if I turn it off and in five or ten minutes I turn it on, it boots well. If I turn it off and half an hour later I turn it on the computer makes the beeps and doesn’t boot well.
I don’t know what else to try or to do, the problem is still there and I would like to fix it as soon as it is possible.
I would appreciate your help.
Maybe I missed it, but what operating system are you using? Isn't this a problem (for those watching the thread) that could have something to do with RAM Throttling? It seems people have this setting Disabled
In the DRAM settings I'd also suggest running the memory at SPD - Enabled
make sure the ECC is Disabled
And I would drop the memory divider to 533MHz just to see what effect this has.
Also what FSB speed are you running? 266MHz 333Mhz?
By the way WELCOME and your English is excllent, very, very, concise post :)
I have the same problem. I have 2 drives in RAID0 on the SATA 1 & 3 connectors, a burner on the Intel IDE port, and added 2 IDE drives on the JMicron IDE ports.Quote:
Originally Posted by Clump
Everything works fine until the system goes into 'Standby' (sleep) mode. When it resumes, the drives on the JMicron IDE port have disappeared! The only way I can get them back is to reboot.
I had the 1305 BIOS, switched to the 1503 BIOS and same issue.
I did the Microsoft 'hotfix' and it did nothing.
Any suggestions?
Try reading the Manual - Asus does recommend SATA1 for Optical Drive...Try your RAID at SATA3+4. Maybe that will fix something, but dont expect too much. Jmicron is crappy - u better disable it and buy ANY cheap PCI-IDE-Controller if u really need the IDEdrives.Quote:
Originally Posted by OldCoot
iLL
Ha ha I like that line Premium Hardware + bad luck= Bad Luck..Quote:
Originally Posted by AndrewZorn
That's great..
I hear ya, that's what usually happens to me aswell..DAMN:mad:
Hi Liquid3D, I'm using Windows XP Professional. Ok I'll disable RAM Throttling (I hope it works :)).Quote:
Originally Posted by Liquid3D
I've tried before with SPD enabled and the problem persists :(Quote:
Originally Posted by Liquid3D
I'll try to change the memory divider too. I'm running FSB at 266Mhz (I haven't done any overclocking on the computer).Quote:
Originally Posted by Liquid3D
Thanks!!! :)Quote:
Originally Posted by Liquid3D
I trotally missed the SATA RAID respsones that looks to be much more of an astute assesment :)
thorlak, it sounds like the only thing that's changed is your MEMORY... you might want to check whether memory voltage is set to "Auto" in BIOS. If it is, overwrite the Auto setting with the rated voltage of your memory. Keep SPD enabled too. Let us know if this works.
Ever since installing a new PSU (an 800W Tagan Dual Engine) and updating to BIOS 1506 I have observed something that did not happen before. When I overclock my E6600 to 3ghz (with the memory voltage at 2v to get 4-4-4-12 at DDR2-833/835) the CPU fan seems to kick in at full pelt when being overclocked, which is something that never happened when I first overclocked many moons ago under BIOS v1101 or something.
Of course it is nice to see the cold boot problem seems to have been sorted but this constant whirling of the fan at full pelt can't be good - even when overclocked the CPU is not too hot and particularly not so given that winter is fast approaching. Is there any way to make the CPU speed lower for idle operations or am I stuck with it like this when overclocking?
I don't know why this is happening. You can use Speedfan, an excellent freeware application to control the fans connected to your motherboard. You can easily find and download it if you google for it.Quote:
Originally Posted by DarthVader
How can you set your memory timings to 4-4-4-12 while overclocking? Don't you have to set memory timings to "by SPD" in order to go past 360-370 fsb? I guess you are using an FSB lower than 360, right?
If it makes you feel any better I bought a P5W64 (~£200), E6700ES (~£400) and modded Vapochill LS (~£450) and with it I'm getting 4.3Ghz so far. Yet, in 3DMark I get pretty much exactly the same score as running it at 3.6Ghz, so I've "wasted" money too. :(Quote:
Originally Posted by AndrewZorn
Aye I've set my FSB to 334mhz and set the memory voltage to 2v (it's corsair stuff so it should do 2.1v with no problems so I hear). I have the timings set manually rather than by SPD.Quote:
Originally Posted by Pol1977
Sorry to jump into this, but why are you assuming a 360 or lower FSB with those timings? I'm running fine at 400 FSB with 4-4-4-12...Quote:
Originally Posted by Pol1977
As far as I know Dual Engine for the 800W model (indeed all Dual Engine models) refers to Dual transfomers, one transformer handles the 3.3V 12V1, and 12V3 "taps". The other transformer the 5V, 12V2, and 12V4 "taps". The rail assigment may be such that raising the memory voltage and Vcore taxes the 3.3V, 12V1 and 12V3 activating the cooling fan.Quote:
Originally Posted by DarthVader
As an experiment I'd try leaving the overclock as is, and simply dropping the memory voltage as well as memory timmings on SPD and Auto respectively. If this stops the fan spin-up then you've found your no longer over taxing those particular rails, hence that transformer and it's simply a design aspect of the Tagan Dual Engine. The 800W certainly isn't their lowest power rated Dual Engine model, however; they may have made an error in simply reducing power without re-assinging transformer tap assigment as wattge dropped. In other words that combination may be fine for the 1KW model, but at lower voltages certain rails might have been assigned to different transformers for a more effective design.
I'm getting a little in-depth on the PSU design, however; the fact the fan is spinning up is a result of a temp increase somewhere in the PSU. If it were a temp increase from other components in your system in which the PSU was then drawing that heat air through it, the fan RPM increase wouldn't be immediate, but have a slight delay.
Most memory sticks have what they call "safety settings" programmed in their SPD. My GSkill 6400 4-4-4-12 ram "works at 2.0-2.1 V at those settings but has SPD programmed to 5-5-5-15 for people not knowing how to set manual voltage and timings and for crappy motherboards" (quote from GSkill technical support).Quote:
Originally Posted by kaiguy
So, I assumed that to have 4-4-4-12 you must be able to set the timings manually, something you cannot do if your FSB is higher than 360-370 MHz with the BIOSes floating around at the moment.
If your SPD is set at 4-4-4-12 there is no reason you can't run at these settings and have 400MHz FSB :)
Thanks for the explanation. I'm actually surprised then that my sticks have the 4-4-4-12 setting in SPD, but I guess it's possible. I have my BIOS set for manual timings at 400 FSB, but maybe it's just reverting to SPD, or, somehow, I'm just lucky with the revision pf my mobo.Quote:
Originally Posted by Pol1977
Food for thought, I guess.
MY sticks(Crucial Tracer) are not SPD programmed for 4-4-4-12 400mhz, but are EPP profiled for such(SPD is 5-5-5-18 400mhz/4-5-5-18 333mhz). maybe that's what make the difference...I have no issues running manual timings @ any FSB speed, unless on 4:5 divider
An indepth discussion indeed. ;) I had actually tried overclocking to the current values with my older Tagan 480W and the CPU fan going at full pelt was an issue with it too (this was with BIOS v1503 at the time I think). With the current one I did try leaving the memory voltage on auto and leaving the memory timings slightly higher as a result (at 5-5-5-12) but the CPU fan was still firing away at full pelt for some peculiar reason.Quote:
Originally Posted by Liquid3D
So, you're saying that your P5W DH Deluxe boots up OK with 400MHz FSB and manual memory timing settings? What BIOS are you using?Quote:
Originally Posted by kaiguy
There are many posts in this forum from people who can't boot with manual timings and FSB higher than 360-370 MHz. I am one of them :PQuote:
Originally Posted by cadaveca
What is your FSB and what BIOS are you using?
good up to 450FSB, 4-4-4-12 bios 1305/1407/1503/1506. I run 416 daily, although now that board is dead. 450 does not work right w/ 1503 tho.
My whole explanation is absolutely WRONG!!!!! I misread your POST I thought it said your were concerned about the PSU fan spinning up not the "CPU" fan therefore disreagrd what I said or apply as if the problem you were discussing was the PSU fan spinning up everytime you etc etc etc.Quote:
Originally Posted by DarthVader
As far as the CPU fan in the BIOS you may have the Fan speed contollers enabled. What I do is disable all such controls. I want my fans to spin full-out because i want my hardware to run as cool as possible regardless of the noise these fans make.
My P5W DH Deluxe boots up perfectly. If you look at my signature you can see the particulars (400FSB, manual RAM timings, etc.). I'm currently using 1506 but it worked fine with 1503 and 1407. I also don't need to have my memory voltage very high at all nor my vcore. My only issue with my setup is the idle and load temps (cpu 46-50 idle). Aside from that, no instability, no boot problems.Quote:
Originally Posted by Pol1977
They are indeed enabled at the moment. It is a highly bizarre problem though because it never used to happen. Perhaps it is a change to the BIOS in one of the updates that is causing this sort of behaviour?Quote:
Originally Posted by Liquid3D
I think I experienced something similar to you. When I was running at stock speeds, the auto-adjusting fan setting in the BIOS worked very well. Once I overclocked (not sure at what point it started happening) the CPU fan just went at max speed. I ended up disabling the automatic fanspeed control and went with a fan adapter to control the speed manually.Quote:
Originally Posted by DarthVader
The problem is I don't know if it was due to a change in a BIOS version, or when the FSB went above a certain speed... so I'm little help there. But the overclocking is what is causing the erratic qfan control.
I also have an E6400 and BIOS 1506 and am running @3.2GHz with 400 FSB and all that on Auto voltage except the MCH which has to be 1.65V (one step up from auto/normal which is 1.55V). My CPU idles at 28C now that it's getting colder but doesn'd go above 32C on idle even when it's warm (the Thermalright 120 Ultra and lian-li PC-V2000BPlusII case help there). And it's whisper quiet with 5 120mm fans @ <1000rpm and a 140mm PSU fan.Quote:
Originally Posted by kaiguy
But I still can't find a way to boot my system at 400MHz FSB without SPD enabled. It doesn't matter what timings I put in manually, even if I put the same ones that are on the SPD. As stated some posts before, it is some memory subtiming issue which we hope will be fixed in some future BIOS.
Why you can run at 400MHz FSB and be able to set your timings manually at the same time is a mystery to me.
Not too sure. The type of RAM could play a role, no? Maybe these Corsair sticks are set for SPD at those settings (but I doubt it as it would likely be a bit relaxed--I should probably change it to SPD to see what CPU-Z shows it as). Or, as I said before, maybe it could be the board revision? I know that the serial number on my first board was 68 or 69MOAG, but this one is 6AMOAG--which, from what I have read, no one has mentioned on this thread. The voltage settings are manually set for the minimum on everything except vcore (1.25) and vdimm (2.0).Quote:
Originally Posted by Pol1977
I think I need to try the 120 Ultra or the new Zalman. These temps with AS5 are really irritating me. But the room where the computer is located is consistently warm.
I couldn't set RAM timings @ 400FSB with REV 1 OCZ Platinum XTC 6400(SPD only), I RMA'd the RAM for the REV 2 stuff and can set all the timings I want @ 400+FSB.
Personally I'm convinced it will be something that has changed in the BIOS. After all if it was not happening back in the day when overclocking it in theory has no reason to do so now.Quote:
Originally Posted by kaiguy
Makes sense, but I could not for the life of me get the automatic fan control to work properly when overclocking. So I'd say its more likely that a BIOS update is what created this anomoly, rather than a setting (you really only have, what, two settings for speed control?).Quote:
Originally Posted by DarthVader
Indeed. Hopefully it will be fixed in a later BIOS update.Quote:
Originally Posted by kaiguy
Beta 1602 on Asus FTP site dated 11/8/06 not sure what it updates or fixes.Quote:
Originally Posted by DarthVader
I just got the board, it seems like this CPU cover thing doesn't come off. Do I have to live with this thing resisting the transfer of heat the whole time? Can I take it off? Can I keep it on without my temps being slaughtered by it?
Hmm, I got the cover thing off, now the stock heatsink takes the CPU out of the socket whenever I take it off.. the socket doesn't hold the CPU in anymor.
I feel lost from AMD-DFI land. :(
Oh great, there's a plastic piece on top of the metal cover that I had to take off, not remove the whole metal piece that kept the CPU in. Now I can't get the metal piece back on.
I'm not sure what you're referring to... What CPU cover thing? The metal bracket that holds the CPU down?Quote:
Originally Posted by Thasp
Or, if you are referring to the pad on the heatsink/fan, that is the not supposed to be removed. It is used instead of thermal grease.
Whatever it is, though, the CPU should be firmly in place. Mine is practically bolted to the board with the metal bracket.
Added: Oh, the plastic guy. Yup, you remove that not the whole bracket. Hopefully you can get it back on, but that's going to be a problem.
I just got it back on, put the CPU on. I feel like I'm putting together my first pC again - boy intel is a pain in the ass!
I just put he heatsink on. I felt like I was fighting it the whole time, I really hope I didn't crack the motherboard doing this. We'll see tonight though. :D
Well mine is one of the first and its always booted at 400MHz with manual timings way back to 0701.Quote:
Originally Posted by Pol1977
Hello axemanrio, I've tried setting voltage to the rated voltage of my memory and keeping SPD enabled too. This haven't solved the problem :(, I think I've tried everything... I finally will make use of the motherboard warranty and change it for a new one.Quote:
Originally Posted by axemanrio
Thanks for the help:)
Unfortunately I can't run my DDRII-1000 (PC2-8000) beyond 533MHz when I clock past 400FSB.
It's sad really given the fact I have an E6400 which does 425FSB (so far) without touching Vcore, check out the numbers below 1.29V Vcore at 425FSB
What MHC voltage are you using for 425MHz FSB?Quote:
Originally Posted by Liquid3D
Thasp
Read the instructions of how to fit the CPU that comes with the CPU and your motherboard. Without following these you may have broken something quite expensive. I've just come over from AMD too and found this installation of the CPU very easy but only after I'd read up on it....
ChrisC
well bios 1503 let me boot up at 3.54ghz, all mobo vcores at max, cpu vcore at 1.5125v - got an e6600.
I feel as if ive finally hit a wall... namely fsb, ive got everything done, hyperpath disabled, spd set i just think that this motherboard doesnt like fsbs of ~400+ unless its modded... or maybe im just too scared to put anymore vcore in my cpu?
I see 1602 is officially out @ asus.com now. How is this compared to the 1503 beta? Does it have the boot bug?
No boot bug in any bios after 1503 . :)Quote:
Originally Posted by jagt
Ok right. How is 1602 anyway? I'm not sure if I can be assed to flash if there's no difference. Running 1503 now.
I'm doing 3.6 GHz right now at 1.5v on my e6400. 57c according to asus, 70 according to TAT. What temp can I be at before I enter the "your CPU will be dead in two months" zone?
There should be no problem as long as you stay bellow the TJunction temp (shown in Core Temp, normally 85C for your processor). But no one can guess what long term overvoltage will do to your processor.Quote:
Originally Posted by Thasp
3.6 on a E6400 means a 450MHz FSB... What motherboard settings allow you to go this high on the FSB?
I'll take detailed pictures later.
chipset voltages, I am at 1.4/1.65/1.2 on the bottom of the screen, I forget which is which, but that is the order when reading up to down.
CPU is 1.5v, I am using the lowest divider. I read around and disabled a few things, I set the PCI express to 100 and pci to 33.
I expected this to be a lot harder.
What is the stock voltage on the e6400? With all the software errors and vdroop and speedstep I have no clue what the voltage is actually supposed to be for this CPU. My opteron ran at 1.575v up from 1.35v regularly. Is this a more insane boost than the opteron, or less?
p.s.
451 FSB ;) :D
Bah, froze after 30 minutes of encoding.
1.525v now.
I just got the CPU waterblock today, an MP-05. I don't have anything else yet so I can't use it. I like how it came with a baggie of tons of screws, nuts, and washers but not instructions or any piece of paper. Very smooth. :)
It's 1.3V :)Quote:
Originally Posted by Thasp
what is better with the new 1602 bios, anyone get higher clock?
WTF I got same issue after installed 8800 GTS and posted for that in 3dteam !!Quote:
Originally Posted by irenic
OMG i'm not alone ! :p: :woot: :toast: :banana:
so what clue to solve that funny pb ? :confused:
It seems that some G80 users , like me too , got lags during posting and under bios settingsQuote:
Originally Posted by Brahmzy
could you tell me please what bios settings do you have ?
tried 1602 bios without success :(
sounds weird !!
Weird indeed - I found that when I had my XFX 8800 GTX in the machine there was a delay (sometimes quite long) anytime I wanted to go into the hardware monitor in the settings.Quote:
Originally Posted by Mykou
However, I have noticed something unusual about this new BIOS version. PC Wizard 2006 is reporting my clock speed (at idle) of around 1.6ghz (for a stock speed of 2.4ghz) and around 1.8ghz for an overclock of 2.7ghz (at least at this speed the fans don't kick in at full pelt like at 3ghz). I probably should check if it has enabled speedstep in the BIOS actually. Oh and has the hardware monitor in the BIOS been simplified with this latest update?
EDIT - Turns out speedstep was set to automatic but even disabling it does not seem to help. Has anyone else found that their CPU speed is being reported as lower than it should be?
EDIT2 - By the way does having speedstep enabled have any effect on gaming? Is it best to leave it disabled?
Quote:
Originally Posted by Thasp
Here's a really short answer - you would be surpised to learn how many of the transistors which make up a CPU can be damaged and your CPU will still work. Heat is of course the enemy, however; voltage can and will damage a CPU regardless of how cool you chill that processor. Cooling is good, but if your constantly raising Vcore to acheive faster speeds, transistor damage will occur.
When your dealing with with a semi-conductive environment i.e. the CPU core, your dealing with an environment which is just that, "semi-conductive" meaning regions will allow voltage to travel on through unless they are "treated" or rather "doped". A CPU is essentially a series of capacitance switches otherwise known as transistors. These are tiny electrical switches which must be charged at all times since they have no moving parts. Indeed at this size it would be impossible to build a CPU with millions of tiny physical or mechnaical switches.
Capacitence switches are the foundation of your CPU and miilions of these switches not only provide the redundance needed for speed but for more complex instructions. So why is it all these switches and the channels along which current travel are housed in a semi-conductive material? For the same reason you wouldn't be able to construct a processor with millions of tiny mechancial switches, you couldn't construct mechanical switches that small nor could insulate wires betwen the switches that small. Thereforee a CPU is made using polished wafers which then have those tiny "lines" or "traces" imaged into it through what is known as Lithography. Once these patterns are projected onto the silicon that silicon is then doped with materials which insulate ensuring the current travelling along the "lines" travels where is should, similair to insulating wire in reverse. During fabrication, a process known as Voltage Threshold Dpoing occurs in which certain levels of dopant are used to confrom with certain voltages.
Another technique you may have heard of is SOI (Silicon On Insulator) where a layer of glass insulates areas around gates and helps prevent phenomenon such as Static Current Discharge and/or gate leakage. But the bottom line is some current inevitably leak's since tolerances are so close. Where talking about traces, interconnects and switches on a molecular level, ergo leakage is ocurring. This is where a substantial portion of the heat generated by your CPU derives from.
Whenever you overclock you're pushing your CPU beyond what the deisgner intended and those tolerances are already very tight. A CPU is profitable only if it is "squeezed" for every MHz or instruction you can get out of it, and then they expect a certain amount of transistor failure and lekage to occur which largely determines product warranty. After all no matter how "smaller faster cheaper" things become it's not logical to build a care that lasts forever or a CPU that lasts 20-years. It help's to remember your CPU is very close to a maximum operating threshold which is closer to failure (failure defined as malfucntion).
Now the whole point of overclocking or what I personally call "Purist Overclocking" is basically taking advanatge of the Fabruication and binning process in the CPU world. As you may or may not know when a CPU or Core is designed it costs approximately 2-billion dollars to constrcut a Fab facility for that single core, since lithography equipment and priocess changes with every die shrink. So the manufactuer deigns a core from which many thousands are made, aLL exactly the same. In other words a Allendale based Core Duo E6300 is essentially the same cpu as a Conroe Core Duo X6800. What the manufacturer does is disable half the cache, and lock the upper multiplier. There are then several methods to make your chip go faster, raise FSB or Rasie voltage and FSB.
The difference between a purist and everyopne else is that the pursit doesn't raise Vcore. In my humble opinion that's not overclocking so much as over-volting and the latter does much more damage to the chip. What the purist is doing is simply taking advantage of the binning process. When Intel produces all those CPUs they bin the X6800's from the E6300's thorugh a series of basic burn in tests, if the "batch" was ideal all pass and have the potential to become X6800's if not they find perspective bins based on more tests. The Purist OCer comes along and buys the least expensive chip[ yet gets the most expensive core unfortunately minus 2MB of cahe, can;'t do much about that. However understanding the Fabrication process is where overclocking began, as overclocker's became more sophisticated they began to identify (through forums) which "steppings" were the best overclockers. They also did this based on which Plant they came from; for example Costa Rice vs. Malaysia. Hence overclocking NOT over-volting was born. I am a Pursit I am most happy when I find a chip that keeps on going without rasing Vcore. I loathe rasing Vcore in fact.
For so many reasons I can't begin to explain them all here, the Conroe core is an exceptional overclocker, for example; Intel was building ineffcient CPUs for so long that their SOI, Lithography, and other assorted technologies to control leakage, keep tolerances tight and produce igher trnasistor counts were basically overkill when mated with the architecture of Conroe. You take all the advances made by necessity to CPU fabrication and incoperate them into a short pipelined, high IPC Core and you have their best decision in ten years. Your Conore is effectively an amalgmation of a PIII with mobile technology fabricated with todays lithopgraphy and interconnect methods and they all come together to produce some of the best performers ever. The previous Net-Burst technology which made so many millions off the ignorance of the consumer who believed speed = performance actually forced AMD into creating the PR Performance Rating system and delayed chips like this for about seven years. Conroe at around the 2.0GHz mark is now a "normal" IPC chip in todays technological world, hence the drop in speed and increase in performance.
So the purist overclocker comes along and knows if the "batch" (stepping) is good then all he/she needs to do is purchase the least exopensive model to get the fastest chip's speed, unfortunately they loose 2MB (in this case) of disabled L2 cache. this is the Overclocker's wet-dream. now when we choose the to raise Vcore we have chosen another road and this is basically a method which "tricks" the Vcore suinuwave to meet the higher voltage peaks at a faster speed. This is difficult to explain but basically traising Vcore damges transistors and floods channels and transistors with more voltage then they were designed to accomodate. This increases static current discharge, capacitance and the opportunity for current to jump across channels. Once current begins to behalf outside the designated chasnnels and leaks into the semi-conductive strata all Hell breaks loose and your CPU is NEVER the same. It doesn't repair itself and all the cooling in the world won't prevent current from doing what it does.
IN conclusion you will most likely NEVER see a dead CPU, failure occurs on the same principle a CPU is built upon, micro-processes x milions. What does happen is a file may open more slowly or such and such an execution will will see errors. it will be a slow bleeding to death by a million tiny cuts. The number of erros in your CPU will slowly increase and once they occur they are not as I said above, repair themselves. This is why I would never buy a used CPU, or any type of semi-conductor, albeit memory, VGA card etc. You si,mply don't know what the end-user did with it. Especially people who sell chips that broke world record. If they clocked that high with so few purists left you can be sure the thing was dmaaged with voltage and has millions of transistors dmaged already. it won;t be noticiable becuase it rarely is you simply get frustrated with all thsese littel erros and the hours you spend trying to fix other things is a waste of your time.
The moral of the story, toomuch voltage is the enemy. So don't use Vcore to run your CPU faster, find it's best speed at default Vcore and be happy. Only in that way can you minimize the number of transistors which would otherwise be destroyed. As In Neurophilosophy which is my area of study, the CPU is the perfect model for the human brain for many reasons. Their both essentially made of organic material, they both operate on low levels of electric current, they both have long and short term memory and they both work on a binary system. On/off that's how synpases fire between synaptic clefts and how your cognitive processes occur. Just as you get to a certain age and brain cells don't regenerate, your CPU is born at that age, ergo dmaged transistors will never regenerate. So don't let your CPU get too hot by pushing voltage into it, that's analagous to feeding your brain GHB. It may feel good for a while but too much will overheat it destryong neurons or transistors, so don't turn your CPU into a "burn-out."
I dont see any problems with these 2 pieces of hardware since I put them together, running 1506, no lag or any weirdness.Quote:
Originally Posted by Mykou
thanks , gonna try 1506 and seeQuote:
Originally Posted by Rattle
Mykou, i'll wait for ur result..
Quote:
Originally Posted by Liquid3D
Temperature kills things, not a little 30% increase in voltage , although voltage can kill, just not at such a low % increase.
I have read up on quantum tunneling and the effect abuse has on our CPUs before I bought the opteron. I just wanted to make sure I wasn't doing a 40% overvolt or that 65 nm CPUs died when overvolted by 0.15v before I did anything, thanks for the clarification.
>So don't use Vcore to run your CPU faster
Be glad you are on the P5W then, the P5B Dlx mobo I am now testing out increases Vcore on its own if the BIOS is set at AUTO. Even at idle, I have seen it go up to 1.5V bringing the temps up too...
The way AUTO Vcore setting is implemented on the P5W vs the P5B Dlx is just so different.
Hi, i need some help.
I have a P5WDH with an E6600, seasonic S550+, 1 raptor 150 Gb, 1 maxtor 320 Gb sata, a X1900XT, all in a coolermaster wavemaster.
Yesterday i unmount everything to change the rad on the SB for a zalman nbf47 and to plug a plextor 760SA.
I remount everything in the case and then: no display on boot, no post only the fans working and the led on the motherboard.
I cleared the cmos during 4 hours (did it two times)
I tried different things, unmount everything, unplug everything except the graphic card, and the memory, unmount the motherboard ---> nothing still the same problem!!
And when i try to boot with no memory or no graphic card, i don't even have a bip from the motherboard !!!
Do you think the mobo is faulty (it was working perfectly well two days ago, at 3.2 G with a zalman 9500 and default voltage)
ps bios is (was 1402)
Irenic there is now bios which solve our pb ..just tested 1506 and same thingQuote:
Originally Posted by irenic
have to wait for official bios release
I had a problem like this once, no beep, no post. I took the motherboard out of the case and tried taking out components one by one. I got the board to post after removing the cpu heatsink and changing the thermal paste. I don't think the problem had anything to do with the thermal paste, but I think the cpu must not have been properly seated in its socket.Quote:
Originally Posted by fmilin
Try removing the cpu heatsink as well as the cpu, and put them back on. This board needs to have the cpu in to be able to post.
every board needs the cpu in order to post :/
i tried to remove the cpu, but all the same.
I just received my new board and it's working, so i am now sure the board was faulty but i don't know why !!??
And you know that how? Are you a trained EE? A 30% increase is a lot. I for one wouldn't want my TV to run w/ 300+V instead of 240V for which it was desgined.Quote:
Originally Posted by fhpchris
Quote:
Originally Posted by Timmaah
you guys reckon ive hit a cpu wall or a fsb wall?
Guys please keep on topic and stop the fighting or I'll close this thread:) Thanks!
Let's see if this is a good one.Quote:
P5W DH Deluxe BIOS 1602
Fix CPU frequency incorrectly display problem on AiBooster
Enhance HDD compatibility
Link To Download Page
Asus is really cranking out the BIOS updates...
errr :stick:Quote:
Originally Posted by njkid32
ban them, don't close my thread :nono:
My latest disappointment is HD performance.
Everything here is faster, but HD performance really goes up and down here. I can go from a steady 27 MBps on LAN transfer down to 2, on a clean reboot when the computer is doing nothing. I thought it was a network issue until I tried writing video data to disk, it took twice as long as it did before.
I did a clean reinstall of windows a day after getting this system all set up. I am going to try different sata ports later, but this is no fun for now. :(
Have you set the drive to enable 32 bit transfers in the bios?Quote:
Originally Posted by Thasp
I never even knew there was such a setting for HDs. I never saw that on my DFI NF4-DAGF before. I will look for it, but if it is not set to enable 32 bit by default, then I do not have it set.
thasp what bios did you get 450 fsb with?
1503, it was stock.
Orthos gives me 83c at full load.. this is insane. I need my watercooling to get here soon. :(
Can I use winflash with this mobo like I did with my old DFI? The mobo manufacturers must be schizophrenic - it's 2006, but they still use floppy disks for BIOS flashing.. ? :(
Quote:
Originally Posted by Thasp
If you have a usb thumb drive it's the best way to flash with this mobo. Load the bios file in the usb drive and stick it in the computer. Reboot into bios and go find EZ Flash in there and it takes you to your usb thumb drive, and load the bios file from there.
How about a UMS MP3 player? I don't have a thumb drive.
Can I put the files onto a CD and go from there?
I dunno about that, maybe try it? If the pc recognize it as a removable disk type and you can store the bios file in there, I dont see why it wouldnt work. I used to do the cd method with the nv boards, but I havent try it with this one yet. Just get a usb flash drive, it's very convenient for loading basic drivers and stuff, also there are tons of them for like under $20
http://www.newegg.com/Product/Produc...ubCategory=522
I could buy a flash drive, but buying one to flash a BIOS is the wrong reason to - it should never require extra money to do what the product should be capable of doing on its own.
I load drivers by keeping my OS and data partitions seperate. I go into a setup folder, have programs and drivers in a seperate folder.
I will have to try a CD.
Do newer BIOSes let you set RAM subtimings? That's the only reason I care about this. I am annoyed at my :banana::banana::banana::banana:tacular RAM's inability to do more than boot into windows and freeze at 2.2v, 5-5-5-15 DDR900. It's 3-4-4-12 DDR 677 2.2v right now which passed lots of memtest. Even though the timings are tight for DDR2, it burns me that I'm running it below stock because I can't use the next-up divider of DDR900 with any kind of timings. :(
I've done it using my mp3 player.Quote:
Originally Posted by Thasp
Hi, I've just flashed the bios using my usb memory stick and found that it didn't reset the bios settings. Thus I will not be using this method and will go back to using Asus Flash, which does...
ChrisC
Have you pulled your heat sink and verified full contact with the surface of the CPU?Quote:
Originally Posted by Thasp
I had the same issue with my Big Typhoon (replaced with Ultra 120 now) after removing it found I had only 50% contact!
sparx
flashed to 1602 and the bios lagginess is still there.. :(
Twice. I put thermal paste on the CPU in a conservative layer, then put a small drop in the middle, like I usually do. Then I put it on, twisted it a tiny bit, and screwed it on very tightly.Quote:
Originally Posted by Sparx
I tried that BIOS option and moving it over one sata port, it's still 27 MBps on LAN transfer and even faster writing video, and sometimes 1-2 MBps.. :(
I have a xfx 8800 gtx card, and it's a delay in post with this videocard.Quote:
Originally Posted by DarthVader
I have 1503 bios.
I presume this issue applies to all 8800 GTXs rather than just the XFX one?Quote:
Originally Posted by Cornel
Are you get the 80c temp read from Core Temp Ver-0.93?Quote:
Originally Posted by Thasp
sparx
hey guys i need help. this is my first post so im not very experienced. I have read all 80+ pages and have tried everything but still cant get over 360fsb.
x6800
p5wdh mb
corsair 5400pro with those lights on them.I have already fried my new corsair8500c5 trying to get them to run at 1066. Can somebody just post a link of all bios settings i need to set to get over 360fsb please...would help me alot thanx
fAh.....so its not just me...lol........damned irritating the BIOS lag :(Quote:
lashed to 1602 and the bios lagginess is still there..
Also.....when I turned on PC today the BIOS was reporting 'overclocking or voltage overclock failure'.......press F1 to go to Setup??
I don't have the thing overclocked!!........just rebooted and all was OK?.......anyone had this?
Quote:
Originally Posted by Severance
Its a ram timing problem (safty feature)
when you boot at FSB 360+ the asus 975 boards
are more stricter with SPD timmings if your pushing high FSB
so when you set manual memory timings then save and reboot
it no's your ram cant run @ those speeds so it gives you a black screen
so now it then wants you to set auto SPD but the problem now is your memory is running way slack timmings
look at your SPD tables and i bet you they suck
so the only way to solve it, is buy ram that has aggrasive timings in the
SPD tables like the teams xtreem,Cellshock, Gskill, and you will boot 400 fsb with manual timings no problem.
i have found this out with my pw5 64 workstation
changed ram and no more 360 fsb bug
can boot @ 400 fsb 1.1 with memory
hope this helps alot of folks i spent ages trying to solve this.
so your saying that if i enable spd i can go over 360? cus iv tried that too.
Quote:
Originally Posted by nanoprobe32
Syne24,Irenic and I got the same lag issue during post .Quote:
Originally Posted by scaramonga
:banana: :banana:
Quote:
Originally Posted by Severance
look @ your SPD tables with CPUz and post a screenshot
if i could just get it to boot with auto mem timings that would be ok cus i could just use memset to lower them? but i cant get it past 360fsb using auto spd
Quote:
Originally Posted by nanoprobe32
yes i see that timings are very loose when using auto spd. But why wont it boot?
Quote:
Originally Posted by Severance
your way above your ram limit, it problay would have to run
cas6 to hit over 360
even the corsair XMS2-8500C5 have very poor SPD timings
http://img293.imageshack.us/img293/9144/cpuzei9.png
Team Xtreem PC6400
http://img152.imageshack.us/img152/3792/spdyd0.jpg
cellshock PC6400
http://img91.imageshack.us/img91/4263/cellshockih9.jpg
so guess out of those 3 rams that will run into the 975 bug
corsair compatable spd timings myth
the Team Xtreem PC6400 and cellshock wont have the FSB bug over 360
simply because their programmed correctly.