View Full Version : Anybody else have this experience with a koolance case?
Hi all,
I've been messing around with my main machine, trying to get the fsb and chip speeds as high as possible. I did the volt mod to my ak31 rev 3 board, and started trying to really put the juice to my duron 950. I noticed that when I didn't have the fsb speed very high, I could hit almost 1300 MHz with the chip, and when I kept the chip speed way down (like underclocked kind of down), and the cpu voltage along with it, I can crank the hell out of the fsb. But not both at the same time. I also noticed that when I did attempt to do both, the machine would hang during the boot process, and the hard drive would sound like it was turning off and back on quickly. I experimented by pulling the power plug out of the hard drive and then quickly replacing it, and sure enough, it made the exact same sound. My case is a koolance, with all of the associated current draw of the 3 fans on top and the two pumps. I've therefore concluded that my problem is that I'm out of power supply, and that when I crank up the fsb AND the cpu voltage, the psu just can't handle it. I didn't foresee this because I'm running a comp usa 500W psu, but I guess everything has its limits. Has anyone else had this happen? If so, how did you solve the power supply problem? I guess I could use an external regulated supply to run the pumps, but I'd really like to keep everything self-contained within the case. That's part of why I bought a koolance case.
WOW you actually disconected power to the H/D .Feel lucky you did not corrupt the data on the disk.
It sounds like the usual ocing issue.You have a high fsb and low multiplier.or vise versa.
I prefer 160x a lower multiplier cause the box is way faster with the fsb raised.
<WOW you actually disconected power to the H/D .Feel lucky you did not corrupt the data on the disk.
It sounds like the usual ocing issue.You have a high fsb and low multiplier.or vise versa.
I prefer 160x a lower multiplier cause the box is way faster with the fsb raised.>
Yeah, hell, it had been apparently doing the same thing with the hard drive power itself plenty without my intervention. Like you said, I'm running with a high fsb now instead of a high cpu speed. But, I have to actually UNDERCLOCK the cpu to get this high fsb speed. That is obviously not good long term, so hence the need for more power.
It ius perfectly fine to lower the cpu multiplier,I dont know where you got the Idea it is bad.
Example.AMD Itself introduced the c series Athlon.It was a step above the b series.
The difference was the c series was guranteed to run at 266 fsb where the b series was rated at 200 fsb.
What amd did was reduce the multiplier on the c series to 9x133 giving you the 1.2 gigs,the b series was set to 12x100 giving you the 1.2 gigs.
From what I remember as long as a member had a 133a chipset the b series ran perfect at the lower multiplier.
So you havbe no problems setting it lower,It is a common way to overclock !
<It ius perfectly fine to lower the cpu multiplier,I dont know where you got the Idea it is bad.>
Hookay, lemme reiterate this again. To raise the fsb, I HAD TO LOWER THE MULTIPLIER UNTIL THE CPU WAS UNDERCLOCKED. As in, the cpu runs at BELOW STOCK SPEED! Stock speed WILL NOT WORK with the high fsb on this machine. Now, it seems to me that both overclocking the fsb and overclocking the cpu (especially with the volt mod) require more power from the psu. When I do both of these things at the same time, the machine generally will not boot, and the symptom of the hard drive turning off and on again manifests itself. My claim is not that low multipliers are bad, it is that I should not be forced to underclock my cpu to get a high fsb, and that I believe this situation to be due to an overtaxed power supply.
LET ME RESTATE NO **** SHERLOC
THAT IS HOW YOU OVERCLOCK A PROCESSOR !!!!!
ARE YOU NOT HERING ME OR WHAT..... YOU HIT 1300 FROM 950 IT IS OVERCLOCKED BY WAY OF FRONT SIDE BUS !!!!!!!!!
CAN YOU SET IT TO 10X130 WHY YOU WOULD I DONT KNOW. IT IS BETTER AT 9X144 WAY FASTER.
IF YOU ARE TRYING TO HIT LIKE 1450 GET REAL THAT CHIP AINT GONNA DO 10X145 IF YA POWER IT WITH AN ARC WELDER.
AT 8X160 IS FASTER THAN 9X144 DONT YOU UNDERSTAND THIS.Overclocking the bus creates more system speed than upping the multiplier.And it is so rare to be able to crank booth up and get stable results.oh did you unlock that processor ?
LMAO
<LET ME RESTATE NO **** SHERLOC
THAT IS HOW YOU OVERCLOCK A PROCESSOR !!!!!
ARE YOU NOT HERING ME OR WHAT..... YOU HIT 1300 FROM 950 IT IS OVERCLOCKED BY WAY OF FRONT SIDE BUS !!!!!!!!!
CAN YOU SET IT TO 10X130 WHY YOU WOULD I DONT KNOW. IT IS BETTER AT 9X144 WAY FASTER.
IF YOU ARE TRYING TO HIT LIKE 1450 GET REAL THAT CHIP AINT GONNA DO 10X145 IF YA POWER IT WITH AN ARC WELDER.
AT 8X160 IS FASTER THAN 9X144 DONT YOU UNDERSTAND THIS.Overclocking the bus creates more system speed than upping the multiplier.And it is so rare to be able to crank booth up and get stable results.oh did you unlock that processor ?
LMAO>
Ok Chief,
Ya really might wanna calm down with the laughing thing. You have obviously mistaken me for someone who has not overclocked before. Let that stand corrected for the record.
Fact: The top speed a cpu can hit does not depend on the fsb. Yes, I understand the simple multiplicative relationship between the multiplier and fsb. Those two numbers multiplied together determine the resulting cpu core speed. Granted, if you can choose between two different multiplier/fsb combos that yield the same cpu core speed, you should obviously choose the one that has a lower multiplier and higher fsb. I learned that WAY on back experimenting with my 800 MHz slot A athlon and goldfingers card. That basic fact is not in dispute. The way I overclock is to determine the fastest core speed the cpu in question can hit by raising the multiplier (and core voltage as necessary) until I arrive at the fastest cpu speed I feel the machine is stable at and has no overheating problems.
Then, I lower the multiplier back down to default or even below, and start raising the fsb. Obviously, I do this to eliminate cpu speed as a limiting factor while raising fsb speed. I apply much the same criteria to this part of the process as I did to the cpu part. Stability is essential. It's a slow process for me, involving a number of different stability tests that I try at each fsb speed. I eventually reach a point where I feel I have maxed out the stable fsb speed on the particular machine in question.
Now, once I have individually determined the max speed of both the cpu and the fsb, I then set the fsb at the max speed, and get the cpu core speed as close to the max value I obtained earlier as I can with the multiplier. This ALWAYS results in a stable machine for me, usually with the cpu running at a little below the previously-determined max speed due to the fact that the multiplier generally does not place the cpu core speed at exactly where I determined it was stable. So, I choose the next lower multiplier for stability.
Please note that I have always ended up with two things being true about the machine: The fsb is running at my previously-determined max stable speed. The second thing is that the cpu is running ABOVE stock speed, usually a little below the previously-determined max stable speed.
However, this machine is breaking those rules. I can stably hit high fsb speeds, or high cpu speeds. But not both. Note that I'm not saying I expect to hit high fsb speeds and high MULTIPLIER settings, but rather that I expect the high fsb and higher-than-stock cpu speeds. I've been able to get these two things plenty of times before. This obviously occurs at a lower multiplier setting than before the fsb speed was increased beyond default.
Here is what I have: A 950 duron (unlocked) on a shuttle ak31 rev 3 board, and 512 MB of pc2100 memory that has proven its ability to run WAY beyond 133 MHz. (I had been using 768 MB, but I took 256 out because I've learned that it's useless in win98. It will go back in when the overclocking process is complete and I switch to win xp.) The memory is stable at 160-something MHz, I can't remember right now off the top of my head. The memory has proven this many times. The cpu is stable at 1200-something MHz, again I can't remember exactly what right now. But, with the memory cranked up to the high speed, the cpu is stable at 891 MHz or so, AND NO HIGHER. Remember, this is a duron 950. See what I'm saying? Why should it now be incapable of even hitting DEFAULT speed?
Not only does this not make theoretical sense, it goes against all of my previous experience. I'm not the most experienced overclocker in the world, but I've done it enough to know when something is obviously out of whack. So, I realized that all of my previous overclocks were done in aircooled situations, without the current draw of the two pumps in the koolance case, not to mention the 3 80 mm fans.
My theory is that both overclocking the fsb and overclocking the cpu take more power than running both at default speeds. I'd imagine overclocking the fsb takes far less extra power than overclocking the cpu, but I think it takes some extra. Now, add to that the 3 80 mm fans and the 2 pumps, and I'm theorizing that you have enough extra current draw to strain the power supply. I'm also theorizing that the way that strain is showing itself is dips in the voltage supplied, which may be disrupting the hard drive. Thus, causing the hard drive to react the way it did.
When trying to run both the cpu and fsb at high speeds, I experienced the hard drive problem. So, after the power drain idea occurred to me, I bumped the cpu speed AND VOLTAGE back down. Of course, this considerably lowered the power drain of the cpu. Sure enough, stable again.
So, we're back at square one as per my original post-I have a situation where I need more power. I strongly suspect that when I get more power, the cpu will suddenly be stable at overclocked speeds while running at a high fsb speed.
Well I am whatever
Replace this power supply,but to think you can get to 1330 with that 950 duron is ok by me,I personally think your issues are instability related directly to the demands put on the processor.
I cant see a guy that seams to be so up on stuff pulling a plug from a rumming HARD DRIVE lol {Had to laugh}sorry
If the fsb and multiplier get so out of whack like you are setting yours anything is liable to happen.That puny duron cranked to the max may never see 65 watts of comsumption.So this is why I state the issue is you are trying to go to high with these settings and even tapping into the United States enegry reserv aint gonna help ya ,I gotta LOL again I am rolling on my back.
I gotta go I cant see the screen any more tears,,, blinding me.
<Replace this power supply,but to think you can get to 1330 with that 950 duron>
Hey man,
WHERE do you see me say ANYTHING about hitting 1330? Looking back, I see that in my first post, I mentioned that I had hit almost 1300. At NO point did I EVER mention 1330. YOU magically came up with that number, not me.
Also, is it that unreasonable to wonder why the cpu won't even run at default speed with the fsb jacked up? Wouldn't you wonder?
On the hard drive thing: It's a machine not being used day-to-day at all. I don't care if unplugging the hard drive corrupts data, that's why I'm not using it for anything except experimenting. Again, IF YOU'LL ONLY READ MY EARLIER POSTS, you will indeed see that the hard drive has already been displaying identical symptoms to those produced by unplugging/reconnecting the hard drive. SO, we now have a situation where a hard drive that's not storing anything even remotely crucial is being experimented with, producing the same symptoms AS IT WAS ALREADY DOING! Ding Ding! (Please look to the last two sentences and reread them. The pertinent point about the hard drive IS CONTAINED IN THEM.)
Can the reasoning I'm employing by unplugging the hard drive POSSIBLY be more obvious? If so, someone please let me know. I am officially tapped out on explanations.
Now, in regards to your condescending attitude: I (and many other people, I suspect) don't appreciate that sort of thing, especially when pertinent main points are being disregarded in an apparent attempt to make fun of someone. I can't think of any other reason for your treatment of the subject at hand. Cheap shots from the peanut gallery aside, I will obtain an additional source of power and see if my theory is right. I will post the results, whether I'm right or wrong. If I'm right, I will be able to achieve a higher-than-default cpu core speed while still running the fsb overclocked to the range I have been able to when the cpu was underclocked. If not, oh well. In either case, little sly remarks and "laughter" are unnecessary.
dont mean any harm I am silly it is late.I am sorry
You Mentioned 1300 so I ran with it.
What exactly did you set it to ??
Is default 7.5 or is it 9.5 ?
default fsb 133 or 100 ??
I see your reason for the hard drive.But was shocked to see ya unplug it and plug it back in.Can yoiu get voltage on the 5 volt line.
Also I was confused because the processor and hard drives are on toatly seperate power lines at different voltages.Cpu 5v H/D 12v.This is why I question the ocing and power problem you are describing.Again I am sorry to have upset ya.Ment no harm just figured we had an argument going.
<dont mean any harm I am silly it is late.I am sorry
You Mentioned 1300 so I ran with it.
What exactly did you set it to ??
Is default 7.5 or is it 9.5 ?
default fsb 133 or 100 ??
I see your reason for the hard drive.But was shocked to see ya unplug it and plug it back in.Can yoiu get voltage on the 5 volt line.
Also I was confused because the processor and hard drives are on toatly seperate power lines at different voltages.Cpu 5v H/ D 12v.This is why I question the ocing and power problem you are describing.Again I am sorry to have upset ya.Ment no harm just figured we had an argument going.>
To answer the questions, default on that cpu is 9.5, since a duron runs at a 100 MHz fsb.
I measured both the 12V and 5V lines, and got no pronounced dips in either. But, I was measuring on an old digital multimeter, the kind that does not have an analog bar on the display to reflect instantaneous changes in the measured voltage. So, I figured that the rather slow-changing numbers on the display might miss a quick voltage dip that could mess the hard drive up. An analog meter or an oscilliscope would have been ideal to measure this under load, but I have neither available down here. (I'm a navy reservist who has been activated for a year, so I couldn't take all my gear with me to my duty station) So, this is where the hard drive experiment came in.
On the separate voltages for the hard drive and cpu, I hadn't thought of that at all. I knew that they are on different power supply buses, but it just didn't occur to me when I started working on this problem. I'm still going to try another source of 12V because 500W is still the total power output rating of the supply. This leads me to believe that I may be exceeding the total rating, even if one or more individual circuits are not being overloaded. It's kinda hard to imagine me having a situation where a 500W supply would EVER fall into question as to whether or not it has enough capacity. If someone had told me 6 months ago that I'd have that big of a supply and end up wondering if I needed more, I wouldn't have believed them. But, that was before liquid pumps, volt mods, and multiple 80 mm fans became involved.
Pats ya on the back buys ya dinner !!Navy guy !!! Cool kick the asses of any one who looks like trouble lol/
Ok lets start with this tHE PROCESSOR BEING A DURON WILL ONLY EVER USE UP TO 45 WATTS.DRIVES MAY USE DEPENDING ON WHAT IS IN THERE 50 WATTS.ADD ON CARDS AND 5 FANS 10 WATTS.
Even if you had 3 more fans and several pumps add what 20 watts.I can promiss you that your system is on the low side of 200 watts of consumption.{The above numbers are most likely high}
I really cant figure it to be power related.What are you trying to set this at when it gets flaky.Is the mother board a 266 fsb one or a 200 fsb one ?
You stated you have raised pro to gets its max what was that
you said you raised the fsb to get its max
WHAT are these Settings ????
Pats ya on the back buys ya dinner !!Navy guy !!! Cool kick the asses of any one who looks like trouble lol/
Ok lets start with this tHE PROCESSOR BEING A DURON WILL ONLY EVER USE UP TO 45 WATTS.DRIVES MAY USE DEPENDING ON WHAT IS IN THERE 50 WATTS.ADD ON CARDS AND 5 FANS 10 WATTS.
Even if you had 3 more fans and several pumps add what 20 watts.I can promiss you that your system is on the low side of 200 watts of consumption.{The above numbers are most likely high}
I really cant figure it to be power related.What are you trying to set this at when it gets flaky.Is the mother board a 266 fsb one or a 200 fsb one ?
You stated you have raised pro to gets its max what was that
you said you raised the fsb to get its max
WHAT are these Settings ????
Ok, here's the settings I reached. With the fsb at 100 MHz, which is stock for a duron, I was able to get the 950 duron STABLE at 1270. Now, this is with 2.064 volts applied to the processor core if I remember right. That generates a LOT more heat than default voltage. Also, you must keep in mind that this is a watercooled setup, so I probably wouldn't have been able to do this with aircooling. With the cpu running UNDERCLOCKED (remember, I had earlier found that this was necessary when running high fsb speeds), I have been able to get the fsb stable to 165 or 166 mhz (can't remember which) with 256 MB of crucial pc2100 ddr ram. Here's another really funny thing that made me suspect power drain problems: When running the fsb as the high speeds, UNDERVOLTING the cpu helped me increase fsb speeds. That's right-the default voltage of the duron is 1.60, and I ran it at 1.55 while cranking up the fsb. It worked-I was able to get a higher fsb. I notice that my memory gets warmer to the touch when running high fsb speeds, so that tells me that the memory does indeed use more power at high speeds.
To answer your questions about the motherboard, it is a shuttle ak31 rev 3. This is a really popular board with athlon/duron overclocking people because it's stable and allows all kinds of nice overclocking adjustments from within the bios without messing with jumpers or dip switches. As for your question about whether it's a 200 or 266 MHz fsb board, you can set the fsb to anywhere from 100 to 166 MHz in 1 MHz increments from the bios. So, since it uses DDR memory, you can reach memory speeds from 200 MHz to 333 MHz. You can actually go higher than that with this motherboard using cpuFSB, which is a program that lets you adjust fsb speed from within windows. I have actually hit 170 MHz with this board while using only 1 stick of 256 MHz crucial ram, but it was not stable. So, I backed it down and did not include that speed in any of my posts.
On the cpu power consumption thing, I have to disagree with you on that. A cpu will uses more power as the core speed is raised, and it uses more power as the core voltage is raised. I have verified both myself by watching the cpu temps while raising the speed a little at default voltage, and then lowering the speed to default and raising the voltage. Doing both at the same time really raises the power comsumption of the cpu, and thus the heat it produces. There are people out there who have calculated heat radiated by cpu's in watts, and it's entirely possible to get an athlon to radiate 150 watts of heat. Now, we all know that durons don't radiate as much heat as athlons, but when you start running a 950 at 1270, and raising the core voltage from default ( 1.60 volts ) to 2.064, I think it's a safe bet you've exceeded 45 watts of heat. Also, I think those liquid pumps probably take a lot more power than fans-after all, they're pumping liquid through narrow tubes instead of air. That's one of the things I was hoping someone could tell me-exactly how much power the pumps take.
dO YOUR DELF A FAVOR GO LOOK AT HEAT SINK REVIEWS.
No AMD outputs 150 watts it would melt.
75 to 85 watts is max output even overclocked to the max.
The duron would need house current to reach 60 watts.You are way off on out puts of the processors.
Now because the duron starts at 9.5 there is very little head room to go beyond 133 fsd on that processor if you leave it at 9.5
This is what I have been telling you all along.Unless you set the bus to 120 or lower you wil never get above 10.5 multiplier.
You can debate all you want but it is silly.10.5x133=1.4 gig
Aint no way you are getting there.I can see 8x160 or 165 but to then think you are underclocking and should be able to raise the processor is pure dreaming.150x9.5 aint gonna happen so I dont understand your insistance you have a probleem that can be solved.Your problem is a wall that processor has hit the wall and you cant do anything about it.
And how you set multiplier and then fsb to understand limits is good,BUT simply foolish to think these 2 figures can work together.If you hit 160 x 8 YOU ARE NOT UNDERCLOCKING THE PROCESSOR.You are OVERCLOCKING IT NO MATTER WHAT THE STOCK MULTIPLIER IS.
And please show me where someone got an output of 150 watts from and amd processor.This is impossible.It is also way way unlikely a duron can even approach the watts of an Athlon.which are usually 60 watts to 79 watts even overclocked.
<ow because the duron starts at 9.5 there is very little head room to go beyond 133 fsd on that processor if you leave it at 9.5
This is what I have been telling you all along.Unless you set the bus to 120 or lower you wil never get above 10.5 multiplier.
You can debate all you want but it is silly.10.5x133=1.4 gig
Aint no way you are getting there.I can see 8x160 or 165 but to then think you are underclocking and should be able to raise the processor is pure dreaming.150x9.5 aint gonna happen so I dont understand your insistance you have a probleem that can be solved.Your problem is a wall that processor has hit the wall and you cant do anything about it.
And how you set multiplier and then fsb to understand limits is good,BUT simply foolish to think these 2 figures can work together.I.>
Ok, lemme bust it on out:
current setup on the computer we're talking about is:
system clock (fsb): 161.64
multiplier: 5.5
OK, NOW GO BACK AND REREAD THOSE NUMBERS. NOW MULTIPLY THEM. WHAT DO YOU GET? I GET 889.02. THE CPU IS A DURON 950.
889.02 IS LESS THAN 950.
SO, THE CPU IS NOW UNDERCLOCKED.
I think part of the misunderstanding between us is that you're thinking I want a high multiplier. I don't-I could care less what the multiplier is, as long as the cpu speed is more than 950. Note that in my many earlier posts, I never once said that hitting any certain multiplier is a goal of mine.
Now, before you start freaking out and thinking I'm stupid for underclocking my cpu, remember what I posted earlier about why I did it. I notice in your last post you said, and I quote,
"If you hit 160 x 8 YOU ARE NOT UNDERCLOCKING THE PROCESSOR.You are OVERCLOCKING IT NO MATTER WHAT THE STOCK MULTIPLIER IS"
End quote.
Ok, once again, I must ask where you got 160 x 8 from? It wasn't from me. If you'll read my earlier posts, you see me mentioning the fact that I lowered the multiplier until the cpu was underclocked, EVEN AT THE HIGH FSB SPEEDS. That's the key here. So, even at the 161.64 MHz fsb speed, THE CPU IS IN FACT UNDERCLOCKED. If you still doubt that, refer to what I wrote earlier in THIS post, the part with my actual, real-life right-now numbers on that machine. If necessary, do the multiplication problem yourself. I believe you will see that I really DO have the processor underclocked.
<Now because the duron starts at 9.5 there is very little head room to go beyond 133 fsd on that processor if you leave it at 9.5>
Who said ANYTHING about leaving it at 9.5? I never did. You may have assumed that I was leaving it at 9.5, but that would be in incorrect assumption. Remember, unlocking a cpu also gives you multipliers BELOW default, not just above.
<And please show me where someone got an output of 150 watts from and amd processor.This is impossible.It is also way way unlikely a duron can even approach the watts of an Athlon.which are usually 60 watts to 79 watts even overclocked.>
Ok, I will now bust it on out. On the heat output of the duron, here we go:
On overclockers.com, there is a link to "utilities". This is on the left side of the page. One of the utilities is called "radiate". This is a program, I think written by someone that's a forum member (I'm not sure) that calculates the heat dissipated by different processors. It allows you to select which processor you're using. PLEASE PLEASE PLEASE download and install it. I think your eyes will open far, wide, and fast. I set the voltage and cpu core speed for my duron. Note that the program won't let me select as high of a core speed as I actually ran, and I set the core voltage at 2.06, where I was actually running 2.064. So, BOTH of these differences will actually cause the radiate program to report LESS than what my chip was actually radiating in heat. Here's the result:
89.2 Watts. Amazing, isn't it? Just a few months ago, no one would probably have dreamed that we would be pushing so much heat out of these chips. YOU NOW SEE WHY I HAVE A WATERCOOLING SETUP. However, I BEG you to just download the program yourself. Try it. If you put in the numbers, you'll see for yourself.
Wanna guess how much my old athlon 1.33 was putting out at the same voltage at 1586 MHz? (Yes, it did hit those numbers, and did it stably. I got rid of it because I built my in-laws a machine with that chip for a birthday present)
OH MY GOD! 112.7 watts! AMAZING! Nah, not really. Just look around some of the posts in the amd cpu part of the forums, you'll see that this sort of thing is relatively common.
Seriously though, you're WAY off on your perception of how much heat amd cpus can radiate. Please note that I'm not calling you stupid at all. I'm just pointing out that you might want to catch up to speed on some of these things. As for me, here comes some more power to take the load off the 500W power supply.
In fact, as an afterthought, put in the specs on an athlon running at DEFAULT. Both voltage and speed. Guess what an athlon 1.4 gig running at default generates? I'll let you do this one yourself.
QUOTE I think part of the misunderstanding between us is that you're thinking I want a high multiplier. I don't-I could care less what the multiplier is, as long as the cpu speed is more than 950. Note that in my many earlier posts, I never once said that hitting any certain multiplier is a goal of mine. {so try 7x150}ANd forget it runs at 161fsb cause 161 is impossible for your system}THat processor is made for 200fsb and you want it to run at 320 Thats crazy dude
Read this post again you stated you push clock to max then you push fsb to max and this is a QUOTE:
Now, once I have individually determined the max speed of both the cpu and the fsb, I then set the fsb at the max speed, and get the cpu core speed as close to the max value I obtained earlier as I can with the multiplier. This ALWAYS results in a stable machine for me, usually with the cpu running at a little below the previously-determined max speed due to the fact that the multiplier generally does not place the cpu core speed at exactly where I determined it was stable. So, I choose the next lower multiplier for stability.
Obviously this statement by you is out the window.In this post you stated you leave the clock at default or very close then raise the fsb to find the stable max.IT IS obvious 161x9.5 never happened.I am guessing maby 9.5x140 is all you got if that.
Then I am thinking you lowered the clock and got it to 150x something then 155 ect and now 161 but lowering the processor each time.AND THEREFORE THROWING OUT YOU TRIED TESTED METHODS YOU STATED WORK SO WELL FOR YOU.
If you think that processor needs more power go get the 600 watter for that 45 watt processor,,,better yet disconnect everything but drive and vid card and see what happens.......Even with your set up I gurantee I draw way more watts than you do nomatter that case you use.And I also use Comp USA 500 watter.
< {so try 7x150}ANd forget it runs at 161fsb cause 161 is impossible for your system}THat processor is made for 200fsb and you want it to run at 320 Thats crazy dude
Read this post again you stated you push clock to max then you push fsb to max and this is a QUOTE:
Now, once I have individually determined the max speed of both the cpu and the fsb, I then set the fsb at the max speed, and get the cpu core speed as close to the max value I obtained earlier as I can with the multiplier. This ALWAYS results in a stable machine for me, usually with the cpu running at a little below the previously-determined max speed due to the fact that the multiplier generally does not place the cpu core speed at exactly where I determined it was stable. So, I choose the next lower multiplier for stability.
Obviously this statement by you is out the window.In this post you stated you leave the clock at default or very close then raise the fsb to find the stable max.IT IS obvious 161x9.5 never happened.I am guessing maby 9.5x140 is all you got if that.
Then I am thinking you lowered the clock and got it to 150x something then 155 ect and now 161 but lowering the processor each time.AND THEREFORE THROWING OUT YOU TRIED TESTED METHODS YOU STATED WORK SO WELL FOR YOU.>
I'm not sure I understand where you're coming from. I never SAID 161x9.5 happened. I SAID 161 happened, and I SAID 1270 MHz happened. I never said they happened AT THE SAME TIME. I never even implied that. Read a couple of my posts back and see the numbers I have ended up with. They are:
fsb is 161.64
cpu multiplier is 5.5
And yes, my method on this machine has failed to yield both an overclocked cpu and an overclocked fsb AT THE SAME TIME. You're not telling me anything by stating this (at least I think that's what you're telling me in your last post), this is the whole reason I posted originally, that and to get other people's experiences with koolance cases.
<ANd forget it runs at 161fsb cause 161 is impossible for your system}THat processor is made for 200fsb and you want it to run at 320 Thats crazy dude>
So what, now I'm a liar? How are you gonna tell me something won't work when I set the fsb speed in the bios myself, and looked at it using cpuid from within windows? Are you psychic? I'll tell you what, I'm in Montgomery (Alabama) right now because I have a couple of days off in my work schedule. When I get back down to Pensacola (where I'm stationed), I'll take a screenshot off the computer we're talking about and send it to you. It will be of cpuid clearly identifying the processor in question as a duron, and it will show the fsb speed. It will also show the 5.5 multiplier I'm currently running.
<If you think that processor needs more power go get the 600 watter for that 45 watt processor,,,better yet disconnect everything but drive and vid card and see what happens.......Even with your set up I gurantee I draw way more watts than you do nomatter that case you use.And I also use Comp USA 500 watter.>
Ok, here we go again. Did you not read one of my last posts? Did you not see where you can go to get the program I told you about called radiate? I notice you COMPLETELY ignore everything I posted about that. Or did you think I was making that up too, kinda like I was making up the fsb speed I'm running on that machine?
I'm not trying to be an a**hole here, but it seems like you're either telling me I'm a liar, or telling me I don't know how to do things like set the fsb speed.
You made a good point a little earlier in your last post. You mentioned that my duron is a 200 MHz fsb chip (which I think I told you myself a few posts ago) and then pointed out how ridiculous it is to expect it to go to 161 MHz. That DOES sound ridiculous, so it got me to wondering. I'm gonna post a question asking how this can be working for me, and see what I come up with. I ask that you read the responses I get, so we can BOTH be educated on this. It's obvious that neither you or I have all the answers in this situation.
diehrd,
This is what I posted in the forum "amd motherboards". The name of the thread is "basic fsb question". I ask you to please look at it and see what kind of responses I get. Feel free to join in on that one too, if you'd like to.
<Hi all,
I'm involved in a debate on another forum within overclockers.com, and I'm running my 950 duron machine at a fsb speed of 161. (actually a little over 161, according to cpuid) It has been pointed out to me that what I'm doing is impossible by another forum member, because my processor is setup for a 100 MHz fsb, since it's a duron. This got me to wondering how I'm doing this. I mean that literally, because of the huge disparity between what the fsb rating on the duron is and the fsb I'm actually running.
It has also been pointed out to me by the same member that my duron can't possibly be radiating more than 45 watts of heat. This is despite the fact that I have run that duron on 2.064 volts, and have gotten the core speed to 1270 mhz. Please note that this fsb speed and core speed have never happened at the same time, even with the really low multiplier that this would require. The fact that these two things have never happened simultaneously is part of why I made the original post that started this debate.
Anyway, any of you reading this might want to tune in to the lively debate happening. It's under the forum "cases and power supplies", and the thread is called "Anybody else have this experience with a koolance case?" Maybe some of you can weigh in with your knowledge and experiences on this one.>
Henry Rollins II
01-31-02, 07:11 PM
This is a very entertaing thread.
A question for mr. v8440: Do you like the koolance case? Itīs about the same price as a Lian-Li and a decent HSF, which make me find it intresting.
regards,
Henry.
<A question for mr. v8440: Do you like the koolance case? Itīs about the same price as a Lian-Li and a decent HSF, which make me find it intresting. >
Yes, I certainly do like it. The case itself is ok, but the cooling system is excellently done, such that a first-time watercooler like myself almost can't mess it up if they follow the directions. No air cooling I've tried has even come close to handling the heat produced by my old athlon 1.33 when it was running almost 1600 MHz at over 2 volts. I wouldn't trade it for anything.
Quote
Yes, I certainly do like it. The case itself is ok, but the cooling system is excellently done, such that a first-time watercooler like myself almost can't mess it up if they follow the directions. No air cooling I've tried has even come close to handling the heat produced by my old athlon 1.33 when it was running almost 1600 MHz at over 2 volts. I wouldn't trade it for anything.
Seems to me you should have kept that OLD 1.33 at ALMOST 1.6
<Seems to me you should have kept that OLD 1.33 at ALMOST 1.6>
Giving you the benefit of the doubt that you aren't trying to be sarcastic, yes, it was a good processor. I built my in-laws a machine using it, since I plan to replace it with an xp chip. Money constraints have kinda put that one off though.
Ok, here is the screensave of the impossible fsb speed on my duron machine. The server wouldn't accept the photo as an attachment at any reasonable resolution, so I had to resize it. This reduced its quality a lot, but it should still be intelligible.
I apologize for the lack of quality of the above screenshot. I'm obviously not a graphics expert! Anyway, if anyone has a way to resize a jpg without totally destroying any imitation of quality, I'd like to know what it is. The program I used to generate the image doesn't seem to offer settings to control the original size of the saved image.
I notice that there have been no more posts to this thread since I posted my terrible looking screenshot. Diehrd, you still there?
vBulletin® v3.8.7, Copyright ©2000-2013, vBulletin Solutions, Inc.