What do I need to get Pent D to 4Ghz...

[Puni Puni Poemy]

I just want to know what I need to get a Pentium D to 4Ghz on Air. Thats all, I dont want to spawn all these arguements.

I dont care if I have to raise the Voltage. I dont care if I need a 80$ 8lbs Heatsink and Noisy Fan... Just need to know what can do it (on air)

Its not that I dont have the money for water, but I am not at that stage yet to trust me Multi Thousand dollar box with putting water inside it. I just dont have the ability to do stuff like that yet.

[deltarealm]

I was like you about water. Made a start with water cooling using a Corsair COOL kit, reasonable price, easy to put together, radiator bolts to back of case, tiny pump, res take one 5" drive bay. Worth a thought.
If anything with air, your be doing your best to stay under the thermal trottle threshold. Add overclock plus voltage, you will be there in no time.

[ZX7891]

Delta, good air coolers will be better than that piece of crap.

[Lithan]

So? the point is, its at 4.1 on stock volts, meaning, that it MAY be stable at 4 with the same volts, or if not, maybe one step up in voltage. with a good heatsink the lil extra voltage shouldnt cause enough heat to be concerned about. I have a friend who is runnign a single core 3.2 at 3.8 on stock voltage. it will boot all the way up to about 4.15 on stock volts.

Methinks someone doesn't put two and two together very well.

In my experiences, 4 ghz with an intel chip should be easily doable with good aircooling.

And how many of those experiences are with P4-d's? Since those would be the only relevant experiences to the question at hand.

That said. A X2 @ 2.8Ghz (actually possible and Commonplace in fact) on air would TROUNCE a p4d @ 4ghz (not been done) on air.


But the relevant question is why are you even looking at dual core. What do you want your rig to do? If you're in a situation where dual core is actually beneficial, then either spring for X2, or settle for a hyperthreading single core cpu. More likely however, you don't need dual core now and should just wait for X2's to drop in price or Intel dual cores to stop sucking

[mr_knowitall15]

keep in mind that its a prescott that we're talking about which already push air cooling on single core. prescott + dual core generates upwards of 210 watts of heat, thats 3x what my p4 is rated for. now overclock it. this is the reason few motherboards will support the chip. the mofsets get mightly toasty.

where the hell did you get 210 Watts from. Im positive the thermal design was supposed to be 130W, compared to teh single core prescotts 110....

[mr_knowitall15]

Methinks someone doesn't put two and two together very well.




And how many of those experiences are with P4-d's? Since those would be the only relevant experiences to the question at hand.

That said. A X2 @ 2.8Ghz (actually possible and Commonplace in fact) on air would TROUNCE a p4d @ 4ghz (not been done) on air.


But the relevant question is why are you even looking at dual core. What do you want your rig to do? If you're in a situation where dual core is actually beneficial, then either spring for X2, or settle for a hyperthreading single core cpu. More likely however, you don't need dual core now and should just wait for X2's to drop in price or Intel dual cores to stop sucking

wtf are you talking about...ive tried to help this guy out with his question, ive provided links to people who have done it, so dont say it cant be done. Its negative people like that that just bring down forums like this. *cough (evilunderpantsgnome) Cough cough*. And lithan, all youve condributed is a bunch of wasted space because you seem to feel the need to come over to thius section and rant on about AMD. Did Puni ask "what does it take to get an X2 to xxxx Mhz??? NO! did he ask anything about, "should i get a pentium D or athlon X2? NO! he asked what hed have to do to get a D to 4 ghz, and ive tried to show him something usefull in regards to his question and some morons that are either uninformed, or biased and feel the need to try to persuade other to think their way, are ruining this thread. so people like lithan, do us all a favor and stay in the AMD section if all youre going to do is bash intel. go youselves. Cheers

[Puni Puni Poemy]

lol

Well thanks. I understand that it will be tough and might not be able to do it, but I am going to try. I think a good cooler like an XP-120 or something might work out well with a Good Solid 120mm Fan. We will see what I can get...

[Delirious]

Ok

First rule of computing, never count on getting a "lucky" chip and build your system around it. Your best bet would be to go with some decent watercooling to be sure of good and stable cooling, and can be had for much less than a vapochill if you do some of the work yourself.

Also you wont get a Pentium D to 4ghz on stock cooling. Keep dreaming.

The Pentium D 830 which runs at 3.0ghz per core sounds like a good choice. These chips have been getting spectacular overclocks depending on cooling. I know in the Xtreme Overclocking section, there is one running @ ~5ghz under DI.

Also next year Intel's desktop Dothans hit. Id love to see a 4800+ beat a Dual Core 3ghz Dothan with 2mb's of cache each taking advantage of dual channel ddr2 and in 2007 they gain a memory controller. Intel really set out its plans, AMD got too confortable with their K8 architecture, and now they will have to up Intel somehow.

How can u compare technology that wont even be out for another year to tech thats out now? Of course it will probably be better if not match it, but Amd will have something else out by then as well.

[Lithan]

wtf are you talking about...ive tried to help this guy out with his question, ive provided links to people who have done it,

No you haven't. He asked for 100% stable. You told him you've seen multiple people getting 4ghz with stock cooling and volts. That's not help, that's false hope.

so dont say it cant be done.

"To get a P4D to 4ghz on air would not be easy at all from all the reports I'm seeing. And if you get a lucky chip that can, it wouldn't be on quiet air. Wait for another revision."-Lithan

Yep. Not only was what I said completely wrong. But you are correct that I said it couldn't be done. Oh how can I hope to sway his judgement when my AMD fanboyism comes up against your mighty reason.

Its negative people like that that just bring down forums like this. *cough (evilunderpantsgnome) Cough cough*.

In the same way it's people who lock their doors who cause all home burglaries, right?

And lithan, all youve condributed is a bunch of wasted space because you seem to feel the need to come over to thius section and rant on about AMD.

I mentioned AMD once in four posts here. Might I add, after this post,

The way i see it, its all up to personal preferance Puni. All things considered, Amd is prolly winning performance wise in most applications (at least at the high end, havent seen much as far as the other AMD dual core chips go yet...), but intels are fun to OC IMO because of the large gains that can be possible, plus their prices are much more attractive than AMDs right now. With AMds CHEAPEST dual core running at $532, the next step up is $729, then $824, then $1065 for OEM version of the top chip, ($1099 for retail boxed version). where Intels chips are $248 at the low end, $331 for the middle speed chip, and $551 for the top chip speedwise, escept the EE, which is $1015. So considering that you can buy Intels 2nd to top chip, for about the same price as AMDs entry level dual core, you can see the difference in price. However, like i mentioned, if the lower price and overclockability of Intel is enough to outweigh AMDS performance crown is up to the buyer.

Where not only do you appear to have an unhealthy fixation on AMD, you also argue for P4D by leaving out key facts. Intel P4D boards cost $100+ more than X2 boards. There is a gigantic advantage with regards to board choice and options with AMD. AMD will have a $345 X2 out in two weeks. AMD's 4200+ beats the tar out of intels 3.2ghz P4D at stock and overclocked (even in your fantasy world where they are getting 4ghz @ stock vcore on stock cooling).

Did Puni ask "what does it take to get an X2 to xxxx Mhz??? NO! did he ask anything about, "should i get a pentium D or athlon X2? NO! he asked what hed have to do to get a D to 4 ghz, and ive tried to show him something usefull in regards to his question and some morons that are either uninformed, or biased and feel the need to try to persuade other to think their way, are ruining this thread. so people like lithan, do us all a favor and stay in the AMD section if all youre going to do is bash intel. go youselves. Cheers

So, in short: Don't point out when you mislead someone because it hurts your widdle feewings?

As for the heat output, you are right about TDP 130watts, yet such is not the full story.

"Processor power dissipation simulations indicate a maximum application power in the range of 75% of the maximum power for a given frequency. Therefore, a system designed to the thermal design point, which has been set to approximately 75% of the maximum processor power would be unlikely to see the thermal control circuit active and experience the associated performance reduction"

Now throw in overclocking and we are obviously well into 200watts+ being a reality.

[mr_knowitall15]

god youre missing the point. all i said was, its possible. it may not happen on every chip, but it can be done. THATS my point. Just because a lot of people said it was impossible, i found a link.

Where not only do you appear to have an unhealthy fixation on AMD, you also argue for P4D by leaving out key facts. Intel P4D boards cost $100+ more than X2 boards. There is a gigantic advantage with regards to board choice and options with AMD. AMD will have a $345 X2 out in two weeks. AMD's 4200+ beats the tar out of intels 3.2ghz P4D at stock and overclocked (even in your fantasy world where they are getting 4ghz @ stock vcore on stock cooling).

what the hell do you mean an "unhealthy fixation"? all im trying to do is state what ive read about the chips on sites liek anandtech, and what ive seen done in forums.
Quote:
Its negative people like that that just bring down forums like this. *cough (evilunderpantsgnome) Cough cough*. whoops i meant zabomb, sorry undrpantsgnome...

[Zebo]

As long as I am not throttling under load I am ok with wahtever i have to do.

What type of cooling do you recommend? (air cooling)

Need phase, period!

Air?
You will throttle stock with intels dual cores let alone overclocking.
Proof 1
Proof 2

Originally Posted by X-bit Labs
Pentium D processors do not allow monitoring the temperature of both cores, this feature is implemented only in dual-core server processors. So, to our great disappointment we cannot find out what actually happened with the second core. In our case, only the second core got overheated and went to thermal throttling, while the first core kept working normally.

Proof 3, notice the heart monitor effect on CPU load, that's throttling

If you OC to 3.8 you are using 338W since a 670 uses 169W and now you got two of them! (This is assuming you don't add volts) No air cooler can handle these temp output w/o throttle.

[Lithan]

It's possible for your cab driver to be a terrorist. That doesn't condone writing down his name and going to the police and telling them you've seen him commiting terrorist acts on several occassions. That is the distinction between presenting a slim chance as a possibility, and making wild presumptions, then declaring them to be documented fact.

[dippyskoodlez]

It's possible for your cab driver to be a terrorist. That doesn't condone writing down his name and going to the police and telling them you've seen him commiting terrorist acts on several occassions. That is the distinction between presenting a slim chance as a possibility, and making wild presumptions, then declaring them to be documented fact.

dual cores almost, or DO throttle with the stock cooler...

I think its safe to say the true chances of getting 4ghz air is pretty rare...

[Charles Wirth]

Hitting 4Ghz with 820 and 830 is easier than you may think and it can be done with stock cooler but I reccomend a Zalman.

The key to 100% stability is good case air flow, and how hard is that??

With watercooling 4Ghz can be reached with no vcore increase.

screw what the reviews say, we know better than to setup with 200FSB and 5-4-4-x timing so perfomance marks at that speed are a complete joke when compared to an X2 with 2-2-2-5 timing. The reviewers knew this when the reviews were done. I often see them using ratio's too and that also hurts performance.

Load that X2 with 2GB of memory and it narrows the gap even more.

Face it, we dont have to run our rigs like the reviewers dumbed down config. Even running 266FSB 1:1 3-2-2-4 is a huge improvement over 200FSB 5-4-4-x and that is a no brainer with stock heatsink and stock volts.

Arbitration has lower latency than a HT chip. I dont have a like to back that up but I do know directly from Intel early on when pros and cons were thrown out for dropping HT.

[Lithan]

Strange that they can run 4ghz with stock cooling since according to you fugger, the reason toms test had their 3.2ghz P4D hitting 90*C+ and crashing was due to the fan being a stock 820 fan. Now you say you can hit 4ghz with an overvolted 820 on stock 820 cooler?

Where's your source or evidence of this? No one else has seen a P4D on stock cooler @ 4ghz. And that's not only reviewers, that includes several Xtreme (and other forums) members with these chips who can't get 4ghz on top of the line air cooling, much less stock.

AMD's are going to gain as much FSB (and likely a good deal more) as intels and many reviewers are using 3-2-2 ram for the intels (for example Anand's Linux compairison). So your claims there are unfounded. Speaking of which, what testers used ddr2-400 4-4-4 in their tests? Most used ddr2-667 4-4-4 or ddr2-533 3-2-2 (which at the time were the fastest speeds available without overclocking, and obviously the most fair compairison possible).

I don't know about your source for the Hyperthread versus Dual core claim, but tests of P4D versus P4EE's tend to disagree with you there.

Also what are you basing your claims that P4D's deal with two gigs any better than X2's do. I've yet to find a system of any platform that will run two gigs 1t at low latencies into the mem speeds where these chips would obviously be going.

[zabomb4163]

Strange that they can run 4ghz with stock cooling since according to you fugger, the reason toms test had their 3.2ghz P4D hitting 90*C+ and crashing was due to the fan being a stock 820 fan. Now you say you can hit 4ghz with an overvolted 820 on stock 820 cooler?

Where's your source or evidence of this? No one else has seen a P4D on stock cooler @ 4ghz. And that's not only reviewers, that includes several Xtreme (and other forums) members with these chips who can't get 4ghz on top of the line air cooling, much less stock.

AMD's are going to gain as much FSB (and likely a good deal more) as intels and many reviewers are using 3-2-2 ram for the intels (for example Anand's Linux compairison). So your claims there are unfounded. Speaking of which, what testers used ddr2-400 4-4-4 in their tests? Most used ddr2-667 4-4-4 or ddr2-533 3-2-2 (which at the time were the fastest speeds available without overclocking, and obviously the most fair compairison possible).

I don't know about your source for the Hyperthread versus Dual core claim, but tests of P4D versus P4EE's tend to disagree with you there.

Also what are you basing your claims that P4D's deal with two gigs any better than X2's do. I've yet to find a system of any platform that will run two gigs 1t at low latencies into the mem speeds where these chips would obviously be going.

fugger has had pentium D chips for a very long time. he got them well before they were released to the public. screw tom's hardware. fugger knows what he is doing. you've seen his 400fsb stable thread, right?

[Lithan]

Fugger also uses Phase change. He might know the chips, but I doubt he knows much about the chips when it comes to air cooling. Also, it's pretty obvious he's exaggerating the impact of tighter timings on everything except a few popular synthetic benchmarks.

[zabomb4163]

http://www.xtremesystems.org/forums/...ad.php?t=66875

[Lithan]

Pretty much confirms what I'm saying. Noone's getting stable at 4ghz with any less than w/c and even then it's hardly "easy". I'm not trying to be an ass about it, but I don't understand why even though he's getting 3.73ghz on air with what's probably a pretty damn good example and certainly the user isn't holding it back, he's here saying that 4ghz on air is easier (than it seems). And that it's possible with stock cooling. I tend to suspect that he's saying 4ghz shots aren't hard to get on air, or maybe even 4ghz sp1m. But what we are talking about here is 100% stable in everything. And 266mhz is a big leap, especially considering that he likely wasn't running stock fan to get that (though he may be).

[Zebo]

Pretty much confirms what I'm saying. Noone's getting stable at 4ghz with any less than w/c and even then it's hardly "easy". I'm not trying to be an ass about it, but I don't understand why even though he's getting 3.73ghz on air with what's probably a pretty damn good example and certainly the user isn't holding it back, he's here saying that 4ghz on air is easier (than it seems). And that it's possible with stock cooling. I tend to suspect that he's saying 4ghz shots aren't hard to get on air, or maybe even 4ghz sp1m. But what we are talking about here is 100% stable in everything. And 266mhz is a big leap, especially considering that he likely wasn't running stock fan to get that (though he may be).

Air??

Who even cares if it's 4ghz stable if the chip is turned off half the time? Show me two SS one with 2 instances of SPi 32M @ 3/4 done, and one done with throttlewatch open the whole time during both SS.

I'll make a few predictions based on what I've linked earlier
1. The system probably will not complete the benches and if it does,
2. No air solution won't throttle at any speed above 3600Mhz.
3. The scores will be signifigantly different above 3600Mhz.

[boshuter]

LMAO...... I read this whole thread and then had to make sure I wasn't on Anandtech with all the AMD fanboys crapping on every Intel thread thats posted

If you took out all the replies from people who have absolutley no experience with the subject at hand other than what they have read "on the internet".... this thread would only be 2 replies long..

[Zebo]

And you could'nt even add links.. pretty worthless post other than to fan some flames.

How about you, or anyone, show 20-30min loaded of PentiumD OC air goodness w/o throttle?

yea.. "show me"

[Lithan]

LMAO...... I read this whole thread and then had to make sure I wasn't on Anandtech with all the AMD fanboys crapping on every Intel thread thats posted

If you took out all the replies from people who have absolutley no experience with the subject at hand other than what they have read "on the internet".... this thread would only be 2 replies long..

Yep everyone who tells you that P4's can't run 50million ghz and cure cancer in minutes is an amd fanboy. Of course by your logic that means that amd fanboys are the only ones who know what they are talking about, since you haven't found a p4d @ 4ghz, 100% stable on air.

Amazing how the fanboys are crapping on the thread for actually giving the guy an honest answer. Meanwhile the 'others' are BS'ing him left and right about how 4ghz is easy on air and P4-D's are made of candy. You know what, if he wants to listen to that and buy a p4d in order to get 4ghz on air, more power to him. Let him learn the hard way that people without a leg to stand on just call anyone who argues their baseless, insane arguements a fanboy of whatever competes with the product they are selling.

[Nasgul]

bla, bla, bla, bla.....

It's time for you to

Here, enjoy.

Pathetic. You think you know, you don't even know that the correct name is not "P4D", it's PENTIUM D.

Intel Processors are much, much...better than amd (advanced Mexican devices and hector ruiz) any given day.
Don't belive?
Here, go spend your time here:
http://www.xtremesystems.org/forums/...ight=amd+death
http://www.xtremesystems.org/forums/...ad.php?t=66550
http://www.xtremesystems.org/forums/...ad.php?t=66648
http://www.xtremesystems.org/forums/...ad.php?t=67261

You will have to laugh at this one http://www.xtremesystems.org/forums/...ad.php?t=68708 so much for the great pos ever.

While you're at it, go find out why the turd-ion cpus suck so much: http://www.gamepc.com/labs/view_cont...urion64&page=1

When it comes to Intel, you don't know anything. And to your dismay the Pentium 4 CAN run at over 50+GHZ,
http://i6.photobucket.com/albums/y218/Nasgul/Insane.png.

[Lithan]

One wonders if an AMD troll behaved like Nasgul, how long it'd take for him to be banned.