Got my 950 65nm

[larva]

When they shorten the pipes and add a memory controller maybe diehrd will once again use an Intel

Perhaps, but will you be any more willing to consider only diehrd cares?

 

[Okagi]

As we found in our first article on Intel's 65nm chips, power consumption did indeed go down quite a bit; however, it's still not low enough to be better than AMD. It will take Conroe before Intel can offer a desktop processor with lower power consumption than AMD's 90nm Athlon 64 line.

Yeah, but look at the leap in power consumption for the Presler going from 3.40Ghz to 4.25Ghz. That's quite substantial. I know people will want to take the Presler further than 4.25Ghz....

Presler is still a power hog, which translates directly into more heat from Intel.

Here's the link:

http://anandtech.com/cpuchipsets/showdoc.aspx?i=2578&p=4

 

[xTrEmEoVrClOcKr]

I just want the Cedarmills for raw speed, I couldn't imagine a 65nm single core on phase change.

 

[hUMANbEATbOX]

And hey new people or people looking for new chips should see both sides of the coin . .

so if you really need to point out both sides of the coin so badly, maybe make your OWN thread in the general cpu section clarifying it all for us. this is not the right thread for that, i'm sorry.

 

[S_Wilson]

Hiya Shawn, whats going on mate?

Anandtech say this: (http://anandtech.com/cpuchipsets/showdoc.aspx?i=2658)


Do you guys who have both (AMD and Intel chips), see this? Frankly, I don't know what to say about this, because my Opty 165 runs hot at 48C (2.7GHz, 1.5V) under load. So, where is everybody getting their Thermal comparision metric from?

Hi there S_N!

Power consumption and heat dissapation are two different things and Intels are higher on both accounts due to circuit leakeage at higher frequencies. Same thing happens to AMD past a certain point.

Temps reported by software are, many times, not comparable between identical systems much less systems based on different CPU's. The best use for the temp reporting is to see if changes you make to the cooling on your system benfitted your system.

Look at the encoding and other multimedia benchmarks on the Anandtech review, not much of an improvement over the 1Mb brethren. So what purpose does adding extra cache actually serve? Maybe this would be useful if the pipelines were LONGER.

2MB per core is the same as the 6xx series so I am sure the normal non-XE dual cores will see some benefit since there wont be hyper threading and we will be supplying abundant bandwidth.

So, have Intel moved away from the inefficient Netburst scheme?

The current mobile chips are not Netburst. Somewhere during Q3, Conroe will be released which is based on the current mobile chips and will be 65nm process. The core architecture will feature much shorter pipes among many other things. This will be Intels new direction and Netburst will ride off into the history books.

 

[speed bump]

This thread is about 65nm Intel dual cores however as they relate to AMD I had an interesting idea.

a 3.2ghz prescott beats a 2.2ghz athlon a 2.2ghz athlon 64 beats a 3.2ghz prescott. Both are very true facts. However why are they true? Becuase the A64 has an inagrated memory controller. Know considering the benefit that this has on a athlon platform consider what benefit this would have on a netburst platform.

Also some other things to consider are netburst was a huge step foward when it came out becuase it pushed the ability to use special instructions and it allowed huge boosts in performance when properly used. Although becuase of archiac OSes these have not been properly used.

However more to the point first of all diehrd you need to realize that he is using an Nvidia chipset motherboard, next to the orginal nforce3 is the worst chipset that Nvidia has ever built. His results with this CPU rival some of the best I have ever seen from this platform. If I put his CPU in my mobo I can guarentee that I could hit 5ghz if my ram could handle it.

Also Intels do get huge boosts from Die shrinks. Consider the highest northwood overclock on 130nm is 5ghz. the highest 90nm prescott is 7.5ghz. Thus far I have seen dual core CPUs on stock mach 2(not LN2 such as the previous 2) at 6.1ghz. So I am willing to bet you will see 8ghz fall pretty easily after Jan 6 which is when I hear the single cores will drop.

Also the four people that have 65nm CPUs that I know of this second at XS to have air results and 2 more have were people that have more experiance overclocking that most people here will ever experiance and those two would be Mikeguava also a member here and on of the current top single card 05 record holders and FUGGER owner of XS, who with chilly1 built the first cascade cooler as well as helped push the realm of extreme cooling more than anyone else. BTW both mikeguava and FUGGER bought there, mike from someone at XS and FUGGER from an Ebay auction.

Finally when you talk about the 5ghz celery remember that was on stock air and superpi stable, the best I have ever seen on 90nm stock air is 4.7ghz complete suicide barely even got a screenie. Now 5ghz on water and an expensive chip is possible but only with 660s and 670s. Which BTW cost up to 5 times more than that celery.

 

[batboy]

Cedar Mill and Presler is an intermediate step and not a new architecture. Basically, it's similar to the Prescotts except for the finer 65nm cores. Once Intel gets up to speed with the new die size, they will release the Conroe which will abandon the outdated Netburst architecture. So, as long as you know that, why not buy one of the new processors that happen to be a no brainer upgrade for those of us with i955 chipsets? The new CPUs will overclock higher than their current counterparts, so why not? I for one am not willing to leave the comfort of Intel chipsets even if AMD processors are slightly better right now. Looks like AMD is 6 months to a year behind the 65nm technology. Anyway, I agree, we are all starting to hijack this thread.

 

[Okagi]

Looks like AMD is 6 months to a year behind the 65nm technology.

So they are "behind". What's the difference if the "behind" products are outperforming the "ahead" products. Buying now would be like throwing your money down a hole. This next lot of 65nm has not evolved enough to make purchasing it a viable step. I say be sensable and wait for Conroe. Hopefully Intel will deliver the goods by then.

 

[S_Wilson]

When they shorten the pipes and add a memory controller maybe diehrd will once again use an Intel

And yes someone started a thread showing there 65nm chip.And the sad part is it performs no better then the 90nm

You seem to be the only person in this section that does not realize the cores are virtually identical only one is on a smaller process, so they will perform the same. Are you being dumb like a fox?

..And i did not see this user get to 5 gigs on air matter of fact I think he posted 4.2 gigs stable
Which is where the 8oo series chips performed.And why I posted to point this out..The chips cost more perform on the same level..I like many seen a 5 gig Celey..I have also seen 5 plus gig 90nm intels..

I don't recall anyone saying 5Ghz on a Pressler was going to be the norm.
Breaking 4Ghz with an 8xx chip is more the exception than the rule. I have played with several and the best one got just under 4Ghz with a ton of voltage and it absolutely saturated my watercooling system with heat. I believe based on research and experience that the average for an 8xx is probably 3.6Ghz and for a 6xx is about 4.1Ghz. I am not talking about phase change as I believe that refrigeration is for food and beer only. Now if the average person is able to get their Pressler to 4.2Ghz and Cedar Mill to 5Ghz with average cooling methods then those will be some decent gains, won't they?

And hey new people or people looking for new chips should see both sides of the coin . .

Maybe you should review your posting style in the Intel area and see if it worthy of your status.

 

[batboy]

Buying now would be like throwing your money down a hole. This next lot of 65nm has not evolved enough to make purchasing it a viable step. I say be sensable and wait for Conroe. Hopefully Intel will deliver the goods by then.

I like throwing my money down holes. I want to upgrade now. I have a system that merely requires me to drop in a new Presler, so why not? I like being a pioneer. Initial reviews are hitting 4.25 gig from these new Preslers using stock cooling. I say that's a big difference from the current 8XX Intel dual cores. The other thing people are forgetting is 2X2M=4M of cache. I don't expect everyone to drive the same kind of car as I do nor do I expect everyone to buy the same computer components. So, chill out.

 

[hyperasus]

I just realized that even if it was Conroe that was being released this week.....there would be haters in here telling us how bad Intels new chips suck. Maybe Intel could have done better then Presler, but meanwhile back at the farm presler is by for the best option us Intel guys have. Runs cooler, faster, clocks MUCH better, and has more cache. I'll be getting my presler asap regardless of what all the X2 users think them. Presler is here and we can't change that. Now lets see what they can do.

 

[3DFlyer]

I hope when Intel shortens those pipes, they don't go overboard like AMD did, and limit the clocks. I doubt this will happen. From all the reports I have read, the clocks are going to be affected very little. It is not a given that short pipes mean lower clocks. It does in AMD land, but that's because of the lack of research and development, and the fact that they lose sight of what they are trying to accomplish. They are "Intel Driven" instead of "goal driven". If they worried less about what Intel does, and focused on what they are doing, they might be taken seriously. They still do not offer a complete solution, and I've seen no reports that they are even working on the problem.

Intel on the other hand is on a mission and they are succeeding.

Fact- The New Pressler delivers cooler running, clocks to 5 on air and 6 on phase, delivers what was advertised all along.

People who say this will be the exception instead of the rule are only fooling themselves. These are ES chips that are being tested, they are NOT refined retail models. The retail models will be better as they always are with Intel CPU's. The future steppings will continue to get even better than that...as they always do with Intel CPU's.

There may be a few that are blinded by AMD's ramblings on their website and all the BS they spew regularly, but I can assure you that it will be quickly quelled as these CPU's are hitting the market in a few days.

When Serious OCF members get ahold of these we will see clocks higher than ever before...just like we did with the 6xx series.

 

[larva]

I hope when Intel shortens those pipes, they don't go overboard like AMD did, and limit the clocks.

AMD clocks aren't purely archictecture-limited. Intel is still the silicon king, and will likely always maintain a clock speed advantage as a result, long pipeline or short.

But really, this is quite enough of the AMD vs Intel rhetoric here. We have a general CPU section, and are inclined to allow even spirited debate of the relative merits in that section. This is indeed the Intel section, and I would prefer (well, let me re-phrase that, I insist upon) a lot less of that content here. If you want to explain why this Intel technology isn't compelling vs other Intel technologies, fine, but I'm drawing the AMD/Intel line at precisely this point in the sand.

 

[3DFlyer]

AMD clocks aren't purely archictecture-limited. Intel is still the silicon king, and will likely always maintain a clock speed advantage as a result, long pipeline or short.

But really, this is quite enough of the AMD vs Intel rhetoric here. We have a general CPU section, and are inclined to allow even spirited debate of the relative merits in that section. This is indeed the Intel section, and I would prefer (well, let me re-phrase that, I insist upon) a lot less of that content here. If you want to explain why this Intel technology isn't compelling vs other Intel technologies, fine, but I'm drawing the AMD/Intel line at precisely this point in the sand.

The General Section is really the AMD Fanboi section, and back about 2 months ago, I decided to avoid it like the plague. I am starting to believe this section is heading that way, and may have to avoid it too.

I also don't post "rhetoric" or get into debates. I resigned from the "debate committee" a long time ago. Debate is pointless. Factual infoprmation sharing is where it's at.

 

[larva]

The General Section is really the AMD Fanboi section, and back about 2 months ago, I decided to avoid it like the plague. I am starting to believe this section is heading that way, and may have to avoid it too.

Honestly, that's just the pot calling the kettle black.

 

[RangerXLT8]

I hope when Intel shortens those pipes, they don't go overboard like AMD did, and limit the clocks. I doubt this will happen. From all the reports I have read, the clocks are going to be affected very little. It is not a given that short pipes mean lower clocks. It does in AMD land, but that's because of the lack of research and development, and the fact that they lose sight of what they are trying to accomplish. They are "Intel Driven" instead of "goal driven". If they worried less about what Intel does, and focused on what they are doing, they might be taken seriously. They still do not offer a complete solution, and I've seen no reports that they are even working on the problem.

Intel on the other hand is on a mission and they are succeeding.

Fact- The New Pressler delivers cooler running, clocks to 5 on air and 6 on phase, delivers what was advertised all along.

People who say this will be the exception instead of the rule are only fooling themselves. These are ES chips that are being tested, they are NOT refined retail models. The retail models will be better as they always are with Intel CPU's. The future steppings will continue to get even better than that...as they always do with Intel CPU's.

There may be a few that are blinded by AMD's ramblings on their website and all the BS they spew regularly, but I can assure you that it will be quickly quelled as these CPU's are hitting the market in a few days.

When Serious OCF members get ahold of these we will see clocks higher than ever before...just like we did with the 6xx series.

This man speaks the truth and I'm going to be one of the OCF members to OC one of these puppies on air.

 

[3DFlyer]

Honestly, that's just the pot calling the kettle black.

Oh really?!? Then maybe it's time I move on to another forum.

If that's the way people feel here, then i don't need it. I ain't. If you really belive that, then just see if the fanboi flames diapear once I am gone...which is right about now. So long.......

 

[Who]

I hope when Intel shortens those pipes, they don't go overboard like AMD did, and limit the clocks. I doubt this will happen. From all the reports I have read, the clocks are going to be affected very little. It is not a given that short pipes mean lower clocks. It does in AMD land, but that's because of the lack of research and development, and the fact that they lose sight of what they are trying to accomplish. They are "Intel Driven" instead of "goal driven". If they worried less about what Intel does, and focused on what they are doing, they might be taken seriously. They still do not offer a complete solution, and I've seen no reports that they are even working on the problem.

Intel on the other hand is on a mission and they are succeeding.

Fact- The New Pressler delivers cooler running, clocks to 5 on air and 6 on phase, delivers what was advertised all along.

People who say this will be the exception instead of the rule are only fooling themselves. These are ES chips that are being tested, they are NOT refined retail models. The retail models will be better as they always are with Intel CPU's. The future steppings will continue to get even better than that...as they always do with Intel CPU's.

There may be a few that are blinded by AMD's ramblings on their website and all the BS they spew regularly, but I can assure you that it will be quickly quelled as these CPU's are hitting the market in a few days.

When Serious OCF members get ahold of these we will see clocks higher than ever before...just like we did with the 6xx series.

Their clocks are going to be affected. Woodcrest, which is based on Conroe and Merom, will launch at only 2.5-2.93 GHz. I doubt that any Conroes will launch much faster, if at all, than that.

 

[larva]

Oh really?!? Then maybe it's time I move on to another forum.

I'm sure you've been through your share.

If that's the way people feel here, then i don't need it. I ain't. If you really belive that, then just see if the fanboi flames diapear once I am gone...which is right about now. So long.......

To use an old saying, don't let the door hit ya where the good lord split ya.

 

[Jame Gumb]

I'm glad guys like you are pioneers, batboy....you're the ones that pave the way and then share your specs and info. That's what makes this site great.

Dang 3D left....... I pour some out for a homie tonight.