Coolaler shows us just how pathetic the i7-4960X will be...

[Ace123]

reading some of the posts on here is hard. How can people constantly berate Intel for releasing the fastest commercially available processor because they arent releasing it on "THEIR TERMS". who gives a crap whats under the hood of a processor if its destroys everything the competition puts together.
The question on the minds of real enthusiast is, Does the processor work for its intended purpose, how does it perform, and how well does it overclock. I dont know about you guys, but my 3930k has helped me take 9 world records at 5.2ghz on water.
I invite everyone who is complaining to pool your money, start a processor foundry and make them "correctly" since it seems intel cant get the job done right to make everyone happy. In the mean time im going to be rushing home giggling like a school girl the day the fedex guy drops mine off at the house so I can trash on some more scores.

The title is shameful

[lutjens]

reading some of the posts on here is hard. How can people constantly berate Intel for releasing the fastest commercially available processor because they arent releasing it on "THEIR TERMS". who gives a crap whats under the hood of a processor if its destroys everything the competition puts together.
The question on the minds of real enthusiast is, Does the processor work for its intended purpose, how does it perform, and how well does it overclock. I dont know about you guys, but my 3930k has helped me take 9 world records at 5.2ghz on water.
I invite everyone who is complaining to pool your money, start a processor foundry and make them "correctly" since it seems intel cant get the job done right to make everyone happy. In the mean time im going to be rushing home giggling like a school girl the day the fedex guy drops mine off at the house so I can trash on some more scores.

The title is shameful

What does it matter if it beats anything AMD has to offer? Well, Intel should immediately make the Extreme chip into a 4C version...it would still beat the 8350...

Technology has always been about pushing the envelope. You boast about your world records on water...imagine what a dual 12C system under chilled water could do. Some people are satisfied with a middle of the road processor, as it meets their needs. The 3930K clearly meets your needs and I'm glad you're satisfied with it. It doesn't meets the needs of everyone. Some want the absolute best that a company can offer and are willing to pay a very premium price for it. This is something that isn't being offered at the moment by Intel. The top Xeon is an ideal processor, except for the locked multiplier that is locked simply because Intel wants to. I have no problem with Intel harvesting dies and selling them as desktop chips, but they need to unlock the top Xeons.

The benchmarks we have seen from Ivy Bridge-E are indeed pathetic...a maximum of a ~10% increase, with most increases being in the 5% range. Intel should, at the very least, have an 8C Extreme chip that would truly be a worthwhile upgrade for people already using LGA2011 systems. Make the regular consumer chip 6C, but a 6C processor no longer qualifies as Extreme.

[cx-ray]

I would have bought an Ivy-E 8 core. Now I'm not upgrading at all until perhaps Haswell-E if that turns out to be decent. Then again, the potential sales to those who would upgrade to Ivy-E from SNB-E might be so low that Intel can't be bothered with it.

[Kain665]

You're making assumptions about the ML370 G6, much like assumptions made elsewhere in your argument. Assumptions about how others use hardware are often surprisingly inaccurate. I'm running GTX690s in my ML370 G6 using PCI-E power adapters that are an option from HP. The ML370 G6 has two additional power connectors for use with video cards, HP being the only server manufacturer to my knowledge that provides such connections. A third is avilable as well, but only if I forgo the use of the third drive enclosure. SLi is possible with the use of an SLi hack and even though this is an "enterprise" server, it plays games surprisingly well...

And provides memory and I/O expansion abilities that consumer machines can only dream about...

Good to know. I put together a machine last November on Intel's reference W2600CR2 platform in their reference case and had to jump through many hoops to get a PDB that supplied 2x PCI-E 6 Pin and 2x PCI-E 8 Pin connectors.

That's the thing...both myself and others would be willing to buy said 16 core CPU, and pay handsomely for it, but Intel refuses to produce one, for any price. All they would need to do to satisfy many people would be to unlock the top Xeons...those who want to use the feature can and those who don't can run the chip at default.

Can we can show of hands of who's willing to spend north of 5 grand for a pair of unlocked 10c Xeons?

[lutjens]

Good to know. I put together a machine last November on Intel's reference W2600CR2 platform in their reference case and had to jump through many hoops to get a PDB that supplied 2x PCI-E 6 Pin and 2x PCI-E 8 Pin connectors.

How'd you manage to get a modifed PDB from Intel? I considered that platform, but rejected it based on the lack of PCI-E power.

Can we can show of hands of who's willing to spend north of 5 grand for a pair of unlocked 10c Xeons?

I would for a pair of unlocked Xeons...but in reality all Intel needs to do here is unlock the existing top bin chips (based upon the SKU stack previously posted, would be the 2.7GHz 12c chip). This chip'll doubtlessly be approx $2.5-3k as it is, so even if only a few people buy them on the basis of being unlocked, the end result is that they'll sell more of their top, high-margin chips. No new SKUs would be required and those who want to play will have no choice but to ante up for them. Intel sells more chips, makes more money, and those who want to overclock these chips will be able to. Win win all around from what I can see...

[aussie-revhead]

I would have bought an Ivy-E 8 core. Now I'm not upgrading at all until perhaps Haswell-E if that turns out to be decent. Then again, the potential sales to those who would upgrade to Ivy-E from SNB-E might be so low that Intel can't be bothered with it.

If Haswell-E turns out like Haswell then we are all in trouble, do we have to wait another generation with our fingers crossed hoping for a decent enthusiast cpu? Haswell arch will be too hot for 8 unlocked cores .... hurry 2015 ....

[informal]

I doubt we will see 8 Haswell cores for high end desktop anytime soon.

[Ace123]

If Haswell-E turns out like Haswell then we are all in trouble, do we have to wait another generation with our fingers crossed hoping for a decent enthusiast cpu? Haswell arch will be too hot for 8 unlocked cores .... hurry 2015 ....

I'm having trouble figuring this post out. If haswell-e turns out like haswell we will all have the opportunity to purchase intel's newest processor thats faster than the previous generation. Unless you think you are "entitled" to something more than that?

[naokaji]

I'm having trouble figuring this post out. If haswell-e turns out like haswell we will all have the opportunity to purchase intel's newest processor thats faster than the previous generation. Unless you think you are "entitled" to something more than that?

Would Haswell-e beat Haswell? no doubt about it, but how much faster will haswell-e be when it won't be compared to haswell but rather the haswell successor?

[Ace123]

It will probably compare to sandy-e as to ivy bridge

[Zaxx]

I'm not a member of the extreme cpu club (obviously) but the current enthusiast short coming is also visible by looking at the non-E launches. The performance increase from Sandy to Ivy to Haswell have all been modest at best. This kept me from pulling the trigger on a Haswell upgrade cuz I usually go for every-other launch following Intel's 'tick-tock' cycling...not this time. The marginal gain just isn't enuff to justify buying a new mobo etc. Being a storage fanatic, I also wanna see what the Z97 chipset has in store...sata express if we're really lucky?

[cx-ray]

It could be Intel wants to refocus on their HEDT platform by making the mid-range a little less attractive. Simply put, overclocking in the future is going to cost you more.

Not too shabby if this is accurate:
http://vr-zone.com/articles/intel-co...led/37832.html

[informal]

Would Haswell-e beat Haswell? no doubt about it, but how much faster will haswell-e be when it won't be compared to haswell but rather the haswell successor?

What Haswell's successor? If you are referring to broadwell it won't launch on desktop next year but sometime in 2015, probably aligned with Haswell-E.

Also I doubt Broadwell will bring anything noticeably better IPC-wise VS Haswell. If Haswell brought us such a miniscule IPC increase over IB and it had all those "uarchitectural bullet points" (most for AVX2/FMA/TSX which are useless for legacy code), Broadwell's chance to wow us more than that are slim. Probably a straight shrink to 14nm, a clock bump and fer percents of better IPC if that. Haswell-E will have massive caches (L3 especially), a thing Broadwell won't have. So Haswell-E will looks very impressive no matter what you use to compare it with.

[safan80]

all socket 2011 support vt-d, they want to force you up the line if you want to overclock and have vt-d. that said, amd and NV will both have on card vitalization, and nothing uses vt-d now so there is no real point in it as you only need it for graphics.

http://software.intel.com/en-us/blog...or-directed-io
http://xen.xensource.com/files/xensu...d_Nakajima.pdf

VT-d is used by Virtualbox, VMWare, and Xen.

VT-d is not restricted to graphics, although that's the biggest hope for it.

[Aberration]

Diminishing returns. Until the next major breakthrough, we can expect similar results for all successive products.

[aussie-revhead]

Diminishing returns. Until the next major breakthrough, we can expect similar results for all successive products.

So what will be the next breakthrough? Worthy of its own thread I think....

Let's hope more software and next gen games will start using more than 4 cores, then we might see some serious headway in the CPU race.

[Aberration]

So what will be the next breakthrough? Worthy of its own thread I think....

Let's hope more software and next gen games will start using more than 4 cores, then we might see some serious headway in the CPU race.

I don't think we know that, any better than we knew the transistor was coming after vacuum tubes. I can take guesses, but most is just trying to improve what we have been doing for decades. Even FinFet, exotic materials, its all just trying to improve what they did before. They are squeezing everything they can out of it. Remember the gains we used to see? If they could make something better, they would. I, like everyone else, sure wish we would see more parallel processing, but the fact is that is also prohibitively expensive.

[Ace123]

I still think the reaction is astounding considering the level of chip we are talking about. If we were talking about a 10% gain over a k6-2 350 I could see the disappointment. what we are talking about here, is a 10% gain over the fastest commercially available enthusiast processor on the planet.
Consider this: If your processor scores.... 10 points. a 10% gain over that is only 1 point. If your processor scores 100 points, a 10 gain of that is 10 points........ If your processor scores 10000 points, a 10% gain over that is 1000 points.
That being said lets say a 3770K scores 10,000 points, and a 3930k scrores 15000 points. Put on another 10% performance gain, that processor now scores 16500 points. So that enthusiast grade processor that was only 50% faster than the next closest grade processor is now 65% faster than the next closest grade. Thats a huge jump in performance.
People are freaking out because 10% doesnt sound like a huge gain in power because we are only talking about 10%, thats only 1 our of 10. You have to look at it in perspective. If your 300000 dollar house lost 10% value over night... thats Huge. Sandybridge-E is huge, and a 10% gain over that hella good.

[schmuckley]

You can see more pics here http://extreme-box.com/index.php/not...ue-el-i7-3970x

only 10% more....we need AMD reaction now

Amen to that..I'm very patient..but geez..It's been since 2009 they really had anything worthwhile
I'm hoping they're working on something over there.AMD is so much more fun.

[lutjens]

They don't want to make an unlocked 8 core CPU because to offer such a chip at the typical "Extreme" price point of $999 would force them to offer 6 cores at a lower price, which would lower their profit margin and divert dies with 8 functional cores, that could instead be sold as Xeons, into the i7 line (again hurting their profits).

A lower priced 6 core Ivy-E would also make it much more attractive to purchasers that would've normally gone with the i7-4770K...and force Intel to lower the price of it (and consequently the rest of the lineup) to restore its attractiveness.

I fully understand all of this and considering that Intel is a for-profit corporation, it's entirely understandable, if not agreeable to us enthusiasts. But Intel can do what it want to their so-called "HEDT" platform...it's not a concern.

What I want is for the E5-2697 v2 to be unlocked...and I don't care what they want to charge for it. It would cost nothing for them to unlock this chip and they might sell a few more of them to boot, which being the one of the most profitable chips, would make them even more profit. If they only want to unlock the top 4-way chip, that'll work too...I don't care, as long as one fully functional chip gets unlocked.

Fat chance of that happening though...