New 22-nm Intel Ivy Bridge Chips 37% Faster Than Previous Generation, Use 50% Less Energy 11 Comments

If it won't work on Linux as #8 says, you're in trouble. Low-end PCs are disappearing. Intel processors are being used in servers, servers run Linux. I'm building a high-end Ivy Bridge machine now - we'll see how this goes. I used a 965X and 980X for the last two. These are database/analytic machines. Have you seen the specs for Nviidia Kepler K110? I have an 2070 in one machine and I can do analytics like lightning (heterogeneously parallel). I intend to put a mini-supercomputer on the desk of every quant, broker, actuary and marketing analyst in America. Where are you going?

Intel should make more RND. One major drawback is: in the silicon chip as silicon has no direct band gap, we experience more heat dissipation as the density of transistor is increasing day by day. We should reckon that more heat generated, more unsteadiness and shorter the life, and cooling is also a hassle and hulking problem. In another sense we need the lightening speed, the silicon technology has already crossed its limit, and no better speed is approximately impractical.

So Intel should think fabrication of chips in other semiconductor. Although the process technology of the above slightly different. Like Gallium Nitride (GaN) Gallium Arsenide (GaAs) or AIN. In future we hope with the aid of Nano technology, compound Semiconductors will leads all. This is the Endeavor to embrace the Tera hertz speed along with exceptionally robust device with very low power and heat dissipation.

[...] sourceforge [...]

Will not work with Linux at all. The graphic drivers are not yet ready and never will for Linux. The output will always be VGA like with terrible tearing of composites and disgusting video tearign with absolute no sync to vblank.

The performance increases from real-world applications are a lot smaller than that, around 10% as stated above. A 37% increase would be a massive leap ahead and too much to ask.

No Way.

i can accept certain benchmarks 37%+ performance improvements BUT overall the cpu performance improvement will be at most 10%.

the only major jump i saw was in the transition from net* pentium based to corexduo based.

take similar cpus (same cache size, same core numbers etc) and see.

check tomshardware etc.

or http://www.cpubenchmark.net/high_end_cpus.html

from :http://hardware.slashdot.org/story/12/04/23/1620243/intel-officially-lifts-the-veil-on-ivy-bridge

"In CPU performance, the new Ivy Bridge Core i7 is only marginally better than last generation's Core i7-2700K. Essentially, Ivy Bridge is not the fantastic follow-up to Sandy Bridge that many enthusiasts had hoped for, but an incremental improvement".

i

Congrats to Intel, BUT... this isn't proper 3D technology.

In true 3D processors, memory and other components are stacked vertically, with vertical channels allowing interconnects. The aim is to provide faster access to memory, ancillaries and to integrate a lot more onto a single processor. It is actually a lot more than FinFETs, which is basically what this announcement is about. Despite the fact that the first true 3D processors were built in the 1980s, although there are some other claims out there and a few companies have announced plans to build them we aren't there with this sort of large scale in 3D yet (even if Angela Merkel seems to think that we are!)

I would have stuck images in here but this won't let me!!

f

I went from a Q9650 to an i7 and was very impressed. If there is an actual 37% increase with Ivy Bridge I will be jumping for joy, especially if there is an accompanying decrease in required electrical power. I'm getting to the point of having to get rid of computers in order to have enough power to add new ones. I agree with David Rathbone that Microsoft is aging poorly and would like to see Intel (and others) do a better job of supporting Linux. I would like to be able to use the onboard graphics and just use nVidia cards for CUDA processing.

[...] New 22-nm Intel Ivy Bridge Chips 37% Faster Than Previous Generation, Use 50% Less Energy | Go Paral.... Share this:StumbleUponDiggRedditLike this:LikeBe the first to like this [...]

To late for all the millions of smart phones!

All the 86 based code has moved on.

Better if Intel could look at better use of memory than sending it all throgh the bottle neck in the CPU.

Regardless of how many cores.

Still good to see 10nm Chips in the future. With 3D, I bet the development tools for 3d is slower than the hardware.

After all ts all running on Microsofts ageing concepts.

The most important thing for the future is validation that 3D FET technology, at 22nm, is manufacturable at this scale and really does reduce power consumption. The improved graphics engine may not be so important in high end desktop systems, and the speed improvement is nice but may not be as significant as the "37%" claim implies - i'd like to see speed measured doing real work. If benchmark results all show 37% speed increase THEN I'll be impressed.