News
 

Bookmark and Share

(18) 

Intel Corp.’s highly-anticipated new-generation Core i7 Extreme-series microprocessors will not be as revolutionary as they could be, according to newly published specifications of the code-named Ivy Bridge-E chips for desktops. Intel plans to release three new chips for high-end desktops (HEDTs) in the second half of 2013, but they will be barely faster than predecessors, if the unofficial information is correct.

Intel plans to release three “Ivy Bridge-E” central processing units this year: Core i7-4960X, Core i7-4930K and Core i7-4820K, according to a slide that resembles those from Intel’s roadmaps published by Chinese VR-Zone web-site. Two of the new microprocessors will feature six cores and one will sport four cores. The new products will feature a little higher clock-speed compared to existing Core i7 “Sandy Bridge-E” microprocessors and will continue to feature 130W thermal design power.

In case the slide is legitimate and specifications published by the web-site are correct, then it appears that Intel decided not to increase core-count of the new HEDT-class microprocessors, even though it was technologically possible to do so. Moreover, given the silicon inside the new central processing units, it was also hard for Intel to push frequencies up significantly. Keeping in mind the fact that Ivy Bridge micro-architecture is not considerably faster than the Sandy Bridge, the Intel Core i7-4800/4900 processors will not be able to provide substantially higher performance than the existing Intel Core i7-3800/3900 chips. Still, those, who want to get even a slight performance boost can install the new chips into existing LGA2011 mainboards.

According to unofficial information, the new Ivy Bridge-E/EP design, which is set to emerge in the second half of 2013 and to power various microprocessor products, will have physical twelve cores and 30MB level-three cache, up significantly from eight cores and 20MB L3 cache inside Sandy Bridge-E/EP. The Ivy Bridge-E variation of the chip will be the base for high-end enthusiast-class Core Extreme processors as well as workstation-class Xeon chips, whereas the Ivy Bridge-EP will be used for Xeon chips for dual-socket severs.

Even though the slide, which presumably comes from Intel’s roadmap, points to availability of Intel Core i7-4800/4900 “Ivy Bridge-E” processors in Q3 2013, the web-site itself reports that the products will be available in November. Given such significant inconsistency, the information about availability should be taken with a grain of salt.

Intel did not comment on the news-story.

Tags: Intel, Core, Ivy Bridge, Ivy Bridge-E, Ivy Bridge-EP, Xeon, HEDT, LGA2011

Discussion

Comments currently: 18
Discussion started: 03/29/13 02:36:59 PM
Latest comment: 04/20/13 01:06:49 AM
Expand all threads | Collapse all threads

[1-12]

1. 
The only thing extreme is the price...
11 2 [Posted by: beenthere  | Date: 03/29/13 02:36:59 PM]
Reply
- collapse thread

 
Even if I made $50K a month, I'd still buy the 4930K and overclock. To me the idea of paying double for 100mhz clock bump and extra 3mb of L3 cache (that doesn't even matter) is simply a waste of money. The entire strategy behind LGA2011 is borked. When Intel will release a 6-core IVB-E, it'll be slower in games and some office apps that a 4-core Haswell due to lower IPC. The flagship platform should always be fastest in all cases and benchmarks against the mainstream platform. LGA1366's strategy/positioning actually made sense. LGA2011 continues to be behind Intel's mainstream platform by nearly a year. It is completely illogical that Intel's flagship platform uses last gen CPU architectures. It could be somewhat justified if Intel bumped up the cores to 8 with IVB-E but they won't even do that.
5 2 [Posted by: BestJinjo  | Date: 03/29/13 11:54:44 PM]
Reply

2. 
You would think if the Xeons were moving to 12-core then Intel could easily release an 8-core desktop while still justifying the workstation premium.

Intel really needed to have skipped Ivy Bridge and moved straight to Haswell. To justify the massive premium that is the 'Enthusiasts Level' there needs to be cutting edge technology inside that can't be found anywhere else yet. Instead they are trying to sell a generation and half old technology as state of the art.

If Haswell overclocks well then I think Ivy Bridge-E is going to have massive problems.
1 1 [Posted by: loadwick  | Date: 03/29/13 04:22:32 PM]
Reply

3. 
the only big improvements to ivy bridge arch over the sandy bridge arch was the gpu. other then that there is little diff performance wise on the cpu side.
3 1 [Posted by: SteelCity1981  | Date: 03/29/13 05:29:21 PM]
Reply
- collapse thread

 
Which echoes my point as there is other GPU in these so why not jump straight to Haswell which still by all accounts isn't much of a CPU jump over Sandy Bridge either but at least the premium end of the market will catch up to the midrange, which to me just sounds so ludicrous when you think about it!
1 1 [Posted by: loadwick  | Date: 03/29/13 11:13:43 PM]
Reply

4. 
Intel's marketing goes into overdrive.
6 1 [Posted by: linuxlowdown  | Date: 03/29/13 06:14:41 PM]
Reply

5. 
show the post
1 4 [Posted by: tbaracu  | Date: 03/30/13 12:12:06 AM]
Reply
- collapse thread

 
I don't know if something was edited, but the 12 and 8-cores are from "unofficial information" (so not from the slide and table) and they talk about 30 and 20MB caches (which are not in the table). They also mention Xeons at the end of the paragraph.

And if you look at the Xeon lineup (ark.intel.com), you'll see 10-cores with 20 VCPUs and 30MB cache.
4 0 [Posted by: mathew7  | Date: 03/30/13 01:29:58 AM]
Reply

6. 
It's all irrelevant really. These are just niche products.
4 1 [Posted by: Skidmarks  | Date: 03/30/13 12:34:47 AM]
Reply

7. 
show the post
0 3 [Posted by: caring1  | Date: 03/30/13 01:32:18 AM]
Reply
- collapse thread

 
It means the upgraded CPUs will barely improve performance by a significant margin over existing SB-E offerings (i.e., 5-10% max).

3 1 [Posted by: BestJinjo  | Date: 03/30/13 02:45:01 AM]
Reply

8. 
12 physical cores and 30MB l3 cache seems to be something of a MAJOR upgrade hopefully it would start at 4GHz and boosting to 5Ghz or more and at least target at DESKTOP pricing so that PC FANATIC as myself can afford to buy one. if the spec is going to be server and targeted corporation in pricing than I´m going to keep using this FX8350 with 8 cores for a long time. PC that is still running fine does not need to be tossed into the trashcan. my son collects lots of those pentium and 486 to run windows 3.1 and DOS. those GWBASIC is essential in learning those most basic computer programming at least in my time in high school. those apple 2e were the major things at the high school because Apple seemed to be giving out those free to school for a time in the 80? because our computer class in high school is filled with those 2e with GREEN color text screen and it got BASIC interpreter programing. we were having FUN just to write a little 3 line code to get 2 number and the third line would do something ab out those two input. i like the one that will repeatedly write a one line message to flood the monitor. our computer teacher would hate me when i write something funny about some people in the class. BUT my last line of code changes the message line so that the final screen hides those funny message earlier about some people. it makes the teacher went MAJOR MAD that she loses the offensive message at the end. just have to delete that 3 lines of codes quick enough so i stay away from trouble.
1 2 [Posted by: idonotknow  | Date: 03/30/13 02:48:08 AM]
Reply

9. 
Why to rush and improve performance? AMD is far behind anyway.

See now funboys how the lack of competition hurts us all?
5 1 [Posted by: nitro912gr  | Date: 03/30/13 05:06:28 AM]
Reply
- collapse thread

 
Exactly. That's why the idea that Intel "must provide incentive for people to upgrade" is ludicrous. Intel themselves have stated that the average PC user now upgrades once in 5 years. Intel is no longer interested in releasing new CPU architectures that are an incentive to upgrade from the previous Tock (i.e, SB to Haswell). They have realized that people upgrade much less frequently which means the best strategy to maximize profits is keep die sizes low (that's why Haswell is still a quad core), while minimizing the amount of $ spent on improvements (which is why IPC barely goes up 10% from SB). In reality, Intel is targeting Core 2 Quad/Nehalem users to upgrade to Haswell, not SB or IVB users. Why should Intel give us a 6-core Haswell at $320, or increase IPC by 17-20% like it was the case with Core 2 Duo to Nehalem/Lynnfield? They wouldn't.

The fact that a $320+ i7 4770K is still a quad-core after Q6600 was $300 around summer of 2007 is proof that progress on the CPU side has come to a near halt. Not only that but most tasks outside of professional apps are now SSD or GPU limited.

Take Far Cry 3 as an example. A system with a 2008 Nehalem overclocked to 3.75ghz provides almost identical performance to a 2012 3770K @ 4.5ghz when paired with a GTX680:
http://www.guru3d.com/art...e_review_benchmark,7.html

All the fun for PC gamers now rests in SSDs and GPUs, and possibly 4K 30 inch monitors at some point. For most PC gamers with a Core i7 2600K @ 4.5ghz+ there is no incentive to upgrade until Skylake.
5 1 [Posted by: BestJinjo  | Date: 03/31/13 12:31:07 AM]
Reply

10. 

This is what happens when the market rejects an CPU type.
With a very small specialized market and no compelling reason to improve performance also with a stock problem why bother.
If there was an O/S which required a performance jump and/or Software which require a better CPU then there would be a reason to improve., but as of now there is nothing to spur on development combine this with complete lack of demand for high powered consumer CPU's is it any wonder no effort is put into these.
3 0 [Posted by: tedstoy  | Date: 03/30/13 06:19:59 PM]
Reply

11. 
What a poor title for this article. They will barely improve significantly?

So they will still improve significantly, but just barely. The improvement has just fallen into the category of 'significant'. So they are a significant improvement.
1 0 [Posted by: texmurpy01  | Date: 04/01/13 09:25:33 AM]
Reply
- collapse thread

 
+1 yes, the wording is a bit whack.

Two opposing adjectives straddling a phrase, and both of them seemingly wanting to affect it.

"Barely significant" would have been a lot better.
0 0 [Posted by: jihadjoe  | Date: 04/20/13 01:06:49 AM]
Reply

12. 
I'm playing Battlefield 3 with E6700 from 2006 and GTX580!

http://www.youtube.com/watch?v=R986xthlqvE
0 0 [Posted by: TeemuMilto  | Date: 04/07/13 07:24:13 AM]
Reply

[1-12]

Add your Comment




Related news

Latest News

Thursday, August 28, 2014

12:22 pm | AMD Has No Plans to Reconsider Recommended Prices of Radeon R9 Graphics Cards. AMD Will Not Lower Recommended Prices of Radeon R9 Graphics Solutions

Wednesday, August 27, 2014

9:09 pm | Samsung Begins to Produce 2.13GHz 64GB DDR4 Memory Modules. Samsung Uses TSV DRAMs for 64GB DDR4 RDIMMs

Tuesday, August 26, 2014

6:41 pm | AMD Quietly Reveals Third Iteration of GCN Architecture with Tonga GPU. AMD Unleashes Radeon R9 285 Graphics Cards, Tonga GPU, GCN 1.2 Architecture

Monday, August 25, 2014

6:05 pm | Chinese Inspur to Sell Mission-Critical Servers with AMD Software, Power 8 Processors. IBM to Enter Chinese Big Data Market with the Help from Inspur

Sunday, August 24, 2014

6:12 pm | Former X-Bit Labs Editor Aims to Wed Tabletop Games with Mobile Platforms. Game Master Wants to Become a New World of Warcraft