Final Words

I see three reasons why you'd want the Core i7 3820:

1. You need PCIe 3.0 today and/or you need more PCIe lanes than a Core i7 2600K can provide,

2. You need tons of memory bandwidth for a particular application,

3. You want a 2600K but you need a platform that can support more memory (32GB+).

If you fall into any of those categories, the 3820 gets the job done. It's easily as fast as the fastest LGA-1155 Sandy Bridge without adding significant power consumption or really being limited on the overclocking side either. The 3820 admittedly targets a niche, but it does so without any real trade offs. If you land outside of the 3820's niche however, you're better served by the 2500/2600K at a lower total platform cost or a 3930K/3960X if you're running a heavily threaded workload and can use the extra cores.

What About Ivy?

By the time the 3820 is available for purchase early next year, Ivy Bridge will be just about a quarter away. For desktop users Ivy Bridge is really only going to bring lower power consumption and a better integrated GPU. If you're seriously considering anything in the SNB-E family, the latter isn't going to matter and the former will be of arguable value. I do expect that we'll see a drop-in upgrade path to IVB-E at some point in early 2013 if you're concerned about platform longevity, although Intel hasn't officially committed to such a thing. It's pretty safe to say that you'll be on your own after IVB-E however, Haswell should be a fairly large departure from IVB-E in a lot of senses.

For everyone else, if you need a desktop system today - the LGA-1155 Sandy Bridge is still a viable option. There's always something better around the corner but I have no issues recommending either that you buy now or you wait for IVB. If you can wait, you'll be getting a cooler CPU with better integrated graphics and faster Quick Sync.

Power Consumption
Comments Locked

84 Comments

View All Comments

  • Hrel - Thursday, December 29, 2011 - link

    So, shouldn't there be a price cut coming on 2700k, 2600k and other similar CPU's? If not Intel is being a dick again.
  • SlyNine - Thursday, December 29, 2011 - link

    Well no, There is no quick synk ( I believe) no IGP. it should be a cheaper chip.
  • SlyNine - Thursday, December 29, 2011 - link

    Without a IGP I guess there couldn't be quick synk. So forget the (I believe) part
  • Tetracycloide - Thursday, December 29, 2011 - link

    So they're adding an IGP most users of the k series CPUs don't want, need, or use, charging them for it, and that means no, they're not a dick? I'm not following. I mean if this is really the reason to keep the current pricing on those SKUs then there's really no other option: Intel is being a dick again.
  • Tetracycloide - Thursday, December 29, 2011 - link

    Well, in fairness to Intel they're being a dick by offering this product at a pretty astonishingly low price for what's on offer which just makes their other SKUs look overpriced by comparison.
  • SlyNine - Thursday, December 29, 2011 - link

    How so, many people want Quick synk which REQUIRES the IGP. So explain how Intel is being a dick?

    Besides, Like there is a way for Intel to remove the IGP. Be practical or go home.
  • SaltwaterC - Wednesday, January 4, 2012 - link

    Food for thought ... http://ark.intel.com/compare/52213,52276,52277,522...
  • chizow - Thursday, December 29, 2011 - link

    I haven't followed SB-E that closely but I figured the 4-core derivative would be a badly cut down (reject) version of the 8-core chip, basically the dregs of the wafer basically cut down to half usable.

    Can't believe they went with an entirely new die that still benefits from the greater 2.5MB L3 amount per core. It should actually perform better than the 3930K in cases that use 4 cores or less.....amazing. Intel couldn't have made this chip only to become a $280 desktop part....are they planning to use it for their low-end servers as well?

    This also really provides that budget friendly high-end part for X79, basically the successor to the i7-920 for entry level enthusiasts. X79 still has its glaring omissions (USB 3.0, limited SATA6G, no Thunderbolt etc) but the 3820 makes it look better compared to SB and even IB.

    $285 is an amazing price point.....I fully expect to see it for $199 at Micro Center a month from now. :) I wonder if there will be a K edition.....
  • Kevin G - Thursday, December 29, 2011 - link

    I was figuring that the 4 core version was a cut down 8 core version too. I suspect that there will still be some of the 8 core dies being sold at 4 core just to get rid of inventory (OEM only parts perhaps?).

    I figure the reason for a native 4 core die is so that they can lower power consumption even further. Even with a completely disabled core, there is still some additional power consumption due to the internal ring bus. Speaking of which, I wonder if the reduction in hops in the internal ring bus will have any noticeable impact on performance anywhere (lower L3 cache latencies ect?).

    The other reason for a native 4 core die is likely for high clock speed or ultra low power Xeon parts. There is a little known 4.4 Ghz dual core Xeon for socket 1366. It wouldn't surprise me if all the chips that test for high clock speeds are binned for a similar Xeon and not Core i7 3820's. :(

    Though if we're lucky many of the X79 motherboards will work with the socket 2011 Xeon's.
  • Denithor - Thursday, December 29, 2011 - link

    Low end servers would be my guess as well. All that memory bandwidth, PCIe lanes, perfect for applications where massive quantities of RAM (along with high speed access to it) is more important than raw cpu power.

    Seems like they ran an article on Facebook's server farms where FB uses AMD systems for RAM caching even though the cpu performance was much, much lower in the same power envelop simply because the AMD systems had more RAM sockets. This would seem like a direct replacement for those kind of systems.

Log in

Don't have an account? Sign up now