Talk:Ivy Bridge (microarchitecture)

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

List of Ivy Bridge Processors[edit]

Comment: I moved this section from the article for now, as it contained some false information.

  • The 2011 release dates are obviously wrong
  • Some of the processor names are wrong (according to the linked article, there is a "Core i5-3470T", not a "3470")
  • The TDP is wrong, according to the linked article
  • I don't really trust the other info, given the above errors
  • In the features the info says "Multiple 4K video playback", I'm not really sure of this, although the pre-marketing said these things, you cannnot find any reference in www.intel.es about that feature, so we don't really know if that is true.

Thue | talk 23:29, 29 November 2011 (UTC)[reply]

2011 release date probably just an error by the author, it should be 2012. There is a Core i5-3470T(dual core 35W) and also a i5-3470(quad core 77W) so he is not wrong. TDP of 77W is correct, just that the list is not complete. While the info seems to be correct, there is a possible that specs will change before launch.(user:xxxxxls2) 171.207.50.210 (talk) 02:20, 30 November 2011 (UTC)[reply]
What is your source for the i5-3470 info? Thue | talk 10:48, 30 November 2011 (UTC)[reply]

1Processors featuring Intel's HD 4000 graphics are set in bold. Other processors feature HD 2500 graphics or no graphics core (Graphics Clock rate indicated by N/A).

Desktop Processors[edit]

[CPU World: Rumoured Ivy Bridge processors]

Target
segment
Cores
(Threads)
Processor
Branding & Model
CPU Clock rate Graphics Clock rate L3
Cache
TDP Release
Date (Y-M-D)
Price
(USD)
Motherboard
Normal Turbo Normal Turbo Socket Interface Memory
Performance 4 (8) Core i7 3770K 3.5 GHz 3.9 GHz 650 MHz 1150 MHz 8 MB 77 W Q2 2011 $ LGA
1155
DMI 2.0
PCIe 3.0^
Up to dual
channel
DDR3-1600
3770 3.4 GHz $
Mainstream 4 (4) Core i5 3570K 3.4 GHz 3.8 GHz 6 MB $
3570 3.4 GHz Q3 2011 $
3550 3.3 GHz 3.7 GHz Q2 2011 $
3470 3.2 GHz 3.6 GHz 1100 MHz Q3 2011 $
3450 3.2 GHz 3.5 GHz Q2 2011 $
3330 3.0 GHz 3.2 GHz 1050 MHz Q3 2011 $

^ Requires a 75/77-series Motherboard

Requested move 2011[edit]

The following discussion is an archived discussion of a requested move. Please do not modify it. Subsequent comments should be made in a new section on the talk page. No further edits should be made to this section.

The result of the move request was: page moved. Vegaswikian (talk) 00:23, 18 December 2011 (UTC)[reply]



Ivy Bridge (computer processor)Ivy Bridge (microarchitecture) – This is the proper name for computer processor designs.Jasper Deng (talk) 05:49, 11 December 2011 (UTC)[reply]

The above discussion is preserved as an archive of a requested move. Please do not modify it. Subsequent comments should be made in a new section on this talk page. No further edits should be made to this section.

Requested move 2012[edit]

The following discussion is an archived discussion of a requested move. Please do not modify it. Subsequent comments should be made in a new section on the talk page. No further edits should be made to this section.

The result of the move request was: page not moved. Arbitrarily0 (talk) 16:20, 11 February 2012 (UTC)[reply]


Ivy Bridge (microarchitecture)Ivy BridgePrimary topic. 126,000 views this month plus 2500 from its old title. Ivy Bridge college is currently still in its university's article which has 3000 views this month. Marcus Qwertyus 22:24, 4 February 2012 (UTC)[reply]

  • Oppose Popularity will decline when the next microarchitecture comes along. —Ruud 22:44, 5 February 2012 (UTC)[reply]
  • Oppose per Ruud. Who views the Intel Core article when searching for "Core"?Jasper Deng (talk) 22:45, 5 February 2012 (UTC)[reply]
  • Oppose – no need to claim primary on such an ambiguous term. Dicklyon (talk) 22:53, 5 February 2012 (UTC)[reply]
The above discussion is preserved as an archive of a requested move. Please do not modify it. Subsequent comments should be made in a new section on this talk page. No further edits should be made to this section.

Move to Sandy Bridge (same architecture)[edit]

I suggest this page is removed into a section in the Sandy Bridge article, as Ivy Bridge is just a die shrink of Sandy Bridge, not a new architecture. This is how it is done on other Wikipedia articles about Intel architectures as well. 83.108.199.200 (talk) 03:10, 12 February 2012 (UTC)[reply]

No. It was originally part of that article, and it worked horribly, with each section having two sub-sections (one for Sandy Bridge, one for Ivy Bridge). This is much cleaner. Besides, there are differences enough between the two (see the article) to justify a separate article. Thue | talk 00:12, 15 February 2012 (UTC)[reply]
I'm okay with either way. But i see nothing wrong with having two sub-sections. Also, because both are separated, quite a number of stuff are duplicated just to let ivy bridge have a new page. 175.156.205.23 (talk) 02:40, 15 February 2012 (UTC)[reply]
You say there is lots of duplicated content - but I can't see any duplicated content? Thue | talk 21:30, 15 February 2012 (UTC)[reply]
CPU specification is a duplication, and also sandy bridge has a successor section just because both articles are separated. Other than that, a lot of the titles are repeats. Not as much as i previously thought, but its still a lot. But i'm not against separating anyway, because pages load faster if they are smaller. — Preceding unsigned comment added by 175.156.193.131 (talk) 12:09, 16 February 2012 (UTC)[reply]
My main argument is that a user will usually want to read about either Ivy Bridge OR Sandy Bridge, depending on which specific CPU he wants information on. But not both. In that case it is much cleaner to have the Ivy bridge content separate. Thue | talk 18:40, 16 February 2012 (UTC)[reply]
I dont see any problems making the Sandy Bridge article a lot like the Nehalem (microarchitecture) article, where the die shrink of Nehalem, Westmere, is a section. 83.108.199.200 (talk) 18:53, 15 February 2012 (UTC)[reply]
Westmere was a simple tweak to Nehalem. Saying the same about Ivy Bridge to Sandy Bridge is like saying a Camaro is a Ram 1500 with a few tweaks. Complete and utter nonsense. Ivy Bridge has too many additional features and capabilities to make it compatible with the SB article. It is so radically different. SoranoGuardias (talk) 17:50, 24 April 2012 (UTC)[reply]
Picking up this conversation again: Both the Sandy Bridge article and the Ivy Bridge article currently mix information about the specific CPU and about the microarchitecture. For all other microarchitectures, we have one article that describes the overall features, and then individual articles about the CPU implementing it, see Nehalem (microarchitecture), with individual articles for Clarksfield, Lynnfield, Jasper Forest, Bloomfield, Gainestown and Beckton. This is more confusing here, because "Sandy Bridge" refers to three separate things: the common microarchitecture (including sandy bridge, sandy bridge-ep, sandy bridge-en, ivy bridge, ivy bridge-en and ivy bridge-ex), the subset of chips that are 32 nm, and the original desktop/laptop chip. The Ivy Bridge name (and article) refers to both the shrinked microarchitecture and all of the chips that have been released to date (but not the upcoming server chips). I think we should move the CPU core specific stuff out of both these articles, to match what we do elsewhere (even though the names are a bit confusing). The microarchitectural differences between Ivy Bridge and Sandy Bridge are relatively small, and I don't think it makes any sense to have separate articles. Anyone interested in the microarchitecture (as opposed to the specific CPU) would be better off having a single article. Arndbergmann (talk) 12:53, 6 September 2012 (UTC)[reply]
i agree there is a bit of a mess of cpu listings, i dont agree with merging tick tock methods together though since each has something different to offer to the table and is worth having its own article, what are your suggestions with the cpu listings? we could just split the list so it only displays what each microarchitecture has, people confuse Sandy Bridge (microarchitecture) as being the main article for the entire Architecture, so i see what you are saying and am willing to help. Matthew Smith (talk) 13:17, 6 September 2012 (UTC)[reply]
I disagree - Ivy Bridge is a tick+, not just a tick on the sandy bridge architecture. The current situation of having the list of Ivy Bridge processors on the sandy bridge page is a horrible mess, making this article lacking in important information, and teh sandy bridge one overly long and complicated. Thue (talk) 18:10, 6 September 2012 (UTC)[reply]
you mis understood, im against the merge too, and agree with the mess it will make, what im talking about is the ivy bridge cpus that are also listed on sandy bridge. Matthew Smith (talk) 18:26, 6 September 2012 (UTC)[reply]

I am moving the list of Ivy Bridge processors back to the Ivy Bridge page where it belongs. The list of Nehalem CPUs is on the Nehalem(microarchitecture) page, the list of Westmere CPUs is on the Westmere(microarchitecture) page, and the list of Sandy Bridge CPUs is on the Sandy Bridge(microarchitecture) page. For some reason Arndbergmann decided to break this style and move the Ivy Bridge CPU list to the Sandy Bridge(microarchitecture) page.67.190.6.143 (talk) 13:09, 8 September 2012 (UTC)[reply]

Out of Date??[edit]

Could User Jasper Deng specify which info is out of date? As of what i know, other than details that are already stated here, only info regarding motherboards are revealed/leaked. 175.156.216.71 (talk) 01:02, 26 February 2012 (UTC)[reply]

Currently the full details are under embargo; Intel has revealed a few details but not the whole deal yet. Suffice it to say it's worth waiting one more month to do a proper cleanup of this page. — Preceding unsigned comment added by 87.112.21.178 (talk) 18:34, 19 March 2012 (UTC)[reply]

Intel_Ivy_Bridge_(microarchitecture) Redirects to Sandy Bridge[edit]

When you do a search for "Ivy Bidge", this article is listed followed by a redirect to Sandy Bridge (redirect from Intel Ivy Bridge (microarchitecture)). Think it should be better redirected to this article. — Preceding unsigned comment added by 217.5.181.210 (talk) 19:16, 29 March 2012 (UTC)[reply]

Intel moved Ivy Bridge to April 23[edit]

This should be included in the article; cheers. — Preceding unsigned comment added by 62.31.233.11 (talk) 20:48, 13 April 2012 (UTC)[reply]

I see no indication that ivy bridge shipped on april 23rd. I can't find anyone selling it anywhere. — Preceding unsigned comment added by 24.47.224.175 (talk) 03:53, 24 April 2012 (UTC)[reply]
Agreed. There is no indication that Ivy Bridge has been released anywhere. Considering the mocking nature of the source article, I'd consider this information ambiguous at best. SoranoGuardias (talk) 17:45, 24 April 2012 (UTC)[reply]
Confirmed. Intel's website clearly states the processors are going to be launched 4/29. Adjusting article. SoranoGuardias (talk) 18:01, 24 April 2012 (UTC)[reply]
4/23, at 12:00 PM EST, was the paper release and NDA lift for Intel's Ivy Bridge desktop processors. Retailers are not permitted to sell the processors until the 29th (30th for some stores due to 29th being a Sunday and time zones). Legends0 (talk) 23:49, 25 April 2012 (UTC)[reply]

Reliable?[edit]

No sources other than this appear to discuss Ivy Bridge-EX to that extent, and much appears to be speculation. Is this worth adding to the article?--Jasper Deng (talk) 20:01, 29 April 2012 (UTC)[reply]

Speculation on Ivy Bridge's heat woes.[edit]

This section suggests that Ivy Bridge's heat problems are due, at least in part, to the use of thermal paste instead of fluxless solder. The "three" references are, in fact, one reference, because the other two just reference the same article at overclockers.com. However, after a little searching, I found this article which claims the change to thermal paste was not at fault. I'd like to see some more information on this, but it seems they did their homework, unlike overclockers, which is mere speculation. Sure, mathematical formulas might be great idea to demonstrate the problem, but not if they don't accurately reflect the real world. Thoughts? --Manguene (talk) 06:06, 7 May 2012 (UTC)[reply]

The problem with the article you mention is that the testing was done by removing the IHS and applying TIM directly to the die, then mounting the CPU cooler to the bare die. A more accurate way of measuring the impact of the Intel thermal paste would be to remove it, apply an aftermarket TIM that is known to perform well and then replace the IHS. I believe this would provide a more "apples to apples" comparison by taking the heat spreader out of the equation. In this article the author replaced the stock paste with two different TIMs and saw significant improvement in load temps, as well as maximum clock speeds. I agree with you that we need more information (and it doesn't help when article authors frame their speculation as facts), I just wanted to point out that there is some evidence out there that the thermal paste is at least partially to blame for Ivy Bridge's heat issues. Genrldisaray (talk) 03:51, 28 May 2012 (UTC)[reply]
Looks like the whole section is a bit dodgy. Someone with an IP address added a claim that similar problems were seen on the Prescott die shrink. As Prescott was a new architecture (IIRC it added DEP, certainly its own page thinks it is different) I have my doubts over the rest of it, so I have removed the entire sentence: "Similar phenomena occurred when Intel released the 90 nanometer Prescott core Pentium 4 processors - a die shrink of the 130 nanometer Northwood processors." (as edited by 72.161.136.211 at 06:57, 8 May 2012.). Most of the rest of that edit has already gone. Number774 (talk) 13:11, 27 June 2012 (UTC)[reply]

Tray or box price[edit]

Intel recently changed their pricing to display to two parts: tray and box. This affects only desktop processors because mobile processors are all tray. Currently all prices follow tray but i would like to recommend box instead. After all a lot of people coming here to look at the price are those buying boxed processor(s). 175.156.219.155 (talk) 10:01, 14 June 2012 (UTC)[reply]

I agree that box prices would be useful to most people. But not all desktop processors have a box price, and you also want make desktop processor prices comparable to server and mobile prices. So to avoid a mess of mixed prices, I suggest sticking with the tray price. Thue | talk 11:45, 14 June 2012 (UTC)[reply]

On a related topic, it is very rare to find tray chips for sale. I would find a column differentiating tray (OEM) chips from retail boxed chips (which are readily available) useful. The current table is useful for comparisons between all possible chips, but confusing for amateur builders looking to compare only those chips which are intended for retail sale. — Preceding unsigned comment added by 41.242.99.76 (talk) 06:12, 31 December 2012 (UTC)[reply]

I would say remove prices altogether. See guidelines WP:NOPRICES. For example, we avoid using the trademark and registered symbols when using trademarks such as Xeon Celeron, etc. on the basis that Wikipedia is not supposed to be a commercial site. Best would be to just link to the Intel Ark or other pages that have more up to date info anyway to avoid it being dated. W Nowicki (talk) 20:31, 28 August 2013 (UTC)[reply]

Requesting new pic[edit]

I'm requesting a new logo pic, although the Core i7 pic is correct for this generation of chip, it does not represent the entire lineup so if anyone can find a good one and post it up it would be greatly appreciated.Matthew Smith (talk) 03:09, 20 August 2012 (UTC)[reply]

Complete listing table[edit]

Just noticed it was moved over to the Sandy Bridge page. Not sure what the point is, given that Ivy has its own page in the first place.--Azul120 (talk) 17:01, 6 September 2012 (UTC)[reply]

Agree - revert. This article has been lobotomized for no good reason. Sandy Bridge and Ivy Bridge is different enough to warrant separate articles. We already discussed that above. Thue (talk) 18:03, 6 September 2012 (UTC)[reply]
I don't mind the "brand name table" (Core iX, Pentium, etc.), and would even kind of welcome it for the pages from Nehalem to Sandy Bridge.--Azul120 (talk) 21:48, 6 September 2012 (UTC)[reply]

Yeah, there seems to have been an enormous amount of successive micro-edits from Arndbergmann and Matthew Anthony Smith in the past couple days (although from what I've seen Matthew Anthony Smith's edits mostly seems to be reverting vandalism and small bits of cleanup). It seriously worries me that Arndbergmann has cut the article from 28,814 bytes to 15,553 bytes over the course of 3 minutes without discussing any of the changes in this article's talk page beforehand (he did post on the talk page 17 minutes after he cut the article in half), especially when the changes he made have already been discussed and agreed against. I don't want to make the edit as I haven't contributed much to this particular talk page, but I strongly support the position that Arndbergmann's edit be reverted. Charwinger21 (talk) 11:12, 7 September 2012 (UTC)[reply]

Agree - revert. The Ivy Bridge page is now worse than useless. Furthermore, adding the list of Ivy Bridge processors in the middle of the Sandy Bridge page just confuses the issue.67.190.6.143 (talk) 12:37, 8 September 2012 (UTC)[reply]

Ivy Bridge Celeron CPUs[edit]

I see entry-level i3 and Pentium CPUs based on Ivy Bridge, but no Celerons. Does Intel have plans for Celeron CPUs based on the Ivy Bridge architecture? And when will they become available for purchase by end users? — Preceding unsigned comment added by 222.165.42.62 (talk) 15:47, 7 October 2012 (UTC)[reply]

I have been watching the news for them, but haven't seen anything. I assume they will come out eventually. Thue (talk) 16:34, 7 October 2012 (UTC)[reply]

Mobile i3 link goes to regular i3s[edit]

I'm not entirely wiki code literate, so I thought I would bring this up.Azul120 (talk) 00:41, 12 January 2013 (UTC)[reply]

3350P Max Turbo is 3.3, not 3.5[edit]

But I won't fix that, this table is a *&$§% --84.183.240.90 (talk) 12:02, 16 February 2013 (UTC)[reply]

Where did the IB logo come from?[edit]

Always wondered. --Azul120 (talk) 02:26, 16 January 2014 (UTC)[reply]

Yeah, that interested me, too. According to the logo's Wikimedia Commons entry and this article, that's some kind of an internal logo created by the Intel's Ivy Bridge development team, and it was published here. Are there any similar logos for later generations of Intel CPUs? — Dsimic (talk) 02:49, 16 January 2014 (UTC)[reply]

QPI category for Xeon?[edit]

I think we should add QPI speeds for Xeon processors for both SB and IB. I have two ideas: changing the existing bus category, or just adding an extra column for the bus speeds. --Azul120 (talk) 19:59, 23 January 2014 (UTC)[reply]

Sounds good to me! Maybe the following table layout would be good enough, illustrated by its heading:
Target
Segment
Cores
(Threads)
Processor
Branding & Model
CPU Clock rate Graphics Clock rate L3
Cache
TDP Release
Date
Price
(USD)
Socket Interconnect Interfaces Memory
Normal Turbo Normal Turbo Type Speed
... or something more simple? Thoughts? — Dsimic (talk) 20:53, 23 January 2014 (UTC)[reply]
I'm only a little lost on the Type part. Other than that, sure. --Azul120 (talk) 01:36, 24 January 2014 (UTC)[reply]
Had in mind something like "n× QPI 1.x" for that column. — Dsimic (talk) 01:58, 24 January 2014 (UTC)[reply]
Ahhh. Might make Interfaces redundant. --Azul120 (talk) 03:10, 24 January 2014 (UTC)[reply]
But, the Interfaces column should contain non–CPU interconnect interfaces, like DMI and PCI Express. With the new Speed column, bundling it all together would be quite confusing, if you agree. — Dsimic (talk) 03:22, 24 January 2014 (UTC)[reply]
True. The Type/Speed columns would only apply to processors with QPI. But I suppose that could be covered with n/a. — Preceding unsigned comment added by Azul120 (talkcontribs) 20:40, 28 January 2014 (UTC)[reply]
Exactly, "N/A" would go into these columns for CPUs intended to be used in single-socket systems. — Dsimic (talk | contribs) 00:17, 29 January 2014 (UTC)[reply]

How should we sort out the rest of the Xeon processors?[edit]

The E7 line just made its appearance, and I'm wondering how we should sort the processors at this point: by going with separate groupings for 2P, 4P and 8P, or to do a combined DP/MP group, like in the Westmere page. The latter would carry the advantage of having all of each (E5/E7) together. --Azul120 (talk) 02:24, 20 February 2014 (UTC)[reply]

My vote goes to putting them together. — Dsimic (talk | contribs) 02:30, 20 February 2014 (UTC)[reply]

FCLGA 2011, LGA 2011-1 etc.[edit]

I'd say we need to describe the differences between FCLGA 2011, LGA 2011-1, LGA 2011-3 and similar variations of the LGA 2011 socket, preferrably as part of the LGA 2011 article. Thoughts? — Dsimic (talk | contribs) 04:43, 27 February 2014 (UTC)[reply]

I think FCLGA2011 and LGA2011 are basically different names of the same thing. Perhaps look up the 2011-1 document for more info on that socket and start there. 2011-3 I believe applies to Haswell-E(P/X) and Broadwell-E(P/X). — Preceding unsigned comment added by Azul120 (talkcontribs) 20:54, 27 February 2014 (UTC)[reply]
Some time ago I've tried to find what are the exact differences, but available sources seem not to know that either. Any chances for you to have a look into that, please? I'd say that would be a really good addition to the LGA 2011 article. — Dsimic (talk | contribs) 20:59, 27 February 2014 (UTC)[reply]
I know that the major difference is chipset support. LGA2011 runs on the X79 chipset with support for DDR3 SDRAM, while LGA2011-3 runs on the X99 chipset and supports DDR4 SDRAM. For these two reasons, the socket and corresponding CPUs are also keyed differently and are physically incompatible with one another. I am, however, unable to provide a reliable source for this info. Jchap1590 (talk) 15:45, 11 June 2014 (UTC)[reply]
EDIT: maybe this can help - Processors Intel Haswell-E and LGA2011 platform-3 Overview and Details Jchap1590 (talk) 15:52, 11 June 2014 (UTC)[reply]
Hello there! That's an awesome reference, thank you very much! I'll add that info into the LGA 2011 article, it will be very useful. — Dsimic (talk | contribs) 05:13, 14 June 2014 (UTC)[reply]
Done, please have a look at new section LGA 2011 § Generations. — Dsimic (talk | contribs) 06:28, 14 June 2014 (UTC)[reply]

Transistor counts for other 3 dies?[edit]

Have those ever been documented? --Azul120 (talk) 00:10, 23 May 2014 (UTC)[reply]

I was looking around everywhere for this information and it doesn't seem to be publicly available. The only transistor counts listed seem to be for i5s (Ivy Bridge-HE-4, 1.4B) Jchap1590 (talk) 17:11, 17 June 2014 (UTC)[reply]

Sever Processors list incomplete[edit]

I notice the Xeon E3 v3 Family processors are not included in this page. Is there an intention to add this information at any point? I believe these SKUs were released almost a year ago. Jchap1590 (talk) 15:34, 11 June 2014 (UTC)[reply]

External links modified[edit]

Hello fellow Wikipedians,

I have just added archive links to one external link on Ivy Bridge (microarchitecture). Please take a moment to review my edit. If necessary, add {{cbignore}} after the link to keep me from modifying it. Alternatively, you can add {{nobots|deny=InternetArchiveBot}} to keep me off the page altogether. I made the following changes:

When you have finished reviewing my changes, please set the checked parameter below to true to let others know.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 18 January 2022).

  • If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
  • If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—cyberbot IITalk to my owner:Online 05:25, 26 February 2016 (UTC)[reply]

Graphics model column is wrong[edit]

Following this article: https://en.wikipedia.org/wiki/List_of_Intel_graphics_processing_units

The graphics model column is wrong. Pentium and Celerons always have "HD Graphics" and not "HD Graphics 2000" or 3000.

We can check this on Intel website, for example for the Pentium G860: https://ark.intel.com/es/products/53492/Intel-Pentium-Processor-G860-3M-Cache-3_00-GHz

There you can see, Graphics model: Intel® HD Graphics.

On the other hand, on the i3 2105: https://ark.intel.com/es/products/55448/Intel-Core-i3-2105-Processor-3M-Cache-3_10-GHz?q=2105

There you can see, Graphics model: Intel® HD Graphics 3000

So the table column should be fixed, it's misleading showing pentiums and celerons having way better graphics that they actually have....

So the entire table is wrong, and its wrong for every single architecture going back to Nehalem (not sure if more). — Preceding unsigned comment added by 81.184.102.1 (talk) 22:52, 4 January 2017 (UTC)[reply]

Move discussion in progress[edit]

There is a move discussion in progress on Talk:Haswell (CPU) which affects this page. Please participate on that page and not in this talk page section. Thank you. —RMCD bot 05:31, 11 February 2017 (UTC)[reply]