Talk:Coffee Lake

Move discussion in progress
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 05:32, 11 February 2017 (UTC)

"and can support DDR4-2400 MHz memory (native) in dual channel mode." - so can Kaby Lake..
That section just says the same than Kaby Lake, it can support dual channel DDR-2400 like Kaby Lake can do, it should be removed since it is not a difference. — Preceding unsigned comment added by 81.184.108.227 (talk) 22:05, 12 March 2017 (UTC)

Coffee Lake will use what socket?
Anyone care to change the Socket Type from LGA 1151 to LGA 1151 V2 — Preceding unsigned comment added by BusriderSF2015 (talk • contribs) 20:16, 8 June 2017 (UTC)

It's a rumor
Intel didn't confirmed nothing, you guys could be spreading fake news. — Preceding unsigned comment added by 177.124.246.132 (talk) 15:43, 4 August 2017 (UTC)

WCCFTech is not a valid source for anything
They post pure speculation and unsubstantiated rumours and nothing else. Not up to Wikipedia standards for sources. — Preceding unsigned comment added by 212.2.165.234 (talk) 09:03, 18 August 2017 (UTC)

How many PCIe lanes?
How many PCIe lanes does the architecture support? 49.145.135.61 (talk) 15:02, 20 August 2017 (UTC)

Intel's marketing department has gone mad
For some reasons the 8th generation low power mobile CPUs are called Kaby Lake Refresh so I've no idea where to list them. I've listed them here anyways. Artem-S-Tashkinov (talk) 09:34, 21 August 2017 (UTC)


 * I suggest that a better location would be in a new Kaby Lake Refresh section of the Kaby Lake article because they are clearly not Coffee Lake, regardless of whether they are 8th generation. 49.145.135.61 (talk) 14:26, 21 August 2017 (UTC)
 * I am moving them over to the Kaby Lake article as they are the Kaby Lake microarchitecture rather than Coffee Lake. Dbsseven (talk) 15:01, 21 August 2017 (UTC)
 * Basically, Intel sells Kaby Lake CPUs as Coffee Lake (Core i3 7000 use Kaby Lake quad-core dies), they clearly share the same Skylake architecture, contrary to Skylake-X which uses a considerably different memory subsystem despite using the name...2A01:CB11:13:D700:B507:27C2:B9AB:77F4 (talk) 19:39, 19 October 2017 (UTC)
 * I still believe that these mobile CPUs should be listed in both articles, because technically they are the 8th generation CPUs. Artem-S-Tashkinov (talk) 19:56, 21 August 2017 (UTC)
 * I would strongly disagree, as this article is about the Coffee Lake microarchitecture and multiple sources have definitively stated the product released today are not Coffee Lake microarchitecture but Kaby Lake microarchitecture.  The generational aspect might be more relevant on the List_of_Intel_Core_i7_microprocessors page. Dbsseven (talk) 20:25, 21 August 2017 (UTC)
 * It may be a good idea to put some information on the Coffee Lake page saying that the processors announced today are NOT Coffee Lake processors, but Kaby Lake, so that people seeing the new 8th gen processors don't have to waste time putting info about the new refresh processors into the Coffee Lake page. JackMacWindows (talk) 21:49, 21 August 2017 (UTC)
 * Good idea. Revised your addition, with cites. Thanks! Dbsseven (talk) 22:24, 21 August 2017 (UTC)

CPU turbo clock rate table - sources?
Where are the sources for those 2/4/6 CPU turbo clock rate for Turbo Boost 2.0 table? that data should be sourced. — Preceding unsigned comment added by 81.184.198.64 (talk) 00:24, 26. Sep. 2017‎ (UTC)
 * Turbo clock rates for 2/4/6 cores are unsourced and not (!) specified by Intel and since I don't want to fight with anonymous editors who revert my edits I'll leave this link here instead. Let other WP editors decide for themselves. Artem-S-Tashkinov (talk) 21:08, 5 October 2017 (UTC)
 * It is unclear whether the turbo clocks in that Anandtech article is reliable. It conflicts with Intel's specification in at least one case: Intel states that the max turbo i7-8700T is 4.0 GHz, but the table in the Anandtech article says it's 4.1 GHz. Before we fill the Wikipedia table with numbers from there, I think it is the best to find an independent source to confirm — Preceding unsigned comment added by 38.109.87.242 (talk) 15:01, 19 June 2018 (UTC)


 * There are turbo clock rates in this German article, which confirms Intel's specs. WikiChip also confirms Intel's specs, but WikiChip doesn't count as reliable source.  &mdash; Pizzahut2 (talk) 15:47, 19 June 2018 (UTC)

Proposition to semi-lock the coffee lake article. Thoughts?
Too many edits by unregistered users; a lot of edits adding Kaby lake Refresh CPUs. — Preceding unsigned comment added by Eddmanx (talk • contribs) 19:28, 6. Okt. 2017 (UTC)


 * Well it's not vandalism. Maybe this will do:
 * * https://en.wikipedia.org/w/index.php?title=Coffee_Lake&type=revision&diff=804119982&oldid=804104055
 * --Pizzahut2 (talk) 21:46, 6 October 2017 (UTC)

Stop readding the chipset section
The Z370 chipset matches the Z270 chipset almost exactly sans DDR3 support, which means no new information (besides there's a whole article for LGA 1151). The second gen chipset is nothing but rumors and as such falls under the WP:CBALL policy. Artem-S-Tashkinov (talk) 15:31, 9 October 2017 (UTC)

Is there a reason why i9 is not listed in the table?
The i9s are part of this architecture, so shouldn't they be listed here? Cecoppola (talk) 00:04, 19 October 2017 (UTC)


 * You mean these?
 * Skylake (microarchitecture)
 * List of Intel Core i9 microprocessors
 * They are Skylake-X, so not Coffee Lake.
 * Pizzahut2 (talk) 01:17, 19 October 2017 (UTC)

Coffee Lake marks a shift in the number of cores for mainstream desktop processors,
Is this articel written by Intel itself? Because AMD has published 6 core CPUs month before Intel. This means that AMD CPUs mark the shift, not this CPUs. — Preceding unsigned comment added by 2003:73:2e54:1200:3cd0:b3e9:5b9a:9c3b (talk) 22:26, 30. Okt. 2017‎ (UTC)


 * Added "Intel's". Btw Ryzen aren't AMD's first 6 core CPUs for mainstream, see Phenom II.--Pizzahut2 (talk) 23:53, 30 October 2017 (UTC)

Recent Intel bug affecting the processors
Should there be a link indicating that this processor would be affected by the recently announced http://www.bbc.com/news/technology-42553818 <!— Template:Unsigned --> — Preceding unsigned comment added by Trajano (talk • contribs) 17:37, 3 January 2018 (UTC)


 * Is there any confirmation that the bug affects Coffee Lake chips? NJA (t/ c)  10:39, 4 January 2018 (UTC)


 * It affects every Intel chip from Atoms to Xeons from the last decade or so. The problem is—if it warrants a mention in this article, then it should be in the articles for all the affected chips. Curly "JFC" Turkey 🍁 ¡gobble! 11:07, 4 January 2018 (UTC)

Btw this is not an Intel only bug. "These vulnerabilities affect many CPUs, including those from AMD, ARM, and Intel, as well as the devices and operating systems running on them." Google security blog Though AMD said their CPUs are only partially affected, see the variant/response matrix in AMD's statement. --Pizzahut2 (talk) 15:43, 4 January 2018 (UTC)


 * This seems like something that really ought to have it's own article to give the full scope. A mention/link here would then be appropriate. Otherwise the content would be repeated over all of the affected products. Dbsseven (talk) 15:55, 4 January 2018 (UTC)


 * There are two articles about it now. Spectre (security vulnerability), Meltdown (security vulnerability) --Pizzahut2 (talk) 16:51, 4 January 2018 (UTC)

After reading more, it seems there are either two "attack scenarios" (for the same bug) or two different bugs altogether. Either way, these two scenarios or bugs are called Meltdown and Spectre, and Meltdown seems to be Intel only. --Pizzahut2 (talk) 16:51, 4 January 2018 (UTC)


 * There are three bugs: Meltdown, which affects only Intel chips going back to 1995, with the exceptions of Itanium and pre-2013 Atoms; and Spectre 1 and Spectre 2, which affect everyone, including non-x86 chips (such as ARM), although AMD says their chips are not susceptible to Spectre 2. Curly "JFC" Turkey 🍁 ¡gobble! 23:37, 4 January 2018 (UTC)


 * Update: Some ARM CPUs are (partially) affected by Meltdown as well. --Pizzahut2 (talk) 13:56, 6 January 2018 (UTC)
 * ,, ,  There is a discussion on the AMD talk page about finding the right language for these exploits. As you all have clearly been interested in these topics, your input there would be helpful in finding consensus. Dbsseven (talk) 22:36, 13 January 2018 (UTC)

"the second 14 nm process refinement following both Skylake and Kaby Lake"
Broadwell was the first 14 nm process, then Skylake was the first refinement, and Kaby Lake was the second, meaning Coffee Lake is the third refinement.

209.150.231.43 (talk) 03:52, 9 January 2018 (UTC)
 * No. Broadwell and Skylake used the same 14nm process. Skylake was a new microarchitecture. Kaby Lake was the first refinement of that architecture and process (14nm+), and Coffee Lake the second (14nm++). See here and also Intel's [Tick-tock model]. Dbsseven (talk) 18:02, 9 January 2018 (UTC)

Further Coffee Lake processors to add
There is a new processor announced 8130U: https://newsroom.intel.com/news/new-8th-gen-intel-core-i3-processor-expands-performance-options-thin-light-laptops-2-in-1s/ Could this be added to the table?

More processors are listed here (which are sourced from the intel website): https://www.anandtech.com/show/12392/intel-adds-new-cfl-cpus-to-database Could these also be added to the table?

5W1TCH3D (talk) 22:54, 14 February 2018 (UTC)

Removing Mention of B0 Stepping Note
I believe this is not necessary and all Products in this page are Codename Coffe Lake not Kaby Lake any of it. 🥇 BUS  riderSFUser (talk • contribs) 00:34, 5 April 2018 (UTC)
 * Processors Core i3-8100 and Core i3-8350K with stepping B0 actually belong to "Kaby Lake-S" family

Graphics output standards detail inaccurate and potentially misleading
Coffee Lake processors natively support DisplayPort 1.4 (which is twice the bandwidth of DP 1.2) and needed for 4K HDR. DisplayPort 1.2 would be available via its embedded HDMI 1.4. Motherboard manufacturers can fit an LSPCON to convert on the board from DisplayPort 1.4 to HDMI 2.0a/b. On Motherboards with DisplayPort, a DisplayPort 1.4 to HDMI 2.0a (or 2.0b) adapter can be used to achieve support for 3840x2160@60Hz with HDR on a 4K HDR TV/Display. Most 2017 and later 4K HDR TV's ship with HDMI 2.0b. Only HDMI 2.0a and later support HDR. HDMI 2.0 does not support HDR. This is significant detail, especially for 4K UHD HDR enthusiasts and it doesn't seem to be easy to find this information in one place. Intel 7th generation and later CPU's are specified as being required to gain access to Netflix 4K content on Windows 10.

I would suggest at least changing the second sentence to say "The integrated graphics on Coffee Lake chips support DP 1.4 (with HDMI 2.0b via Motherboard-based LSPCON or external adapter) and HDMI 1.4".

Ref: https://newsroom.intel.com/newsroom/wp-content/uploads/sites/11/2017/09/8th-gen-intel-core-overview.pdf

Ref: https://www.intel.com/content/dam/support/us/en/documents/graphics/HDR_Intel_Graphics_TechWhitePaper.pdf

Example CPU spec ref for HDMI 1.4: https://ark.intel.com/products/126688/Intel-Core-i3-8100-Processor-6M-Cache-3_60-GHz#tab-blade-1-0-4

Example adapter: https://www.club-3d.com/en/detail/2442/displayport_1.4_to_hdmi_2.0a_hdr/

Ref: https://help.netflix.com/en/node/55763

Mcnaugha (talk) 12:59, 9 July 2018 (UTC)

Add column Release date
I wonder why the tables do not provide the Release date of each processor. Who knows? Do you mind if we add it? --Oliver H (talk) 22:19, 31 July 2018 (UTC)


 * I object. 1) Hardly anyone needs it 2) It'll be roughly the same for all CPUs and thus it'll make tables unnecessarily wider 3) Not really relevant. You could add it as a text somewhere in the article though again it's of very little usefulness if any. Artem-S-Tashkinov (talk) 10:31, 1 August 2018 (UTC)


 * However to compare the Release prices we need to have an idea on the Release dates, isn't it? I really appreciate having this information. I am missing the Release date while comparing the different products, and I am sure I am not alone in this case. On the other Wikipedia articles, the tables provide a column Release date. I think we should do the same here: add the column Release date for Coffee Lake. --Oliver H (talk) 15:21, 7 August 2018 (UTC)


 * If you absolutely need it, add it as YYYY QQ, e.g. Q1 2018 (the first quarter of 2018). No one really needs to know the exact day and also, it's not like Intel always released new products in the channel upon the official release. Artem-S-Tashkinov (talk) 09:56, 8 August 2018 (UTC)

Arbitrary highlighting of certain CPU models
I propose that we stop doing that altogether. There's no data which CPUs are more popular and which are not and then this information needs to be maintained and updated, and then in the future these CPUs won't be sold at all and this highlighting will be absolutely meaningless. Artem-S-Tashkinov (talk) 23:50, 10 August 2018 (UTC)

Xeon E-2100 family per-core turbo frequencies
As of 09-Nov-2018, the Xeon E-2100 "Workstation processors" table contains only a single "Max Turbo clock rate" column. However, pre-N-core turbo frequencies are known for long time.

Anandtech published this info in july --- https://www.anandtech.com/show/12199/intel-launches-coffee-lake-xeon-e-entry

And Intel itself published the same numbers in "Intel® Xeon® E-2100 Processor Family Datasheet" --- https://www.intel.com/content/dam/www/public/us/en/documents/datasheets/xeon-e-2100-datasheet-vol-1.pdf (see page 10). Bolkhov (talk) 08:01, 9 November 2018 (UTC)

DMI
The Skylake article contains this statement:

On the variants that will use a discrete Platform Controller Hub (PCH), Direct Media Interface (DMI) 2.0 is replaced by DMI 3.0, which allows speeds of up to 8 GT/s.

The Intel page for the i5-8600 used in the new mac Mini specifies "8 GT/s DMI3" so I'm guessing this is also true here (though it's not stated). &mdash; MaxEnt 18:57, 10 November 2018 (UTC)

Celeron G4930 is listed under both 8th gen and 9th gen sections
Celeron G4930 is listed under both 8th gen and 9th gen sections:

https://en.wikipedia.org/wiki/Coffee_Lake#Desktop_processors_(Coffee_Lake_S)

https://en.wikipedia.org/wiki/Coffee_Lake#Desktop_processors

Is this correct?


 * No, it's a refresh, should be removed from the first (8th gen) table. Artem S. Tashkinov (talk) 23:22, 1 June 2021 (UTC)

i3-8100B GPU Max frequency
The i3-8100B does not have CPU turbo, but GPU Turbo (Max clock rate) is listed as "-" in the table. Does this mean that it does not boost GPU frequency? Or unknown if it does? Intel Ark suggests "Graphics Max Dynamic Frequency = 1.05 GHz" for this processor, if that information is correct. — Preceding unsigned comment added by 91.127.83.227 (talk) 11:40, 12 August 2022 (UTC)