Jump to content

Talk:Meteor Lake: Difference between revisions

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Content deleted Content added
Prose
 
(10 intermediate revisions by 6 users not shown)
Line 1: Line 1:
{{WikiProject banner shell |class=Start|1=
{{Old prod|nom=Artem S. Tashkinov|nomdate=2023-07-17}}
{{WikiProject banner shell |1=
{{WikiProject Intel }}
{{WikiProject Intel}}
{{WikiProject Technology }}
{{WikiProject Technology |class=Stub}}
{{WikiProject Computing |importance=Low}}
{{WikiProject Computing |class=Stub}}
}}
}}

{{old move|date=16 July 2023|destination=Meteor Lake (Microarchitecture)|result=not moved|link=Special:Permalink/1166443270#Requested move 16 July 2023}}
{{old move|date=16 July 2023|destination=Meteor Lake (Microarchitecture)|result=not moved|link=Special:Permalink/1166443270#Requested move 16 July 2023}}


Line 59: Line 57:


: That the Meteor Lake architecture will include an NPU in some processors is a verifiable fact in reliable sources, e.g., [https://www.pcmag.com/news/the-meteor-lake-npu-meet-intels-dedicated-silicon-for-local-ai-processing The 'Meteor Lake' NPU: Meet Intel's Dedicated Silicon for Local AI Processing] and a section in the AnandTech article [https://www.anandtech.com/show/20046/intel-unveils-meteor-lake-architecture-intel-4-heralds-the-disaggregated-future-of-mobile-cpus/4]. So it is worth a mention in this article. We don't have more than the initial announcement and Intel marketing materials, however--no independent reviewers have tested the engine, there isn't any software using it, etc., so going in depth beyond basic facts that are beyond dispute would be premature--wait for reliable sources to review the NPU in depth. So I agree a brief statement about the NPU is warranted. But I would caution you about attacking other editors--it is a violation of civility [[WP:CIVIL]] and could get you banned if you persist in doing so. Such ad hominem attacks will also convince no one of the point you are trying to make--only citations of relevant content backing up your assertion in reliable sources will do that. --<code>&#123;&#123;u&#124;[[User:Mark viking|Mark viking]]&#125;&#125;&nbsp;{[[User talk:Mark viking|Talk]]}</code> 16:33, 22 September 2023 (UTC)
: That the Meteor Lake architecture will include an NPU in some processors is a verifiable fact in reliable sources, e.g., [https://www.pcmag.com/news/the-meteor-lake-npu-meet-intels-dedicated-silicon-for-local-ai-processing The 'Meteor Lake' NPU: Meet Intel's Dedicated Silicon for Local AI Processing] and a section in the AnandTech article [https://www.anandtech.com/show/20046/intel-unveils-meteor-lake-architecture-intel-4-heralds-the-disaggregated-future-of-mobile-cpus/4]. So it is worth a mention in this article. We don't have more than the initial announcement and Intel marketing materials, however--no independent reviewers have tested the engine, there isn't any software using it, etc., so going in depth beyond basic facts that are beyond dispute would be premature--wait for reliable sources to review the NPU in depth. So I agree a brief statement about the NPU is warranted. But I would caution you about attacking other editors--it is a violation of civility [[WP:CIVIL]] and could get you banned if you persist in doing so. Such ad hominem attacks will also convince no one of the point you are trying to make--only citations of relevant content backing up your assertion in reliable sources will do that. --<code>&#123;&#123;u&#124;[[User:Mark viking|Mark viking]]&#125;&#125;&nbsp;{[[User talk:Mark viking|Talk]]}</code> 16:33, 22 September 2023 (UTC)

:: Additional information can be found at page 21 at [https://download.intel.com/newsroom/2023/ai/2023_Innovation_Day2_Keynote.pdf Intel Innovation 2023 Day 2 Keynote Presentation Slides]. <!-- Template:Unsigned --><small class="autosigned">—&nbsp;Preceding [[Wikipedia:Signatures|unsigned]] comment added by [[User:Rjluna2|Rjluna2]] ([[User talk:Rjluna2#top|talk]] • [[Special:Contributions/Rjluna2|contribs]]) 13:27, 26 September 2023 (UTC)</small>


{{reflist-talk}}<!-- write your message above the reflist-talk template if you are adding a comment in this section -->
{{reflist-talk}}<!-- write your message above the reflist-talk template if you are adding a comment in this section -->
Line 77: Line 77:


:Desktop processors are already mentioned in the description (the last sentence). There's no further information to add. [[User:Artem S. Tashkinov|Artem S. Tashkinov]] ([[User talk:Artem S. Tashkinov|talk]]) 08:55, 26 September 2023 (UTC)
:Desktop processors are already mentioned in the description (the last sentence). There's no further information to add. [[User:Artem S. Tashkinov|Artem S. Tashkinov]] ([[User talk:Artem S. Tashkinov|talk]]) 08:55, 26 September 2023 (UTC)

:Meteor Lake will '''not''' be on desktop platform (LGA 1851), here are four sources to back it up: <ref>https://www.anandtech.com/show/21076/intel-meteor-lake-soc-is-not-coming-to-desktops-well-not-technically</ref><ref>https://arstechnica.com/gadgets/2023/09/intel-reiterates-next-gen-meteor-lake-cpus-arent-coming-to-most-desktops/</ref><ref>https://www.digitaltrends.com/computing/intel-meteor-lake-desktop-not-socketed/</ref><ref>https://www.windowscentral.com/hardware/cpu-gpu-components/intel-wont-make-meteor-lake-desktop-chips-after-all-at-least-not-in-the-way-youd-expect</ref>
:I think the "Meteor Lake will be on desktop" confusion comes from mini-PCs, all-in-ones, and other compact desktops that will feature a Meteor Lake mobile processor. You could consider it to be a desktop application of Meteor Lake, but it's still actually using a ''mobile platform''. —&nbsp;[[User:AP 499D25|<span style="background:#1F6295;color:white;padding:1q 5q;border-radius:10q;font-family:Franklin Gothic, Verdana">AP&nbsp;499D25</span>]] [[User talk:AP 499D25|<span style="color:#1A527D">(talk)</span>]] 01:48, 7 November 2023 (UTC)

{{reflist-talk}} <!-- add your comments in the Meteor Lake desktop section above here -->

== AVX-512 support ==

According to GCC source code Meteor Lake has the same ISA extension support as Alder Lake and Raptor Lake - https://gcc.gnu.org/git?p=gcc.git;a=blob;f=gcc/common/config/i386/i386-common.cc;h=988805a3aeddce9952a3256b279e081877909c7a;hb=refs/heads/releases/gcc-13#l2089

It does not support AVX-512, and this should be reflected in the General Information section. [[Special:Contributions/188.47.113.184|188.47.113.184]] ([[User talk:188.47.113.184|talk]]) 16:33, 29 December 2023 (UTC)

:AVX-512 is not available on Alder Lake, Raptor Lake and Meteor Lake. I'd support a mention in the article. [[User:Digital27|Digital27]] ([[User talk:Digital27|talk]]) 03:20, 30 December 2023 (UTC)

== Previous Generation Branding ==

At the [[Meteor Lake#Branding|Branding]] section, the previous generation branding should not be included in this article. Please advise. [[User:Rjluna2|Rjluna2]] ([[User talk:Rjluna2|talk]]) 16:45, 5 April 2024 (UTC)

== Homegrown AI Tools Shorten Design Cycles from Weeks to Hours ==

Here how these engineer work with [[Intelligence amplification|Augmented Intelligence]] on this generation microprocessor as described at [https://www.intel.com/content/www/us/en/newsroom/news/homegrown-ai-tools-shorten-design-cycles-to-hours.html#gs.8a8jmw Homegrown AI Tools Shorten Design Cycles from Weeks to Hours]. [[User:Rjluna2|Rjluna2]] ([[User talk:Rjluna2|talk]]) 15:43, 21 April 2024 (UTC)

== Prose ==

I'm '''really not''' a fan of prose being used for describing platform features:
* It takes a ton of space
* It requires very careful and mindful reading because the information therein is quite important
* Various pieces of information are easy to miss or misread
* It's not in line with all the existing articles

Perhaps it doesn't belong to CPU/GPU related articles. The same applies to the [[Lunar Lake]] article. Please leave your comments on the [[Talk:Lunar Lake#Prose]] page. [[User:Artem S. Tashkinov|Artem S. Tashkinov]] ([[User talk:Artem S. Tashkinov|talk]]) 12:44, 18 June 2024 (UTC)

Latest revision as of 12:44, 18 June 2024

Orphaned references in Meteor Lake

[edit]

I check pages listed in Category:Pages with incorrect ref formatting to try to fix reference errors. One of the things I do is look for content for orphaned references in wikilinked articles. I have found content for some of Meteor Lake's orphans, the problem is that I found more than one version. I can't determine which (if any) is correct for this article, so I am asking for a sentient editor to look it over and copy the correct ref content into this article.

Reference named "RCP" in group "lower-alpha":

  • From Intel Core: Price reflects Recommended Customer Price (RCP) rather than MSRP. RCP is the cost per unit, in bulk sales of 1000 units or more, to OEMs, ODMs, and retail outlets when purchasing from Intel. Actual MSRP is higher than RCP
  • From Comet Lake: Price is Recommended Customer Price (RCP) at launch. RCP is the trade price that processors are sold by Intel to retailers and OEMs. Actual MSRP for consumers is higher
  • From Raptor Lake: Price is Recommended Customer Price (RCP) at launch. RCP is the trade price that processors are sold by Intel to retailers and OEMs. Actual MSRP for consumers is higher.

I apologize if any of the above are effectively identical; I am just a simple computer program, so I can't determine whether minor differences are significant or not. Feel free to remove this comment after fixing the refs. AnomieBOT 16:01, 14 July 2023 (UTC)[reply]

Requested move 16 July 2023

[edit]
The following is a closed discussion of a requested move. Please do not modify it. Subsequent comments should be made in a new section on the talk page. Editors desiring to contest the closing decision should consider a move review after discussing it on the closer's talk page. No further edits should be made to this discussion.

The result of the move request was: not moved. Per strong opposition as there isn’t a page for the lake yet. If a page is created and disambiguation is needed the qualifier should be lowercase. (non-admin closure) - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 02:24, 22 July 2023 (UTC)[reply]


Meteor LakeMeteor Lake (Microarchitecture) – Meteor Lake is also a lake in Canada Maxim Masiutin (talk) 01:06, 16 July 2023 (UTC)[reply]


The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Orphaned references and Intel's references

[edit]

I've corrected the references and added Intel's references to address the issues raised in the request for deletion. --Maxim Masiutin (talk) 15:54, 21 July 2023 (UTC)[reply]

NPU section dispute

[edit]

Lack of references in this section is not a valid reason for content delete. Valid reference is in CPU section to whole external article. Deleting whole section content just because editor dont understand what is NPU despite reference it clearly saying Neural Processing Unit is pure vandalism imho. Duplicated reference in NPU section is not needed imho but was added to protect this section content against editor vandalism. Purpose of this section is not duplicating content of sources external references or teaching how important is NPU in contemporary processors. --159.205.25.200 (talk) 10:36, 22 September 2023 (UTC)[reply]

Once again NPU section was vandalised by Artem S. Tashkinov with remarks : "The linked TPU article contains 0 (zero) readable information (pictures are pictures, they are promotional materials) about the NPU, what it is, how it used and why it's important for the user. Proper citations are still required. " Remarks above are obviously false. Citation from linked TPU article : "The NPU contains four basic hardware components, the Global Control (a crossbar) alongside a memory mapping unit (MMU) and DMA; a tiny scratchpad RAM; and two NCEs (neural compute engines).

Each NCE contains a programmable DSP, and the inference pipeline. The inference pipeline contains the main number crunching resource, the MAC array. These are a set of logic components that accelerate matrix multiply and convolution (MAC), INT8 and FP16, with up to 2,048 MACs per cycle. Besides the MAC array, there is fixed function hardware for data conversion and array activation. Each NCE contains two VLIW programmable DSPs that supports nearly all data-types ranging between INT4 to FP32. At a hardware level, AI inference acceleration is memory intensive, and so, Intel deployed a localized SRAM to work like a scratchpad. This memory intensiveness is also why the NPU is located on the SoC tile, sharing the same fabric as the memory controllers."[1]

As you can see there are lot of readable information in disputed article but they yet not readable for Artem S. Tashkinov --159.205.20.243 (talk) 14:34, 22 September 2023 (UTC) --159.205.20.243 (talk) 14:58, 22 September 2023 (UTC)[reply]

That the Meteor Lake architecture will include an NPU in some processors is a verifiable fact in reliable sources, e.g., The 'Meteor Lake' NPU: Meet Intel's Dedicated Silicon for Local AI Processing and a section in the AnandTech article [2]. So it is worth a mention in this article. We don't have more than the initial announcement and Intel marketing materials, however--no independent reviewers have tested the engine, there isn't any software using it, etc., so going in depth beyond basic facts that are beyond dispute would be premature--wait for reliable sources to review the NPU in depth. So I agree a brief statement about the NPU is warranted. But I would caution you about attacking other editors--it is a violation of civility WP:CIVIL and could get you banned if you persist in doing so. Such ad hominem attacks will also convince no one of the point you are trying to make--only citations of relevant content backing up your assertion in reliable sources will do that. --{{u|Mark viking}} {Talk} 16:33, 22 September 2023 (UTC)[reply]
Additional information can be found at page 21 at Intel Innovation 2023 Day 2 Keynote Presentation Slides. — Preceding unsigned comment added by Rjluna2 (talkcontribs) 13:27, 26 September 2023 (UTC)[reply]

References

  1. ^ W1zzard (2023-09-19). "Intel Meteor Lake Technical Deep Dive". TechPowerUp. Retrieved 2023-09-19.{{cite web}}: CS1 maint: numeric names: authors list (link)

Intel Innovation 2023: Empowering Developers to Bring AI Everywhere

[edit]

Intel discusses about these forthcoming Intel Core Ultra processors at Intel Innovation 2023: Empowering Developers to Bring AI Everywhere. Rjluna2 (talk) 20:15, 22 September 2023 (UTC)[reply]

Intel Innovation 2023: Technologies to Bring AI Everywhere

[edit]

Intel discusses AI at the forthcoming of this generation processor at Intel Innovation 2023: Technologies to Bring AI Everywhere. Rjluna2 (talk) 20:37, 22 September 2023 (UTC)[reply]

Meteor Lake Desktop

[edit]

Meteor Lake CPUs have been confirmed for desktops in 2024.

This page needs editing to reflect that. 92.11.253.208 (talk) 00:08, 26 September 2023 (UTC)[reply]

Desktop processors are already mentioned in the description (the last sentence). There's no further information to add. Artem S. Tashkinov (talk) 08:55, 26 September 2023 (UTC)[reply]
Meteor Lake will not be on desktop platform (LGA 1851), here are four sources to back it up: [1][2][3][4]
I think the "Meteor Lake will be on desktop" confusion comes from mini-PCs, all-in-ones, and other compact desktops that will feature a Meteor Lake mobile processor. You could consider it to be a desktop application of Meteor Lake, but it's still actually using a mobile platform. — AP 499D25 (talk) 01:48, 7 November 2023 (UTC)[reply]

AVX-512 support

[edit]

According to GCC source code Meteor Lake has the same ISA extension support as Alder Lake and Raptor Lake - https://gcc.gnu.org/git?p=gcc.git;a=blob;f=gcc/common/config/i386/i386-common.cc;h=988805a3aeddce9952a3256b279e081877909c7a;hb=refs/heads/releases/gcc-13#l2089

It does not support AVX-512, and this should be reflected in the General Information section. 188.47.113.184 (talk) 16:33, 29 December 2023 (UTC)[reply]

AVX-512 is not available on Alder Lake, Raptor Lake and Meteor Lake. I'd support a mention in the article. Digital27 (talk) 03:20, 30 December 2023 (UTC)[reply]

Previous Generation Branding

[edit]

At the Branding section, the previous generation branding should not be included in this article. Please advise. Rjluna2 (talk) 16:45, 5 April 2024 (UTC)[reply]

Homegrown AI Tools Shorten Design Cycles from Weeks to Hours

[edit]

Here how these engineer work with Augmented Intelligence on this generation microprocessor as described at Homegrown AI Tools Shorten Design Cycles from Weeks to Hours. Rjluna2 (talk) 15:43, 21 April 2024 (UTC)[reply]

Prose

[edit]

I'm really not a fan of prose being used for describing platform features:

  • It takes a ton of space
  • It requires very careful and mindful reading because the information therein is quite important
  • Various pieces of information are easy to miss or misread
  • It's not in line with all the existing articles

Perhaps it doesn't belong to CPU/GPU related articles. The same applies to the Lunar Lake article. Please leave your comments on the Talk:Lunar Lake#Prose page. Artem S. Tashkinov (talk) 12:44, 18 June 2024 (UTC)[reply]