Talk:Jaguar (microarchitecture)

From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Computing  
WikiProject icon This article is within the scope of WikiProject Computing, a collaborative effort to improve the coverage of computers, computing, and information technology on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
 ???  This article has not yet received a rating on the project's quality scale.
 ???  This article has not yet received a rating on the project's importance scale.
 

Anybody have any sources for the "8 in PlayStation 4 and Xbox One"? As far as we know, there are articles floating online claiming the consoles are "8 core", " 2x 4cores" and "4 x 2cores", if there are no reliable sources I will remove the core count mentions when mentioning the consoles.--202.66.131.118 (talk) 01:10, 15 July 2013 (UTC)


Since there are no references to xbox one being quad core, and it is not clear whether the 8 cores are on a single die or not, I am removing the reference to the core count used in the PS4 and xbox. --202.66.131.118 (talk) 06:49, 16 July 2013 (UTC)

Info about G-series: http://www.brightsideofnews.com/news/2013/4/24/amd-launches-kabini-based-g-series-soc.aspx202.66.131.118 (talk) 10:31, 16 July 2013 (UTC)

Consoles (cpu clock speed correction)[edit]

the ps4 cpu clock speed is described as 1.6-2.75ghz, the only reference to 2.75ghz i found was a "reported" devkit "maximum clockrate". (it is not even specificated if is cpu clock, its just a fcc document that mostly covers and regulates emi and rf emissions limits)

it is important to state that is normal to a devkit to deviate from the consumer hardware it targets (i.e. ps3 devkits had twice the amount of ram, more clockspeed would help debugger performance for example)

the actual correct clockrate seems to be 1.6ghz (confirmed by sony, and in sync with the playstation 4 wikipedia article that points to the sony statement) — Preceding unsigned comment added by 177.83.7.20 (talk) 22:35, 9 October 2014 (UTC)

GPU architecture[edit]

As can be read here: Graphics_hardware_and_FOSS#ATI.2FAMD there are a couple of GPU microarchitectures, the latest being VLIW5, VLIW4 and GCN (Graphics Core Next). It would be nice, if the architecture would be named in this and also in other articles, e.g. Jaguar (microarchitecture), List of AMD Accelerated Processing Unit microprocessors, etc.

Semsi Paco Virchow (talk) 18:42, 27 July 2013 (UTC)

Athlon/Sempron Kabini's?[edit]

Should the Athlon 5350, 5150, Sempron 3850 and 2650 be added to this page? As I understand they have Jaguar architecture as well. — Preceding unsigned comment added by 137.120.181.25 (talk) 08:52, 8 April 2014 (UTC)

Yes, providing you can find a reference that confirms they are based on Jaguar.BananaLanguage (talk) 09:20, 8 April 2014 (UTC)

External links modified[edit]

Hello fellow Wikipedians,

I have just added archive links to one external link on Jaguar (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 or failed to let others know (documentation at {{Sourcecheck}}).

Question? Archived sources still need to be checked

Cheers.—cyberbot IITalk to my owner:Online 01:31, 27 March 2016 (UTC)


PS4 Neo APU Specs[edit]

There is no consensus in this RfC to include the reported but unconfirmed PS4 Neo APU specs here.

However, Tigraan (talk · contribs) mentioned a compromise wording:

"on day Foo, Bar said spec X would be Y". It gives an information of reasonable use without being crystalbally, and even if the page goes unmodified until the end of time it will be a correct assertion, so what is the problem exactly? Please do not answer "page size"

LaughingVulcan found this approach appealing; this resolved LaughingVulcan's concern about using the "preliminary" wording. No other editors commented on Tigraan's proposal, so I encourage editors to open a new RfC to discuss it.

Cunard (talk) 22:31, 25 September 2016 (UTC)

The following discussion 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.

I undid the deletion of the reported but unconfirmed PS4 Neo APU specs. The sources are reliable but the final specs are not verified, as the deleter noted. However, I believe it is better to include the reported values and revise as they are confirmed. Totally unknown values are marked properly. I have returned the specs but added a "verify credibility" tag for clarity. Dbsseven (talk) 00:52, 20 August 2016 (UTC)

You are citing reported but unconfirmed developer kit specifications. These often use substitute components that are comparable to the end product. For example, the specifications for the first Orbis dev kits were dissimilar to its commercial form. We need to wait until reliable sources report on final hardware. — TPX 08:30, 20 August 2016 (UTC)
The source cited, Ars Technica is a reliable source. Waiting for only final hardware specifications for CPU/APU design is not entirely reasonable. This limits articles to only fully released products with itemized specifications. Even when a product is announced the specs may change (ie. the Xbox One GPU speed), or may never be confirmed by the manufacturer (ie. Apple A processors specs). Within CPU/APU articles it is common to include the specs of upcoming products from reputable journalists and cite them as such. The article/specs can be revised as new information becomes available. If you have a different way to note the specs as unconfirmed, that would be most constructive. Dbsseven (talk) 22:19, 20 August 2016 (UTC)
We agree that the Ars Technica is a reliable source for the first wave of development kits, but not final hardware. Therefore, the source does not support the inclusion of final hardware specification in the table. We need to wait until early next month. — TPX 23:21, 20 August 2016 (UTC)
But the Ars article says nothing about development kit. It says "codename for the console is "NEO," and it even provides hardware specs". So the assertion that this is a development kit specs is yours, not the sources. Dbsseven (talk) 15:22, 22 August 2016 (UTC)
Ars Technica credits Giant Bomb for the story and specifications. That is because Giant Bomb received developer documentation that is believed to be authentic. Eurogamer also obtained the same documents seen by Giant Bomb [1] which formed the basis of the Ars Technica article and many others. That documentation provides a target render for developers, but nowhere does it say the hardware is fixed and final. On the contrary, there have since been reports saying that Sony are looking to increase the performance of Neo in response to Microsoft's system that will arrive one year later. We really can't profess to know what is inside the final hardware, or the clockrate, until we have reliable sources that examine the final hardware in detail. We don't have long to wait. — TPX 19:58, 22 August 2016 (UTC)
I do not believe the point here is the wait. Even announced specifications may be not final, and some specs never be announced. I realize the first report form EuroGamer is based on developer specs, but that does not necessarily preclude a reputable source from confirming them if they happen to be correct. What should the standard for inclusion be: final hardware examined by a third party, itemized specs from Sony, a product announcement? I believe it was inappropriate to completely delete the section if is correctly cited with current information. The specs can be revised as more information becomes available. Dbsseven (talk) 20:31, 22 August 2016 (UTC)
Word from an official source is good. Or a reliable source that conveys the same information. Even a third-party product teardown reported by a reliable source is sufficient. But it must be final hardware and confirmed, otherwise we are going around in circles. — TPX 22:26, 22 August 2016 (UTC)

I am trying to find an consensus for a standard of inclusion. I believe the disagreement is between:

  • Include final and verified specs from the manufacturer, perhaps conveyed by a reliable source. Include specs from a credible third party with physical access to the final product.
  • Include reported information of products prior to release from a reliable source. Revise as information from reliable sources becomes available.

I agree the standards you have set are certainly sufficient. But they unnecessarily exclude products not yet announced but documented by reliable sources. I believe "final hardware and confirmed" is an unreasonably high standard. This would mean waiting for release date on every product. Manufacturers have changed specs between announcement and release, or after release even. Or they may never release specs.

As an example: Wikipedia includes upcoming events articles. These certainly cannot be final articles on those events, and yet there is no disagreement on their inclusion and revision. Dbsseven (talk) 23:34, 22 August 2016 (UTC)

Pictogram voting comment.png 3O Response: I don't think specs of unfinished versions or rumored products are encyclopedic, even if verifiable. See WP:CRYSTALBALL. Rhoark (talk) 00:49, 23 August 2016 (UTC)
  • Comment both views have merit. I doubt that either course of action would lead to much harm, especially in the longer run. I recommend that we take the more tolerant route, but if there is serious resistance, breathe deeply, think peaceful thoughts and wait a couple of weeks. JonRichfield (talk) 11:16, 29 August 2016 (UTC)
  • Pictogram voting comment.png RFC Response: IMO a reasonable compromise is to report with attribution: "as preliminarily reported by Ars Technica". Staszek Lem (talk) 17:25, 1 September 2016 (UTC)
  • I am also via RFC and I agree with the above. Marking it as preliminary would be useful. Jerod Lycett (talk) 04:30, 4 September 2016 (UTC)
  • Comment. Completionist view. Five years from now we sincerely hope the product will have been deployed and enjoyed, the exact specs should be known, and another editor will come along and notice the lack of specs and add it in – trust me, you guys and gals who report on products and specs are nothing but thorough when you come across missing information.  :) Now substitute, “day after release,” for, “five years from now.” Does the encyclopedia or the reader suffer greatly with the lack of information until that point? (And that is a question, not a rock-solid conclusion…)
The problem with marking it ‘preliminary’, in general: The day it hits the streets, the specs are no longer preliminary, whether we know they are accurate or not. While, as above, one might assume that a reliable editor may come along and remove ‘preliminary,’ thoroughly check those preliminary specs to the released product, and correct if necessary… There’s no guarantee of that. If the specs were never there until release, no direct harm is done by the lack. If they are there pre-release, and not corrected at release, you may have erroneous reporting due to spec changes on top of keeping an outdated “preliminary,” until an editor gets around to noticing it. I don’t have diffs for you, but it is something I see happen in other areas of WP coverage. (“A court decision is expected in May, 2010. [article never reported if the decision occurred.] X reports the album will have a run time of 3 and half hours. [article doesn’t say whether the album dropped or not, let alone actual run length.]”) You have potential very minor harm if the article is not updated. Just MVHO. LaughingVulcan 13:30, 5 September 2016 (UTC)
But the alternative is not between dated information and up-to-date information. If the "preliminary" comment was not there, there would be nothing at all. Is it really better to have nothing, rather than a sourced though dated information, that may push the reader in the correct direction to confirm or infirm the mention? "Yes" is not an unreasonable answer, but I disagree with it. (By the way, Template:As of is very useful for these sorts of things.) TigraanClick here to contact me 13:50, 8 September 2016 (UTC)
Yes, I see your point. To me it is all but a coin flip, hence making a comment rather than a !vote on it. I think if my choices are dated information and no information, I lean towards no information, actually. But since that's a risk no matter what unless you get an article to have a long running steward who keeps up with any changes in production as well. While I won't vote for it, I think giving the date with attribution does go a long way towards letting the reader know just how out of date a reference may be. Good answer! LaughingVulcan Grok Page! 22:18, 9 September 2016 (UTC)
  • Include with attribution and date, i.e. "on day Foo, Bar said spec X would be Y". It gives an information of reasonable use without being crystalbally, and even if the page goes unmodified until the end of time it will be a correct assertion, so what is the problem exactly? Please do not answer "page size"... TigraanClick here to contact me 13:46, 8 September 2016 (UTC)
  • Comment: I agree with JonRichfield and Staszek Lem's statements above. Both views are logical, yet the content dispute persists, therefore either do nothing and wait or state information with attribution and date. Best, FoCuS contribs; talk to me! 13:28, 9 September 2016 (UTC)

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.

PS4 Neo TMU/ROP/core specs[edit]

Multiple IPs have consistently been modifying the PS4 Neo APU specs without citation. In particular the TMU and ROP count, and the CPU details. While the latest edit claimed they were accurate, they are not in any of the cited sources. If there is no citation, I do not believe this information should be included. These values cannot be asserted/included without citation. Should we ask for semi-protection? Dbsseven (talk) 21:45, 22 September 2016 (UTC)