Talk:List of AMD Phenom microprocessors

From Wikipedia, the free encyclopedia

Contents

[edit] Proposed split

I put the Athlon 6-series processors on this page because they are not based on the Athlon 64. They are based on K10. Therefore, since this page is a list of K10 based processors, they should be on this page. Athlon 64's are based on K8, which is a different uArch. Also, in the future when you propose a split, it would be nice to put something on the talk page. -- Imperator3733 01:44, 21 October 2007 (UTC)

Guess you won't mind my messy presentation, probably even my awkward grammar ;)

Anyways, the dual-core processors will be sold under the brand as "Athlon X2", that's enough rationale to move the entries to that page/section, right? So they're based on a brand new microarchitecture, and so what does that have any impact on me and my future purchasing plan with AMD/Intel processor? The SSE128 FP Units or the L3 cache or SSE4a or the 32-bit prefetching? What are the importance of an microarchitecture to an average end-user (non-techno geeks and probably not the ones who edit in-depth technical details on processor pages, like Bulldozer and Nehalem) or general public (not all but still a majority) as customers/future users? Do they know what in the earth is that K10 *new* microarchitecture representing? Isn't Wikipedia intended for audiences as them? Here is not a repository for technical specification manuals for BIOS developers or system integrators as hosted on AMD developer website or Partner pages!

I put it this way, from AMD naming schemes since PR ratings (<sarcasm>For instance, Athlon 64 X2 7600+1 for AMD 3.8 GHz dual-core processor, comparable to 7.6 GHz Intel Pentium D, that's double the performance</sarcasm>) we know that most of the people probably do not care about the specifications of the processor they are buying, they use relative comparisons (in forms of "figures") to identify the performance of EACH processor, it was used to be clock speeds (also known as the MegaHertz war, ended by Pentium D 3.8 GHz processors, but 'not for overclockers'); now it's the number of cores (FYI: core wars, as Nehalem 8-core plus 80-core Tera-scale project vs. Sandtiger - native 8 Bulldozer cores), the model numbers ("Intel Core 2 Extreme QX9770" vs. "AMD Phenom 9780 processors1"), and the power consumption figures - TDP/ACP (Xeon E74701 105W TDP vs. Opteron 8370SE1 105W ACP, same number on paper, right? Guess which consumes less power? Place you bets!) that matters. Thus a product of "AMD Phenom 9790 processor1" (4-core, 105W ACP) is far better than "AMD Phenom 7780 processor1" (3-core, 105W ACP) and is "in theory" better than "Intel Core-whatever QX9770" (4-core, 136W TDP, Ouch!) offerings by comparing these aforementioned numbers, right? Then the stress of the *new* microarchitecture becomes diminished by the customers' tendency of these kind of comparisons and thus the whole microarchitecture thing becomes meaningless.

You may argue that the microarchitecture of a processor may affect actually performance which is reflected by benchmark results, but actually how many people who wanted to build a decent computer for mostly Internet surfing and simple Solitaire games will be expected to see "complicated" benchmark results? Users not only compare products using those superficial figures, but also with their general usage experience (e.g. Slow/fast? Lags/smooth? Hot/cool? Noisy/quiet? Ugly/eye-catching? Unsatisfied/contended? Unhappy/Happy? etc.), so what's a set of very objective benchmark result which is affetced by a slight change (or stepping or errata) in microarchitecture related to those people who use subjective feelings to compare things? No, actually, so what's the difference of 10.324s vs 10.697s for 1M SuperPi? And how about 130.5 fps vs 128.6 fps (1.9/128.6 = 0.0148±0.00005 second)? Will the system slow down to an "unacceptable level" due to this slight change?

As a sidenote, the AMD K10 article still exists in Wikipedia just because at the time of starting the stub, the microarchitecture has NO marketing names whatsoever and there have to be a place for rumors/speculations/prelimilary informations (like Sandy Bridge, thus two article AMD K8L and AMD K10 appeared - thanks to the Inq confusion, then at a later time AMD K8L merged with AMD K10), the AMD K10 article should be officially gone (read "redirected") at the time of the release of Phenom 9000/7000/whatever 6000 series processors (according to the usual pratice of AMD CPU architecture articles, I know Intel ones will remain whatsoever), and sections will be merged/moved to Phenom article and other articles (Phenom FX, probably Phenom X4/X3/X2, depending on final market branding if including core count), same as the other articles about the AMD microarchitecture (AMD K7 redirects Athlon, and AMD K8 redirects Athlon 64, except that AMD K9 is "cancelled" thus is remained as reference being cited), and finally there should be a List of AMD Phenom microprocessors, so renaming this article and split the table entries to List of AMD Athlon X2 microprocessors should be the easiest way of solving this branding problem and I don't see the point of insisting the stress on AMD K10 microarchitecture here, as:

  1. this does not meet what the usual pratice of AMD microarchitecture articles and
  2. the majority of the general public is less concerned by the cause, process (K10 CPU microarchitecture), instead

they expect more about the final outcome (Phenom processors).

That's my opinion. Feel free to flame me now, and have a nice day.

  • 1 Please note that the above model numbers (except officially announced ones) are for not intended for serious reading.

--202.40.157.145 07:14, 23 October 2007 (UTC)

First, I do not intend to flame you. My point is although Kuma will be sold under the Athlon X2 brand, it is more similar to the Phenom processors. It is quite unfortunate that AMD is already using the Athlon X2 brand, because otherwise I would have no problem with spliting this into List of AMD Phenom microprocessors and List of AMD Athlon X2 microprocessors. Currently, List of AMD Athlon X2 microprocessors redirects to List of AMD Athlon 64 microprocessors. Maybe what we could do is change List of AMD Athlon X2 microprocessors to non-redirect and have a {{main}} tag on the section about the Brisbane (K8-based) Athlon X2 processors pointing to the List of AMD Athlon 64 microprocessors section on The Athlon X2s. The rest of the article could then have the Kuma processors, and the Phenom processors could be in List of AMD Phenom microprocessors. This page could then redirect to the Phenom article, and have a note at the top of that page saying that the Athlon 6-series processors are at List of AMD Athlon X2 microprocessors. What do you think? -- Imperator3733 15:21, 23 October 2007 (UTC)
While this maybe true for identifying the uArch of the processor, but the same goes for K8 and K10 series of Opteron processors, the Athlon X2 processors came from Athlon 64 X2, AMD decided to scrap the "64" label to downplay the significance of 64-bit computing, so the brand is inherented from K8 generation to K10 generation when AMD uses the brand for dual-core processors. I think Wikipedia put more focus on the product brandings instead of the technical side of the product, while I am not saying that the technical stuff would be avoided, but instead it occupies only for a small portion. And I think following the same pattern as previous articles is best for readers, I have no objection for the split of the list in this article and the list in Athlon 64 processors list to the List of AMD Athlon X2 processors, "if and only if" the Athlon X2 will bacome a "larger product line" (e.g. more than 10 models), otherwise appending to the last subsection is my choice. --202.40.157.145 02:59, 24 October 2007 (UTC)
Kuma has now been moved to List of AMD Athlon X2 microprocessors. I'll do the others soon. -- Imperator3733 16:24, 24 October 2007 (UTC)
This article has now been renamed to List of AMD Phenom microprocessors. -- Imperator3733 17:48, 24 October 2007 (UTC)

[edit] Projected Prices

First, I clearly had the column headers labeled as "Projected Price at Introduction", not just as "Price at Introduction". This implies that these are not necessarily the price at introduction, just that that is what is the current information says.

Second, its not like the sources are The Inquirer or The Register. Two credible sources, TG Daily and DailyTech both have the same prices listed. Those articles were published on November 6 and both are referenced. I can not be sure of the accuracy of the the information, but it is the current information and is from credible sources.

Please address any concerns that you have with those columns here first. Thank you. -- Imperator3733 01:37, 13 November 2007 (UTC)

I'm not opposing the addition of ANY which may help others to understand the subject better, even being crystal clear about the subject matter. I'm also not opposing the style being used in Intel lists, having a launch price and product codes or something else, to be applied here. But if it's project price range, then it's not worthless to be here, 1) possible updates to the price without leaking the news, making the list here outdated and 2) changes to a launch price is very common to a company for several reasons and 3) everyone can say anything about the final numbers which presented in a certain range, a number range is just as pointless predicting future product specfications (The example you gave, the Inquirer and the Register often guess from "current trends", and PC Watch (Japanese site) used to guess about every single thing, including the how the CPU components are configured). The only thing notable is the "final" or what I say "official" launch prices. Other stuff reported somewhere around the web about the prices, although sourced, are not up-to-date (note the reports dated November 6, 2007, not a small possibility that the prices have been changed internally) and untrusyworthy. As for the Intel lists, they have the launch price but either they are sourced very well (leaked official roadmaps, e.g. "Intel Desktop Processor Roadmap Performance") or have been put up after launch, we have concrete proof. AMD, on the other hand, have the "keep the mouth shut" attitude towards future products (even product launches), making all other sources, even though those reports primary sourced Taiwanese mobo maker, unnamed AMD employee(s) or retailers, once again untrustworthy. All we need here is concrete proof that the officials admitted those informations, but words from those sources? "nope, I'll pass". A reference to an online press coverage with screenshots about the price list made in a sales presentation or a confirmation from AMD spokesperson would be nice, otherwise everything unofficial are posed to be speculations by a certain extend.
My suggestion: Put up the Launch Price after launch prices are "officially" announced. --202.40.157.145 05:16, 13 November 2007 (UTC)
P.S.: A quote in TGDaily report: "While we always have to "take these pre-launch prices with a grain of salt", it is already clear that AMD will be aggressive in its pricing..."
P.P.S.: I'll try to make my replies shorter, sorry for your inconvenience.
Okay, we'll wait until the prices are officially announced. -- Imperator3733 15:55, 13 November 2007 (UTC)

[edit] 9600 Black Edition release date

Was the 9600 Black Edition really launched on November 19, 2007? I thought that it came later than that. Maybe I'm mistaken, but I think all previous Black chips were released after their non-Black counterparts. -- Imperator3733 (talk) 01:23, 27 January 2008 (UTC)

Official release seems to be December 19, 2007. [1] (german) --Denniss (talk) 13:12, 27 January 2008 (UTC)

[edit] Separate article for future chips

I just moved all the sections with {{futurechip}} labels to List of future AMD Phenom microprocessors. -- Imperator3733 (talk) 01:41, 27 January 2008 (UTC)

[edit] Black Edition Notes

Somehow the article needs to denote the difference that the "Black Edition" brings. Previous attempts to denote the difference have been erased. It is very relavent to the article and should be included somehow. Quovatis (talk) 16:17, 20 February 2008 (UTC)