Talk:Socket A

Why no love for non AMD chipsets?
We're totally missing the VIA KT133A chipset here. It was most well known chipset of the day, SDRAM, yet faster than AMD's poor DDR chipsets — Preceding unsigned comment added by 114.23.109.150 (talk) 10:53, 26 November 2021 (UTC)

First 1 GHz x86 platform?
I'm pretty sure it was on Slot A and not Socket A. AMD broke the 1 GHz barrier when they released the 800, 900 and 1 GHz versions on March 6th 2000. I think Socket A didn't debut until June that year. DrFod

-- Very true, according to articles on review sites, the first 1 GHz CPU (Athlon) debuted on Slot A. http://www.tomshardware.com/cpu/20000314/index.html
 * You're right. It was not Socket A, but it was Slot A to have the first 1 GHz CPU. Racecar56 (talk) 00:14, 30 November 2009 (UTC)

AMD recommendation
I change the order of the heat sink recommendation as AMD recommends 300 grams not 10.6 ounces. As the 10.6 is derived from the published 300 grams recommendation, it should be in brackets not vice versa. Originally the article only had the 300 grams until an anon added the 10.6 ounces and put the 300 grams in brackets too so even if we stick with the status quo, it should be 300 grams first not vice versa. Nil Einne 07:53, 19 June 2007 (UTC)

Article is unreferenced
While going through the article history, I noticed more then once someone had added an unreferenced tag which was later removed. Although I probably wouldn't have bothered to add an unreferenced tag myself, it is normal practice that any tags added should be kept unless they are inappropriate. For example, removing a POV tag if the person who added it is has not explained the POV problems is okay. However you should not remove an unreferenced tag (Template:Unreferenced) if the article is clearly unreferenced as this one is. There appears to be some confusion about references on wikipedia by some editors. There is no such thing as an article which doesn't need references. I suggest editors who are still confused check out our policies such as Verifiability (which is linked to from the tag). Most of the claims on this article are simple factual claims which can easily be referenced and which anyone with a bit of knowledge of this area can confirm are true but this doesn't mean they don't have to be referenced. Nil Einne 08:07, 19 June 2007 (UTC) In closing, I will get a life - who cares which comes first and which is in brackets? (love, phobos2deimos) —Preceding unsigned comment added by 64.203.60.149 (talk) 21:34, 27 September 2008 (UTC)

It was referenced a while back, removed the references tag. -someone else-

Any upgrades avaible to me?
i have a 2500+ socket A. The support link says that it only goes up to 3000+ which is not true. i tried to look on newegg for socket A CPUs, but none came up. PLEASE HELP!!!! —Preceding unsigned comment added by 72.75.242.56 (talk • contribs)
 * This article and the athlon article both state 3200+ as the top end though you may need to check if your board supports the required FSB and if you use a mobile or otherwise unlocked chip the required multiplier setting. Afaict none of theese chips are in production anymore so you will probablly have to either buy secondhand or old-stock. If I were you i'd seriously consider a new pc or at least motherboard/ram/cpu combination. Plugwash 22:21, 28 June 2007 (UTC)

Athlon 3200+ is 2216 MHz, not 2333 MHz
I have corrected this error twice, yet someone keeps reverting it. The Athlon XP 3200+ runs at a stock speed of ~2216 MHz, NOT 2333 MHz. 400 MHz FSB x 5.5 = 2200 MHz +/- 20 MHz. Please keep it at the correct speed. Also to note, the Athlon XP 3200+ cannot run at 2333 MHz because how are you supposed to get an odd clock speed with an even bus speed? It doesn't make sense.Ggigabitem (talk) 22:07, 25 January 2008 (UTC)
 * The standard 3200+ is 2.2 GHz but there's a special OEM model with 166 MHz FSB and 2.33 GHz. --Denniss (talk) 11:08, 26 January 2008 (UTC)


 * The OEM version the 333 MHz version is so rare that it should be mentioned separately from the 400 MHz version as it would cause confusion. Ggigabitem (talk) 22:40, 26 January 2008 (UTC)

Die cracks?
"Indeed, they were so small that many users ended up with cracked processors while trying to remove or attach heatsinks for their fragile processors. This made installing non-standard or non-certified heatsink solutions a risky business. " Does anyone have an actual source of widespread issues with users cracking the dies/substrates when installing a heatsink? I put citation needed tag on this. — Preceding unsigned comment added by Kasm279 (talk • contribs) 10:30, 2 October 2013 (UTC)
 * Dies were not really fragile but they also lacked a heatspreader which could also act as protective cover. With standard heatsinks there was no problem with broken dies unless you made errors while mounting them. More problems originated from overly heavy heatsinks and/or heatsinks which are clamped or screwed-on with too much force. --Denniss (talk) 22:06, 2 October 2013 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 1 one external link on Socket A. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Corrected formatting/usage for http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/23792.pdf

When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at ).

Cheers.—cyberbot II  Talk to my owner :Online 10:01, 2 April 2016 (UTC)

Does it use FSB, or HT?
Chipset table confusingly states "FSB/HT". It's got to be either one or the other. I highly doubt it supports both, as the pinout/wiring of HT would be quite significantly different to that of FSB.

Were there any chipsets made for Socket A that actually support HT?

Otherwise I think it's best HT is removed so it's just "FSB" only. AP 499D25 (talk) 13:17, 10 November 2022 (UTC)