Jump to content

Talk:Apple A4

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Explanation for the Cablevision anon-IP

[edit]

If you read the reference article at the EE times [1], you will find that you would have been correct to say that the Intrinsity (now fully subsumed into Apple corporation) designed CPU core has been found to be identical to the core in Samsung S5PC110A01 chip. But the rest of the components on the SoC die are different in placement and functions. To repeat the actual die is different, not just the packaging which includes the RAM chips -- so it is correct to say its a new design by Apple. Moreover this fact is covered about two paragraphs down with "The same Cortex-A8 CPU core used in the A4 is also used in Samsung's S5PC110A01 SoC.". Feel free to word it better. -- KelleyCook (talk) 18:50, 7 July 2010 (UTC)[reply]

You requested the original research, but in fact I want to read it from you. Apple a4 is OEM samsung chip without any differences. Apple don't hold an arm license so, the word "designed" is at least inappropriate. Apple company bought Intrinsity with a portfolio after this chip first release. EE times article doesn't state specific changes.
Moreover EE times gives not s5pc110 image, but s5pc100 or s5pv210 without 540 power vr graphic IP and name it as s5pc110 for some unknown reason.
The analysis of the A4 gives that the core is the same, but the chip is not.[2][3] There are differences between A4 and Samsung's chip, and there is zero evidence that these changes are not by Apple's design. Further: Apple might have handed a detailed specification of what they wanted to Samsung and Apple can very well have several employed engineers at Samsung directing the design effort. In both cases the result would Apple's design but constructed and manufactured by Samsung. No one in the industry is contesting Apple's claim that they designed the A4, and your claim and original research does not override uncontested trusted sources. -- Henriok (talk) 08:55, 8 July 2010 (UTC)[reply]
Yes I see. But I'm saying that these s5pc110 is not a core itself, Samsung's customized ARM cortex A8 is a core. For me it obviously appears that the chip that was preparated at www.ubmtechinsights.com is not s5pc110, but s5pc100 or s5pv210. And anyway, chips are not designed, neither customized overnight. For other arguments, please see my comments in PoP section
What YOU are saying is completely irrelevant to Wikipedia. What you say is original research and won't fly. You state that A4 is identical in every respect to S5PC110 when there is ample evidence that this is not the case. I find it EXTREMELY hard to believe that UMB TechInsigts would have made this analysis using the wrong chip. Either way Wikipedia can under no circumstance take your or my word before the word of UMB TechInsigts. -- Henriok (talk) 11:15, 9 July 2010 (UTC)[reply]

Apple also hadn't any chip IP prior intrinsity acquisition. And probably still don't have Arm V7 license http://www.arm.com/products/processors/licensees.php . This A4 is a stock oem chip, please change word designed to something different. Apple is not a chip making company and microchips doesn't appear out of thin air. —Preceding unsigned comment added by 64.134.103.135 (talk) 21:12, 7 July 2010 (UTC)[reply]

You claim, without any source to back it up, that Apple probably doesn't have an ARM v7 license. Let me show you an article that say that they probably do have an ARM architectural license: [4] And this was in 2008, way before they bought Intrinsity. The fact that Apple does not show up on ARM's list of licensees proves nothing. Your claim is unsourced original research and won't fly on Wikipedia while citing EETimes would. -- Henriok (talk) 08:55, 8 July 2010 (UTC)[reply]
Please see my comments in PoP section.

PoP

[edit]

Pop is a feature of a board assembly and not a chip. The board on which this chip comes is not a "microchip" (comment on 21:08, 6 July 2010). Nothing prevents a manufacturer to simply die chips in simm manner instead stacking them.

You do not seem to know much about the technological background of package on package chip modules. Yes there IS a reason why this high-tech method to package the SoC and SDRAM together in one module is used, other than simply the size of the end-result. Its mostly done if the total number of interconnects make alternatives too unwieldy or impossible, even with BGA there can simply be too many interconnects needed. In this case the very wide bus width and bus-speed between the CPU and the SDRAM would make a more conventional approach unpractical or even impossible. And yes the consensus is that Apple (as they themselves have announced) are the designer of this PoP SoC. Unless you find reliable sources clearly stating otherwise claiming otherwise is not allowed on WP. Also Apple has a long standing relation with ARM, in fact they were crucial to ARM in their early period. Claiming "apple doesn't hold a license to ARM" without giving factual evidence of that statement is also simply WP:OR, if you cannot state reliable sources. Simply said, you cannot know that that is true, as there is at least one secret license holder which is widely thought to be Apple. Mahjongg (talk) 23:16, 7 July 2010 (UTC)[reply]
it seems that you don't know technical side of the issue. Yes, PoP reduces number of interconnections, but simply, any BGA microchip can be welded side by side. DIMM installation is nothing new. The only notable feature is that pinouts are aligned and preprogrammed for DDR chip.
You seem to not understand that also pushing out all the address lines, and specially the 64 bit high speed databus would simply become impractical, and even if it would be done that way, the resulting (much) longer interconnects would also negatively affect the bus-speed. Remember that the base die is already a SoC, not simply just a processor, that means that interconnects for all peripherals, and the LCD must also be provided. If this would have been attempted without PoP technology the resulting chip would need so many balls that it would become impractical, to route them all out, and the chip probably wouldn't even fit, in a product like the iPad, let alone an iPhone, especially if you also wanted to put normal DIMM modules next to it. Theoretically it could be done, practically not. Also I simply don't see the relevance of your claim, except to push a non nPOV. It certainly doesn't give you an excuse to remove the fact that the A4 is a PoP module from the article Mahjongg (talk) 23:30, 8 July 2010 (UTC)[reply]
I don't know by what kind of experts thou articles are written. When we refer to SoC, there are a complete assumption that we are referring to a microchip. And there are absolutely no problem to route out even an ennumerous amount pins graphic chips on video cards even with a few layers PCB, not to mention SoCs which are designed the way to reduce all unnecessary pins.
I don't know why I keep bothering to answer an anonymous poster who doesn't bother to sign hist posts, and has no argument worthy of removing the FACT that the A4 is a PoP SoC. If you have a reliable source that says that the A4 is NOT a PoP, then you have an argument, until then you have spawn enough nonsense already. Your only drive seems to be to "prove" Apple is not the designer of the A4, well you fail to do so, but that actually doesn't matter, what matters is that you fail to provide reliable sources that agree with your OR which is clearly not neutral. Mahjongg (talk) 15:04, 9 July 2010 (UTC)[reply]
I know that this is an old thread, but you have the burden of citation the wrong way around. If you believe that the A4 is a PoP SoC then the burden is on you to provide the reliable sources. (See WP:Burden). The anon user has correctly challenged your assertion because you have offered no references (though I will concede the point that he should have signed his posts). 86.183.25.124 (talk) 18:03, 4 April 2011 (UTC)[reply]

ARM licensees: http://www.arm.com/products/processors/licensees.php "Apple has a long standing relation with ARM, in fact they were crucial to ARM in their early period." ??? Only as I know, once back in 198x they were rivals. As both were trying to make cheap PC alternative. Also, even if apple had arm license long time ago, they had to relicense it for V7 architecture, and specifically for V7a. And simply apple is not a microchip engineering company anymore. —Preceding unsigned comment added by 24.47.174.169 (talk) 08:38, 8 July 2010 (UTC)[reply]

Apple was a co-founder or ARM Holdings, together with Acorn Computers and VLSI Technology. Apple remained a large share holder until the late 1990s when Apple sold off ARM stocks in increments to keep business running. That, however, doesn't make them an ARM licensee. In that you are right but Apple probably has a broad ARM architecture license[5], and knowing Apple that wouldn't necessarily be something that they want ARM to put in their marketing. Or Samsung for that matter. Not showing up on ARM's list proves absolutely nothing. -- Henriok (talk) 10:38, 8 July 2010 (UTC)[reply]
Don't know what you are meaning, but as I know there are no such thing as a "broad ARM architecture license". Every time ARM comes to a new solution they change everything, including licensing. Even though, we are not talking about the core itself, but SoC as a whole. Acquiring a whole IP package wouldn't have ended with one acquisition of a small core customization specialized engineering company. Samsung like have hundredth of licensed IP pieces per chip. And the article you pointed out gives a P.A. Semi and not recently acquired Intrinsity; P.A. Semi design PPC chips and SoCs, and not ARM basedd chips. Simply, what you have to know is that Apple is not a chip engineering company, they license chips and there are absolutely no way any company can come with a chip overnight.
I'm no expert in ARM licensing but I guess ARM is and quoting from their Financial Report issued in July 2008: "Architecture license signed for current and future ARM technology with strategic OEM" and Cnet who listened to the earnings call used the words "signed an architectural license with the company, forming ARM's most far-reaching license for its processor cores". All this is in my referenced link. I haven't made anything up. If this un-named OEM was Apple, they have had _at_least_ two years for development of the A4, and that's far longer than your "overnight" suggestion. As far as Apple being a chip design company, they certainly were from the 1970s up to 2006 and by buying PA Semi in 2008, they are again. The two years in between they might or might not have kept their chip design competence, we don't know. My guess is that they kept some based on the ample customizations of their products even in they do use standard parts to a very large degree. -- Henriok (talk) 10:59, 9 July 2010 (UTC)[reply]

The anonymous poster

[edit]

Why do we still argue with him, he doesn't even know the difference between Acorn and ARM, and even bases his expert status on his intimate knowledge of them being the same. Lars T. (talk) 14:31, 9 July 2010 (UTC)[reply]

This is really besides the point. Wikipedia policy demands that he provide WP:reliable sources. Until this is done, the contributions can't be put in the article. -- KelleyCook (talk) 14:49, 9 July 2010 (UTC)[reply]
[edit]

Hello fellow Wikipedians,

I have just modified one external link on Apple A4. 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:

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

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

  • If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
  • If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 14:59, 16 October 2016 (UTC)[reply]

[edit]

Hello fellow Wikipedians,

I have just modified 2 external links on Apple A4. 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:

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

  • If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
  • If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 03:01, 8 July 2017 (UTC)[reply]