Chipgate: Did you get the good A9 or the crap A9 in your iPhone 6s/Plus?

“With the release of the iPhone 6S, it has come to light that Apple is splitting the manufacture of its A9 processor between Samsung, its hated rival and at the same time business partner, and TSMC, the leading chip manufacturing company in Asia,” Andy Patrizio reports for ITWorld. “The split is said to be 60/40 in favor of TSMC.”

“As it turns out, the chips from the two firms are not identical. The site Chipworks, which specializes in teardowns and reverse engineering devices, notes that Samsung uses 14nm while TSMC uses 16nm,” Patrizio reports. “The TSMC chip is just a bit larger – 104.5 square millimeters vs. 96 square millimeters for the Samsung chip – but it shouldn’t be much of an issue because it’s still the same A9 processor design from Apple. If anything, it should favor Samsung because it has the smaller chip, right?”

“Apparently not,” Patrizio reports. “One Reddit user posted tests of a pair of 6S Plus phones and found the TSMC chip had eight hours of battery life vs. six hours for the Samsung… A second set of tests showed the phone with a TSMC chip is ever so slightly faster than the Samsung chip in the benchmark tests and the Samsung chip runs a few degrees hotter.”

MacDailyNews Take: Apparently Samsung’s chip stamping operation takes after their phones.

Read more in the full article here.

“There’s no way to tell if you’re getting stuck with a Samsung A9 processor when you purchase your iPhone 6s in stores or online, but you can find out whether you got saddled with a crappier processor with some help from a couple of apps,” Buster Hein reports for Cult of Mac.

“Free iOS app Lirum Device Info Lite is the easiest way to find your A9 processor info,” Hein reports. “If you have a Mac, you can check your device info using the iMazing app.”

Read more in the full article here.

MacDailyNews Take: Looks like Samsung didn’t steal all of TSMC’s trade secrets, after all.

Thank Jobs in heaven, all of our iPhone 6s Plus units have TSMC A9 processors inside! Unfortunately, if these benchmark reports are true, some Apple customers are going to feel cheated over this. Seriously: Markedly less battery life, runs hotter, and it’s a total crap shoot whether you’re buying one or not? Seriously, Apple? Who at Apple thought selling two differently stamped processors as the “A9” was a good idea and do they still draw a paycheck? As shareholders, we’d like to know the answers. If these processors are really not that different, then Apple needs to issue a statement, with proof, ASAP. If this is FUD, nip it in the bud (for a change).

If this isn’t FUD or a mistake by the testers, then this is like Ford selling a “Mustang” with a welded hood. Some of them come with 5.2L V8 engines and some of them come with 3.7L V6 engines, but Ford just calls them both “Mustang Engines” and you don’t know which one you’ve got until you can get it to the mechanic to pry open the bonnet.

The iPhone 6s/Plus with a TSMC-stamped A9 processor is simply more valuable than iPhone 6s/Plus units stuck with crappy Samsung-stamped A9 chips. We’re not sure what can be done about that at this point, with millions sold (recall?), but we bet some lawyers are drafting up class action lawsuits as you read this. And, you know what? If these tests can be widely replicated, they’re probably right to be doing so.

Now, who’s stamping out the iPad Air, iPad mini, and iPad Pro chips, Apple? We’d like to know BEFORE we shell out for them. In fact, we’d like to run Lirum Device Info on that iOS device in the Apple Store, so we know what we’re getting. We despise supporting thieving Samsung in any fashion, anyway.

Apple Return Policy: For any product in its original, undamaged and unmarked condition, simply return it with its included accessories and packaging along with the original receipt (or gift receipt) within 14 days of the date you receive the product, and Apple will exchange it. We advise doing so, over and over, if need be, until you get the best processor in your iPhone 6s or iPhone 6s Plus. Buy it. Open it. Install Lirum Device Info. Run it. If you have a TSMC phone, you’re good to go. If not, return the inferior iPhone and repeat the process. Customers shouldn’t have to do this, Apple. Customers should be satisfied they have the best of the best, Apple. This doesn’t seem like a satisfactory buying experience, Apple.

Again, if these processors are really not that different, then Apple needs to issue a statement, with proof, ASAP.

Bottom line: Samsung makes crap and, if these processors really prove to be so different, Apple really screwed the pooch this time.

UPDATE, 5:22pm EDT: In a statement to TechCrunch, Apple claims iPhone 6s/Plus’ A9 battery performance only varies 2-3% between TSMC and Samsung variants in “real-world usage.” So, if you’re shooting a time lapse video, a feature that Apple has promoted and offered for years, and your iPhone 6s with a Samsung-stamped A9 runs out of battery significantly faster than another iPhone 6s with a TSMC-stamped A9, that’s not of concern to Apple nor is it supposed to be of concern to the customer? That’s a specific case of “real-world usage” that is expressly offered and promoted by Apple, no less. Surely the lawyers will take note of that fact, too.

We’re not satisfied with Apple’s statement and Apple customers should not be either.

SEE ALSO:
Apple claims iPhone 6s/Plus’ A9 battery performance only varies 2-3% between TSMC and Samsung variants in “real-world usage – October 8, 2015
Samsung stole trade secrets from TSMC to win Apple A9 stamping deal – August 26, 2015
TSMC sues former ex-employee over leaking trade secrets to Samsung – February 9, 2015
TSMC overtakes Samsung in FinFET, confident they will land Apple A9 orders – October 7, 2014

[Thanks to MacDailyNews Reader “TheloniousMac” for the heads up.]

62 Comments

    1. http://www.macrumors.com/2015/10/07/tsmc-samsung-a9-battery-tests/

      Benchmarks have suggested there is little if any difference in chip performance between the two A9 variants, but some recent battery testing (via Engadget) is hinting at the possibility of significantly better battery life for models with the TSMC-manufactured A9 chip.

      We should caution that data points remain few at this time and controlling for variables to accurately focus the comparison only on the differences in the A9 is difficult, but these limited tests are generating significant amounts of interest in our discussion forums and elsewhere.

      Perhaps the most dramatic result comes from a reddit poster who compared Samsung and TSMC versions of the iPhone 6s Plus using the battery life test included in Geekbench 3, finding the TSMC version lasting nearly two hours longer than the model with Samsung A9 chip.

      1. I used the iMazing app for Mac that is linked in the article above. It downloads from the developers site. 15 day free trial. Works like a charm.
        Happy to see that my 6S is the TSMC variant.

        1. Yes this applies to 6s and 6s Plus. It’s the model of the CPU chip. The model number you listed is the model of your Apple phone which isn’t the right thing to look at.

          I had to download a free app called BMSSM to find my model of CPU. (The Lirium info lite app mentioned in the MDN post didn’t work for me)

        1. Try letting the battery drop to the point where the phone shuts itself down, wait about ten minutes and charge it up to 100%. That calibrates the the charging circuitry and does some resets that there are no other way of doing.

          Hope that helps.

  1. Message to all those who should not be allowed to write about, or offer opinions about tech (including MDN): The conclusions reached in ITWorld article stem from ONE (1) COMPARISON and exactly TWO (2) DATA POINTS. From this ridiculously small sampling, everyone is going to rush to judgment and, yes, to court.

    Ahh, life in America these days. Who needs critical or analytical processes, or real data, to reach a conclusion? If the story fits my ideology, or my preconceived notions, I believe it and run with it.

    As for me, I seriously doubt that Apple would use two processors whose performance was so disparate.

    1. By the way, MDN, Apple HAS issued a statement on this.

      Here is the reality of a device such as an iPhone: It is composed of an aggregation of components each of which has a production tolerance. Depending on how these variations line up on any given device, there will be small differences in overall performance.

      Apple’s own testing/sampling says that this overall variance is about 2% to 3% for the iPhone 6s. That sounds about right.

      Move along… there is nothing to see here.

      1. Yes, a wholly unsatisfactory statement, to which MacDailyNews responded:

        So, if you’re shooting a time lapse video, a feature that Apple has promoted and offered for years, and your iPhone 6s with a Samsung-stamped A9 runs out of battery significantly faster than another iPhone 6s with a TSMC-stamped A9, that’s not of concern to Apple nor is it supposed to be of concern to the customer? That’s a specific case of “real-world usage” via a feature that is offered and promoted by Apple, no less. Surely the lawyers will take note of that fact, too.

        We’re not satisfied with Apple’s statement and Apple customers should not be either.

    2. MDN is so full of crap on this one. Apple has to make tens of millions of these devices rapidly. You bitch when they don’t have enough to fill all of their orders immediately. You bitch when they use Samsung chips. You bitch when other people complain about Apple products. And now you bitch because every single iPhone is not perfectly identical. And you make that snap judgment based on a couple of rough data points, even though Apple claims only 2 or 3% difference. You friggin’ hypocrites. How many times in the past have you told other complainers to listen to Apple and mocked them for class action lawsuits. Now you appear to be trying to instigate one.

      What is wrong with you guys?

        1. I made the same leap and I’m blown away! I got a real kick out of the Geekbench 3 results, the 6s single core result was faster than the 5s multi core result! Compared to my 4s, more than 10 times faster on both single and multi!

          The 6s is going to be a tough act to follow.

  2. If they tested 10 of each, I might believe there was a real difference.

    At least in the article I clicked to, it was one of each. Unless it can be shown that phones with A9’s from TMSC do not vary that much, no conclusion can be drawn.

Leave a Reply to hkiphone09 Cancel reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.