Tests show iPhone XS LTE speeds best iPhone X, can't match Galaxy Note 9 or Pixel 2

Posted:
in iPhone edited October 2018
Though a handful of early iPhone XS adopters are complaining about less-than-desirable cellular reception, a new test reveals Apple's latest flagship smartphones boast much improved LTE performance over last year's model. However, the XS models, which rely on Intel modems, are still behind handsets that use battle-tested Qualcomm silicon.

iPhone XS LTE Performance


Conducted by PCMag in partnership with Cellular Insights, the evaluation pitted iPhone XS and XS Max against top industry performers including Samsung's Galaxy Note 9 and Google's Pixel 2. Last year's Intel-powered iPhone X, using the chipmaker's XMM7480 modem, was also included for reference.

For 2018, Apple opted to ditch Qualcomm in favor of Intel's next-generation XMM7560, an LTE modem capable of communicating with networks run by the four major U.S. wireless carriers. The latest iPhones are also the first to sport 4x4 MIMO antennas, boosting top-end speed and enhancing performance in areas where cell signals are weak.

Testing involved creating a 20MHz channel of Band 4, employed by AT&T, T-Mobile, Verizon and major Canadian carriers, with Rohde & Schwarz equipment capable of supporting 4x4 MIMO configurations. Cellular Insights attenuated an LTE signal from -85 dBm, a strong signal, and slowly decreased reference power until the phones ceased to download data.

Overall, iPhone XS Max performed nearly on par with the Galaxy Note 9 and Pixel 2, both of which sport Qualcomm modems in the X20 and X16, respectively. At -85 dBm, all three smartphones achieved nearly 400 Mbps download speeds. The XS exhibited a precipitous drop in performance at -86dBm, while the Pixel 2 and Note 9 saw nearly identical dips at -89 dBm and -90 dBm, respectively.

Apple's phone stopped receiving data at -128 dBm, while the Note 9 ceased reception at around -131 dBm. Interestingly, the Pixel 2's X16 modem continued to function beyond the measured range, though performance degraded to approximately 10 Mbps.

As can be expected from two nearly identical smartphones, Apple's XS and XS Max put in largely similar performances. The larger iPhone XS Max pulled ahead at certain signal levels, but the slight deviation would go unnoticed in a real-world scenario, the publication said.

iPhone XS Max iPhone XS LTE Performance


Apple's latest iPhone, while trailing a bit behind competitors using Qualcomm modems, is markedly improved compared to it predecessor. Lab testing saw the iPhone XS Max handily outperform the iPhone X at all signal levels, nearly doubling the 2017 iPhone under good conditions. The gains are primarily attributed to 4x4 MIMO antennas, though fine tuning of Intel's new XMM chip seemingly helped iPhone XS Max perform with relatively weak signals.

Real-world testing bore similar results, with iPhone XS Max seeing an average download speed increase of 6.6Mbps on all U.S. carriers, according to statistics provided by speedtest specialist Ookla. AT&T was most improved with an 8Mbps bump in performance. In Canada, overall download average deltas jumped to 20.2Mbps, with Telus being the network most greatly benefitted from the changes Apple made this year.

Comparatively, Samsung's Note 9 displayed average LTE download speeds of 43.2Mbps during the week of Sept. 24, while iPhone speeds averaged 38.9Mbps for the same period. Again, download figures massively improved in Canada, where the Note 9 averaged speeds of 97.7Mbps to XS Max's 85.4Mbps, the study found.

Finally, the publication notes a bump in iPhone X performance when upgrading from iOS 11 to iOS 12. Notably, download speeds on an iPhone X running iOS 12 neared performance levels seen by Qualcomm versions of the device running iOS 11.

The test results arrive roughly one week after early iPhone XS and XS Max customers began to complain of poor cellular and Wi-Fi reception. Today's findings -- lab tests in particular --
suggest Apple's hardware is not to blame, leaving hope that a firmware update will solve the sporadic issue.
«1

Comments

  • Reply 1 of 26
    claire1claire1 Posts: 510unconfirmed, member
    It's time Apple builds their own modems but wow looking at that graph you couldn't tell who's faster than who they're so close. 
    watto_cobra
  • Reply 2 of 26
    d_2d_2 Posts: 118member
    (for once) Intel should be commended for stepping up their game, as that’s quite a jump in performance from last year to this year

    also, it would be interesting to compare the power requirements of each company’s modem 
    Solibackstabgeorgie01MplsPjbdragonwatto_cobra
  • Reply 3 of 26
    SoliSoli Posts: 10,035member
    d_2 said:
    (for once) Intel should be commended for stepping up their game, as that’s quite a jump in performance from last year to this year

    also, it would be interesting to compare the power requirements of each company’s modem 
    I wish we could get a good read on that for the different bands and different distances from a tower. I'm not even sure that's feasible but I'd love for it to happen.
    watto_cobra
  • Reply 4 of 26
    The difference is indeed negligible.

    I stay stick with Intel as Intel can improve performance over time and even beat Qualcomm.
    I hope other phone manufacturers switch to Intel as well.

    tmaychabigwatto_cobra
  • Reply 5 of 26
    foggyhillfoggyhill Posts: 4,767member
    So, were the frack can you actually use this in an actual app or actual web site ?
    That's like saying my car can run 200 mph while the other one can only run 195 mph.... Who the hell cares. You're not going to go 200mph, or even 100mph ANYWHERE..
    edited October 2018 claire1rezwitscrossladtheodore007williamlondonStrangeDaysjbdragonhammeroftruthboltsfan17watto_cobra
  • Reply 6 of 26
    The difference is indeed negligible.

    I stay stick with Intel as Intel can improve performance over time and even beat Qualcomm.
    I hope other phone manufacturers switch to Intel as well.

    I agree. Especially with help from Apple itself.
    watto_cobra
  • Reply 7 of 26
    No issues with LTE on XS Max on Verizon, but definitely issue with Wifi performance. Phone seems to prefer 2.4ghz on an eero mesh network, even with network and phone restarts, resetting network settings, enabling band steering, etc. Will you be testing Wifi performance too? Do you think this is also a software issue? I’m running the iOS 12.1 beta with no improvements. 
    watto_cobra
  • Reply 8 of 26
    I am wondering why this was not discovered earlier by all the beta testers of IOS 12 with the iPhone X if there is indeed such a difference on weak LTE signals?
    williamlondon
  • Reply 9 of 26
    claire1claire1 Posts: 510unconfirmed, member
    backstab said:
    The difference is indeed negligible.

    I stay stick with Intel as Intel can improve performance over time and even beat Qualcomm.
    I hope other phone manufacturers switch to Intel as well.

    I agree. Especially with help from Apple itself.
    Looking at those performance jump numbers, they look very Apple like. 
    watto_cobra
  • Reply 10 of 26
    radarthekatradarthekat Posts: 3,842moderator
    “...can’t match...”

    Really?  That’s how you’d characterize that graph?  I’d say “...virtually matches...”   people read that headline and the colloquial interpretation is that Apple falls meaningfully short of Qualcomm’s offerings.  
    unbeliever2thtcornchipMplsPchiaGrimzahnchabigjbdragonmaltzwatto_cobra
  • Reply 11 of 26
    “...can’t match...”

    Really?  That’s how you’d characterize that graph?  I’d say “...virtually matches...”   people read that headline and the colloquial interpretation is that Apple falls meaningfully short of Qualcomm’s offerings.  
    Right? Especially when the text states "Overall, iPhone XS Max performed nearly on par with the Galaxy Note 9 and Pixel 2".
    watto_cobra
  • Reply 12 of 26
    flydogflydog Posts: 1,123member
    “...can’t match...”

    Really?  That’s how you’d characterize that graph?  I’d say “...virtually matches...”   people read that headline and the colloquial interpretation is that Apple falls meaningfully short of Qualcomm’s offerings.  
    Agree.  The headline of this article is extremely misleading, but "virtually matches" wouldn't have garnered as many clicks. 
    radarthekatchiachabigjbdragonwatto_cobra
  • Reply 13 of 26
    Soli said:
    d_2 said:
    (for once) Intel should be commended for stepping up their game, as that’s quite a jump in performance from last year to this year

    also, it would be interesting to compare the power requirements of each company’s modem 
    I wish we could get a good read on that for the different bands and different distances from a tower. I'm not even sure that's feasible but I'd love for it to happen.

    And this is the core issue right here. "Other tests".

    Remember waaaaay back when I said my daughters classmates father was an engineer at Apple? A cellular radio engineer.

    I know I posted this before as well, but it's worth bringing up again. Cellular Insights came out of nowhere around the time the infamous iPhone 7 LTE tests were done and showed the Intel version slower than the Qualcomm version. I asked him about this article and he made an interesting comment to me. To paraphrase again, he wondered how some no-name blogger was able to secure specialized cellular test equipment that's very expensive (well into 6 figures). He also commented on why they would use such equipment and then only conduct a single test out of the numerous tests it's capable of.

    Cellular Insights sure appears to be a Qualcomm funded shill. They don't really seem to do anything, and their articles are either praising Qualcomm or bashing Apple/Intel.
    tmaycornchipradarthekatRayz2016StrangeDayswatto_cobra
  • Reply 14 of 26
    foggyhill said:
    So, were the frack can you actually use this in an actual app or actual web site ?
    That's like saying my car can run 200 mph while the other one can only run 195 mph.... Who the hell cares. You're not going to go 200mph, or even 100mph ANYWHERE..
    That's why the Y axis goes from a maximum speed of 400mbps down to ZERO, while the X axis is based on signal.  Everyone can use this information.  If you compare the X to the XS, you can see that with a weak signal of -126dBm, you wouldn't get any data connectivity on the X, while on the XS you'd still be getting about 10mbps.
  • Reply 15 of 26
    radarthekatradarthekat Posts: 3,842moderator
    “...can’t match...”

    Really?  That’s how you’d characterize that graph?  I’d say “...virtually matches...”   people read that headline and the colloquial interpretation is that Apple falls meaningfully short of Qualcomm’s offerings.  
    Right? Especially when the text states "Overall, iPhone XS Max performed nearly on par with the Galaxy Note 9 and Pixel 2".
    You’ll note I was calling out a specific phrase, in the headline.  
    watto_cobra
  • Reply 16 of 26
    MplsPMplsP Posts: 3,925member
    “...can’t match...”

    Really?  That’s how you’d characterize that graph?  I’d say “...virtually matches...”   people read that headline and the colloquial interpretation is that Apple falls meaningfully short of Qualcomm’s offerings.  
    My thoughts exactly. I would say there is no meaningful difference between the Samsung, the Pixel and the XS. It also clearly shows that Intel is a viable option when it comes to cellular modems. Exactly what Qualcomm needs!
    chabigjbdragonwatto_cobra
  • Reply 17 of 26
    Yes, those graphs are close enough.
    chabigStrangeDaysjbdragonwatto_cobra
  • Reply 18 of 26
    BarBar Posts: 2unconfirmed, member
    The problem with this article is that it doesn’t address real world use. My iPhone XS max has faster download speeds than my iPhone X for sure. But the problem arises when signal strength isn’t 100%. My iPhone X definitely gets better reception indoors. Also what no one wants to address is why all the betas leading up to public release work fine and then all of a sudden the public release has issues. This has happened with iOS 11 and now iOS 12. 
    watto_cobra
  • Reply 19 of 26
    Right? Especially when the text states "Overall, iPhone XS Max performed nearly on par with the Galaxy Note 9 and Pixel 2".
    You’ll note I was calling out a specific phrase, in the headline.  
    Yes, I am agreeing with you that the headline is misleading and doesn't line up with the article content.
    StrangeDayswatto_cobra
  • Reply 20 of 26
    GeorgeBMacGeorgeBMac Posts: 11,421member
    I'm wondering how much of the difference between the X and the Xs is due to the improved chip and how much is due to the improved antenna?

    (I also wonder what is the meaning of life and other unanswerable questions)
    watto_cobra
Sign In or Register to comment.