Intel sees excellent Q2 2018 results amidst news of 10-nm delays

We've reported on AMD's Q2 2018 results just yesterday and noted that things are looking rosy for the red team. Now the blue giant published its quarterly results, and the basic diagnosis is “that's a ton of money.” The company raked in $17 billion in revenue, up 15% year-on-year, and made $5.3 billion in operating profit, up 37% since the same quarter last year.

Net income got a 78% boost from Q2 2017, up to a massive $5 billion. Stockholders should be extra-happy with the earnings per share of $1.05, an 82% rise year-on-year. Intel's ever-important gross margin saw a very slight drop of 0.2% on the year to a still-impressive 61.4%.

It's not all good news, though. Intel let slip that its only expects system with 10-nm chips inside in the second half of 2019. After we last discussed the issue, that bit of news comes as no surprise, but it nevertheless bothered investors enough that stock prices took a tumble. After all, arch-rival AMD seems to be firing on all cylinders with its roadmap, while any next-generation Intel architectures are still waiting for the 10-nm node to yield well enough to ramp. Should AMD and other Intel rivals ship products on 7-nm-class processes from TSMC and GlobalFoundries, it would mark the first time in recent memory that Intel didn't enjoy a chip process advantage.

  Q2 2018 Q2 2017 Change
Revenue $17 billion $14.8 billion up 15%
Operating income $5.3 billion $3.8 billion up 37%
Net income $5.0 billion $2.8 billion up 78%
Gross margin 61.4% 61.6% down 0.2%
Earnings per share $1.05 $0.58 up 82%

According to Intel's report, every one of the company's major business segment saw revenue growth. In heartwarming news to PC enthusiasts in general, Intel's Client Computing Group brought in $8.7 billion, a 6% rise compared to Q2 2017. The company remarks that it saw demand for its “highest-performance products” in this segment and predictably mentions gaming as one of the underlying reasons for that growth.

The Data Center Group brought in $5.5 billion in revenue, or a whopping 27% year-on-year growth. The Internet of Things group is still a relatively minor player in the Intel mix with $880 million in revenue. However, that amount represents a historical record and a 22% rise since Q2 2017. Intel's non-volatile memory and Mobileye businesses likewise had record quarters. Income from the Non-Volatile Solutions Group rang in at $1.1 billion (up 23% year-on-year), while revenue from Mobileye went up by 37%.

As for future projections, Intel believes it'll bring in $18.1 billion in revenue in Q3 2018 (a 6% rise versus the current quarter), plus or minus $500 million. The company is counting on a 32.5% operating margin in that quarter and on earnings per share of $1.05.

Comments closed
    • jarder
    • 1 year ago

    Bit late to the party here, but having had time to digest these results I can see how wall street thinks that it is actually bad news.

    Because Intel is so heavily invested in it’s fabs and as we know it takes many billions to kit out a new fab, not having a new process node to move to is actually going to cause a massive rise in profits for intel, in the short term at least, and at the expense of it’s long term outlook.

    This had me thinking, at what point does intel abandon it’s 10nm process and move straight to it’s own 7nm process. Sure there would be some risk associated with such an unprecedented move, but it might be their best way to get that process node advantage back.

      • NoOne ButMe
      • 1 year ago

      who says 7nm is ready?

        • K-L-Waster
        • 1 year ago

        Over and above that, migrating a chip design to a different process isn’t simply a case of taking the existing trace layout and shrinking it. Even if the process itself is production ready, getting the chips that in the pipeline working on it would be non-trivial.

          • NoOne ButMe
          • 1 year ago

          yeah, although given the massive delays, I wish Intel had been willing to port Icelake (or at least cannonlake, even as a “tick”)to 14nm.

          Oh well. What is done is done.

        • jarder
        • 1 year ago

        I’m sure it’s not ready, yet, but they have surely been working on it in the background.

          • NoOne ButMe
          • 1 year ago

          they pulled engineers off of 7nm to try to fix 10nm… fixing 10nm isn’t going to well.

    • jihadjoe
    • 1 year ago

    Some things are made to last, but 14nm (++++) is forever.

    • Ushio01
    • 1 year ago

    How the hell does Intel make less net income than Facebook.

      • Alexko
      • 1 year ago

      Well, Intel does not sell a bajillion ads.

        • JustAnEngineer
        • 1 year ago

        As far as we know, Intel isn’t providing your personal information to foreign governments working to manipulate U.S. elections, either.

          • chuckula
          • 1 year ago

          You’re lucky that our total lack of CEO prevents us from taking you up on your offer!

          • blastdoor
          • 1 year ago

          Well, there’s the problem!

          But seriously… it is absurd.

      • K-L-Waster
      • 1 year ago

      People buy hardware once every 3-6 years.

      People visit FB and get hit by ads several times a day.

      • Sahrin
      • 1 year ago

      Spent it all on altera.

      Kidding aside, they make something physical and Facebook doesn’t.

      Just like Mocrosoft’s margins are way higher than Apple’s, because Apple makesphysical stuff, and Microsoft only (mostly) makes software.

    • Chrispy_
    • 1 year ago

    Squire: “Your excellency, the kingdom is on fire and the people are revolting against us”
    Intel: “It’s great to be king”

      • JustAnEngineer
      • 1 year ago

      “The people are revolting.”
      “You said it. They stink on ice!”
      [url<]https://www.youtube.com/watch?v=h0iAcQVIokg&t=00m12s[/url<]

    • elites2012
    • 1 year ago

    cant wait till the real report comes out.

      • K-L-Waster
      • 1 year ago

      Is that you, Charlie?

    • chuckula
    • 1 year ago

    It’s so sad seeing Intel exploit the mining market for a quick buck. Live up Q2 2018 with your non-diversified product portfolio while you can!

    In a few days I’m pretty sure we’ll all have [b<]32[/b<] reasons why there won't be a Q2 2019 earning report from Intel.

      • xeridea
      • 1 year ago

      I don’t see them getting rich from miners… The only thing they make that miners buy is 2 core Celerons for $40.

      They are getting far less dominant though, with Ryzen doing so well, and getting behind on fabrication, when not long ago AMD CPUs were subpar, and competing foundries were a full node behind.

      • Srsly_Bro
      • 1 year ago

      And 64 #Examples why Ice Lake is already a failure.

        • chuckula
        • 1 year ago

        I’m trying to decide if I want to call Ice Lake a failure or claim that it’s canceled outright.

        There’s just too many opportunities here.

          • Gadoran
          • 1 year ago

          Ice Lake is a good baseband for a fast 7nm shrink. So it is not a failure.

            • NoOne ButMe
            • 1 year ago

            is that in 2024, or 2025?

            Sorry, can’t resist…

    • ronch
    • 1 year ago

    I see Intel’s profits are on the ry.. ahh… WTH.

      • chuckula
      • 1 year ago

      You laugh, but Intels [b<]increase[/b<] in profits over Q2 of last year is $1.2 billion. AMD's total profits were $116 million. So if AMD literally increases its profits by [b<]1000 percent[/b<] in Q2 of next year, it will not quite have caught up with Intel's year-over-year increase in profits.

        • K-L-Waster
        • 1 year ago

        You missed a grand slam there, Chuck: Intel’s increase in profits exceeded AMD’s total revenue for the quarter….

          • chuckula
          • 1 year ago

          THIS IS THE 2018 PRO-AMD VERSION OF CHUCKULA HERE!

          I’m also not carping about the luxury weekend that AMD is throwing for most of the tech press pre-Ripper 2 either.

          Although I do reserve the right to use it against the usual suspects who will accuse literally every tech website of being in a pro-Intel conspiracy for the crime of accurately reporting the results of the 9900K vs. the 2700X in a couple of months.

        • ronch
        • 1 year ago

        Um, that was an attempt at sarcasm towards how everyone is saying AMD is on the Ryze or AMD sales are Ryzen.

    • derFunkenstein
    • 1 year ago

    [quote<]Stockholders should be extra-happy with the earnings per share of $1.05, an 82% rise year-on-year. [/quote<] Narrator: [url=https://www.google.com/search?q=INTC&rlz=1C1EJFA_enUS768US768&oq=INTC&aqs=chrome..69i57j0l5.936j1j7&sourceid=chrome&ie=UTF-8<]They weren't[/url<].

      • blastdoor
      • 1 year ago

      I seem to recall rhetorically asking how long it would take for Wall Street to comprehend that Intel is falling behind in process technology. I guess now we know!

        • chuckula
        • 1 year ago

        Considering the news for this quarter was actually better than for last quarter is shows that Wall Street is behind the curve really.

        The Q1 2018 report just said 10nm production was ramping up at some point in 2019, which could have meant that no products would actually be on sale in 2019.

          • uwsalt
          • 1 year ago

          But the company did leave more room for optimism in its comments on the 18’Q1 earnings call, as compared to Q2.

          [url<]https://seekingalpha.com/article/4166652-intel-intc-q1-2018-results-earnings-call-transcript?part=single[/url<] Then, Krzanich said the following in his opening remarks: [quote<]We continue to make progress on our 10-nanometer process. We are shipping in low volume and yields are improving, though the rate of improvement is slower than we anticipated. As a result, volume production is moving from the second half of 2018 into 2019.[/quote<] Further questions about the 10nm delays prompted this exchange: [quote<]Stacy Aaron Ragson – Sanford C. Bernstein & Co. LLC: So as the volume production pushes out into 2019, given you understand the yield issue supposedly, is this a first half pushout, or does it push out into the second half? And when it actually does ramp, do you think it actually will be the current 10-nanometer process that's shipping, or will that be slipping out to 10-nanometer plus potentially? Brian M. Krzanich - Intel Corp.: So I'm just going to correct you. You said that supposedly we have the solutions. We do understand these, and so we do have confidence that we can go and work these issues, Stacy. Right now, like I said, we are shipping. We're going to start that ramp as soon as we think the yields are in line. So I said 2019. We didn't say first or second half, but we'll do it as quickly as we can based on the yield. The last part of your question about whether will it be a 10 or 10-plus-plus or 10-plus I think was your question, the yield improvements that we're making are just that, more focused on yield.[/quote<] While there was some hedging in that response, the general sense in the market at the time, thanks in part to the repeated emphasis on already-shipping 10nm product, was that the delay in volume production on 10nm was more likely (or at least equally likely) to 2019’1H rather than 2H. That's a more than reasonable interpretation of volume production "moving [b<]from the second half of 2018 into 2019[/b<]." At worst, both possibilities were priced in. As a result, the comments on the 18’Q2 earnings call regarding 10nm were more certain – in a negative way – and therefore more jarring: [url<]https://seekingalpha.com/article/4190920-intel-intc-q2-2018-results-earnings-call-transcript?part=single[/url<] In his opening remarks, Swan said this: [quote<] we continue to make progress on 10-nanometer. Yields are improving consistent with the timeline we shared in April, and we expect systems on shelves for the 2019 holiday season.[/quote<] Naturally, that prompted questions for more clarity on timing, including this exchange: [quote<] Vivek Arya - Bank of America Merrill Lynch: Thanks for taking my question. Bob, on 10-nanometer progress, any color on what these are doing? And systems you mentioned on shelf for second half 2019, I assume that's mostly PC Client. Any sense of when we can think of timing for your server products on 10-nanometer also? Venkata S. M. Renduchintala - Intel Corp.: Hi, this is Murthy. I'll take that one. We continue to make progress on 10-nanometer. Yields are improving consistent with the timelines we shared in April. And yes, you're quite right. The systems on shelves that we expect in holiday 2019 will be client systems, with data center products to follow shortly after.[/quote<] To the extent this timeline was known internally at Intel at the end of 18’Q1, it wasn’t conveyed (clearly, if at all) in such terms at that time. So I don’t think it’s so much that the market was behind the curve, so much as it simply reflected the (optimistic) uncertainty. That uncertainty was reduced, in a negative direction, on the call, hence the drop in share price. I think there is probably some (justifiable) concern that this will translate into some delay on ramping the 7nm node as well. With that in mind, I’m hesitant to say that the stock is oversold on the news, but, all else being equal, I would be surprised if it didn’t regain some (but not all) of those losses over the next few weeks. Given the struggles with 10nm, the bigger picture question for Intel will be how quickly and successfully it is able to ramp volume production on 7nm. On the one hand, it appears that Intel made a mistake in deciding to delay adoption of EUV until 7nm. The additional production steps and complexity now required at 10nm are undoubtedly a significant factor in Intel’s difficulties at that node. On the other hand, Intel will be using EUV at 7nm. And 7nm is being developed partly in parallel with 10nm. While the difficulties at 10nm will negatively impact the timing and available resources for ramping 7nm to volume, Intel will have an opportunity to regain some of the time and ground it lost in getting to 10nm. Certainly, though, Intel is a ways off from its late-aughts mantra of “One Generation Ahead” in process technology. That’s a pretty big deal in terms of its position in the marketplace, including in the increasingly important (and competitive) enterprise and datacenter segments.

            • Gadoran
            • 1 year ago

            Are you sure the 10nm delay is a real issue?

            Right now Intel enjoy the performance and the power consumption of 10nm+ thanxs to 14nm++.
            The only disadvantage is die size but the yields are likely far higher than early TSMC 7nm process.

            In this moment the reduction of 30% in area is not enough to justify the process shift.

            Look at AMD, only high core count SKUs for server, nothing for consumer at the horizon…..TSMC is unable to give high clock speed.

            Rome will be a product of nice only.

            • blastdoor
            • 1 year ago

            You could just as easily be making the case for Intel to have stuck with 22nm, rather than moving to 14nm. When the 14nm process was first introduced it was only used for a small, low power Core M chip. Intel stuck with 22nm for a while for the larger, more power hungry chips.

            But you have to start somewhere. If you don’t, you never progress.

            • NoOne ButMe
            • 1 year ago

            What 30% shrink?

            Using the closest comparible part, Broadwell 2C+GT3 (48EU) is a 48% shrink (70/133). Call it a 45% as EU count is down to 40 in Cannonlake 2C+GT2.

            Even higher given that this is not discounting the fact that memory controllers and I/O tend to not shrink very much at all.

            It isn’t close to the 63% (1/2.7) Intel is talking about (which are not possible for a CPU), but it is far more substantial than 30%.

            • Gadoran
            • 1 year ago

            Ummm a better comparison is high power Haswell GT2 vs. Skylake GT2, it was a 31% shrink with features (Intel way to speak). Without features we are in the 35/40% range nothing more.
            Only with Airmont Intel was able to give a 0.62/0.63X shrink, but it was a low power part with high density libraries.
            Qualcomm say that TSMC 7nm shrink less than hoped so it is not unlikely even Intel will shrink less than expected in high power cpus. Unfortunately the RC issue in wires becomes an hard thing to face expecially on high power SKUs (but now even in phone SoCs).

            Anyway it is all a matter of costs and profit. Likely right now a 14nm++ SKU is more profitable than a 10nm+ one at the same performance and power consumption.

            I think customers care nothing of the process node if the performance is at he same level.

            • NoOne ButMe
            • 1 year ago

            The 30% that blastdoor referenced was from Skylake/Kaby Lake 2C+GT2 versus the Cannonlake 2C+GT2.

            Intel is getting a “mere” 2x or similar shrink from 2.7x. Not inconcievable yeah?

          • NoOne ButMe
          • 1 year ago

          The current fiscals were better.
          The long term outlook was worse, and from listening to the call, and history of blunders, it’s hard to get confidence in Intel’s 10nm projections.

          Uncertainty causes downwards pressure.

          Intel faces lots of uncertainty about it’s big cash cow. It still hasn’t turned it’s non-CPU “data-centric” product lines into profit centers.

            • Gadoran
            • 1 year ago

            Bet the big cash flow could be hurted more by a suboptimal process around??

            Intel was right to advertise, after all there is AMD and its Epyc server line.

            Still i think Intel is pretty happy, AMD nice succes has destroyed all the hopes about an ARM victory in server space. It is a win win situation. The second source is here, nobody cares seriously anymore about ARM on datacenters, this really matters in the long term.

            • blastdoor
            • 1 year ago

            Although it sounds strange on the face of it, I guess you’re right that Intel would rather lose business to AMD than lose it to ARM.

            I’m not sure I’d count ARM out altogether, though. So far the right company hasn’t come along to carry the ARM banner forward into the server space, but that doesn’t mean it won’t happen eventually. Maybe it happens vertically with a company like Amazon deciding to design their own processors for AWS.

      • techguy
      • 1 year ago

      That’s not how the markets work. Buy the rumor, sell the news.

        • derFunkenstein
        • 1 year ago

        I can’t not make an Arrested Development reference, but I’m aware of the old axiom.

Pin It on Pinterest

Share This