Slashdot Mirror

← Back to Stories (view on slashdot.org)

Intel Unveils New Coffee Lake 8th Gen Core Line-Up With First Core i9 Mobile CPU (hothardware.com)

Posted by msmash on from the good-timing dept.

MojoKid writes: Intel is announcing a big update to its processor families today, with new 8th Gen Coffee Lake-based Core chips for both mobile and desktop platforms. On the mobile side of the equation, the most interesting processors are no doubt Intel's new six-core Coffee Lake parts, starting with the Core i7-8750H. This processor comes with base/max single-core turbo boost clocks of 2.2GHz and 4.2GHz respectively, while the Core i7-8850H bumps those clocks to 2.6GHz and 4.3GHz respectively. Both processors have six cores (12 threads), a TDP of 45 watts and 9MB of shared Smart Cache.However, the new flagship processor is without question the Intel Core i9-8950HK, which is the first Core i9-branded mobile processor. It retains the 6/12 (core/thread) count of the lower-end parts, but features base and turbo clocks of 2.9GHz and 4.8GHz respectively. The chip also comes unlocked since it caters to gaming enthusiasts and bumps the amount of Smart Cache to 12MB. Intel is also announcing a number of lower powered Coffee Lake-U series chips for thin and light notebooks, some of which have on board Iris Plus integrated graphics with 128MB of on-chip eDRAM, along with some lower powered six-core and quad-core desktop chips that support the company's Optane memory in Intel's new 300 series chipset platform.

73 comments

  1. Meltdown and Spectre compatible? by Anonymous Coward · · Score: 5, Insightful

    I am curious if these chips will break compatibility with the previous Meltdown and Spectre data sharing apps. Have they made changes to this feature set?

    1. Re:Meltdown and Spectre compatible? by Anonymous Coward · · Score: 0

      Hmm, that is the question.

    2. Re:Meltdown and Spectre compatible? by tigersha · · Score: 3, Funny

      The coffee lake was caused by a Meltdown of a valve in the Santa Clara Starbucks, so no, I suspect not.

      --
      The dangers of excessive individualism are nothing compared to the oppressiveness of excessive collectivism
  2. Have they fixed Meltdown and Spectre? by Anonymous Coward · · Score: 4, Interesting

    So do these new chips have Meltdown and Spectre hardware fixes?

    1. Re:Have they fixed Meltdown and Spectre? by Anonymous Coward · · Score: 5, Informative

      No. Only workarounds in microcode that reduce performance.

    2. Re:Have they fixed Meltdown and Spectre? by 93+Escort+Wagon · · Score: 5, Funny

      No. Only workarounds in microcode that reduce performance.

      They also contain Intel's patented Deflect-Towards-AMD technology.

      --
      #DeleteChrome
    3. Re:Have they fixed Meltdown and Spectre? by Billly+Gates · · Score: 4, Informative

      No. Only workarounds in microcode that reduce performance.

      Neither. Branch prediction is so Central to the CPU architecture that it can't be disabled. MS is working on it's compiler to see if can do special assembly tricks to hide the cache.

      Linux has kernel hacks which attempt to hide the data from the cache which hackers can still overide with skill

      --
      http://saveie6.com/
    4. Re:Have they fixed Meltdown and Spectre? by RoccamOccam · · Score: 1

      My understanding was that Intel's version of branch-prediction applies an access-rights check too late (allowing Meltdown snooping). That doesn't sound like it couldn't be fixed, especially since AMD doesn't have that same problem.

    5. Re:Have they fixed Meltdown and Spectre? by F.Ultra · · Score: 1

      It can be fixed with a new architecture but not with a new microcode (microcode can only change so much).

    6. Re:Have they fixed Meltdown and Spectre? by Anonymous Coward · · Score: 0

      AMD's implementation is still vulnerable to Spectre. I don't think there is anything that can be done about it short of sending CPUs back to the 486 days.

    7. Re:Have they fixed Meltdown and Spectre? by Anonymous Coward · · Score: 0

      It can be fixed with a new architecture

      Without a performance penalty?

    8. Re:Have they fixed Meltdown and Spectre? by hcs_$reboot · · Score: 1

      Anyway, Facebook leaks data on our behalf, so why bother.

      --
      Slashdot, fix the reply notifications... You won't get away with it...
    9. Re:Have they fixed Meltdown and Spectre? by TheRaven64 · · Score: 2

      Meltdown allows you to use timing attacks to snoop data across a system call. This is because Intel used an optimisation where they treated system calls as branches, whereas on AMD chips they resulted in a pipeline stall. The Spectre vulnerabilities work at the same hardware privilege level, though not necessarily at the same software privilege level (for example, you can read past a bounds check in a NaCl or JavaScript sandbox and read memory outside of the sandbox, which gives you the memory disclosure vulnerability that you need to launch a code reuse attack).

      The defence against Meltdown is to unmap kernel memory when in userspace. This means that the CPU would have to speculate past the CR3 update (switch page tables to the userspace mappings) to be vulnerable. Current CPUs don't do that, because it's really hard to do: You need to be able to invalidate TLB fills, because the page tables that you've installed might be wrong. TLB fills as a result of normal speculative execution are fine, because the TLB is always an arbitrary subset of the contents of the page tables, so you don't need to invalidate them. Sometimes this can even give a big performance boost. Apple found a few years ago that they were getting a big speedup because a mispredicted branch was prefetching some data into the cache that they were using later. The mispredicted branch was much cheaper than stalling for the cache fill. In hindsight, I should have realised that Spectre-like attacks were possible when I learned about this.

      The defence against the Spectre variant 1 attack is to add a data dependency where previously there was a control dependency. For example, if you have some code that looks roughly like this pseudocode:

      if (bounds_check(address, offset))
      {
      load(address + offset);
      }

      You turn it into something like:

      check = bounds_check(address, offset);
      if (check)
      {
      load(select(check, address + offset, 0));
      }

      Where the select becomes a conditional move instruction (or some equivalent arithmetic operation). This transformation means that the load now has a data dependency on the result of the bounds check and so won't be dispatched until the bounds check has been calculated. This, in turn, means that there won't be any observable side effects of the load if the branch would not be taken because the instructions inside the conditional will be canceled as soon as the branch is determined to be not taken. This probably has a small performance overhead, because it will introduce pipeline bubbles. I'd be surprised if it were more than 5% though.

      Variant 2 involves poisoning the branch target buffer so that at a specific point in execution the CPU will predict a jump to attacker-controlled code. You can then put timing sensitive instructions at that point and probe register values. The mitigation for this is called a retpoline, where you perform an indirect branch using a return instruction, which then uses the return buffer for prediction and so will predict the address after the last call. This basically forces a branch mispredict, but to a location that isn't controlled by the attacker.

      Some of the proposed hardware fixes involve not sharing branch predictor state across security contexts. This is not ideal, because often that sharing is beneficial. For example, if you an Android app, it's forked from a zyogte process that sets up the VM and pre-loads a bunch of classes. All apps will have the same core system code in the same addresses and can benefit from sharing branch predictor state. Similarly, if you run a server in a pre-fork model.

      --
      I am TheRaven on Soylent News
    10. Re:Have they fixed Meltdown and Spectre? by F.Ultra · · Score: 1

      Without the extreme penalty that there is now with the software solution but they will loose that slight performance benefit that they got over say AMD by cheating with the MMU check. For a context switch heavy application you pay around 30% penalty if the benchmarks are anything to go by, with a new architecture I'm sure that the penalty is below 0.01%

  3. New CPUs come in so they refuse to fix Spectre by Anonymous Coward · · Score: 5, Informative

    In the meantime they have posted that there will be *NO* microcode updates for the older generations as stated in https://newsroom.intel.com/wp-content/uploads/sites/11/2018/04/microcode-update-guidance.pdf (serach for "stopped").

    There are software workarounds, but well... still leaves a bad taste considering they originally wanted to develop new microcode for those generations.

    1. Re:New CPUs come in so they refuse to fix Spectre by Anonymous Coward · · Score: 0

      There are software workarounds, but well... still leaves a bad taste considering they originally wanted to develop new microcode for those generations.

      Do you know for a fact it architecturally possible to implement the workaround for Spectre v2 on those processors via a microcode fix? Many of the STOPPED group are older CPUs (while I do know of a few old Core [2] (Solo/Duo) systems still around, I can't really say I care much about them as they were released ten years ago), embedded devices (which, due to their specific target market, are as likely as not never to see any update for anything anyway), or server specific CPUs (that would have been replaced at least twice by now).

    2. Re:New CPUs come in so they refuse to fix Spectre by Anonymous Coward · · Score: 0

      You (and Intel, in case you don't work for them) should care more, because these older CPUs are everywhere - in companies, hospitals, government institutions, etc.

    3. Re:New CPUs come in so they refuse to fix Spectre by thegarbz · · Score: 2

      Given their previous microcode rollouts I think we can all collectively sigh with relief.

    4. Re:New CPUs come in so they refuse to fix Spectre by Anonymous Coward · · Score: 0

      ...(while I do know of a few old Core [2] (Solo/Duo) systems still around, I can't really say I care much about them as they were released ten years ago)...

      So, you are a short-sighted dullard. Congrats.

    5. Re:New CPUs come in so they refuse to fix Spectre by Anonymous Coward · · Score: 0

      You do realize many companies have a dual depreciation cycle for PCs and other IT equipment, right? They use 3 years for tax purposes and 5-8 for the IT equipment replacement budget.

  4. Will this thing have the AMT hole? by Anonymous Coward · · Score: 0

    In case the subject is lost. One question: Will this thing have the AMT hole? Nothing more.

  5. Obligatory silly name joke by o'reor · · Score: 1

    So, at long last, does this 9th generation Intel Core processor make coffee, as suggested by its name ?

    Come on now, we developers have been waiting ages for this essential feature !

    #CoffeeLake

    --
    In Soviet Russia, our new overlords are belong to all your base.
    1. Re:Obligatory silly name joke by chmod+a+x+mojo · · Score: 2

      It's been out for decades! Just put your perc pot over the molten hole where your Prescott processor used to be before it melted through the motherboard, floor, and planets core. You should have a fresh pot in no time at all.

      --
      To err is human; effective mayhem requires the root password!
    2. Re: Obligatory silly name joke by aliquis · · Score: 1

      It's the first Intel chip which refuses to run Java?

  6. Why bother? by Anonymous Coward · · Score: 2

    It's all window dressing. This is about as exciting as the difference between a 2017 Hyundai Elantra and and 2018 Hyundai Elantra. We are approaching almost a decade since Intel offered anything significantly different or improved.

    1. Re:Why bother? by Anonymous Coward · · Score: 0

      Hey! They added a new feature called Intel Thermal Velocity Boost that allows them to gain 200MHz for .5ns.

      They are still dirty liars about TDP since it's specified at base clocks. So having a CPU with 2.9GHz that can "boost" to 4.8GHz for a nanosecond doesn't give you much unless you like laptops bursting in flame.

    2. Re:Why bother? by 110010001000 · · Score: 1, Insightful

      That is because Moore's Law has been dead for sometime now, because physics. Digital computing is reaching a dead end now. Apple is smart to start creating their new chips because Intel has hit a ceiling with nowhere to go.

    3. Re:Why bother? by Anonymous Coward · · Score: 0

      Hey, maybe THIS will be the chip that TSX actually doesn't get nuked in a microcode patch. TSX is like the most hilarious vaporware, having been launched in every generation since Haswell, but always having something that needs some post-launch update, which inevitably disables it.

    4. Re:Why bother? by sinij · · Score: 2

      Digital computing is reaching a dead end now.

      Not until Netcraft confirms it.

    5. Re:Why bother? by Anonymous Coward · · Score: 0

      Well, they can add more cores, but Windows maxes out at 12 cores, so why bother?

    6. Re: Why bother? by Anonymous Coward · · Score: 0

      Bullshit

    7. Re:Why bother? by TFlan91 · · Score: 1

      To extend your analogy, I would certainly purchase a 2018 Hyundai Elantra over my current 2008 model. But I've heard there were some safety risks with the newer ones, so I'll go with the Tesla (AMD).

    8. Re:Why bother? by Anonymous Coward · · Score: 0

      Moore's Law does say you get better performance, just cheaper and more densely packed transistors. It's still going strong, physics and CPU architectures just mean we're getting rapidly diminishing returns on improving single-threaded performance with more transistors in a small space. The $/transistor is still going down which mean $/performance is going down which is also interesting.

    9. Re:Why bother? by Anonymous Coward · · Score: 0

      As Intel has kicked out large part of their CPU engineering teams and replaced their tick-tock model with tick-sleep-tick one, what else can you expect? This is of course a good thing for the competitors in the long run and in a short term for Intel executives too, as they can swim in quarterly bonuses.

    10. Re:Why bother? by TheRaven64 · · Score: 1

      Moore's Law gets you better performance because more transistors give you better performance. In particular, Dennard Scaling meant that the smaller transistors used less power so you could have more specialised pipelines, more complex specialised instructions, and so on. Unfortunately, Dennard Scaling ran out about a decade ago, so although Moore's Law has given us more transistors, the number that you can power at any given time has stayed almost constant.

      --
      I am TheRaven on Soylent News
    11. Re: Why bother? by Anonymous Coward · · Score: 0

      There's no reason they couldn't at least start layering more cores on and lower power a bit to control cooling. Where's my 20 core / 40 thread cpu?

    12. Re:Why bother? by Billly+Gates · · Score: 1

      Funny I swear I had a 56 core Xeon running Windows server in our MDF

      --
      http://saveie6.com/
    13. Re: Why bother? by Anonymous Coward · · Score: 0

      Not bullshit, you have to get a special license when you have more than 12 cores on modern Windows. It is not a technical limitation but a licensing one. Windows 10 Pro for Workstations

    14. Re: Why bother? by aliquis · · Score: 1

      Graphics cards has done fine.
      Ryzen did too.

      The problem is lack of competition.
      See what Nvidia do now as far as new models go.

    15. Re: Why bother? by Anonymous Coward · · Score: 0

      Your link doesn't support your claim.

    16. Re: Why bother? by Anonymous Coward · · Score: 0

      There: https://secure.raptorcs.com/content/CP9M06/purchase.html
      Ok, it's preorder, but only for a matter of weeks, and it dissipates a lot of Power. But a Power9 with 18 cores and 72 threads is reasonably powerful, and it has 90MB of L3 cache.
      There is also a 22 core/88 threads on the same website, with 110MB of L3 cache, but probably not worth it at almost twice the price.
      None are cheap, but the amount of L3 cache makes Intel's ones look tiny.
      The 8 core/32 threads is available at a reasonable cost, and it has 80MB of L3 cache (the 4/8 cores have 10MB/core, the larger models have 10MB/core pair).

  7. Available at Starbucks! by Ronin+Developer · · Score: 0

    Thought it said Intel Unveils Coffee Latte....

  8. For 3D, CAD And DCC Users This Is Great News by dryriver · · Score: 1, Insightful

    Laptops for content creators and CAD designers have been stuck in 4-Core, 3.3 - 3.7 GHz land for many, many years. 6 cores, 4.8 GHz max and also Optane memory will definitely make a difference in this area. If your last Core i7 laptop was bought 2 - 3 years ago, a new Core i9 laptop should feel much faster in comparison. I would have loved 8-cores instead of 6, but maybe that's coming next.

    --
    Why did the chicken cross the road? Because Elon Musk put an AI chip in its head.
    1. Re:For 3D, CAD And DCC Users This Is Great News by Anonymous Coward · · Score: 0

      Don't think for a second that Intel actually advanced anything in a meaningful way for laptops. It's all PR.

      Those chips might have more cores and higher possible turbo buckets but their TDP is still specified at the base clocks. If you lack powerful cooling the CPU will not hold the higher frequencies for long, especially if you use AVX/AVX2. That 4.8GHz is single-core and not for very long.

    2. Re:For 3D, CAD And DCC Users This Is Great News by Khyber · · Score: 1

      "Laptops for content creators and CAD designers have been stuck in 4-Core, 3.3 - 3.7 GHz land for many, many years."

      That's because CAD has historically been gimped by GPU performance, not CPU performance. Please try your pondering again when you have actually used CAD machines for over 20 years!

      --
      Still waiting on Serviscope_minor to wake up to fucking reality and realize that Jessica Price isn't going to fuck him.
    3. Re:For 3D, CAD And DCC Users This Is Great News by TheRaven64 · · Score: 1

      I expect the 64-128MB of eDDR as a L4 cache will have more of an impact. For FPGA place and route, we've found that the desktop versions with this are about 50% faster than anything else on the market (performance is dominated by the sequential bits of the algorithm and the working set doesn't fit in cache).

      --
      I am TheRaven on Soylent News
  9. What's the point? by Anonymous Coward · · Score: 0

    AMD Ryzen 2 will be better. Heck, Ryzen 1 is better still.

  10. Where the market is by AlanObject · · Score: 1

    We are approaching almost a decade since Intel offered anything significantly different or improved.

    One of the hardest lessons for hardware tech entrepreneurs to learn is that IT and personal-computing buyers, the majority of business for Intel, is not really interested in anything "different." What they want is what they bought last year faster and cheaper and 101% software compatible.

  11. I wonder... by WolfgangVL · · Score: 2

    Every time I see the word "unveil", I first wonder, how long was it veiled to begin with? Then I wonder, how many more iterations are still "veiled", awaiting the perfect time to "unveil", so as to maximize profits?

    I imagine a bunch of marble pedestals with thin white sheets over them, each with red LCD displays counting down...... every now and then an alarm goes off and the room full of monkeys next door starts typing till they come up with a name......

    --
    You are being ripped off every second of every day, so that advertisers can help rip you off even more tomorrow.
  12. Meltdown and Spectre? by Anonymous Coward · · Score: 0

    What about Meltdown and Spectre? Has anyone asked about those yet?

    1. Re: Meltdown and Spectre? by Anonymous Coward · · Score: 0

      If you look at the cached comments above, you'll see your comment was successfully predicted...

  13. Just not in California by ScentCone · · Score: 1

    Doesn't California now consider anything with "Coffee" in the name to be a carcinogen?

    --
    Don't disappoint your bird dog. Go to the range.
  14. Too little too late. by Anonymous Coward · · Score: 0

    I'm switching to Apple desktops that will be using APPLE chips. So long Intel, go suck an egg.

  15. We'd get these in a 5 yrs by postmortem · · Score: 1

    if AMD didn't come up with its competitive lineup. Suddenly intel is all interested in making chips that are beyond regular evolution process of last 5-10 yrs.

    1. Re:We'd get these in a 5 yrs by thelexx · · Score: 1

      They seriously need to check their prices too. I just retired an I3 machine taking only the vid and p/s to the new one, and had a thought about keeping the old one going with a slightly upgraded cpu. Seeing I5's going for only slightly less than the Ryzen 1700X I just went with for my main, I changed my mind quickly. Utterly ridiculous.

      --
      "Gold still represents the ultimate form of payment in the world." - Alan Greenspan, 1999
    2. Re:We'd get these in a 5 yrs by Billly+Gates · · Score: 1

      They seriously need to check their prices too. I just retired an I3 machine taking only the vid and p/s to the new one, and had a thought about keeping the old one going with a slightly upgraded cpu. Seeing I5's going for only slightly less than the Ryzen 1700X I just went with for my main, I changed my mind quickly. Utterly ridiculous.

      Intel commands a premium with it's brand name. Especially among gamers where an i5 is faster than a Ryzen

      --
      http://saveie6.com/
  16. same pci lanes bottleneck by eaglesrule · · Score: 2

    16 pci-e lanes ought to be enough for everyone.

    1. Re:same pci lanes bottleneck by Anonymous Coward · · Score: 0

      There's Xeon for those in need of more...

  17. AVX-512 by Anonymous Coward · · Score: 0

    Coffee Lake shows some modest performance gains over Skylake, but I'm waiting for Cannon Lake, as it will include AVX-512 which should speed up some things quite a bit - in particular video encoding.

  18. Same as intels offering last year... by Anonymous Coward · · Score: 0

    Fully Minix compatible, with hardware VNC server included, and its like a whole 5% faster then.. last years Intel chips. Woo hoo!1

    This is a great product from a trustworthy manufacturer.

  19. 128 MB of VRAM? by Anonymous Coward · · Score: 0

    For INTEGRATED graphics that can't even run Solitaire?

    Has Windows gotten that bloated?

  20. Unlocking by Anonymous Coward · · Score: 0

    It should really be a crime for manufacturers to offer deliberately disabled hardware at a lower price point.

  21. Actually, for the Elantra GT it's significant... by Anonymous Coward · · Score: 0

    the 2018 is a re-design based on the European model.

  22. New line of defective processors by Anonymous Coward · · Score: 0

    This is so exciting. I can't wait to get my hands on these new processors without Meltdown fixed in hardware.

  23. intel needs to up the pci-e lanes / faster DMI by Joe_Dragon · · Score: 1

    intel needs to up the pci-e lanes / faster DMI.

    Most boards link to the CPU-X16 to slots but while M.2 / networking / usb / sound / etc is all on that X4 DMI link just one m.2 card can max out on it's own.

    1. Re: intel needs to up the pci-e lanes / faster DMI by Anonymous Coward · · Score: 0

      There are very few m2 cards that come close to maxing out the pcie x4 link. It's why so many cheaper cards are being released with a x2 link.

  24. That brings up a question: by Anonymous Coward · · Score: 0

    Is the heat issue across the entire chip or mostly just the core running? So if we got a chip with say 10 cores but only need 2 running to handle the active threads, could they turbo boost 2 of the 10 until they get hot, then bounce to 2 cooler ones, etc? Maybe this is how the current speed boost already works???

    1. Re:That brings up a question: by Anonymous Coward · · Score: 0

      Temperature across a chip is quite uniform. Maybe you should put the turbo at the two chips farther apart, but heating one core will heat its neighbour, and at 10 cores or, physical distances between cores are rather small.

  25. How about adding a real feature. by Anonymous Coward · · Score: 0

    Like allowing customers the ability to disable the built in spy system/management engine?

  26. Intel's Most Jittery Processor Yet! by Hallux-F-Sinister · · Score: 1

    Intel Unveils New Coffee Lake 8th Gen Core Line-Up With First Core i9 Mobile CPU

    I wonder how jittery a processor designed with a LAKE of coffee will be. Bet it makes the computer really good at taking a SHIT though, amiright?!?

    --
    Our reign has gone on long enough. Indeed. Summon the meteors.
  27. We WILL get these in a 5 yrs. ;) by Anonymous Coward · · Score: 0

    That's probably when you can actually buy them and hold them in your hands.

  28. Unveil and unload... by Anonymous Coward · · Score: 0

    Intel might have unveiled a new chip, but they also unloaded Wind River Systems...

    https://venturebeat.com/2018/04/03/intel-to-spin-out-wind-river-software-division-to-tpg/

  29. August 04, 2017,.... by AbRASiON · · Score: 1

    https://hardware.slashdot.org/...

    "There's a "z390" (?) is a cannonlake chipset or "PCH" - and it's coming out next year - but that chipset is only for cannonlake processors, except there are (apparently) none of those planned for desktop."

    No sign of this chipset, no sign of an 8 core processor.
    Note this: https://www.intel.com/content/...

    The new chipsets do include the rumoured 'free' "Intel® Wireless-AC MAC"
    As well as a newer USB revision, 3.1 vs 3.0

    So what you end up with is the 'premium' z370 chipset, now a 'gimped' product, lacking several of the new features (I'd also heard Bluetooth 5 on the H370 also?)

    A real mess Intel has got itself into with product launch delays, push forwards, branding changes, just totally sloppy. One would normally assume you get the be-all and end all solution at the top end, sadly not the case.
    (I will say the 6 core CPUs from late last year are a good move forward, if you're wanting to hang on though, in another 12 months we might see 8 core with that free Wifi AC and USB 3.1 support)

    Oh and when is 2.5 and 5 GBit going to become common? I could really do with this soon. Knowing Intel? Not for a while.