Recent PCs have little to fear from Intel’s Spectre microcode updates

The Meltdown and Spectre vulnerabilities still loom over the world’s computer users as companies scramble to mitigate them. Like many PC owners of late, I’ve been biting my nails waiting for Intel to issue stable versions of its Spectre-mitigating microcode updates and for the blue team’s hardware partners to bake them into new firmware for  their products.

I’m still waiting on a microcode update for my aging Haswell desktop, but Dell issued a new firmware update for my Alienware 13 R3 last week that promised some Spectre-cide™ goodness for the Core i7-7700HQ inside. Before I patched that system, though, I wanted to record some pre-patch performance numbers to see just how much I was giving up by fully protecting my system against these vulnerabilities (or as fully as possible, at least).

Part of the fear surrounding Meltdown and Spectre, aside from the potential for the leakage of sensitive data, is the potential performance hit from patching those vulnerabilities. Intel has claimed from the beginning that any performance hit from the Spectre and Meltdown patches would “not be significant,” was “highly workload-dependent,” and might change with time as the mitigations for those vulnerabilities were refined. Early tests from some sources foretold potentially large slowdowns for servers, but just like Intel says, the potential slowdowns range from “minimal” to “measurable” depending on workload, according to Red Hat. Those projections are more for data centers running atop Linux than Windows desktops, though.

Microsoft does have some ideas for the kinds of performance hits we should expect on PCs, and its work suggests any performance changes will be CPU-dependent and operating-system-dependent. Redmond expects that systems with Skylake and newer CPUs might experience “single-digit slowdowns,” while “some owners” of Haswell and older parts should expect to notice the slowdowns from these patches. Those projections are for systems running Windows 10, too. Microsoft advises that older versions of Windows on Haswell and older chips will experience a noticeable performance hit.

With all that in mind, my Kaby Lake laptop running Windows 10 should be the best-case scenario for post-Spectre patch performance. Until Intel issues stable microcode updates for Haswell and older CPUs, we won’t be able to gauge just what the effects of Spectre-cide might be for those systems—and that assumes big PC companies and motherboard makers plan to go to the effort of issuing new firmware for their older products in the first place.

One thing is for certain: this story is far from finished. Spectre and Meltdown represent classes of attacks, not a single bug that can be exterminated with a single patch. It’s important to remember that the microcode updates Intel is issuing are mitigations, not curatives. These patches may make Spectre much harder to exploit, but if there’s one group one never wants to underestimate, it’s security researchers. We may yet see new and novel ways of exposing privileged data through processor side channels. For now, whatever margin of safety Intel’s patches claim to afford seems worth having. Let’s see how much performance we’re trading for those safeguards.

Our testing methods

Our test system was configured as follows:

Alienware 13 R3
CPU Intel Core i7-7700HQ
Memory 16 GB (2x 8 GB) DDR4-2666
Graphics card Nvidia GeForce GTX 1060 6GB
Storage Samsung PM961 512 GB NVMe SSD

There are a lot of moving parts in testing the performance implications of Spectre and Meltdown between operating system updates and microcode changes. Microsoft has already issued the operating-system-level patches necessary to stop Meltdown cold on affected PCs, though, and I let Windows automatically update on my personal system, so the effects of those changes are already in place whether I like them or not. I’m not particularly concerned about the effects of Meltdown mitigations here, either, since Microsoft calls out Spectre patches as the ones with the largest potential performance impact.

Still, I isolated our before-and-after comparisons to just one change: a Spectre-related firmware patch on an otherwise fully-updated Windows 10 machine. I ran my tests back-to-back, immediately before and immediately after I applied the necessary firmware from Dell. That fact means we should be looking at the performance impact of that one patch and little else.

We tested using Windows’ Balanced power plan. To ensure accuracy, we ran each benchmark at least three times and took the median of the results. Our tests were conducted with the system connected to AC power.

Our testing methods are generally publicly available and reproducible. If you have questions regarding our methods or results, leave a comment on this article or join us in our forums.

JavaScript

First off, we have several synthetic JavaScript benchmarks that give us a good idea of single-threaded performance changes. These benchmarks should be broadly relevant for the speed and responsiveness users enjoy when scrolling through Facebook, Twitter, and other anxiogenics in this modern age.


Across all of our JavaScript results, we get mid- to high-single-digit percentage decreases in performance. The fully-patched Core i7-7700HQ falls 6.7% behind its unfettered self in Jetstream, 9.7% behind in Octane, and 7.3% behind in Speedometer, though Kraken doesn’t seem to be hurt much by the new microcode.

On the whole, a geometric mean of the performance changes we saw suggests a 6.1% loss in these benchmarks. That may not sound like much, but in this day and age, that kind of figure can encompass an entire generational change’s worth of performance for Intel CPUs. To be fair, most recent systems feel sufficiently snappy in day-to-day use, but every little bit helps.

 

Feeling out the whole hog with PCMark 10 Extended

I also tried to get a sense of the Spectre patch’s effects on my system using Futuremark’s PCMark 10 Advanced suite. That benchmark’s Extended test gathers data on a range of typical desktop tasks, including app start-up times, video conferencing, web browsing, word processing, and spreadsheet manipulation. It also gives us a high-level view of gaming performance, thanks to its integrated Fire Strike graphics and physics tests.

Even better, Futuremark fully details just what its benchmarks actually test instead of spitting out an opaque index value. Thanks to that information, I can spitball about just what’s caused a dip in performance for a given benchmark. My thanks to Futuremark for providing a license key for this handy software.

While PCMark’s gaming tests are no replacement for a test of several real-world applications with frame-time data behind them, they do let us see whether gaming performance potential changes much, if at all, before and after I apply the Spectre microcode patch. Sorry to disappoint, but few games are actually CPU-bound these days at real-world resolutions and settings. Spectre mitigations are unlikely to have a major effect on gaming performance to begin with.


The PCMark 10 Essentials test encompasses app start-up times, video conferencing performance, and web browsing tasks. The biggest concern among these results is the 13.5% drop in app load time scores post-Spectre patch. That result could mean that some PCs could feel significantly more sluggish than before when users go to load software for the first time in a day. Just how much users actually feel the slowdowns PCMark 10 exposes may depend on how often they put their PCs to sleep versus performing cold shutdowns, though. Windows’ SuperFetch caching could also mitigate some of the pain.

Other I/O-heavy interactions with a system might not be as fortunate, but few of our CPU benchmarks (and few applications in general) are actually storage-bandwidth-bound on the desktop. We may need to explore that topic in greater depth separately.

Outside of I/O-bound work, the Essentials video conferencing score fell just a couple of percent, and web browsing performance remained essentially unchanged. For folks who truly don’t lean on their PCs that much, this test suggests that Intel is correct when it says that most users won’t notice a performance impact from its mitigations.


The Productivity portion of the PCMark suite also turns up some big swings pre- and post-Spectre patch. For spreadsheet jockeys like myself, the 9% performance increase in PCMark 10 is welcome, but folks who spend more time in Word (as represented by LibreOffice Writer) might find their text-entry work a little more sluggish post-Spectre patch. Futuremark’s detailed discussion of its word-processing benchmark suggests a fair bit of saving and loading documents is going on behind the scenes. If that work meaningfully involves the SSD in its evaluation, that could explain the performance drop we see in what should otherwise be a fairly lightweight task.


The Digital Content Creation test simulates photo editing, rendering and visualization, and video-editing work. None of these applications show more than a couple of percentage points’ worth of performance decreases pre- and post-Spectre patching, and that’s well within the 3% margin of error that Futuremark suggests is typical for controlled testing with PCMark. Creative pros that aren’t bottlenecked by I/O bandwidth would seem to be largely safe from the side effects of Spectre-cide.


PC enthusiasts might worry about gaming performance the most among the potential performance drops that might arise from Spectre patches. Few games are I/O-bound, however, and Futuremark’s 3DMark Fire Strike Graphics and Physics tests certainly are not. As with the Digital Content Creation test, the gaming performance potential demonstrated by 3DMark is, for all intents and purposes, unchanged pre- and post-Spectre update.

Conclusions

The best thing I can say about Intel’s Spectre patch is that it doesn’t seem to harm the day-to-day performance of recent PCs much—at least going by Dell’s version for my Alienware 13 R3. As I’ve already lamented, though, per-core CPU performance gains of any kind are worth their weight in gold these days, and even the single-digit losses I saw across most of my light-usage benchmarks sting like a tarantula hawk when single-digit performance increases are all we’ve gotten from the blue team over the past few years.

Perversely, AMD might be able to take some good news out of this mess. Ryzen CPUs weren’t that far behind Intel parts in lightly-threaded workloads to begin with, and the company doesn’t believe its Ryzen CPUs are affected by the Meltdown exploit at all. AMD does expect to offer optional firmware updates for customers concerned about Spectre, but it doesn’t seem to be as concerned as Intel is about the need for a patch. The effects of Meltdown and Spectre patches on Intel systems and the potential performance benefits of second-generation Ryzen CPUs with higher clock speeds could help AMD close the gap even further soon.

For those whose PCs use Skylake or newer CPUs inside, it’s probably OK to go ahead and patch up with Spectre-cide if your motherboard maker or big OEM has a firmware update available. Outside of some potentially troubling behavior in I/O-bound workloads that we need to explore further, the performance trade-off from the patch just isn’t big enough to risk running unprotected. That’s a small bit of relief in what has otherwise been a rough year for PC enthusiasts. It remains to be seen just how much similar patches for Haswell and older PCs are going to hurt performance, but until we get stable microcode from Intel and firmware updates with that microcode from motherboard makers, the unease of running an unpatched system and the potential performance pitfalls of those patches for older processors are going to haunt us. All we can do is wait.

Comments closed
    • wownwow
    • 2 years ago

    People miss the most important point; it’s not the few % of performance but the company’s INTENTIONALLY not following the privilege levels defined by itself, but people accept it, amazing!

    • TheCEOofSEO
    • 2 years ago

    I rushed in too early and flashed my Bios with the first Lenovo patch (which was recalled – and apparently now there is no way for me to revert it) My Laptop fan runs constantly now. I can feel my laptop putting out more heat. Of course I’m experiencing slower encoding times on my Video editing software Adobe Premier Pro (about a 50% slowdown) and the blue screen of death once a day. I want my old laptop back – it was running perfectly…Grrrrr. Anyone else experiencing more heat output – more CPU drag?

    • brucek2
    • 2 years ago

    It strikes me that these modest single digit numbers are about the same as we’ve seen lately in generation-to-generation single core performance.

    In other words, the performance hit will be equivalent to pushing many people back a processor generation.

    I’m not sure what this means for efficiency/power usage (ie laptops), but it’s not unreasonable to expect that one will affect the other?

    • Thbbft
    • 2 years ago

    Have you no high end Intel or Threadripper chips to hand?

    HEDT performance is where the differences will most impact productivity. A current Threadripper 1920 vs. a comparably priced Intel chip face-off with attendant scatter plot would be useful.

    • just brew it!
    • 2 years ago

    Ubuntu’s latest Meltdown/Spectre patches (released last week) completely broke the Nvidia GPU driver on my primary desktop. Backed off to the previous kernel for the time being until I can get that sorted.

    • snowMAN
    • 2 years ago

    “Feel significantly more sluggish” for a 13.5% slow-down is exaggerated, most normal people don’t even notice a difference in speed until it gets to at least 33% slower (or 50% faster).

      • Takeshi7
      • 2 years ago

      That’s not true. For a system that’s easily fast enough it probably doesn’t make a difference, but for a machine that’s right on the edge between usable and unusable, 13.5% can be huge.

    • setaG_lliB
    • 2 years ago

    My only system that seems to be affected by the Windows 7 patch is an older laptop. It’s powered by a C2D T7200 (Merom @ 2GHz, 667FSB, 4MB L2) and has a 5400RPM HDD.

    Windows and application launch time is definitely longer. Audio stutters occasionally when CPU usage is above 85%, which never used to happen. According to Process Explorer, CPU time devoted to servicing DPCs and hardware interrupts has increased a bit since the update was installed. I’m guessing the higher DPC latency is causing the audio stutter.

    I haven’t noticed a decrease in performance on any of my other systems (IVB-E i7, IVB i5, Phenom X6, C2D e8600, Haswell mobile; all running Win7 on Samsung 840 or 850 SSDs).

    • meerkt
    • 2 years ago

    What’s the toll Microsoft’s update(s) take?

    • anotherengineer
    • 2 years ago

    Part of me was hoping they would give the fix a code name, like, hmmm

    Pac-Man

    • Takeshi7
    • 2 years ago

    needs more storage benchmarks.

    • hansmuff
    • 2 years ago

    I lose 25% performance in Visual Studio building a large set of solution files. That’s on a 5600U CPU in a HP EliteBook 840 G2. I feel that difference, and I hate it. But it’s a corporation owned machine, so there’s little I should be doing about it.

      • Flying Fox
      • 2 years ago

      Gives you ammunition to bitch for an upgrade? 😛

    • w76
    • 2 years ago

    My Haswell-era xeon-powered Unraid server, mostly acting as a file server but also running Emby and a couple other services in docker containers, I’ll probably never patch. I don’t even know if I’ll patch my Kaby Lake laptop.

    Why should I take the performance hit? As for the server, it’s linux, nothing is getting installed by mistake, so I don’t see the attack vector this exploit could use. The laptop, well, I’m careful, have been for years and haven’t seen it get infected with anything for years. Basic security principles seem to be the key here.

    This is no different philosophically then choosing the relatively light-weight built-in AV in Windows 10 over much more obtrusive, feature-rich and a resource-heavy products. We all know there are vulnerabilities, but what we use to mitigate against them falls along a spectrum of cost and resource benefit analysis. A lot of people are happy with built-in protection.

    And in this case, I don’t see the benefit outweighing the cost.

      • highlandr
      • 2 years ago

      Running your own, trusted code on a linux server [i<]shouldn't[/i<] give you any trouble. But consider the fact that Amazon lost ~10% of the performance of their entire cloud, and they don't DARE run without patches. That is way worse.

    • sweatshopking
    • 2 years ago

    Thinking my 4790k might exist without a patch as it’s used for gaming exclusively and i regularly format it since i run insider fast previews. The decline doesn’t sound worth it if we are going to see greater drops on haswell

    • freebird
    • 2 years ago

    The only thing we have to “fear” is fear itself!!!

    (and a legal disclaimer that any other unknown/unannounced vulnerability(ies) may cause us to “hacked” at any moment.)

    That spector image should have a wink. ; )

    • ozzuneoj
    • 2 years ago

    Where exactly are the attackers who exploit these vulnerabilities coming from (to get to my CPU) and are there any reports of people actually being attacked using these vulnerabilities?

      • Aether
      • 2 years ago

      The only potential attack vector of which I am aware for personal computers is web site Javascript. However, Chrome and Firefox have made changes that should make timings attacks difficult, so my understanding it that there is not an easy external attack vector for personal systems.

        • ozzuneoj
        • 2 years ago

        Thank you.

        Anyone else have any more information explaining exactly how someone could exploit these vulnerabilities? I understand that cloud services have to do whatever is in their power to prevent any data from being stolen, so they WILL go through these fixes, regardless of the expense, but it still seems nearly impossible that massive corporations would be vulnerable to things like this at all since it requires an attacker to have some kind of access to the system in the first place.

        If every system on the planet that isn’t running a Broadwell or newer CPU is going to be vulnerable to these expoloits, I would hope that it will be mitigated on as many levels as possible.

        If the attacks can come through Javascript or HTML5 and Chrome and Firefox both have fixes that significantly limit the usefulness of exploiting Javascript (and hopefully soon HTML5?), it seems premature for the vast majority of users to permanently neuter the performance of their computers unless they regularly use that specific system to handle highly sensitive information… and even then, I’d hope you weren’t doing things with the system that would expose it to external attacks in the first place. Its not like Spectre didn’t exist before the general public knew about it.

        Its like people only think about security when the media mentions the latest big security flaw in the computers that run the entire world.

        Nothing we do is secure, all of your data could potentially be exposed to someone through some unknown means. That’s how things have been for 20+ years now obviously, since we didn’t know about Spectre before.

        It definitely sucks, but every day we make the decision to take security (or other) risks by weighing the pros and cons, turning on our computers and giving away our personal information to websites we really don’t know anything about. A big scary spectre logo, lots of media attention and another way for lots of people to make tons of money in different markets… doesn’t really change much.

          • Redocbew
          • 2 years ago

          Yeah, attacks carried out through bad passwords and social engineering are almost certainly going to be more common and receive less media attention. Having physical access to a system usually means it’s game over anyway. Without the JIT attack, then some other exploit would have to be used first in order to execute the attack. I suppose it depends on your perspective whether or not the hype is justified.

    • mad_one
    • 2 years ago

    I appreciate the sensible selection of benchmarks, though a couple real games would make a nice addition. A lot of publications tested with compute benchmarks that do not make a lot of kernel calls and thus naturally aren’t affected at all.

      • Prestige Worldwide
      • 2 years ago

      +1 for more games

      • Eggrenade
      • 2 years ago

      I want to look inside the second.

        • mad_one
        • 2 years ago

        You second looking inside the Spectre?

          • Prestige Worldwide
          • 2 years ago

          Let’s inSpectre it

          (•_•)
          ( •_•)>⌐■-■
          (⌐■_■)

        • derFunkenstein
        • 2 years ago

        I looked inside the second, and the second was me!

    • cynan
    • 2 years ago

    My Sandybridge-E system felt noticeably less responsive with the MS Meltdown patch enabled. As far as I’m concerned this makes how minimal the additional performance penalty is for addressing Spectre sort of moot.

      • mad_one
      • 2 years ago

      Older CPUs are affected a lot more since they don’t handle page table changes as well (the feature was added to make virtualisation faster, but ended up helping with the Meltdown patches).

      As far as I understand, current CPUs do not need to clear the TLB when the page table changes.

    • Shouefref
    • 2 years ago

    I’ld really like to see the final comparison of patched Intel’s versus patched AMD’s.
    I expect AMD will patch for Spectre sooner or later.

      • IGTrading
      • 2 years ago

      Why not see results vs. AMD’s competing CPU now ?

      Why is AMD “perversely” anything ?

      Why do I have the feeling that this is a “nice pat on the back for Intel and lets diss AMD” article ?!

      I’m not saying it is, but why do I keep having this feeling of some sort of Intel bias @ TechReport ? 🙂

      I see bias at other websites as well. Not many, but some. Not even all authors, just some.

      Why isn’t TechReport the same as it was back in 2005 … 2010 …. ?!

      I never, absolutely never had such a feeling back then … I was reading it “religiously” and now I find myself questioning almost every article …

        • DancinJack
        • 2 years ago

        Because you’re a shill.

          • chuckula
          • 2 years ago

          How can he be the shill?!?!?

          [b<]I'M THE SHILL![/b<] Unless he's secretly me. #TylerDurden4Life I'm sure IGTrading would love to write a long essay that severely criticizes AMD on the following topics: 1. Lisa Su works too hard and cares too much... DAMMIT! 2. AMD's technology is far far too advanced for you pathetic mortals to appreciate it. Perhaps they should only release old hardware since the world just isn't ready for their innovation. 3. All of AMD's products -- ESPECIALLY THE GPUS -- are just priced too low. 4. Smartphones: Since Adreno is really AMD technology that has never been changed one bit since it was sold to Qualcomm over 10 years ago, should AMD be criticized for not collecting 100% royalties for being the true inventor of all smartphone technology?

            • Redocbew
            • 2 years ago

            His name is Robert Paulson.

          • IGTrading
          • 2 years ago

          Not at all.

          Actually I think that more than half of my assessments conclude with a recommendation of Intel over AMD because our clients’ interests are paramount and we only care about design requirements that are required by the client.

          So yeah, I have my own opinions, but I’m 110% the same boring professional expert that will give you the technical answer with no regards about you liking it or not. Even I don’t like my own statistics 🙂 and I’d be happy to recommend more AMD solutions, but clients want what clients want and the reality is that Intel’s compiler rules x86 application the market.

          Therefore you can stop worrying about me being the boogie man 🙂 I’m just more blunt than the next guy.

        • Jeff Kampman
        • 2 years ago

        1) It’s not an apples-to-apples comparison when AMD doesn’t have comparable updates in the wild yet

        2) This piece repeatedly establishes that we are interested in the state of Intel’s recent CPUs, since that’s what we have microcode updates for

        3) It’s bad for PC owners everywhere when competitive standings regress as a result of performance-robbing bugs instead of advance thanks to technical improvements; cf. the Phenom TLB bug [url<]https://techreport.com/review/13741/phenom-tlb-patch-benchmarked[/url<]

          • IGTrading
          • 2 years ago

          Hi Jeff, thanks for taking the time to reply .

          1) It should be made to be apples-to-apples. Intel needs a patch. AMD doesn’t need it. Therefore test it like it is.

          IMHO, AMD clearly states it doesn’t need any Meltdown patch and Google’s researchers certify this. Google’s lab says that in almost a year of trying, they were not able to expose AMD’s cache content like they did with Intel.

          Therefore we can safely take it that AMD’s CPUs are safe from Meltdown (until we have proof of the contrary) .

          I hope that we don’t need more assurances about this ….

          We have AMD and Google saying the chips should be safe unless there will be a guy way more talented that Google’s team and he manages to achieve what they weren’t able to do , despite trying for months on end.

          IMHO we can make it apples to apple because Intel needs a patch. AMD doesn’t need it. Therefore test it like it is.

          2) Yeah , you clearly say Intel in the title and everywhere. That’s true and uncontested.

          But the real world does have an alternative to Intel so it is not really fair to say : “Intel has little to fear aka 6% performance loss” while a whole generation of Intel chips usual brought less than 5% improvements.

          6% may seem “little” but a losing the performance brought after a year and a half of Intel development work is not “little” in my opinion.

          Intel used to pay and probably still pays OEMs like DELL, HP, Fujitsu to not develop products with AMD components or limit these or not advertise these and let the market believe that there is no alternative to it.

          The press should be informative and inform everybody that there are alternatives that happens to be safer and better in this case.

          This article (and this is just a gut feeling , not an accusation because honestly there is nothing to be accused of) just doesn’t feel right to me. But that’s just me and my comment.

          It’s a gut feeling from a guy sitting comfortably at the office and reading about your hard work. So don’t take it too seriously.

          You did the work. I’m just sitting on the side line and talking about it.

          3) I know about the Phenom TLB and that was a complete mess. Great article by TR by the way!

          The thing is that it should not even be brought up into this discussion because it’s like comparing broom with the former planet Pluto , in my opinion.

          Here is why :

          100% of Intel’s CPU are vulnerable to Meltdown.
          100% of Intel users will lose performance.
          large% of applications are affected by the performance loss, but probably not all.

          AMD’s TLB was present in just the 1st generation of Phenom and solved by a revision and all chips were exchanged on demand. Intel will NOT exchange the CPUs, but apply a patch that brings a performance loss.

          Still, yes, we can say that 100% of AMD’s 1st gen Phenom had the TLB issue. True.

          But NOT 100% of the apps were affected.
          Not even a low% of the apps were affected.
          Not 100% of the users lost performance.
          Not even a low% of the users were affected.
          0% of the computers’ security was affected by the TLB bug.

          With Intel, if you get hacked, the damage is waaaay more costly than the 6% lost of performance.

          IMHO Meltdown = high risk of costly damage + performance loss on Intel

          AMD’s TLB = performance loss (if patched) , stability risk (if scenario is matched), zero risk of hacking and subsequent costly damage.

          It is my opinion that the two can’t really be compared.

          Lets take cars for example :

          Intel Meltdown : Anybody can hack every Intel powered car (100%) and drive you in a pole. You NEED to patch and this WILL erase 6% of your car’s performance .

          AMD TLB : All cars with 1st gen of Phenom with a certain hardware revision, if they drive on a steep 17% inclination hill, while in foggy weather, with temperatures lower than 8 degrees, on a dirt road can experience an engine stall.

          The fix comes with a 10% loss of performance or an exchanged CPU from AMD.

          You don’t drive on such a road, in such conditions, at such a temperature, at such an inclination, then you don’t experience any issue.

          That’s quite the difference IMHO.

          But again, my comments are just comments, mate.

          You did the work. You wrote the article. It is still valid work with useful information for which I thank you, even if I’m whining from the side line 🙂

          Besides, you can be biased without knowing it, without doing anything intentionally.

          Heck, I’m always asking myself (with all the nonsense going through my head) if I’m not leaning too much towards AMD or Qualcomm or whatever …. But we press on.

          Have a good day mate!

            • K-L-Waster
            • 2 years ago

            Dude, why on Earth are you going on about Meltdown? This is a test of the *Spectre* microcode, not the Meltdown Windows update.

            (Oh, right, Spectre is the one that *can* affect AMD, so in order to not be biased we have to pretend it doesn’t exist…)

            • IGTrading
            • 2 years ago

            Short answer : because I’m an idiot 🙂

            Long answer, AMD is only vulnerable to some certain versions of Spectre, but not to others. So still they have the upper hand.

            But yeah, my confusion was major : it is perfectly valid to say that AMD hasn’t yet released the updates.

            Question to self : why the heck are you so stuck on Meltdown mate ?!?! 🙂

            P.S. That’s what I get from commenting at the same time as writing my evaluations.

            • Redocbew
            • 2 years ago

            [quote=”Spectre research paper”<]AMD states that its Ryzen processors have “an artificial intelligence neural network that learns to predict what future pathway an application will take based on past runs” [3, 5], implying even more complex speculative behavior. As a result, while the stop-gap countermeasures described in the previous section may help limit practical exploits in the short term, [i<]there is currently no way to know whether a particular code construction is, or is not, safe across today’s processors – much less future designs.[/i<][/quote<] Emphasis mine. Leaving aside the slight embellishment of calling a branch predictor an "artificial intelligence neural network" this is the whole reason why news about these exploits blew up the way it did. I'm sure chip engineers will be able to deal with the issues learned here in future designs given the time to do so. [quote="Spectre research paper"<]We have empirically verified the vulnerability of several Intel processors to Spectre attacks, including Ivy Bridge, Haswell and Skylake based processors. We have also verified the attack’s applicability to AMD Ryzen CPUs.[/quote<] Furthermore, AMD has backed off their initial statement on the branch target injection attack in Spectre. Now AMD says only that it would be "difficult". Even if that were not the case these exploits clearly caught many people off guard, and to say that AMD has "the upper hand" is disingenuous at best.

            • Redocbew
            • 2 years ago

            The rules of being a shill:

            1st rule: You do not talk about being a shill.

            2nd rule: You [i<]do not[/i<] talk about being a shill. 3rd rule: If someone says "shill", downvotes you, or defeats your argument, then you must respond with a gigantic wall of text in the hopes that lots of words will serve your agenda better than you can. 4th rule: Only [s<]two sides[/s<] one side to the argument. Statements to the contrary will be met with continued protestations of your extraordinary experience as if that somehow matters. 5th rule: One company at a time. 6th rule: No upvotes, no acceptance. 7th rule: Threads will go on as long as they have to. 8th rule: If this is your first time being a shill, then you must continue to be a shill even after it becomes obvious to everyone.

            • chuckula
            • 2 years ago

            I really like how you’ve extended the Fight Club metaphor.

        • K-L-Waster
        • 2 years ago

        Mentioning anything Intel does that bears some semblance of working == Anti-AMD bias

        Mentioning anything about AMD as if it isn’t > 123.594% perfect == Anti-AMD bias

        Not slamming Intel for… well, existing at all == Anti-AMD bias

        Have I missed anything?

          • DancinJack
          • 2 years ago

          You have not, good sire.

          • mertenz
          • 2 years ago

          In my opinion, there is some merit IGTrading post , the “perversely” use caught my attention too.

          In another article :

          “CPU Hash would seem to deliver a resounding win for the Ryzen 5 2500U, but remember that the Zen architecture supports Intel’s SHA Extensions.”

          Semiotically and Psichologically interesting to say the least, writing something like that and not an usual(i.e.):

          “CPU Hash results resounding win for the Ryzen 5 2500U, thanks Zen architecture support of Intel’s SHA Extensions.”

          Of course, the author is free to choose his words, and the data gathered is valuable to me. Thanks.

          I think adding the competition results would enriched the article. Like in that Wasson article Jeff linked about Phenom TLB bug.

          Of course, post like chukulas being the same butthurt, passive-agressive straw man attacks are not shocking to anyone. But they are tiring and boring. And hes got little friends to decrease S/N ratio.

          Thanks God for reasonable people like “just brew it!” (among others), trying to keep de discussion above kinder garten level.

          Thanks again for the article.

    • auxy
    • 2 years ago

    This would have been a lot easier to read if the charts didn’t keep flipping between after/before and before/after. (・へ・)

    Also,
    [quote<]Spectre-cide™[/quote<] (*'▽')b

      • Shouefref
      • 2 years ago

      Spectre-cide is REALLY important! 🙂

      • Duct Tape Dude
      • 2 years ago

      It’s ordered by higher performance first (which is difficult to discern when the numbers are so close), and the colors are dark and light. I think it’s sufficient.

    • derFunkenstein
    • 2 years ago

    That 13.5% dip in app load times pairs with what I thought I was just imagining on my work laptop (Dell Latitude i7-6600U, 16GB, 500GB AHCI M.2 drive). I had just figured I was dreaming that it felt slower to open apps, particularly Visual Studio, and then also opening a pretty large solution. Office apps have felt like they still perform similarly, but it never took long to open Excel in the first place.

    • chuckula
    • 2 years ago

    Thanks Jeff.

    Quick question: Which browser was used to run the JavaScript benchmarks?

      • Jeff Kampman
      • 2 years ago

      Edge.

Pin It on Pinterest

Share This