Ryzen vs i7 (Mainstream); Threadripper vs i9 (HEDT); X299 vs X399; Xeon vs Epyc

Discussion in 'Hardware Components and Aftermarket Upgrades' started by ajc9988, Jun 7, 2017.

  1. Raiderman

    Raiderman Notebook Deity

    Reputations:
    462
    Messages:
    746
    Likes Received:
    1,726
    Trophy Points:
    156
    Honestly, besides my laptop in my signature, I have not purchased an Intel CPU since the Pentium 4 days. It has nothing to do with fanboyism, but rather bang for the buck. I suppose this is why I havent built a desktop in ten years or so, because AMD didnt offer anything that had any bang for the buck. When I looked at Intel, I just thought that they were too proud of their stuff.

    Hopefully AMD takes advantage of the position they are in right now, and continues to build on a fantastic product so we can have lower prices all around. Believe it or not, Intel needs AMD. If there was no AMD, most of us would be sitting on Core 2 cpu's right now. That doesnt sound attractive to me.
     
    ajc9988 likes this.
  2. Support.2@XOTIC PC

    Support.2@XOTIC PC Company Representative

    Reputations:
    478
    Messages:
    3,117
    Likes Received:
    3,471
    Trophy Points:
    331
    We started to see it with Ryzen, the next gen hopefully is that really taking wing. I'm hoping also that they can do the same for NVidia, but I wasn't nearly as impressed by Vega as Ryzen/Threadripper.
     
    Raiderman likes this.
  3. ajc9988

    ajc9988 Death by a thousand paper cuts

    Reputations:
    1,233
    Messages:
    5,083
    Likes Received:
    7,117
    Trophy Points:
    681
    Everyone bashes them versus Nvidia. But, the truth of the market share is that Intel is MORE of a monopoly than Nvidia is, or at least from the number I've seen.. I do think Intel is in trouble. Here is what I posted about them in another thread:


    I listened to the conference call while working today. The questions from the analysts really should be focused on. There was a pop on adoption of server CPUs in the fourth quarter which was uncharacteristic and they are predicting that to be lower than the past FIVE YEAR AVERAGE. They are saying it is because of the sales, but what do you want to bet they did a sales push BEFORE the vulnerabilities came out, after all, they took money from the Intel Inside add campaign and moved it to sales for the server side last quarter. Smells rank to me.

    They also focused attention on the areas that would do well, but as the analysts pointed out in many questions is that they kept saying these divisions were going to be up, but the overall revenues flat. So where is the money suck? This is also why automotive was combined with mobile, to hide the fact of bleeding money on the mobile side of things.

    As to 10nm, they mentioned it, specifically finishing tooling to do mass production (FINALLY) in the second half of the year, with expenditures for tooling as well as some on R&D. We already know that Coffee was pushed to the summer to deal with Meltdown and that Kaby-R (the faux, 6-core coffee) was pushed out the door in limited supply to take it off inventory before the vulnerability was disclosed, which also boosted sales and the scaled down availability is likely why costs where down for Intel in Q4 (when you aren't making as many chips, you don't need to spend on materials, now do you?). But, they, although mentioning 10nm, did try to steer away from talking about it too much, although one analyst tried to pin them down on it toward the end of the call.

    So, what I heard is they are going to have issues with certain segments, although they reiterated puffery like "strong" in describing demand, and general optimism in regards to cloud and other server segments. They did announce a large cloud deployment, so a plus in that partnership in China (which AMD already announced a deal with the same company this past summer or fall).

    To give a better look, I'll need to dig into the 10-K and review the call again after that digging. I want to especially review the amount set aside for litigation. I'll also want to review their definitive proxy statement, which should be coming soon, which gives a really good look at the activities of the executives holdings, aside from seeing their public sales, 10b5-1 sale certs, etc. I'm just that kind of guy! ;-)

    But, 10nm is a nothing burger this year, cannon is a joke, mostly skipped. We have Cascade-X and Coffee next summer with the CPU Meltdown fix (no guarantee Spectre is fully fixed for them or AMD for their April launch, except maybe on variant 2 for the two companies). Ice is when it comes, but Whiskey being talked about and them saying FURTHER development on 14nm suggests that they may have to run that against 7nm release for AMD in Q4 2018 to Q1 2019 for server (hopefully), and H1 2019 on the consumer side. That is going to be rough, especially since that means AMD will have transistor density and efficiency for a couple months to half a year.



    This plays straight into the head to head coming, which will definitely be the fight to watch!
     
  4. Support.2@XOTIC PC

    Support.2@XOTIC PC Company Representative

    Reputations:
    478
    Messages:
    3,117
    Likes Received:
    3,471
    Trophy Points:
    331
    Didn't mean it to sound like bashing, I think it was more that I got what I expected from the GPU side, and then got way more than I expected from the CPU side.
     
    Raiderman and ajc9988 like this.
  5. ajc9988

    ajc9988 Death by a thousand paper cuts

    Reputations:
    1,233
    Messages:
    5,083
    Likes Received:
    7,117
    Trophy Points:
    681
    Well, and wanting more on the GPU side isn't bad, as they obviously need a kick in the seat. I did post about the new heads for Radeon graphics division the other day:

    http://www.pcgamer.com/amd-hires-fo...cs-execs-to-lead-its-radeon-technology-group/
    "Rayfield will serve as senior vice president and general manager of AMD's Radeon Technologies Group, the company's dedicated graphics division from which Koduri departed, and Wang will hold the role of senior vice president of engineering for the same department. Both will report to AMD CEO Dr. Lisa Su.

    "Mike and David are industry leaders who bring proven track records of delivering profitable business growth and leadership product roadmaps," said Dr. Lisa Su. "We enter 2018 with incredible momentum for our graphics business based on the full set of GPU products we introduced last year for the consumer, professional, and machine learning markets. Under Mike and David's leadership, I am confident we will continue to grow the footprint of Radeon across the gaming, immersive, and GPU compute markets."

    Rayfield is a 30-year veteran of the technology industry, having most recently served an executive role at Micron. Prior to that, Rayfield was the general manager of the mobile division at Nvidia, where he led the team that created Tegra.

    As for Wang, this is not his first stint with AMD. He previously worked on AMD's system-on-chip (SoC) efforts, and prior to that, he held various technical and management positions at ATI, ArtX, SGI, Axil Workstations, and LSI Logic. His most recent position before returning to AMD was at Synaptics, where he served as senior vice president of Systems Silicon Engineering."


    This shakeup after Sue took over the head position in his absence might have spurred a recommendation to split responsibilities and hire more people with tighter focus. Hence, one to focus on the engineering and one to take on general management. Both are awesome in their own rights, and the one engineer is an ATI vet. So, this could be very interesting. Both have experience in SOC (one with Tegra, the other with AMD), and both seem to be good fits for where this is going. Especially with AMD potentially working on a multi-die card, which may utilize aspects of those skills.
     
    Raiderman likes this.
  6. Support.2@XOTIC PC

    Support.2@XOTIC PC Company Representative

    Reputations:
    478
    Messages:
    3,117
    Likes Received:
    3,471
    Trophy Points:
    331
    That's good news! I will admit my main concern in the GPU side right now is that someone (preferably both Red and Green teams) need to start making dedicated mining cards so take pressure off the gaming ones.
     
    Raiderman and ajc9988 like this.
  7. ajc9988

    ajc9988 Death by a thousand paper cuts

    Reputations:
    1,233
    Messages:
    5,083
    Likes Received:
    7,117
    Trophy Points:
    681
    They do, but the restrictions make it where you cannot sell them in the US! Not only that, only reason Nvidia is scared of it is they saw what happened when AMD cards hit the resell market when the 2014 crash happened and they know how many of their Pascal cards will be dirt cheap, thereby undercutting future gen sales. But I don't see price relief unless buying at release. Otherwise, it will continue like this for the next year or so.
     
  8. Support.2@XOTIC PC

    Support.2@XOTIC PC Company Representative

    Reputations:
    478
    Messages:
    3,117
    Likes Received:
    3,471
    Trophy Points:
    331
    Wonder how that will affect the PC gaming landscape...I almost see this being good for notebooks, since they're unlikely to be used for mining, they may stay reasonably priced.
     
    ajc9988 likes this.
  9. Papusan

    Papusan JOKEBOOKS = That sucks! Dont wast your $ on FILTHY

    Reputations:
    15,733
    Messages:
    19,929
    Likes Received:
    31,287
    Trophy Points:
    931
  10. ajc9988

    ajc9988 Death by a thousand paper cuts

    Reputations:
    1,233
    Messages:
    5,083
    Likes Received:
    7,117
    Trophy Points:
    681
    So here is the Torvald tirade from last weekend in its entirety:

    From Linus Torvalds <>
    Date Sun, 21 Jan 2018 13:35:59 -0800
    Subject Re: [RFC 09/10] x86/enter: Create macros to restrict/unrestrict Indirect Branch Speculation
    On Sun, Jan 21, 2018 at 12:28 PM, David Woodhouse <dwmw2@infradead.org> wrote:
    > On Sun, 2018-01-21 at 11:34 -0800, Linus Torvalds wrote:
    >> All of this is pure garbage.
    >>
    >> Is Intel really planning on making this **** architectural? Has
    >> anybody talked to them and told them they are f*cking insane?
    >>
    >> Please, any Intel engineers here - talk to your managers.
    >
    > If the alternative was a two-decade product recall and giving everyone
    > free CPUs, I'm not sure it was entirely insane.

    You seem to have bought into the cool-aid. Please add a healthy dose
    of critical thinking. Because this isn't the kind of cool-aid that
    makes for a fun trip with pretty pictures. This is the kind that melts
    your brain.

    > Certainly it's a nasty hack, but hey — the world was on fire and in the
    > end we didn't have to just turn the datacentres off and go back to goat
    > farming, so it's not all bad.

    It's not that it's a nasty hack. It's much worse than that.

    > As a hack for existing CPUs, it's just about tolerable — as long as it
    > can die entirely by the next generation.

    That's part of the big problem here. The speculation control cpuid
    stuff shows that Intel actually seems to plan on doing the right thing
    for meltdown (the main question being _when_). Which is not a huge
    surprise, since it should be easy to fix, and it's a really honking
    big hole to drive through. Not doing the right thing for meltdown
    would be completely unacceptable.

    So the IBRS garbage implies that Intel is _not_ planning on doing the
    right thing for the indirect branch speculation.

    Honestly, that's completely unacceptable too.

    > So the part is I think is odd is the IBRS_ALL feature, where a future
    > CPU will advertise "I am able to be not broken" and then you have to
    > set the IBRS bit once at boot time to *ask* it not to be broken. That
    > part is weird, because it ought to have been treated like the RDCL_NO
    > bit — just "you don't have to worry any more, it got better".

    It's not "weird" at all. It's very much part of the whole "this is
    complete garbage" issue.

    The whole IBRS_ALL feature to me very clearly says "Intel is not
    serious about this, we'll have a ugly hack that will be so expensive
    that we don't want to enable it by default, because that would look
    bad in benchmarks".

    So instead they try to push the garbage down to us. And they are doing
    it entirely wrong, even from a technical standpoint.

    I'm sure there is some lawyer there who says "we'll have to go through
    motions to protect against a lawsuit". But legal reasons do not make
    for good technology, or good patches that I should apply.

    > We do need the IBPB feature to complete the protection that retpoline
    > gives us — it's that or rebuild all of userspace with retpoline.

    ********.

    Have you _looked_ at the patches you are talking about? You should
    have - several of them bear your name.

    The patches do things like add the garbage MSR writes to the kernel
    entry/exit points. That's insane. That says "we're trying to protect
    the kernel". We already have retpoline there, with less overhead.

    So somebody isn't telling the truth here. Somebody is pushing complete
    garbage for unclear reasons. Sorry for having to point that out.

    If this was about flushing the BTB at actual context switches between
    different users, I'd believe you. But that's not at all what the
    patches do.

    As it is, the patches are COMPLETE AND UTTER GARBAGE.

    They do literally insane things. They do things that do not make
    sense. That makes all your arguments questionable and suspicious. The
    patches do things that are not sane.

    WHAT THE F*CK IS GOING ON?

    And that's actually ignoring the much _worse_ issue, namely that the
    whole hardware interface is literally mis-designed by morons.

    It's mis-designed for two major reasons:

    - the "the interface implies Intel will never fix it" reason.

    See the difference between IBRS_ALL and RDCL_NO. One implies Intel
    will fix something. The other does not.

    Do you really think that is acceptable?

    - the "there is no performance indicator".

    The whole point of having cpuid and flags from the
    microarchitecture is that we can use those to make decisions.

    But since we already know that the IBRS overhead is <i>huge</i> on
    existing hardware, all those hardware capability bits are just
    complete and utter garbage. Nobody sane will use them, since the cost
    is too damn high. So you end up having to look at "which CPU stepping
    is this" anyway.

    I think we need something better than this garbage.

    Linus




    So, looking at this, the fix to Spectre is going to be so costly in performance that Intel is leaving it up to people who may not know they have to switch it on to do so to be protected. Don't see this on the AMD side, but am glad AMD said, after the fact, they will be doing a microcode update to close off the possibility of the variant being discussed, even though they said they believe they have a near zero exposure to it. Happy Saturday!!!
    https://lkml.org/lkml/2018/1/21/192
     
    hmscott and Raiderman like this.
Loading...

Share This Page