Live data from Hacker News

Viewing profile — n7pdx

n7pdx

HN member
Joined
Fri, Jan 07, 2022, 9:16 PM UTC
HN karma
87
Public activity
36 items

About n7pdx

No profile information was provided.

Recent public activity

  1. comment
    Comment #35110805

    As toxic as Brian Krazinich was, you can’t really blame one person for the problem at Intel. Yes, he and Murthy initiated the layoffs. But then the orders were executed by mid/seni…

  2. comment
    Comment #33754949

    There’s another side to the Intel 10nm process debacle. Intel designers and architects have relied on a process and manufacturing advantage for years to overwhelm any problems they…

  3. comment
    Comment #33688235

    LOL I remember working on Cannonlake A0. The dozen (not plural) parts we got passed sort but all bricked. Turns out the PLL was randomly losing lock due to variation alone. Also th…

  4. comment
    Comment #33191672

    This will age poorly. Performance king at unconstrained power is utterly irrelevant from a financial point of view, and requires very little engineering effort to boot: shoving amp…

  5. comment
    Comment #33158229

    LOL yeah. He lured a bunch of his old buddies from retirement with fat packages as if those old Intel-lifer hacks had any fresh ideas to contribute. Pat is literally an Intel swamp…

  6. comment
    Comment #33134133

    Less cache, much smaller OOO window, AVX-512. E-core is an area play so Intel can score some multi-thread benchmarketing wins on integer heavy code that don’t have larger memory/ca…

  7. comment
    Comment #32727778

    Bunch of nonsense. The issue is mismanagement and incompetence. Charlie got suckered by Intel management talking points.

  8. comment
    Comment #32628823

    Intel fanboys whine about Apple/Intel CPU comparisons, here is their answer. The takeaway from the story: connecting a bunch of shitty chiplets together makes a shitty SoC. Except …

  9. comment
    Comment #32623868

    LOL. Anyone who works in chip design would know much M1/M2 changed the hardware game. It is actually the dabbler enthusiast talking about how power/perf isn’t important because his…

  10. comment
    Comment #32623741

    They know it matters, they just don’t have the competence to do it since they promoted a bunch of toadies and charlatans into their technical leadership, and also outsourced a ton …

  11. comment
    Comment #32378145

    It isn’t. TSMC has multiple customers lined up with the volume required to ramp these processes to high volume, Intel only has their own designs and whatever foundry customers will…

  12. comment
    Comment #32293025

    Trusting Intel to be the torch bearer of Western semiconductor manufacturing is a grievous mistake to begin with. Intel is crippled with incompetence and corruption, government mon…

  13. comment
    Comment #32288276

    My org lost a few people to Intel. They all did their research and know full well the technical and managerial leadership at Intel are total clowns and that Intel is finished. So t…

  14. comment
    Comment #31982525

    Looks bad. I have two examples to go with: - The Sapphire Rapids validation catastrophe proves they haven’t learned a damn thing about high quality first silicon even after all the…

  15. comment
    Comment #31979848

    lol this will age well. i remember designing CPUs at intel when they had like a ~2.5 gen process lead (literally 4-5 years) and the result was still barely better than competition.…

  16. comment
    Comment #31800936

    You are completely over-complicating things. P6 derivatives stuck with “traditional” Tomasulo algorithm for way too long. Everyone else moved to pointers for arch regs long before …

  17. comment
    Comment #31800871

    No, it wasn’t. Sandybridge started with the Nehalem code base. I was there when it happened.

  18. comment
    Comment #31461158

    Yeah, they are done. Worked there for 15+ years in the CPU design org. The total collapse of the technical cadre in the 2010s will be studied in business schools as a case study of…

  19. comment
    Comment #31389533

    And when it doesn’t, will you apologize for calling everyone else a fanboy because they don’t buy into your excuses? Go compare the perf/power curves of the alder-lake cores (both …

  20. comment
    Comment #31389087

    LOL yeah. More powerful, i.e. it consumes more power. Fun fact: the M1 big core caps itself around the same wattage where most Intel SKU consider the core to be below stock wattage…

  21. comment
    Comment #31388085

    "Dunno why every comment implies there's some design gap that leads to M1 or AMD being more powerful or efficient." That is because there is a huge design gap. Intel architecture a…

  22. comment
    Comment #31298990

    Will give him less than two years. High level individual contributors like him are absolute poison to the Intel bureaucracy.

  23. comment
    Comment #31231260

    Because Intel is too incompetent to design a CPU that can perform at 1 watt, which is what it takes to run 4 cores in a fanless chassis.

  24. comment
    Comment #30738408

    Even if you ignore the software world which has been easily outpacing hardware compensation for the last decade, if you only look at hardware companies, Intel has been either at or…

  25. comment
    Comment #30736826

    Yeah agreed. I was there through the high point, through when the company committed suicide and finally resigned when the corpse started to stink publicly. Blaming everything on th…