Broken Silicon 270 (Video): AMD Zen 5 Ryzen 9000 Review, Intel Nova Lake Issues, Raptor Lake Microcode Test
🕑 Added 2024-08-12 22:06:47 +0000 UTCWe discuss the latest Zen 5 releases & discuss what’s going wrong with Intel’s Royal Core project.
0:00 Growing Pains & YouTube Shorts (Intro Banter)
3:18 Microcode Distribution, Zen 5 Geomeans (Corrections)
5:47 R7 9700X & R5 9600X Release to Polarizing Reviews
20:28 Zen 5 Utilization & Pricing
30:52 Will Zen 6 be on AM5? WTF is up with 800 Series Mobos?
33:13 5900XT & 5800XT Reviews
37:52 Strix Power Scaling & AMD Kraken Whispers
46:34 Intel Reports Disastrous Q2 2024 Earnings
57:54 Beast Lake is Cancelled & Nova Lake is in TROUBLE
1:02:07 Intel 0x129 Microcode Update Tested
1:09:43 Nvidia Blackwell (is not) Delayed due to Design Flaws
1:12:20 Rogue River Forest, Lunar Lake Release Date, RDNA 4 Incoming (Wrap-Up)
1:23:46 GPU market Demands, GDDR6 RTX 5000 Variants, AI Cadence (Final RM)
https://videocardz.com/newz/sk-hynix-confirms-gddr7-memory-will-enter-mass-production-in-third-quarter
https://videocardz.com/newz/intel-core-ultra-200v-lunar-lake-launches-september-3rd
https://www.tweaktown.com/news/99792/amd-is-hosting-mysterious-gamescom-event-on-august-23-new-radeon-announcement-coming/index.html
https://steamdeckhq.com/news/new-steamos-beta-update-desktop-image/
https://www.techspot.com/review/2877-amd-ryzen-7-9700x/
https://www.techspot.com/review/2878-amd-ryzen-5-9600x/
https://www.tomshardware.com/pc-components/cpus/amd-ryzen-5-9600x-cpu-review
https://www.phoronix.com/review/ryzen-9600x-9700x
https://www.techpowerup.com/review/amd-ryzen-9-9700x-performance-smt-disabled/
https://www.tomshardware.com/pc-components/cpus/amds-ryzen-9-5900xt-ryzen-7-5800xt-launch-today-for-dollar349-and-dollar249-respectively-existing-ryzen-5000-is-less-expensive
https://www.youtube.com/watch?v=11FWyDiT8bE
https://www.youtube.com/watch?v=24x_EE_zN2o
https://youtu.be/WyRb1rOZytw?si=ZQKhReOQthTHUVEz
https://www.intc.com/financial-info
https://www.youtube.com/watch?v=b6vQlvefGxk
https://www.reuters.com/legal/intel-is-sued-by-shareholders-alleging-securities-fraud-2024-08-07/
https://youtube.com/live/igDHiEbKGwY?feature=share
https://youtu.be/1bEv74JrHQo?si=sY7evVyoCwq8YhCl
https://www.phoronix.com/review/intel-raptor-lake-0x129
https://youtu.be/SMballFEmhs?si=ZM615s6YtJS0k_V2
https://x.com/mooreslawisdead/status/1820735159085277494
https://www.semianalysis.com/p/nvidias-blackwell-reworked-shipment
https://www.tomshardware.com/pc-components/gpus/nvidia-reportedly-selects-intel-foundry-services-for-chip-packaging-production-could-produce-over-300000-h100-gpus-per-month
https://x.com/OneRaichu/status/1822671710015131772
https://www.pcmag.com/news/amid-layoffs-and-cpu-controversy-intel-postpones-innovation-event
https://videocardz.com/newz/sk-hynix-confirms-gddr7-memory-will-enter-mass-production-in-third-quarter
https://videocardz.com/newz/intel-core-ultra-200v-lunar-lake-launches-september-3rd
https://www.tweaktown.com/news/99792/amd-is-hosting-mysterious-gamescom-event-on-august-23-new-radeon-announcement-coming/index.html
https://steamdeckhq.com/news/new-steamos-beta-update-desktop-image/
Comments
Moore's Law Is Dead
Just release ONE 192-bit die every year to the low end, and aim for it to always be at least 50% faster than their best GPU tile. If they do that, they can use that to scale GPU performance ahead of the weaker APUs (they hit a scaling wall with Broadwell because they weren't ever driving for radically higher performance), and also ensure they have good drivers for all of their APUs. However this would need to be very low volume, and not waste costs on tons of marketing and silly expensive elaborate coolers. ^That's what I think they should do at most...and the truth is if they can't afford to do this, then they should just partner with Nvidia to produce GeForce tiles on Intel nodes. That would be WAY more cost effective, and hell - I'd love to buy a Nova Lake laptop that has a built-in Nvidia tile on the APU.
Lucas
What do you think Intel should do with their floundering gpu business? Just get rid of it and focus on their core business or go full sunk cost fallacy and try to turn this mess into something worth buying?