Navin Shanoy on Intel Ice Lake Server: "Literally we got Ice Lake silicon back just a couple of weeks ago...we're very excited about Ice Lake coming to the market in 2020"
Rusty Toms retweeted
Gearing up for a Zen 2 disclosure. Bring it on, @AMD!
anandtech.com/show/13538/amd…
Jim Cramer, interviewing Lisa Su, asking good questions: $AMD's products are good, why didn't they gain more share third quarter when Intel was struggling so much? cnbc.com/video/2018/10/25/am…
Hey all you @ChipCollective people, when are you going to start using carbon nanotube sheets instead of thermal grease :D
fujitsu.com/global/about/res…
Rusty Toms retweeted
New GeForce Driver Fixes Performance Issue: Threadripper 2990WX Tested dlvr.it/Qjqv2H
Rusty Toms retweeted
GeekBench creator says Huawei is artificially boosting its benchmarks, too. I guess it could get worse. / Huawei Benchmark Boost - Geekbench buff.ly/2x1flaR
Rusty Toms retweeted
#Intel can't supply 14nm Xeon-SPs. HPE directly recommends AMD Epycs as a result. In their own words, with screenshots. $INTC semiaccurate.com/2018/09/07/…
Rusty Toms retweeted
Huawei & Honor's Recent Benchmarking Behaviour: A Cheating Headache anandtech.com/show/13318/hua…
@chiakokhua starting new thread. IMHO, small die sizes are a core of $AMD 's strategy. Assume parity among fabs, how long does it take to get from 30% yield to 90% yield? If AMD dies are 1/4 size of $INTC 's, then when Intel is at 30% AMD is already at 86%, shipping in volume
So just with the small die size, they can grab x months of lead. If they are behind, they catch up, if they are ahead, they leave Intel in the dust.
If they can somehow do this with their graphics (I don't know anything about this), they can get a huge process/cost advantage over $NVDA with their humongous dies
"AMD's first 7nm product is now slated to arrive later this year with the company's next-generation Radeon Vega Instinct GPU, which was demonstrated at this year's Computex and comes with 32 GB of memory." crn.com/news/components-peri… $AMD