Aiming to become the global leader in chip-scale photonic solutions by deploying Optical Interposer technology to enable the seamless integration of electronics and photonics for a broad range of vertical market applications

Free
Message: pigheaded re: Thunderbolt Junk

Apr 22, 2014 07:46PM

yup, apparently he's still playing deaf as a post.

is this a representative example of the resistance PC & Co. face from the industry?

http://www.extremetech.com/computing/181099-next-gen-thunderbolt-details-40gbps-pcie-3-0-hdmi-2-0-and-100w-power-delivery-for-single-cable-pcs#comment-1351430933

I wonder if he'll click the link I shot back at him, might do him some good.

is that what they teach in journalism school now? to make and judge the news, rather than to report it?

sad.

R.

---

pelov_lov

This is truly the only worthwhile thing Intel has developed in the last several years. Their CPU development has nearly stalled, their on-die GPUs are consistently 'meh', and their mobile SoCs are second rate...

Pat D. @pelov_lov

You can blame the CPU development stall on the sad state of the "high end" from AMD---their absolute best consumer CPU, the 9590, uses 220W and can't match a 2600K Sandy Bridge from 3 years ago (i know, I have both.). Oh, and it ships with a Water Cooler.

Heres hoping AMD has another K7 era...we need it to push Intel as much as to keep AMD in the game.

Joel Hruska @Pat D.

You really can't.

You can blame Intel's higher *prices* on the lack of competition from AMD, certainly, but seriously, think about the macroeconomics of this. The PC market sank by about 12% last year. The talk, everywhere, is that Intel will be devoured by a surging ARM as tablets catch conventional laptops.

Do you think Intel *doesn't* want to whip out an amazing CPU with a 20 - 30% speed boost in a single year? Do you really think they'd just ignore the opportunity to knock everyone on their ass and demonstrate who wears the REAL pants in the semiconductor industry?

Intel, generally speaking, sees Qualcomm as its hot competitor these days, not AMD. And it has spent a great deal of money improving its chips to compete in those markets. While it's had very limited success in doing so, the fact that you can buy "big-core" x86 chips that run as low as 12W is because Intel has focused on pushing conventional x86 processors into the lowest power envelopes it can engineer.

RobVanHooren @Pat D.

CPUs, even whole platforms that are legacy silicon: dead, you might want to consider alternatives.

Monolithic GaAs-based optoelectronics are coming like a freight train, at 100x the speed and 80% less power draw. Bolt-on to CMOS, so the fabs don't have to squirm in order to implement.

Good audio interview here (3 days old, sorry).

http://www.midasletter.com/201...

cheers,

R.

Joel Hruska @RobVanHooren

You continue to claim that a large collaboration of buzzwords will magically reinvent technology, despite the fact that nothing you discuss is even on the roadmap.

The ITRS lays out roadmaps for the semiconductor industry. They discuss the current best knowledge for implementation of graphene, CNTs, III-V's, nanowires, SiGe, and a host of other technologies. Everything from nanoprinting to self-directed assembly gets evaluated, discussed, and ranked.

http://www.itrs.net/Links/2013...

The individual chapters are 30-60 pages, and there are 16 chapters. It's literally the best, most comprehensive report on the state of the semiconductor industry that's made publicly available and open to all -- and your much-beloved optoelectronics aren't even a topic of discussion.

No one at Intel, TSMC, Samsung, GlobalFoundries, UMC, Chartered, or IBM is even *talking* about the kind of monolithic optoelectronic logic circuits you argue are right around the corner. Graphene and CNTs, in contrast, are on roadmaps that stretch out to 2030. Do you really think that a dark horse technology based on approaches that have been thoroughly investigated for 40 years is so new, so stunning, so capable of taking the world by storm that no other company has even bothered to *discuss* it at a tech conference?

Optoelectronics and GaAs have been billed as the Next Big thing since the 1980s.

http://bit.ly/QwRrfF

http://bit.ly/1jJmFIL

They're not coming on like a freight train. Hell, Intel is still working to implement silicon photonics as part of a frickin' cable standard.

RobVanHooren @Joel Hruska

you must have missed ERM chapter 6, ERD chapters 4,5, and 6, and pretty much all of the FEP.

you must also like crow.

thanks for being so open minded.

here's another link where you might learn something.

http://bit.ly/1gKqH69

enjoy,

R.

11
Apr 23, 2014 09:28AM
4
Apr 23, 2014 09:51AM
1
Apr 23, 2014 10:30AM
Share
New Message
Please login to post a reply