eplt tech blog
Nvidia has been performing exceptionally well in recent times. Over the past 20 months, the increased demand for AI compute power from chatbots (such as ChatGPT) and Generative AI for…
#Apple makes it seem so simple to just put 2 CPUs into one package and double the performance. Why didn’t #Intel or #AMD do the same? apple.com/newsroom/2022/…
The challenge for #Intel is not about winning the performance race with #AMD or #Apple every now and then. But can they consistently supply to these annual upgrade cycle now…
Not that easy for Chinese companies to win in the x86 or ARM space. Maybe in some IoT and ASIC markets. protocol.com/china/chinese-…
It took #Apple about 10 years to build their own SoC well enough to ditch #Intel. How long will it take for #Huawei to do the same? Will US restrict…
Software should be okay. But there aren’t many good or popular open source processor architectures to base on, and hard to reinvent one quickly. Can’t buy Intel or AMD, still…
Waiting for the first 900pm (picometer) chip process. Or 9 angstrom (Å). We will get there. 😀 zdnet.com/article/samsun…
Will this have serious impact to #Apple SoC roadmap? Not likely. cnet.com/news/apple-los…
Guess AMD and Nvidia can never play nice together. Not just about graphics, but will push other firms to build their own AI chips, making GPU less important. engadget.com/2019/01/07/nvi…
“Just 20 million”, that’s like $20B of revenue. #Pixel or #Essential will be lucky to do just a tenth of that. 9to5mac.com/2017/10/24/20-…
Amazing. Just 8nm now. 35 years go, 286 processer (the first popular PC x86) was 1,500nm (over 200x bigger)! engadget.com/2017/10/18/sam…
Why can’t #Samsung focus on their own chip? Is the #Exynos not good enough? trustedreviews.com/news/samsung-g…