Do you have reading difficulties? RTS window in this slide goes from week 50 2016 till week 04 2017 which is cleaarly inside week 01 2017 when KBL-S officially starts. The only important message for you is that RTS reflects the launch/sales start.
Im not talking about RTS launch, im talking official sales embargo lift dates (retail launch of the product, when we have the reviews and we can buy the actual product). Kabylake official sales embargo lift is Q1 2017 no matter if RTS is Q4 2016. Officially KBL Desktop havent been launched in the retail, sales embargo is not lifted yet, you cannot go to MicroCenter or any other big retailer and buy a KBL-S product in 2016 period.
http://www.bitsandchips.it/english/52-english-news/7854-rumor-even-intel-is-studying-a-new-x86-uarchAccording to our sources – the same who reported the development of Zen two days before official AMD public announcement – Intel is studying a new uArch in order to replace the current x86 uArchs in Desktop and Enterprise market.
Doesn't really seem like a bad idea to me. It's still like half a decade out, so yeah.Intel will throw away some old SIMD and old hardware remainders.
Intel's own compilier, targeting ARM?Intel will be able to develop a thin and fast x86 uArch, and ICC will be able to optimize the code both for ARM and for x86 as well.
This new uArch will be ready in 2019-2020.
If they go alone it sounds like Itanium very hard... But if they join forces with AMD and VIA to create x86-x128 (bits)... That would change everythingIntel's own compilier, targeting ARM?
That seems awful quick for deployment of a new arch, too. Three years? Seems way too quick for Intel, seeing as how long it takes them to develop even a minor revision of their existing arches.
We don't know how long Intel's been working on it.Intel's own compilier, targeting ARM?
That seems awful quick for deployment of a new arch, too. Three years? Seems way too quick for Intel, seeing as how long it takes them to develop even a minor revision of their existing arches.
It's not a new arch if they just throw out all the "useless" instructions to make a core that's more like ARM,specialized perf/power efficient at all it can do and pretty sucky at everything else.Intel's own compilier, targeting ARM?
That seems awful quick for deployment of a new arch, too. Three years? Seems way too quick for Intel, seeing as how long it takes them to develop even a minor revision of their existing arches.
If he has legitimate sources he should be able to provide a code-name for the first 7nm parts, since the three architectures after Tiger Lake do in fact have codenames now (SemiAccurate published them). If he can't even provide a codename, then I call BS.
What are the codenames??!!!If he has legitimate sources he should be able to provide a code-name for the first 7nm parts, since the three architectures after Tiger Lake do in fact have codenames now (SemiAccurate published them). If he can't even provide a codename, then I call BS.
Do we have these code names or is it behind a pay wall? Semiaccurate was wrong the last time afaik, but I'm curious.
Do they continue with the lakes?Sorry, it's behind a pay wall so can't say what they are, but they seem pretty legitimate to me.
The only fact he says "ARM is lacking useful SIMD" proves the guy is clueless.
I was so hyped when Intel hyped up this new SSD tech one and a half years ago (1000x faster this, 1000x greater that, higher density, blah, blah) but in the time since,
How do you go from hyping 1TB in a single m.2 card to 16GB?
Do they continue with the lakes?
Based on an OEM Roadmap they don't.
Sapphire Rapid looked like nonsense to me but Semiaccurate wrote Intel goes Pokemon which is somehow related.
What is Tinsley?Interesting roadmap, how reliable do you think this is?
There is no CNL here and the implication is that Ice Lake is part of Purley rather than Tinsley.
What is Tinsley?
The source is reliable. I can share the source with you if you want, posting the link in here is not a good idea because Intel is hunting.