https://www.theverge.com/circuitbreaker/2017/10/17/16488414/intel-ai-chips-ne...

Intel unveils new family of AI chips to take on Nvidia’s GPUs





When the AI boom came a-knocking, Intel wasn’t around to answer the call. Now, the company is attempting to reassert its authority in the silicon business by unveiling a new family of chips designed especially for artificial intelligence: the Intel Nervana Neural Network Processor family, or NNP for short.

The NNP family is meant as a response to the needs of machine learning, and is destined for the data center, not your PC. Intel’s CPUs may still be a stalwart of server stacks (by some estimates, it has a 96 percent market share in data centers), but the workloads of contemporary AI are much better served by the graphical processors or GPUs coming from firms like Nvidia and ARM. Consequently, demand for these companies’ chips has skyrocketed. (Nvidia’s revenue is up 56 percent year on year.) Google has got in on the action, designing its own silicon named the Tensor Processing Unit to power its cloud computing business, while new firms like the UK-based Graphcore are also rushing to fill the gap.

Intel’s response has been to buy up AI hardware talent. It purchased vision specialist Mobileye this March; the chipmaker Movidius (the firm responsible for the silicon in DJI’s autonomous drones) last September; and deep learning startup Nervana Systems in August 2016. Since then, it’s been busy teasing this line Neural Network Processors, which were previously known under the codename “Lake Crest.” The NNP chips are a direct result of its Nervana acquisition and fold in the company’s expertise to achieve “faster training time for deep learning models.” (Intel says it also took advice from Facebook on the chip’s design — but didn’t give much detail.)

But how much faster exactly? Intel isn’t saying. While Google touted the launch of its latest-generation TPU chips by publishing head-to-head tests against rival hardware, Intel will only say that it’s on track to meet its goal of improving deep learning training speeds by 100 times by 2020. The company is similarly vague on when its NNP chips will be available to customers, though perhaps more details will leak out today. Some time before the end of the year in limited quantities is the expectation.


Yes, Intel spent good money to buy this 200nm "4 chips on a daughterboard" monstrosity that requires its own complete motherboard worth of excetera stuff just so they could say they had a really functional (but slow as dog shite) AI processor.   Intel was catching too much crap for showing off things THAT DID NOT WORK AT ALL -- so they bought themselves something that does work    (but just look at it --- it is glacially slow compared to modern ARM, AMD or Nvidia combined AI processor sets and the programming used for it is really PRE-historic compared to modern ARM standards).

If you look at where the rear motor mount is, it is near the swingarm pivot. Connecting the steering head to the S/A pivot in as straight a line as possible is a key part of making a bike handle. Gotta imagine not using the mount will make the bike handle worse.
My 2 cents.
-Armen

Today, some small news.

Samsung laps Intel yet again, this time by two generations not just one.   Yep, 8nm Samsung is shipping right now and Intel still doesn't have anything 14nm chipset that works.

Samsung laps 14nm Intel by two full generational levels, 10nm and 8nm.   If Intel manages to stay stuck on 14nm for another 6 months, that lap count could go up to 3 - 4 full generations as TSMC's and Samsung's 5nm and 7nm are coming out VERY soon.

EVERYBODY else is at 10nm and kicking down on their next level lower.

Intel is deserving all the abuse anyone wishes to heap on them at this point.    
yep, it really is the same old lighthouse I've been showing you for 4 and a half years now -- man that ice is getting really really thick now.

 Can you tell which side of the lighthouse the lightkeeper likes to pee from?

It much more fun to capture one , that might be the only time I'm more focused than when I'm riding. Just the tiniest distraction and your bit. I think it's more dangerous letting them go than it is catching them. Cause by then the snake is really pissed off and waiting for its chance for revenge. But I totally understand why people kill them , Fear & Ignorance !
Overcoming fear is tough , overcoming ignorance close to impossible. I realize their danger and how easy it is to kill them rather than relocating.
I advise not messing with any snakes unless you know what you are  doing. Even a large non poisonous gopher snake can really put some hurt on you.
Just watch them , most of the time they will be on their way once they see you , some harassment helps too. Most snakes won't leave if you are a threat , too close , just back off a let them be on their way.

My daughter brought over her Savage because it wouldn't start. Gas would puke out the breather hose when cranking. So I checked the oil level and it looked murky in the sightglass. Opened the oil cap and the crankcase was full to the top with gas and oil! Drained it and it had 2 gallons in it. I'm gonna say it needs a new float seat and needle valve? That's why it's important to shut the gas off when parked I told her.  

LANCER wrote on 06/29/17 at 13:47:46:


Is your tank still for sale? I just received my aftermarket tank today, but I don't think I'm going to keep it.

I always got the line from Harley guy's that they bought a Harley" cause they keep their value",that may have been true 15-20 years ago ,but it's a joke now,there's one for sale on every corner .

my first wife   ?

More years ago than I care to admit I topped 100 on a triumph Ton Ten
in England. It seemed fast at that time. But as we all know now not so fast. :'(

FWIW. My bike will do 90 indicated. What the true speed is I have no idea and I don't care much. It feels fine at that speed. Mods, see below.