Holy Crap you are actually quoting _Politifact_ for your counterpoint!

Problem for Intel is that RAM is about 5 bucks per gigabyte. This is about 10x slower than RAM but only 2x cheaper. So if this is targeted at "workstation" users, then in that class a lot of these will have motherborads capable of terabyte or more RAM.

Though the logo was first plagiarised 15 years ago: https://twitter.com/noonanjo/s...

You're gonna hate cool-retro-term https://github.com/Swordfish90...

I vape. I find that vaping requires a more regular "hit" than actual cigarettes, so roughly every hour as opposed to every 1.,5-2 hours or so. I find that the act of walking down and back up the stairs, and getting some "fresh" air, not only helps my productivity through brain rest, but also because somehow, complex coding problems sometimes "come together" better when I'm not in front of the code. Not suggesting anyone addict themselves to nicotine, because you could achieve this without smoking/vaping, but the act of doing something completely different does seem to allow my brain to subconsciously put stuff together somehow.

uses a lot more RAM though.

Why is everyone moaning? Our leaders' policies will likely light us all up like Krakatoa before we hit 40. Might as well light up and enjoy because the way things are going, smoking is probably the last thing that'll kill you.

Electric planes' biggest problem is at takeoff. They need ginormous runways to gather enough speed. Cuts out 90% of small airports.

Even if it doesn't come from AI, surely humans themselves will soon realise that there are too many of us around. Cue reversal of the whole "love one another thing" to the exact opposite - exterminate one another to survive.

Any chance Linux will work on this thing? Looks like a sweet terminal.

Ad hominem attacks on product leadership are an effective marketing strategy. They're the converse of celebrity endorsements. Irritating but effective.

didn't he say "cluster"

It's TFLOP at 16 bits. AT 32 bits it will be 500 GFLOP. Apples for Apples Geforce 980 will do 5 Tflops so 10x more compute. At much less than 10x the price. Actually almost the *same* price. Per watt you'd have a slight advantage at the GPU level but once you consider having to buy 10 boards just to compete that advantage would vaporize too. Then you have to consider memory bandwidth where the 980 will crush this device natively, and of course even more if you consider having to distribute work via ethernet on the cluster.

Have you considered that consultants need to have a machine that's on the go *and* is usually plugged into the wall, client side? That datasets are often huge that can be visualized on a 4k screen? That 32 gig is actually sometimes even not enough? If you're pecking around on Facebook at starbucks, I hear you. But in the real world of high end engineering, finance, bio, 32 gigs is small. I use my computer to pitch machine learning algos to financial clients. I have a lots of competition. There is a growing market for this.

Very disappointed that these machines don't support at least 32 gigabytes of RAM, especially the 15inch. We've been stuck at 16 gig now for like 5 years in this class of machine. I'm hoping the new "retro thinkpad" hopefully out next year will be the machine of choice for developers.