Intel tests definition of insanity with (leaked) typoslab Skylake CPUs The definition of insanity is doing the same thing over and over again, but expecting different results. And it appears Intel is testing that definition – if these internal Intel Powerpoint slides that made their way to a blog called FanlessTech this month are the real thing. The slides suggest that Intel is still plugging …
... might well be making the same mistakes over and over again and expecting different results --- but nearly every form of success other than 'striking it lucky' comes from trying things again and again until they work - whether it's creating and marketing products, kung fu or musical performance.
The quoted definition is a horribly poor one anyway - it only applies if everything concerned is deterministically influenced solely by the user input and nothing else; if any uncontrolled variables are present (yeah, welcome to the Real World) the definition of insanity would be rather to expect the same thing to happen every time...
"The quoted definition is a horribly poor one anyway" -- DropBear
Yeah, it's a particular bugbear of mine. At least it wasn't mis-attributed to Einstein, which is the usual state of affairs with this (mis)quotation. It actually seems to have come from an Narcotics Anonymous pamphlet (page 11 [25 of 68] of this pdf) which contains the more sensible "repeating the same mistakes ..."
Anyway, every time I see it, I have to complain: it's right up there with "everything happens for a reason" and "everyone's entitled to their opinion." I keep thinking there should be a pithy retort along the lines of "you clearly can't tell your perseverance from your perseveration" but I'm still looking for something effective.
Intel likes to sell in the overpriced segment. All well and good for a monopoly, but these days there is quality competition at lower prices. Factor in Apple and Samsung making their own CPUs, and Intel can only sell to third place OEMs. Those OEMs have to aim for the biggest segment for economies of scale rather than Intel's prefered customers.
"Intel likes to sell in the overpriced segment."
WTF ??? Are you from the same tribe that keeps slagging Intel off for selling tablet parts below cost and undercutting AMD in the "entry level notebook" segment?
So wake me up when Samsung, Apple, Qualcomm or Nvidia get a share of the server market that exceeds Intel's share of "mobile" :-)
I think Intel is going to always struggle prising ARM from the mobile market, ironically for the same reason no-one has moved the Intel architecture from PC's. There is too much investment by mobile manufacturers in the ARM eco-system and they like the control ARM provides over the SoC. for example, is Apple going to handover it's special sauce for making chips to Intel for it's iphone? Probably not
However there are other markets where such chips would be great. Industrial control solutions and IoT for starters. . However they are not sexy applications and Intel seem very reticent to use there most power efficient cores in these areas allowing AMD and lately ARM to make headway in these areas.
"However there are other markets where such chips would be great. Industrial control solutions and IoT for starters"
IMO, IF they are serious, they really should be looking pushing RAS features such as ECC etc. The IoT crowd seem to forget that stuff like light switches are *reliable* and offer consistent performance, something not easily replicated by lowest-possible cost hardware. :(
Its about time some of the reliability enhancing technologies used in servers migrated down into 'user' devices.
You can't have been around for long. It was only a decade or so ago that Intel stopped selling Desktop chipsets with ECC support. (Cure 2 Duo CPU era). ISTR that AMD chipsets still do support ECC, although for some reason most MoBo manufacturers disabled it at the board level. Certainly it wasn't many years ago I constructed a microserver around an AMD Sempron rather than anything Intel for this reason. It didn't need speed but it really did need not to corrupt data without anyone knowing that it was failing.
Doesn't happen often, but I have seen filesystems trashed by an undetected RAM failure. I also wonder why O/Ses don't try to protect themselves by running a memory tester on unallocated memory pages, maybe one minute per hour whenever the system is lightly loaded or idle. Nowhere as near as good as ECC, but better than doing nothing and hoping for the best.
It is true that SOC vendors and peripherals makers collaborate using ARM, Linux and Android to rapidly iterate innovation in the simulator and prove it against a suite of millions of apps for several generations between actually etching chips. Ain't gonna get that from Intel.
So they will again sell these for the never popular Windows slab - 19 years failing. And pair it with the sort of low resolution display you can get on a $50 Android slate. And sell it in the $800 and up tab space that belongs to Apple. And of course all the other features will be wrong too.
I wouldn't mind it in a 9" QHD Ubuntu slate, at a price comparable to what I can already get. This will happen when pigs fly.
Edit: and before you pooh pooh the idea, at least it has the advantage of not already having been tried and failing miserably for long enough to barrel a good whiskey.
So... are Intel supposed to stop? I mean, it's what they do - develop CPUs. One could argue that tablets and mobile sales are dropping, but they aren't going to stop are they? PCs still sell regardless of whether the doomsayers label them dead.
What next? Ford to stop developing cars because... insanity?
Once upon a time Windows was 90% of the smart devices market and the argument was that Intel would allocate 90% of their software platform support to it because of market realities, not technical ones. "Pragmatic and hard numbers based" is an ironclad defense for any Intel proposal. Well this year less than 25% of smart devices will have Windows on and yet Intel is still allocating 90% of their platform support to it, defeaturing their hardware for competing OS platforms. Maybe they could reevaluate their position on this issue.
Well, these chips don't appear overnight- I imagine something like this:
2013, Intel boardroom: Wow, these tablet things are really taking off! We need a better chip for tablets, ASAP!
Intel engineer: We'll get right on it!
*
2015, Intel engineer: Well, we had to work like crazy, but here's that chip you wanted for tablets.
Intel boardroom: For what, now?
They'll selling them for Surface Pro. You can call that a tablet if you want, but it is really a laptop with a really poor keyboard since I've never seen one used in any other way.
I doubt Intel will continue with their giveaways of these chips trying to get them into the Android tablet market, especially as the Android tablet market is shrinking.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
