Catch up on stories from the past week (and beyond) at the Slashdot story archive

 



Forgot your password?
typodupeerror
Check out the new SourceForge HTML5 internet speed test! No Flash necessary and runs on all devices. ×

Comment Re:Banish cars from the city center (Score 1) 174

Allow service vehicles, public transportation, cabs and bicycles, and everyone will be happy.

Okay, sure. Just as soon as I figure out how to get my SUV registered as a "service vehicle"...

Don't even try to tell me it won't happen... There's widespread fraud just in handicapped placards, because parking a few feet away is too much hassle for some people. You think the ultra-wealthy will resign themselves to riding the bus with the peasants? Not a chance, they'll find any means to maintain their status, privilege and convenience.

Comment Re:Qualcomm doesn't make chips (Score 1) 90

My understanding is most server farms are connected to dedicated nuclear power plants anyway, so power consumption isn't an issue. Heat dissipation? Yeah, that might be an issue.

Heat and power are the same issue. The conservation of energy means that power in is power out, and the power out is heat that needs to be dissipated. A rule of thumb for data centres is that every dollar you pay in electricity for the computers, you need to pay another dollar in electricity for cooling. If you want high density, then you hit hard limits in the amount of heat that you can physically extract (faster fans hit diminishing returns quickly). This is why AMD's presence in the server room went from close to 100% to close to 0%: Intel was much better at low power.

Comment Re:Intel 10nm != Other Foundry 10nm (Score 1) 90

Intels problem is that it cannot sell FAB time because they are vertically integrated

This is true. Intel will fab chips for other people, but they've had very few customers because everyone knows that the priority customer at Intel fabs is Intel and if yields are lower than expected it won't be Intel chips that get delayed.

Intel builds a FAB and runs its next gen chips off of it for a few years, then they are stuck looking for something to do with the FAB when it is no longer current-gen

This is simply not true. Slashdot likes to think of Intel as a a CPU vendor, but that's actually quite a small part of their business. They make a lot of other kinds of chip and a great many of these don't require the latest and greatest fab technology. This has always been a big part of their advantage over AMD: they have products that will use the fab for 10+ years, so they can amortise the construction costs over that long a period.

TSMC's revenue is now approaching Intel's, and unlike Intel they can keep all their FABs busy making money, so the outlook for Intel is grim without a serious restructuring, which they are doing (see recent massive layoffs, and bullshit marketing about their new "cloud strategy")

This is the important part and is where the ARM ecosystem has an advantage over Intel. No single processor vendor has to compete head-to-head with Intel. As long as the total size of the ecosystem is large enough, the foundries can invest in process improvements.

Comment Re:ARMing servers. (Score 2) 90

AMD had a unique market opportunity to build up a good manufacturing base w/ quality fabs for their CPUs, but didn't. Intel gave top priority to their fabs, and are the standard

AMD spun off their fabs for precisely this reason. Building fabs is insanely expensive and the only way to do is to amortise the cost over a lot of chips. Even at its peak, Intel was producing 4-5 times as many CPUs as AMD and had a load of lower-end products (e.g. network interfaces) that they'd start using the fabs for once they were a generation old. There was absolutely no way for AMD to compete head to head with Intel in fab technology, because they couldn't get the economies of scale.

This does; however, highlight just how bad Intel is at CPU design. AMD has been able to achieve rough parity for decades (and been ahead a couple of times, with the original Athlons and Opterons) in spite of always being at least one process generation behind in fabrication technology.

Comment Re: It takes a LOT of cache and very clever data p (Score 1) 90

Linked lists are just traditionally implemented linked lists. Hash tables are just traditionally implemented hash tables

Linked lists suck for caches, but hash tables don't have to. There's a trend for libraries to provide things like hopscotch hash tables as the default hash table implementation and these are very much cache aware. The real problem is the trend towards languages that favour composition by reference rather than by inclusion, which means that you do a lot of pointer chasing, which is very bad for both caches and modern pipelines.

Comment Re:But... (Score 1) 175

I think that this is less of a problem in Paris, but it's crazy in some other places. I used to live in Swansea, and for the last few years I lived there the cost of a day ticket was less than most fares (which the drivers knew, so they'd give you the day pass if you asked for most things). At £2.30/day, it wasn't too bad, but for 3-4 of you it was often cheaper to get a taxi. We went back a few months ago and it was cheaper for one person to get a taxi for shortish hops than to take the bus and the cost of the day tickets had gone up enough that it wasn't worth it.

The city council had spent millions remodelling the city centre to allow larger bendy busses when I was there. I never saw one more than 50% full and apparently a year or two ago they discontinued them. If they'd spent the same amount of money on more frequent, subsidised, minibuses, treating it as infrastructure that encourages people to do things that raise tax revenue rather than as a profit centre, then they'd have had far more people using public transport. Instead, privately owned bus companies have made a lot of money and it's now at a point where it's cheaper to drive than to take the bus.

Comment Re:But... (Score 3, Informative) 175

Every week I drive to the supermarket and pick up 20-30 kg of stuff[*].

Why do you do this? I haven't done a big supermarket shop in person for over 10 years. It takes 10-20 minutes to drive each way, an hour wandering around the shop, I have to queue for the checkouts, and it's just a horrible experience. All of the major supermarket chains deliver and it takes about 10-20 minutes to do the shop online (5 minutes for a routine shop where I'm just adding stuff from my favourites) and then it's delivered to my door, by a van that's delivering to a dozen other people on the way.

I'm doing pretty well just to walk through the store and COLLECT the stuff. And no, nobody will deliver it, even if I had two pennies to rub together to pay them with.

Delivery from most supermarkets here is free and even from the rest it's far cheaper than the cost of driving there, even if you don't factor in the cost of your time.

Comment as it begins, so it ends (Score 1) 176

I'm not going to blame Fitbit. But they didn't earn any positive karma, either.

Nor am I surprised that the real reason came out of the blue, after it was too late to inform my decision (I would have bought one for spare, while supplies lasted).

Emphasis with Kickstarter is "start". Then it quickly morphs (usually) into the same old close-to-the-vest business wisdom, and you end up with half of the advantage you wished for, and a quarter as much stability as a going concern.

I have zero interest in any other smart watch.

Sigh. It's a sad thing.

Comment three large screens (Score 1) 62

Carousels can suck it.

I've whipped up CSS Userscripts to remove them from web pages more than once, lest I click on one in a moment of weakness (which I always regret 3 s later).

Old motto: there's another fish in the sea. This maxim is true, also, regarding web content. But it often helps to enforce this programmatically.

Comment ctrl-v strikes again (Score 1) 62

Google's main product is advertising and user info (to better target advertising), not search.

And this tired observation moves the discussion forward how exactly?

I've seen this helpful ctrl-v "eyeballs are the product" contribution more than a 100 and probably less than a 1000 times since I joined Slashdot.

Add some useful context? Ah, fuck it. ctrl-v has miles to go before it sleeps.

Comment Re:Qualcomm doesn't make chips (Score 2) 90

You're entirely right that the memory subsystem is 90% of the battle for most server workloads once you exceed ten cores.

For integer workloads with unreasonable parallelism and unreasonable cache locality (that Intel's AVX doesn't already handle almost ideally), I'm sure this design will smoke Intel on the thermal management envelope, a nice niche to gain Qualcomm some traction in the server mix, but hardly a shot heard around the world.

And Qualcomm better be good, because Intel will soon respond with Omni-Path Knights Hill—perhaps also larded with HBM—that could probably take on the same workload between power sprints (less power efficiency in the CPU itself—which isn't always the main power draw—and probably more flexible as part of a tidy one-vendor-rules-them-all server mix).

I'm all for vendor diversity, but let's not get ahead of ourselves thinking that 10 nm levels the playing field, sucking down the data aquifer through a double-wide handful of drinking straws.

Yes, core count matters, but size matters even more when it comes to the hose.

Looky looky, the bow moveth:

Intel announcements for AI: Nervana 100x faster than GPU, Knights Crest & Mill 4x faster, SKL mid-17

Kx Streaming Analytics Crunches 1.2 Billion NYC Taxi Data Points using Intel Xeon Phi

Slashdot Top Deals

Matter cannot be created or destroyed, nor can it be returned without a receipt.

Working...