Please create an account to participate in the Slashdot moderation system

 



Forgot your password?
typodupeerror

Comment: And form talking to our researchers (Score 1) 96

by Sycraft-fu (#48901299) Attached to: Ask Slashdot: GPU of Choice For OpenCL On Linux?

Between a bit better language design and superior support and tools, CUDA is way easier to do your work in. We've 4 labs that use CUDA in one fashion or another, none that use OpenCL. A number have tried it (also tried lines like the Cell cards that IBM sold for awhile) but settled on CUDA as being the easiest in terms of development. Open standards are nice and all but they've got shit to do and never enough time to do it, so whatever works the easiest is a win for them.

On a different side of things, I've seen less issues out of nVidia on CUDA than AMD on OpenCL for video editing. Sony Vegas supports both for accelerating video effects and encoding. When I had an AMD card, it was crashes all the time with acceleration on. Sony had to disable acceleration on a number of effects with it. I had to turn it off to have a usable setup. With nVidia, I find problems are very infrequent.

Obviously this is one one data point and I don't know the details of development. However it is one of the few examples I know of a product that supports both APIs.

Comment: Re:Insurance (Score 4, Informative) 181

by Rei (#48898643) Attached to: Calif. DMV Back-Pedals On Commercial-Plate Mandate For Ride-Share Drivers

That falls into statistically normal usage. Being a commercial driver absolutely does not. Statistically, a commercial driver drives way more than a noncommercial driver, and they're much more likely to be sued, and for more money. It's absurd to argue that they should be able to drive on insurance rates calculated for statistical norms of noncommercial drivers. If you allow that sort of ignoring of statistics then you might as well get rid of all statistical tables period and charge every last person the same rate for all types of insurance.

Comment: Insurance (Score 4, Insightful) 181

by Rei (#48898577) Attached to: Calif. DMV Back-Pedals On Commercial-Plate Mandate For Ride-Share Drivers

Why, exactly, should Uber drivers get to drive passengers using regular non-commercial drivers' insurance? Commercial insurance costs more because people who drive people around for a living are much more likely to cost the insurance companies more money. If you're letting them drive on non-commercial licenses than that means that regular drivers are subsidizing Uber-drivers.

Comment: It's also a load of shit (Score 1) 309

by Sycraft-fu (#48897763) Attached to: UHD Spec Stomps on Current Blu-ray Spec, But Will Consumers Notice?

NTSC stuff is so bad when viewed on a large TV. It is amazing how blurry things look when you flip back and forth between the HD and SD channels. That is part of what lead to the rise of big screen TVs was actually having content for them. With NTSC, a large TV just meant a big blurry image. With ATSC it can mean a nice large image.

Comment: Also (Score 1) 309

by Sycraft-fu (#48897747) Attached to: UHD Spec Stomps on Current Blu-ray Spec, But Will Consumers Notice?

Why shouldn't they continually improve their products? Even with NTSC sets this was done. New ones would be larger, have better focus, more clearly resolve the signal, have better phosphors, and so on. Why shouldn't this continue? They should keep trying to improve their products as technology allows.

None of that means you need to buy a new toy all the time though. You can stick with what you have until it breaks, or until the new stuff is a big enough leap that you wish to own it.

I think a lot of the whining from people comes down to simple jealousy. They'd like to own the new stuff, but cannot afford it, or do not wish to. So they try and hate on it and act like a luddite. You see it practically any time Slashdot has a story on new technology. People complain about it like it is somehow a bad thing that there might be something new.

Comment: And they could probably handle 120fps (Score 1) 309

by Sycraft-fu (#48897721) Attached to: UHD Spec Stomps on Current Blu-ray Spec, But Will Consumers Notice?

Most panels in higher end screens are actually real 120fps panels. However that is just used for 3D and for reduced motion blur. The only set I know that advertises support for 120fps input is Vizio. Others could do it, if they wanted to, however.

As you say, the issue with higher refresh rates isn't in the display technology.

Part of it is just getting people used to the idea I think. We've seen shitty, jerky, frame rates in moves for so long people start to associate that with being "cinematic". People need to get used to the idea that's bullshit and maybe they'll start to like it more.

Hopefully sports and such will get shot at 60fps some day and that may help.

Comment: Oh yay, more about the bullshit clock (Score 4, Insightful) 196

by Sycraft-fu (#48897163) Attached to: Doomsday Clock Moved Two Minutes Forward, To 23:57

You know, when something says that we are so close to destruction for over half a century... well you have to wonder why anyone would put any stock in it. It is a bit hard to reconcile with being on the edge of destruction, and yet everything continuing to not be destroyed.

Comment: It also doesn't really matter (Score 4, Insightful) 140

by Sycraft-fu (#48895419) Attached to: NVIDIA Responds To GTX 970 Memory Bug

Thing thing is if you go and look at benchmarks of the cards in actual games, you find out the 970 wrecks shit, particularly given its price point. The 980 is an overpriced luxury (I say this as a 980 owner) because the 970 gets nearly the same performance for like half the price. The difference with its memory controller just doesn't seem to matter in actual games out there on the market.

And that's the real thing here the the spec head forget: You buy these to run actual software. If it does well on all actual software, then who gives a shit about the details?

Comment: Not necessiarly (Score 1) 180

He may well have been as smart as he thought (I'm not saying that is the case for sure, mind) but turns out others were smart enough, and more knowledgeable in the ways that mattered.

Hans Reiser is a good example. Man is unquestionably very smart. However, he had the geek hubris that I call SMFU, Smartest Motherfucker in the Universe syndrome. He figured he was so much smarter than everyone else, he could easily get away with his crime. Turns out that the police have some smart people too, and those people know a lot more about criminal investigation than he did.

Comment: Right and wrong (Score 1) 180

Right in that yes, they already have a lot of evidence, and are just working to seal the deal. They like to have everything in a row and an overwhelming amount of evidence before going to trial.

Wrong about the contempt thing. If you look it up in the US you find out that the courts have decided the 5th amendment applies to passwords. So you can keep your mouth shut and they can't compel you to hand over a password. If it is locked with something physical like a key fob or fingerprint, that you have to hand over. Basically if something is solely in your mind, they can't compel you to hand that over if it can be used against you.

Comment: Geeks in particular tend to forget this (Score 4, Insightful) 180

by Sycraft-fu (#48879147) Attached to: Silk Road Journal Found On Ulbricht's Laptop: "Everyone Knows Too Much"

The FBI may not be all up to date on the latest technologies and they aren't great at dealing with things purely in the digital world. However they are one of, if not the best investigative organizations in the world. They have a lot of experience investigating crimes of all kinds, often committed by experienced criminal organizations that are quite clever.

So there's a good chance if they are interested in getting you, they will. They are quite literally professionals at it, and they institutionally learn from their experience. You very well may know a lot more about computers than they do, but they almost certainly know way more about criminal investigations than you do.

Comment: Re:Science by democracy doesn't work? (Score 2, Insightful) 435

by Alsee (#48877703) Attached to: Science By Democracy Doesn't Work

if the same data has been used to claim a warming trend and the same data is used to say otherwise I'd call that invalid data.

The same data has been used to claim men landed on the moon, and that the moon landing was a hoax. Therefore all data related to the moon landing should then be ignored. As it's proven flawed on both counts.

-

10 to the 12th power microphones = 1 Megaphone

Working...