Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!


Forgot your password?
Get HideMyAss! VPN, PC Mag's Top 10 VPNs of 2016 for 55% off for a Limited Time ×

Comment [citation given] (Score 1) 171

F9-001: Success
F9-002: Success
F9-003: Success
F9-004: Primary mission success, secondary mission scrubbed due to ISS safety rules
F9-005: Success
F9-006: Success (first v1.1 flight)
F9-007: Success
F9-008: Success
F9-009: Success (first flight with landing legs)
F9-010: Success
F9-011: Success
F9-013: Success
F9-012: Success
F9-014: Success
F9-015: Success
F9-016: Success
F9-017: Success
F9-018: Success
F9-020: Failure, RUD at T+150s
F9-021: Success, first v1.1 FT, first successful landing at Canaveral
F9-019: Success
F9-022: Success
F9-023: Success, first successful landing on droneship
F9-024: Success
F9-025: Success
F9-026: Success
F9-027: Success

One failure. Out of twenty-seven, for a success rate of 96%. Unless you want to count landings as necessary for success, in which case they have a 19% success rate - but by that metric, Soyuz, Proton, Atlas, Delta, Titan, Redstone, Saturn, Ariane, Athena, and Zenit all have 0% success rates, and only Energia-Buran and STS also have a non-zero success rate, with 50% and 98%, respectively.

Over that same period North Korea fired four missiles (claiming to have fired even more but not supported by evidence) and launched two orbital rockets. The missiles may or may not have failed - they fell vastly short of their design range but that may have been deliberate - and both rockets worked, although their payloads may have failed. At least, this is all the info I could find - there's no convenient list of every launch attempt they've made, and I suspect most failed launches are never revealed.

Comment Undecided - for now (Score 2) 185

I've been wanting to get into VR. But every platform has at least one seemingly-fatal problem right now:
Oculus: being evil with DRM, no motion controls yet
Gear: Underpowered, works only with certain phones
Vive: Weakest software support
PSVR: Not out yet, but likely to be underpowered (PS4 is well below Oculus/Vive req. specs, PS4NEO unknown), and likely to contain evil
Cardboard: Decent replacement for the View-Master but doesn't work for actual gaming
LG 360: Literally hadn't heard of it, which means the software is likely weak; also only works with certain phones

Vive has the easiest problems to overcome - just make more games for it. The SteamVR model seems like it could make VR cheap and standardized, which would make them the de facto PC VR platform. And I think I'd be happiest with Valve leading the VR industry.

PSVR has potential. They're going hard on the software side, and they have history of making good consumer hardware. I'm concerned about the horsepower - if they require VR games to be usable on the stock PS4, that's going to hamper game design, making my experience worse even if I get a NEO. And the NEO itself is an unknown.

Oculus has a head-start, and they're doing the best on average. Their software library is okay, their headset hardware is reportedly the best, and the motion controllers are coming soon. They're in a position where they could win. But they don't have enough of a lead for me to bet $700 that they'll win, and honestly? I don't want them to. I was willing to write off the concerns about Facebook buying them, but after that DRM stuff they decided to pull? I can no longer ignore them.

Comment Re:Is it even possible to buy a new 32 bit chip? (Score 1) 378

For computers? Quite some time. There was a one-off Atom netbook chip back in 2008, and before that was Core (the predecessor to the more popular, 64-bit capable Core 2) in 2006-2007 and some of the early Pentium 4s up to 2005. On the AMD side, you have to go back to K7, which stopped being made in 2005. So everything that you'd want to run a desktop distro on is at least eight years old.

Intel did make x86-32-only chips for smartphones until much more recently, but you wouldn't want to run a desktop distro on those, anyways. And it's not like the Linux kernel is dropping support for it, so whatever weird hack project you might theoretically want to make with a bunch of old smartphones is still just as doable.

Comment Re:Taken from reddit comments (Score 1) 147

Agreed. I saw this news elsewhere, and it's pretty clear that the assets are, bare minimum, "traced" from other games. I'd need to see a more technical analysis to know if they were directly ripped from another game (and then modified just enough to not be copy/paste), but that seems more probable than not.

Comment Re:Weak argument (Score 1) 951

We had transoceanic ships half a millenium ago, and it improved quite a bit from those days, but today's tech would be basically recognizable to someone from the 1600s, even if unbelievably large in scale. Metal ships & propellers seem to be the biggest advances (disregarding nuclear fuel sources vs ICEs) and those aren't considered new by any means.

Our ships would be basically recognizable to someone from the 1600s. What about our jetliners? Or our spacecraft? In the 1600s, only experts in relatively narrow fields would have had anything at all to relate those things to. Most would have had no idea such technology could even exist.

Comment Moral imperative (Score 1) 114

I think it is a moral imperative (on all of us) to eliminate falsehoods. I would greatly prefer it if Facebook (and other companies with considerable control over the flow of information) would restrict the spread of objectively and provably false ideas. Perhaps not actually blocking them, but how hard would it be to add a "this has been proven false" message (with citations) to people sharing, say, anti-vax propaganda? And how much benefit would the public gain by it? Quite a bit, considering the anti-vaxxers have caused actual deaths. That's an extreme case for an extreme gain, but less-extreme cases will still have gains.

Most politicians are pretty well-practiced at avoiding statements that could be factually wrong, and would not be too badly affected. Even Trump is pretty good at this - his claim about how much the wall will cost is hard to disprove without actually building the damn thing (argue against, yes - disprove, no). But he provably lies pretty often - his stories about seeing Muslims celebrating in the streets as the WTC collapsed are demonstrably false. Or his claims to have never settled a case out of court, or never declared bankruptcy.

As long as it would be done fairly (ie. all candidates are subject to the same scrutiny) and to a set standard, I think this would be a good thing.

Comment Re:Coal Powered Cars Are Awesome. /s (Score 4, Insightful) 123

Decent starting point, but you stopped thinking far too quickly.

Centralizing power generation (moving from millions of small gasoline engines to hundreds of big oil/coal turbine generators) allows for greater efficiency. Most things work better at scale - you get more power extracted per unit fuel. And it allows you to cost-effectively install better pollution-reducing devices - big scrubbers on the exhaust, to keep particulates and such down. So even if the power grid were 100% fossil fuels, it would still be a gain.

But the grid isn't 100% fossil fuels. In some places those are a minority already - where hydro or geothermal or nuclear dominate. And it decouples the generators from the infrastructure - if all cars ran off batteries, we could switch over to whatever new power method works best, as we invent it. If we had cheap, efficient, clean nuclear fusion, switching to it would be easy if we were on electric cars. Switching from gasoline/diesel engines to fusion engines would require a lot more change to the infrastructure - new fueling stations built, new pipelines for deuterium run, new mechanics trained on new engine types.

Comment Re:Why SLS? (Score 1) 224

I'm American, and I habitually avoid "tonnes" because, when spoken, there's no way to know if it's the "1000 kilograms" metric tonne, the "2000 pounds" American ton (aka short ton, or customary ton), or the "2240 pounds" British ton (aka long ton, imperial ton, or, most confusingly, also called the metric ton). "Megagram" is far harder to confuse with anything else, and even if it's not natural for most people to use, it's obvious what it means - 1000 kilograms, or 1000000 grams. Even Americans can figure out what it means.

So it's used primarily by Americans who wish their country could hurry the fuck up and switch to metric already.

Comment Re:Why SLS? (Score 3, Informative) 224

SpaceX does not "already" have one. They definitely didn't when SLS started.

SLS is big. Really big. 70Mg to orbit with just the base model, potentially up to 150Mg with upgrades. It will be classified as a "Super-Heavy-Lift Launch Vehicle", the same class as Saturn V.

Falcon 9, as currently flown, can orbit 13Mg ("Medium Lift"). The biggest rocket currently flying, Delta IV Heavy, can orbit nearly 29Mg, making it one of two flying Heavy Lift Launch Vehicles. The analogous Falcon Heavy vehicle is specified to orbit 53Mg, putting it on the edge between HLLV and SHLLV.

That said, SLS is an absolute disaster of a project. It reuses almost every part of the Shuttle's powertrain - same engines, same fuel tanks, boosters that are identical except for being 25% longer. It uses an upper-stage engine that's flown since before Saturn. Years of study were spent on related designs, like Ares. And yet it will cost $18,000,000,000 and eight years to design, build and launch one of them? In eight years, SpaceX went from not existing, to building their own rocket using their own engine to launch their own spacecraft. And it's looking likely that, eight years from now, they'll have not one but two of their own super-heavy-lift rockets, Falcon Heavy and the Mars Colonial Transporter launcher.

Comment And SpaceX (and others) are saving NASA (Score 4, Informative) 224

After the Space Shuttle retired, and with all replacement programs canned by Congress, NASA had no way to get astronauts into space, except by hitching a ride with the Russians, and NASA had no native way to resupply the ISS.

The COTS program has already fixed one of those. NASA now has access to two locally-made spacecraft that can fly on either the entirely-American Falcon 9, or the partially-foreign Antares or Atlas V. This gives pretty robust resiliency - a single accident cannot halt the entire system. (Two back-to-back RUDs can do that, though, as we saw).

The CCD program is getting NASA access to two spacecraft capable of shuttling astronauts to low orbit - one built to fly atop basically any lift rocket that can handle the load. Three other spacecraft are in the program, theoretically able to replace either of the two main CCD craft should they fall too far behind schedule or too far over budget - helping to ensure robust access to space.

Where would NASA be right now without them? Well, they could still loft satellites or probes on the high-price ULA vehicles, but they'd probably have to abandon the ISS. Between only having Russia for crew transfers, and only having Russia, Japan and the ESA for resupply missions, they would not have been able to effectively operate the ISS.

The entire cost of COTS, CRS and CCD combined is $12.3B. For comparison, the Constellation program cost $9B, and produced no flyable launch vehicles or spacecraft before it was canceled. SLS will have cost us $18B by the time it makes its first test flight. Considering the commercial programs* have given us multiple, redundant systems, and included the cost of dozens of missions, while SLS is a single spacecraft for a single rocket that will perform a single flight on its $18B budget, I think we're getting a pretty good deal.

* SLS is technically "commercial" as it is being made by several independent corporations. However, the key difference is that it is not competitive. If Aerojet Rocketdyne cannot produce engines at sufficient quality and quantity, or at a low enough price, NASA has no alternative. Same for the boosters and Orbital ATK, or the upper stage and Boeing, etc.. (The other difference is that the COTS/CRS/CCD program rockets are assembled by the contractor, while SLS will be assembled by NASA, but this is not a particularly meaningful distinction)

Comment Re:Why hasn't this been privatized yet? (Score 1) 25

Hey dumbass, this was a private industry launch. NASA has been contracting with independent companies for cargo to the ISS since the Space Shuttle stopped flying - both to Russian space firms, and to several American ones.

For this flight, the contract was to Orbital ATK, but due to the grounding of the Antares 100 lift vehicle (after one exploded in late 2014), Orbital ATK subcontracted the launch vehicle (at their own cost) to United Launch Alliance. The Cygnus spacecraft is also made by Orbital ATK, and is part of the cost.

Total cargo was 3513kg. It was part of an ongoing NASA contract, with precise details not known, but the original contract was $1.9B for eight flights, or $237M per launch. This contract was later extended with three more flights, presumably at similar costs.

This was the heaviest Cygnus launch to date, so dividing the payload mass by the cost gives us a lower bound on cost-per-mass of $67K per kilogram, or $31K per pound. So you're off by at least two orders of magnitude, probably closer to three.

Incidentally, the price-per-kilogram of a Saturn IB launch was $25K/kg (or thereabouts - it never flew to the ISS's orbit, so I had to make some estimates), and the price-per-kilogram of a Space Shuttle launch was $98K/kg (full program cost divided by successful launches) to $28K/kg (individual launch cost). All three of those numbers are adjusted for inflation, BTW, I'm not trying to bullshit on this. Also incidentally, the Falcon 9/Dragon flights to the ISS under the same program have a cost-per-mass of $40K/kg (they're more public with their pricing), although this includes a substantial deorbiting payload, which only SpaceX currently offers in any substantial amount.

And for one last dose of perspective, shipping from L.A. to NYC is $23/kg ($9/lb), for an overnight Fedex shipment of 400kg (the mass of the earliest test CRS flights). You literally think that shipping to outer space should cost no more than twice what it costs to ship around the country. Maybe you should get some perspective before you start bitching?

Comment Re:Aw, come on ... (Score 1) 372

There's some app-tier logic that tends to fuck this up. I myself had to deal with it.

In my case, I had written a set of webservices that took in parameters as POST form variables, and updated records accordingly. Parameters that were not sent were not modified. POST form variables are string-only, so I had originally planned for the empty string ("") to be the value indicating "set this field to null", but that caused problems for the web-tier developer, so (under mild protest) I made it so the string "null" would set a field to null.

I had written it generically, and in such a way that it transformed the set-to-null string into an actual null prior to processing, so even fields which could not be set to null, like name fields, would be modified that way.

And then there was a user named "Au Null".

And then I forced the web-tier developer to figure out how to send an actual empty string.

And then I used JSON forms for all future webservices, when I was given an option, so an actual null data type could be used.

Slashdot Top Deals

The reason computer chips are so small is computers don't eat much.