Re: Geostationary vs. LEO
The Indian launcher is being used first for a geostationary satellite, but it can be used also for LEO launches. The comparison of mass-to-LEO is not unreasonable (some of that mass can then be used to boost to GEO).
2645 publicly visible posts • joined 19 Sep 2007
You haven't seen the expected price tag on an SLS launch then? More than $2bn per launch (20 launches are expected to cost NASA $60bn). Even the Falcon Heavy is thought to cost around $160 million per launch.
To quote the article "the firm cites a Gamespot review to assert that a PC with Optane and a 1TB, 7,200 RPM spinning rust hard was four to eight times faster than a solid state disk at read-heavy tasks data."
Gamespot said no such thing. They made only limited comparisons with SATA SSDs, and you can pick up a 500GB SATA drive for not much more than the Optane + 1TB combo.
But that Gamespot review is mostly comparing against traditional HHDs, with only passing mention to SSDs, and when it does talk about them it makes speed comparisons with SATA units, not M.2.
Optane seems to be at best only a small improvement over SSD, and at worst a disappointment.
The majority of companies that use an ARM processor take an off-the-shelf ARM design, combine it with components from ARM or 3rd parties to form their SoC and hand the results to a fab company (TSMC, Samsung etc) to build. A small number of companies (Apple, Qualcomm and nVidia for example) have Architecture licences, which allows them to come up with their own CPU designs that execute ARM code. The resulting designs can be faster and/or more power efficient than the stock ARM cores.
The silicon design company (PA RISC IIRC) was bought many years ago, well before the first of their internal ARM designs. The designers may be pre Apple, but their work has all been done on Apple's time.
That very much depends on what kind of workload you throw at it. See https://www.hpcwire.com/2015/06/09/ibm-power8-outperforms-x86-on-stac-benchmarks/
The Power CPU has vastly more memory bandwidth, so it may be slower at headline FP numbers but it can chew its way through more data. It's also got much better fault tolerance and recovery. It isn't the no-brainier you seem to think over which you should chose.
A small turboprop can burn 70 gallons of jet fuel per hour. Commercial aircraft tend to need two of them. A single generator, with a battery to handle surge demands and as backup, driving two electric motors, may make a fair bit of sense. Considering you need much less power in a cruise then even more so.
The only reason that .NET on Linux is cut down is that it's living on top of a different OS environment that is incompatible with a lot of the library code. There's no reason they can't compile the full stack if Windows is the host (though they will need a new JIT compiler).
Sorry AC, your nerd mode failed. It had a full TMS9900 @3MHz. TI had intended to use the TMS9995, but had problems with it. They did use the 9995 in the TI 99/2 and TI 99/8, but they never got beyond the prototype stage.
BTW, both chips were NMOS, so had the same hideous power requirements. The 9900 (and I presume the 9995) needed a complex 4 phase clock, which might be where you got the 12MHz number from.
Yes, it had a 16 bit CPU, but they only gave it 256 bytes of 16 bit memory. All the rest was a nasty 8->16 bit multiplexed cludge. TI marketing didn't want it to compete with their lucrative mini computer market, so they deliberately made sure it couldn't.
@Paul - you seem sadly missguided about the power of an Atom X7-Z87xx, they are nowhere near Core i5/i7 in performance. Clock for clock the Core i5-5250U from over a year ago is about 95% faster, and that's only 2 core 4 threads.
They're still extracting the P. Trying to compare an Atom based machine to Core i5/i7s based only on the clock speed and number of cores is a joke. Prices starting at 400 euros for what is basically an updated netbook is another.
On giving the wrong reason for the failure? It failed because of creases in the pressure vessels that caused problems when they deliberately used supercooled helium, not because the helium was too cold. The long term fix is in the design and manufacture of the COPVs, not changing the helium temperature.
to statically do instruction scheduling, that x86 figured out how to do dynamically in hardware didn't help. Newer generations of processor thus required a recompile to work at full speed on later CPUs.
It was always too little, too late, and not enough better than the much cheaper x86/x64 ranges to draw much attention.
The helium was deliberately supercooled in order to increase the amount of fuel that could be loaded. Because it was so cold, any oxygen that found its way between the tank wrapping and the tank (specifically into buckles) formed into ice. That ice was dangerous and caused the explosion.
The supercooled helium was by design, and wasn't the cause of the buckling.
You seem to forget that, having had nuclear weapons used on them, the Japanese have an absolute horror of them. Without the declared threat (which the Japanese people wouldn't stand for) atom bombs aren't any use defensively, so why would they bother, especially given that the rocket isn't any use in lofting anything but the smallest battlefield weapons?
Apple were involved in setting the Nano SIM standard, but weren't the only participant. IIRC the solution chosen didn't match Apple's initial suggestion. Oh, and the Micro SIM dates back to 2003, before the first iPhone, so no Apple didn't push that one through either.
"After being told contradictory information by BT customer services, it transpired he was not eligible for fibre broadband because the cable only went to the cabinet - and not to his home."
Yes, that's the way that BT Infinity works, they have an optic link from the exchange to the cabinet, then VSDL over copper from the cabinet to your home. Nothing said there precludes BT from providing him with a faster connection.
The problem is that if you're more than about 800m from the cabinet then VDSL slows down to similar speeds to ADSL2+, so it's only useful for short runs. This is OK for urban environments, less so for rural.
While Apple have come up with few brand new ideas, what they have invented are lots of smaller parts/software that goes into making existing ideas better. It's the combination of those, plus the way they are integrated into the whole, that makes their devices popular with the public.
To use your analogy, just because BMW didn't invent the car doesn't mean that they can't do very well selling quality cars engineered with their own ideas added to the upper end of the market.
Like GPS it doesn't get anything back from the receivers. The satellites simply transmit the time and their location. The receiver calculates the differences in time of the received signals and does the math.
Mobile phones often use AGPS, which uses the mobile network to help get the initial fix, but then you can be tracked by your mobile signal so I'd be far more worried about that.
Apple doesn't DRM its music. The worst that can happen is that it will replace the music with a 256kbps AAC version. iThing owners haven't needed a PC of any variety since about iOS8. As to the quality of the Windows version of iTunes, I can't comment as I haven't used it for many years now.