Re: When do you break even?
Next Sunday then...
2084 posts • joined 12 Jun 2009
Next Sunday then...
Got flagged up as infringing.
It clearly isn't of course. As 2 seconds work would have shown.
"Linus shows a lack of maturity"
Well, almost every developer I've ever met (including me) shows a lack of maturity in one way or another.
"Linux on desktop"
Is that really what he wants to see, or does he just not care a jot, given the OS is used on the top 10 supercomputers, in every android phone, in huge numbers of household devices?
I'd have to say that as a project manager he has done quite well. Most project managers I have worked with would not have got Linux to the place it is now. Since there is no other open OS project that has got anywhere near close to Linux, it would appear that is quite the success.
You could indeed argue that not enough market research was done, or that the initial batch wasn't large enough.
But you could also argue that a few people running a charity simply don't have the time or funds to run market research, or indeed, have a big enough house to remortgage to pay for the initial 10k batch....
Whilst hindsight is 50:50, you also have to take in to account constraints encountered at the time.
Also worth noting that the ARM cores on the SoC are UK designed, as is the Videocore4 GPU. Made in the Far East, but so are most chips nowadays.
So all round a very British effort.
In fact, some of the new Pi2's are made in China - but those are destined for the Far East plus Americas, not Europe.
They will continue to be made there whilst demand outstrips supply. Sony are making 20k a day In Wales, and it's not enough.
It's a wafer scale package in the power circuitry. Wafer scale means no package, just the silicon die rwith a BGA underneath, soldered direct to the PCB.
Not yet sure what the permanent fix will be. But blutac is clearly the current option for people who encounter the issue.
Prior performance is not an indicator of future results.
So, no children then?
Because last time I tried it was rubbish on Linux machines.
Usual bunch of townie bastards who have enough money to actually live in towns, complaining that the people who grow them their food are sponging of their taxes.
I suggest that instead of living in your nice city you actually visit the countryside, not as a tourist, that's pointless, but to actually see what it's like, before commenting.
I wonder how many would be interested to know the following fact.
Much of farmable countryside is being bought up by pension companies, for above usual selling prices, forcing owners/farmers out or pricing land too high for them to buy enough land to survive. Then they get the land farmed by cheap immigrant labour, putting a lot of locals on the dole.
For YOUR pension. So a lot of people in the Countryside are directly contributing to YOUR fucking pension.
We want some of that money back please. Broadband would be good, as would roads that are not so bad they break car springs (same roads that bring your food to the city)
Bugger, and I'm not even a farmer...
It's a choice on the Pi. You either have it on an SD card, or you....don't....and use Linux/RISCOS/Plan9/BSD instead.
There will be no 'lock' in - it's simply not possible on a Pi since it doesn't have the required bootloader.
The Raspberry Pi Foundation have received no money from MS, it's entirely MS's doing to port their OS to the Pi.
They have had a bit of tech support from the PF, but nothing extensive, and nothing paid for.
I'm not qualified to talk about the security side of things.
Indeed. There isn't any way of embedding a boot loader on Pi, it just doesn't work like that.
Pi's come with NOTHING installed. You then install what OS you want on an SD card. There is no possibility of MS lock in here.
AIUI, this Win10 is the IoT version, so command line kernel stuff only, no desktop.
But I'll find out more this evening...!!
As someone who has actually spoken to Eben Upton about this, let me say again - this upgrade was NOT done at the behest of MS, nor was it done for MS.
It was done because there is now serious multicore competition in SBC space, and because the new chip became available. Note that there has been no increase in price. If this upgrade could only be done by a price increase, it would not yet have happened, MS or no MS.
I'm not convinced they were. Everything you need is pretty much available as ARM side libraries (Framebuffers, 2D bitmap acceleration, 3D acceleration via OpenGLES etc).
Stuff I use Raspi's for that will be better with the new one....
1. Media player (Yes, Kodi works really well, at home, and in the motorhome)
2. Camera work. Quad cores and NEON means much more image processing capability
3. H264 testing (day job). New device is up to 30 times faster at software H264 encode.
4. Scratch for the children. Works much faster
Stuff I can now use it for that was too slow before.
2, H264! See above
3. Compiles (a hell of a lot faster means I don't need to cross compile as much)
4. Decent GUI desktop - this is much faster. Now can have multiple open windows with no slow down.
I'm sure there is loads of other stuff that others do that I don't.
Hasn't done that since the B+ was launched...
MS are following the RPF, not vice versa.
So, no, I do not expect the Pi to become a Windows only device, which is what you seem to be implying. For started, the Win10 being ported is the IoT version - no desktop.
Which is going to make it difficult for the educational side to actually do the works it so far been able to do well with Linux.
Bollocks. The new Pi was not driven in any way by MS. AT ALL.
It's more driven by all the newer, higher powered devices like ODROID and similar, and the fact that Brcm developed a higher performance chip that could be bought for the same/similar money.
No need for conspiracy theories when you can have the above facts for free.
Tried it with headphones, did appear to play back stuff. I suspect will be much better quality over HDMI or USB sound card, was a bit hissy with the headphones.
Cpu usage was about 18% during the playback. Not tried an A7/Neon build.
At $35 worth a punt I think!
Got musescore working, and it does appear to run that demo OK, but I do not have any speakers attached so difficult to tell. CPU usage < 20%
This is the v6 compiled version, so no cortex-a7 or NEON optimisations.
The newer kernels for Pi are moving towards device tree, which should give you the HW abstraction you are looking for.
Doesn't run XP, just Linux (and RISCOS and a couple of others that won't be of interest to you).
But it runs XBMC nicely (all Pi's do), and I use them to stream, from a NAS or from the PVR.
With the new one, you might be able to decode some HW unsupported formats, but probably not HEVC.
Sorry Tom7, tried to install musescore, but failing to install some dependencies. Will try again later.
Someone gave my entirely factual post a thumbs down. I'm intrigued as to why.
I presume a ODROID-C1 buyer who is now looking at their A5 cores?
No, there won't be a Win10 specific version - it will run on the B2
Also, this is just another OS, to supplement what is already there. Not a ship jump to MS.
Let me repeat.
No, you didn't read about this months ago. Today is the first time this board has been publicly discussed.
There was no opportunity - there are no chips with a53's in that would give the right upgrade path.
Had a couple for a week to do testing.
Although the headline figure is 6x performance with the right apps, a recompiled x264 with NEON extensions showed upto 32x increase when encoding H264. That's 'quite good'.
LXDE is MUCH faster as the multi cores off load all the graphics work, and the extra ram makes a big difference. The big L2 cache also improves things. Browser is very usable.
It's pretty impressive for $35.
There is a minor issue with the PiBow coupe, but Pimoroni have fixed it already.
Should that be exhaust rather than jet?
But notwithstanding that, a number of companies have demonstrated vertical launch/landing - Armadillo aerospace, Masten Systems, Blue Origin. SpaceX are the only ones at the stage of trying to return a booster though. I suspect BO might be the next ones.
It's an idea that has come of age.
7.8% still gives a thrust to weight ratio > 1, so not able to hover.
Hydraulic fluid ran out about a minute from landing, it's an open system, so fluid is lost during use. Tanks wasn't big enough (or carrying enough fluid). Grid fins locked hard over when they ran out.
Parachutes have already been tried by SpaceX and sucked, hence they are trying this hoverslam landing approach. Parachutes also add unnecessary weight . Stage already has grid fins and thrusters for maneuvering, plus gimballed engine nozzle, but is incapable of hovering (cannot throttle engine low enough)
tbh, the phrase catastrophic is also wrong. A catastrophic result would mean no data was received in this test flight, meaning it was wasted, whereas there is a clearly a lot of data and in many ways this was a successful TEST.
Next test in three weeks.
I don't wear a smock, or carry a pitchfork.
But I do live in a rural location, because I couldn't afford to live closer to work.
Within 10 miles of me I have a large windfarm, supplying you power, some world class companies supplying you food, sheds, play areas, a CFC disposal yard, friends who supply large amount of lamb for you to eat, huge amounts of potatoes for you to convert to chips, companies that supply computer equipment worldwide, a large biodigestor under construction (power and ethanol I believe). There's is also the Washes, a flood relief system for built up areas upstream, which stops you getting flooded.
All of the above work better with decent broadband, farming especially which is getting more and more reliant on web access.
Please think outside your own mindset before posting.
Musk does apparently play KSP.
The centre core of the F9Heavy will be travelling too fast for return to landing site (without affecting payload too much), and will therefore need to be barge landed, if recovered at all.
It's debatable whether even at terminal velocity the rocket could have sunk the barge (or even punctured the 20mm steel deck). The engines act as crumple zones, and the whole thing doesn't weigh much empty anyway. Top speed of a couple of hundred MPH with no braking burn, but with no braking burn it would miss the barge anyway.
Lights not much help. IR cameras have been suggested, but they would be swamped by the heat from the engine plume.
TBH, the telemetry is what they need, video would be a bonus anyway.
Original landing time would have been in daylight had it not been for the scrub on Tuesday.
All the questions above can be answered with a very quick Google search.
The Merlin engines are said to be capable of 40 cycles without servicing, and Musk has stated that even the servicing after that is only a few replacement parts. They were designed with reusability in mind.
Benefits of reuse are huge, the 1st stage alone costs >$50M. They are aiming for rapid turnaround (a day or so), so that a huge saving from throwing the stage away each launch.
Currently trying to land on barge, intent is to return to launch site (except for F9H centre core which will need a barge landing)
Foggy round our neck of the woods yesterday, quite surprising the number of cars that either have only side or no lights on at all. Note, sidelights are generally inadequate round our area as they get muddy very quickly and are pretty useless in that state.
Pretty sure a car that turns lights on automatically would be of immediate benefit given the parlous state of the driving around here.
Much much too expensive. This is going to be a big piece of silicon, and that means expensive.
And it will be a very expensive lump of silicon.
I'd keep the Jazz. I've been impressed with my 02 plate Civic 1.6s petrol, 220k miles, still going strong. Do all my own servicing, and have done for last 5 years. Probably needs new clutch (still original) and cam belt (on second), but TBH, I'll just run it until is dies a death. No point in trading in.
Sorting out the youngest son's PS3 to work on PSN, with a sub-account of a parent account.
Assumed it was this hack at first, but now fairly sure it's just a completely shit system designed by someone who never actually tried to use it.
Quite. Lots of people claiming the mileage claim is cobblers, yet I have read of quite a number of S drivers in very cold countries who are very pleased with the range performance. I will remain hopeful until tests show it one way or another.
Cold weather also reduces the range of ICE cars as well btw. It really isn't an easy thing to quantify. http://www.fueleconomy.gov/feg/coldweather.shtml
Calm down Bleu.
We get it, you don't like his books. No need to go on about it, you'll end up being more like Baxter than you think.
All writing is theft, in one way or another.