You could do that, but there'd be too much latency for really good players. Flipper buttons are digital but modern games (anything newer than 'Funhouse') scan them every 2ms and react. You can flip them 'lightly' as a result. Also remember that the flipper itself takes a small but non-zero amount of time to rise and fall and that matters (eg. when flipper-passing the ball).
One of my first jobs on pinball was writing life-testing code for a test fixture. Springs and solenoids last for a really long time. I'm skeptical that anything using air pressure would hold up as well.
I don't believe that at all. The PS2 is hard to emulate because it's an exotic design intended for a particular programming style (stream processing) and it took people a long time to understand it. It was also designed to be as powerful as possible for the price, so it sacrifices things like regularity and robustness.
I used to do the 'intro to PS2' chat for new programmers and I would draw the architecture diagram on the whiteboard, starting with the main bus and CPU. They'd be fine at first and as more and more boxes appeared they would get steadily more apprehensive. There are 7 big black books which describe the PS2 hardware, sometimes quite tersely, and there is much, much more you need to know to get the best out of it. I am not surprised at all that emulation has proven a tall order.
I just came upon this document from 1982 which talks about keeping the tank in orbit. Some of the things it talks about are now precluded by safety (eg. having a fueled LOX/LH2 engine in the orbiter cargo bay), but it's worth reading.
By maximum payload I mean the total mass of stuff you can haul into space in the orbiter cargo bay and/or mid-deck and that's 24.4 tons currently. That means that if the orbiter wasn't carrying any cargo at all but instead wanted to lift the tank to the circular orbit, the tank is 8.6 tons too heavy.
I don't know much bigger the SRBs would need to be - the Tsiolkovsky rocket equation would answer that I think but I haven't done the math. There are also constraints on the trajectory (maximum dynamic pressure, maximum G force from acceleration, safe place to drop SRBs, ability to abort in-flight and land safely somewhere) that still need to be observed and they complicate the math further.
When MECO happens the orbiter and tank are in an orbit with a low point of about 35 miles, well inside the atmosphere. The tank separates and burns up half an orbit later. The orbiter's OMS engines fire to circularise its orbit and keep it in space. The external tank weighs 33 tons dry. That's more than the maximum payload for the orbiter, which means that the system as-is cannot save the tank. If it used bigger SRBs and/or had more main engine thrust (plus the fuel/oxidiser for them to burn) that might be a different story.
Saturn Vs were hugely expensive to build and fly. http://www.astronautix.com/lvs/saturnv.htm says $431mil in 1967 dollars, which is just over $2.8bil in 2010 dollars! Saturn 1Bs lifted 18.6 tons for $107mil in 1967 dollars, which is $702mil in 2010 dollars. That's worse performance and cost than STS, which averages around $600mil/flight (Shuttle launch cost estimates are harder to come by than for traditional rockets) for more mass, more crew volume and more mission time. Launching a huge space station on 2-5 Saturn Vs would've allowed design choices that aren't currently possible, but there's no way it would've been cheaper in the general sense.
The Apollo spacecraft was at the end of its life, too. Pure Oxygen, non-standard docking, no solar panels. It needed to be replaced with something no matter what. STS was originally envisioned as fully reusable and somewhat smaller and would've been a good replacement in my opinion. The development budget wasn't there for it and the military was forced to be on board. That's why the cargo bay is 60ft long and it has huge wings. The former was useful on occasion (Kibo only just fit in the bay) but the cross range of the wings was never used. When the military projects were merged with STS, that was part of their trade studies so a vehicle which couldn't do it was unsuitable for their requirements at that time.
At this point I consider NASA first and foremost a red state jobs program. That's not necessarily their fault, but it makes it impossible for them to run a space program per se. If they get downsized to $5-$7bil and made into a pure blue sky R&D and science organisation, maybe that's not such a bad thing. At least it'd stop this game of rocket musical chairs.
At that point you're really building a new rocket which happens to look like a Saturn V. Changing the thrust means changing the thrust structure and the plumbing, the latter of which is highly optimised (read about POGO and their solution to it). Changing the tank and/or skin composition is a radical change, too. Lastly, the avionics would need a modern re-do. The old Saturn V one is the Instrument Unit http://en.wikipedia.org/wiki/Saturn_V_Instrument_Unit and was completely custom, with an analog computer among other things. Of course that can all be done and you're right that modern techniques may be superior, but designing a new rocket is a very expensive thing.
SpaceX's big advantage is their vertical integration. They build everything in house from software to engine nozzles and they do take advantage of volume manufacturing as you discussed. That's why they're so cheap compared to the existing companies.
Lastly, Energia was also higher payload than Falcon Heavy. Like Saturn V it was infeasibly expensive to operate routinely.
It's mostly moved to APIs now. In the PS2 era, the PS2 used register level programming and assembly language but the Xbox used DirectX and the Gamecube used a custom API that was more or less OpenGL. In the current era the Xbox360 has its own API called XGraphics and the Wii and PS3 use OpenGL. The PS3 has another low-level API called libgcm but it's been a while since I was involved with that platform so I don't know the specifics.
The same has happened with handhelds. Through Gameboy Advance it was all register level programming but the DS and PSP have OpenGL-like graphics APIs.
I enjoyed learning to use the PS2 hardware but it was a pain in the ass for full game development. Licensing Renderware saved a lot of time and money. Knowing the hardware well enough to understand PS2 Performance Analyser dumps and help the artists change the assets was plenty good enough.
Virgin Mobile USA sells two Android phones for $150 and $200 and you don't need any service at all from them. If you want phone/data you can pay $25/mo (not a contract, just month to month) but the phone works fine without that. The first thing I did with mine was hook it up to my WiFi, before I'd even ported my number over from my old phone.
I've used TWiki (OSS, all Perl IIRC and aimed at corporate usage) at one job and Confluence at another but not Plone. Confluence is good for non-technical people because it has a pretty good wysiwig editor, but its search was simply wretched. I think we had a lot of 'lost' knowledge in the Confluence DB because nobody knew it was there and the obvious searches didn't show it - I would come across nuggets now and then. If you have the discipline to build index pages, it's probably a good choice if you have a lot of non-engineer type people.
TWiki (and this was a number of years ago so it may have improved) was almost the reverse. Good search, good architecture for plugins, but no wysiwyg so non-technical contributors had trouble with it. They were writing a wysiwyg plugin so that may have now arrived. It was easy to maintain and of the two I would say I like it better.
The article says Portugal is going to roll out a national network of electric vehicle charging stations in 2011. They needed the power infrastructure first.
That's the fundamental misconception I always try to address with non-programmers. Programming APPEARS to be about typing. It is REALLY about thinking.
It is wrong. Copying music without permission is wrong. However, the media companies are pissing up a rope trying to stop it, and the punishments handed out for copying music are totally disproportionate - that's as wrong or more wrong than the copying. However two wrongs do not make a right.
Fake-recycler gets hardware donated for free. They pay $X in collection costs. They pay $Y to ship to China. Chinese company pays $X+$Y+$Z to buy the hardware. Fake-recycler makes $Z profit. Chinese company pays $A to strip hardware to components (copper wire, metal cases, individual chips). Chinese company sells components for $X+$Y+$Z+$A+$B to whoever is buying the wire and so on.
$X probably isn't very much. It's not like it's a delicate operation. $Y is low because there are so many otherwise-empty containers going back to China. $Z doesn't have to be very much for the business to be worthwhile - it's not like it employs a lot of skilled people. $A is very low. There's a large surplus of unskilled labour in China. $B is probably low, but it's not like the company is doing the dirty work itself.
As long as $X+$Y+$Z+$A+$B comes out cheaper than the cost of buying the stuff new everyone's getting paid. I honestly don't know how cheap new copper wire is or exactly what chips can be reused in this way.
Lego Star Wars/Indy Jones/Batman and now Lego Harry Potter is coming. Those games are perfect for a parent to play with their kid. They have a lot of replay value.
I'd also suggest looking at Xbox Live Arcade games. Off the top of my head, Kingdom For Keflings, Pacman Champ Edition, Carcassonne, Ticket To Ride, Puzzle Arcade, Monkey Island (humour might be unsuitable for a 7yo, not sure), Geometry Wars 2 (it's not as hard as the first one and has more game modes).
There's a huge number of downloadable games, the good thing is they all are demos of themselves - if you like it you just buy and it unlocks.
The payload penalty might be problematic. Also, you can't really cluster more than 2 flyback stages due to the size of the wings. If you could use a parafoil and land with skids, that might solve that problem and to be fair western rockets don't really use clustering (Delta IV Heavy being a notable exception).
The PS2 backwards compatibility always used a hardware GS (the GPU). They emulated the EE and the VU1s, but not the GS. I'm not sure the cost reduction was really worthwhile, but it's not something they can just flip back on, on machines that don't have it.
Linux could probably use some work to make it more game friendly - just reading about the maze of libraries/drivers for sound acceleration makes my head hurt - but it's really about sales. Either we can work on supporting Linux which would gain approximately L sales, or we can spend those resources on improving the Windows version of the game and that would gain approximately W sales. Market research says W is much larger than L so it's generally not worth doing. Of course there are special cases so this isn't a blanket statement.
The situation was the same when making PS2 and Xbox games. Porting a PS2 game to Xbox was often straightforward, but Xbox sales were 1/3rd or less of the total (for the games I worked on, anyway) that spending time to support things like custom soundtracks or hard drive caching or high-def framebuffer wasn't worth it. It was better to spend the time optimising the PS2 version or implementing extra general gameplay features that both SKUs would benefit from.
If Canoncial and Red Hat and the other companies did something to change that financial picture you'd see more games on Linux, but I suspect that would not be good business practice for them.
The last time I did a WoW free trial I didn't have to give them any payment information at all. The account just went dormant after two weeks and they told me I could buy the game online to re-activate it if I wanted. I didn't and they never billed me in any way.
Slashdot Mirror
You could do that, but there'd be too much latency for really good players. Flipper buttons are digital but modern games (anything newer than 'Funhouse') scan them every 2ms and react. You can flip them 'lightly' as a result. Also remember that the flipper itself takes a small but non-zero amount of time to rise and fall and that matters (eg. when flipper-passing the ball).
One of my first jobs on pinball was writing life-testing code for a test fixture. Springs and solenoids last for a really long time. I'm skeptical that anything using air pressure would hold up as well.
It would be an interesting curiosity, certainly.
Orthonogality would've been a better word than Regularity, but I only thought of that now.
I don't believe that at all. The PS2 is hard to emulate because it's an exotic design intended for a particular programming style (stream processing) and it took people a long time to understand it. It was also designed to be as powerful as possible for the price, so it sacrifices things like regularity and robustness.
I used to do the 'intro to PS2' chat for new programmers and I would draw the architecture diagram on the whiteboard, starting with the main bus and CPU. They'd be fine at first and as more and more boxes appeared they would get steadily more apprehensive. There are 7 big black books which describe the PS2 hardware, sometimes quite tersely, and there is much, much more you need to know to get the best out of it. I am not surprised at all that emulation has proven a tall order.
There are lots of interfaces that don't use that sort of window model. Console games, handhelds, mobile phones, all sorts of embedded systems...
I just came upon this document from 1982 which talks about keeping the tank in orbit. Some of the things it talks about are now precluded by safety (eg. having a fueled LOX/LH2 engine in the orbiter cargo bay), but it's worth reading.
http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19940004970_1994004970.pdf
By maximum payload I mean the total mass of stuff you can haul into space in the orbiter cargo bay and/or mid-deck and that's 24.4 tons currently. That means that if the orbiter wasn't carrying any cargo at all but instead wanted to lift the tank to the circular orbit, the tank is 8.6 tons too heavy.
I don't know much bigger the SRBs would need to be - the Tsiolkovsky rocket equation would answer that I think but I haven't done the math. There are also constraints on the trajectory (maximum dynamic pressure, maximum G force from acceleration, safe place to drop SRBs, ability to abort in-flight and land safely somewhere) that still need to be observed and they complicate the math further.
When MECO happens the orbiter and tank are in an orbit with a low point of about 35 miles, well inside the atmosphere. The tank separates and burns up half an orbit later. The orbiter's OMS engines fire to circularise its orbit and keep it in space. The external tank weighs 33 tons dry. That's more than the maximum payload for the orbiter, which means that the system as-is cannot save the tank. If it used bigger SRBs and/or had more main engine thrust (plus the fuel/oxidiser for them to burn) that might be a different story.
Saturn Vs were hugely expensive to build and fly. http://www.astronautix.com/lvs/saturnv.htm says $431mil in 1967 dollars, which is just over $2.8bil in 2010 dollars! Saturn 1Bs lifted 18.6 tons for $107mil in 1967 dollars, which is $702mil in 2010 dollars. That's worse performance and cost than STS, which averages around $600mil/flight (Shuttle launch cost estimates are harder to come by than for traditional rockets) for more mass, more crew volume and more mission time. Launching a huge space station on 2-5 Saturn Vs would've allowed design choices that aren't currently possible, but there's no way it would've been cheaper in the general sense.
The Apollo spacecraft was at the end of its life, too. Pure Oxygen, non-standard docking, no solar panels. It needed to be replaced with something no matter what. STS was originally envisioned as fully reusable and somewhat smaller and would've been a good replacement in my opinion. The development budget wasn't there for it and the military was forced to be on board. That's why the cargo bay is 60ft long and it has huge wings. The former was useful on occasion (Kibo only just fit in the bay) but the cross range of the wings was never used. When the military projects were merged with STS, that was part of their trade studies so a vehicle which couldn't do it was unsuitable for their requirements at that time.
At this point I consider NASA first and foremost a red state jobs program. That's not necessarily their fault, but it makes it impossible for them to run a space program per se. If they get downsized to $5-$7bil and made into a pure blue sky R&D and science organisation, maybe that's not such a bad thing. At least it'd stop this game of rocket musical chairs.
At that point you're really building a new rocket which happens to look like a Saturn V. Changing the thrust means changing the thrust structure and the plumbing, the latter of which is highly optimised (read about POGO and their solution to it). Changing the tank and/or skin composition is a radical change, too. Lastly, the avionics would need a modern re-do. The old Saturn V one is the Instrument Unit http://en.wikipedia.org/wiki/Saturn_V_Instrument_Unit and was completely custom, with an analog computer among other things. Of course that can all be done and you're right that modern techniques may be superior, but designing a new rocket is a very expensive thing.
SpaceX's big advantage is their vertical integration. They build everything in house from software to engine nozzles and they do take advantage of volume manufacturing as you discussed. That's why they're so cheap compared to the existing companies.
Lastly, Energia was also higher payload than Falcon Heavy. Like Saturn V it was infeasibly expensive to operate routinely.
It's mostly moved to APIs now. In the PS2 era, the PS2 used register level programming and assembly language but the Xbox used DirectX and the Gamecube used a custom API that was more or less OpenGL. In the current era the Xbox360 has its own API called XGraphics and the Wii and PS3 use OpenGL. The PS3 has another low-level API called libgcm but it's been a while since I was involved with that platform so I don't know the specifics.
The same has happened with handhelds. Through Gameboy Advance it was all register level programming but the DS and PSP have OpenGL-like graphics APIs.
I enjoyed learning to use the PS2 hardware but it was a pain in the ass for full game development. Licensing Renderware saved a lot of time and money. Knowing the hardware well enough to understand PS2 Performance Analyser dumps and help the artists change the assets was plenty good enough.
Virgin Mobile USA sells two Android phones for $150 and $200 and you don't need any service at all from them. If you want phone/data you can pay $25/mo (not a contract, just month to month) but the phone works fine without that. The first thing I did with mine was hook it up to my WiFi, before I'd even ported my number over from my old phone.
It was Hubbard's cult who had that shut down? I didn't know that.
I've used TWiki (OSS, all Perl IIRC and aimed at corporate usage) at one job and Confluence at another but not Plone. Confluence is good for non-technical people because it has a pretty good wysiwig editor, but its search was simply wretched. I think we had a lot of 'lost' knowledge in the Confluence DB because nobody knew it was there and the obvious searches didn't show it - I would come across nuggets now and then. If you have the discipline to build index pages, it's probably a good choice if you have a lot of non-engineer type people.
TWiki (and this was a number of years ago so it may have improved) was almost the reverse. Good search, good architecture for plugins, but no wysiwyg so non-technical contributors had trouble with it. They were writing a wysiwyg plugin so that may have now arrived. It was easy to maintain and of the two I would say I like it better.
The article says Portugal is going to roll out a national network of electric vehicle charging stations in 2011. They needed the power infrastructure first.
That's the fundamental misconception I always try to address with non-programmers. Programming APPEARS to be about typing. It is REALLY about thinking.
It is wrong. Copying music without permission is wrong. However, the media companies are pissing up a rope trying to stop it, and the punishments handed out for copying music are totally disproportionate - that's as wrong or more wrong than the copying. However two wrongs do not make a right.
Fake-recycler gets hardware donated for free.
They pay $X in collection costs.
They pay $Y to ship to China.
Chinese company pays $X+$Y+$Z to buy the hardware.
Fake-recycler makes $Z profit.
Chinese company pays $A to strip hardware to components (copper wire, metal cases, individual chips).
Chinese company sells components for $X+$Y+$Z+$A+$B to whoever is buying the wire and so on.
$X probably isn't very much. It's not like it's a delicate operation.
$Y is low because there are so many otherwise-empty containers going back to China.
$Z doesn't have to be very much for the business to be worthwhile - it's not like it employs a lot of skilled people.
$A is very low. There's a large surplus of unskilled labour in China.
$B is probably low, but it's not like the company is doing the dirty work itself.
As long as $X+$Y+$Z+$A+$B comes out cheaper than the cost of buying the stuff new everyone's getting paid. I honestly don't know how cheap new copper wire is or exactly what chips can be reused in this way.
A lot of this is due to externalities, after all.
Yes but if it's a 64bit int we might wind up with 8 quintillion of them - we'd be neck deep or something!
Lego Star Wars/Indy Jones/Batman and now Lego Harry Potter is coming. Those games are perfect for a parent to play with their kid. They have a lot of replay value.
I'd also suggest looking at Xbox Live Arcade games. Off the top of my head, Kingdom For Keflings, Pacman Champ Edition, Carcassonne, Ticket To Ride, Puzzle Arcade, Monkey Island (humour might be unsuitable for a 7yo, not sure), Geometry Wars 2 (it's not as hard as the first one and has more game modes).
There's a huge number of downloadable games, the good thing is they all are demos of themselves - if you like it you just buy and it unlocks.
Boeing did a study of making a winged Saturn V first stage back in 1962.
http://www.astronautix.com/lvs/winturnv.htm
The payload penalty might be problematic. Also, you can't really cluster more than 2 flyback stages due to the size of the wings. If you could use a parafoil and land with skids, that might solve that problem and to be fair western rockets don't really use clustering (Delta IV Heavy being a notable exception).
The PS2 backwards compatibility always used a hardware GS (the GPU). They emulated the EE and the VU1s, but not the GS. I'm not sure the cost reduction was really worthwhile, but it's not something they can just flip back on, on machines that don't have it.
Full disclosure: I work on an online PC game.
Linux could probably use some work to make it more game friendly - just reading about the maze of libraries/drivers for sound acceleration makes my head hurt - but it's really about sales. Either we can work on supporting Linux which would gain approximately L sales, or we can spend those resources on improving the Windows version of the game and that would gain approximately W sales. Market research says W is much larger than L so it's generally not worth doing. Of course there are special cases so this isn't a blanket statement.
The situation was the same when making PS2 and Xbox games. Porting a PS2 game to Xbox was often straightforward, but Xbox sales were 1/3rd or less of the total (for the games I worked on, anyway) that spending time to support things like custom soundtracks or hard drive caching or high-def framebuffer wasn't worth it. It was better to spend the time optimising the PS2 version or implementing extra general gameplay features that both SKUs would benefit from.
If Canoncial and Red Hat and the other companies did something to change that financial picture you'd see more games on Linux, but I suspect that would not be good business practice for them.
The last time I did a WoW free trial I didn't have to give them any payment information at all. The account just went dormant after two weeks and they told me I could buy the game online to re-activate it if I wanted. I didn't and they never billed me in any way.
It may also be that there was an entirely different game called Project Eden released in the USA in 2002 or so.
Perhaps, but I think your function is misnamed. It should be called "is_b_nonzero".