I still remember when Sony Ericsson used their own special connector (kinda like the old iPhone connector) headphones in the K series phones. Of course could have the privilege of paying an extra $100 or more to get the same exact hardware with an audio jack on a W series phone...
You don't have a clue what you are talking about. POWER has high performance. It's just expensive like heck. But then again their clients are willing to pay the price.
It was the first processor with large embedded DRAM caches that I can think of. It has 4-way SMT, decimal floating point, hardware transactional memory, and gobs of memory bandwidth.
It's not optimized for low-power operation though. It's for people who care about performance and backwards compatibility with legacy IBM hardware.
There are CPUs other than Bulldozer with shared FPU between cores. UltraSPARC T1 (Niagara) is one of them.
Frankly I could care less about FPU performance as all the FP compute bound applications I use are targeted for GPU architectures. I have a Piledriver CPU and for compiling code and running applications it is good enough. It's not like I have to render graphics for Quake 1 with a CPU software rasterizer anymore. On most web servers the FPU will be basically idle. Bulldozer was originally designed for the high end x86 server market and it shows.
AFAIK most OS specific code for Bulldozer is actually workarounds for hardware bugs. I suspected it was going to have bugs (new arch) so waited for Piledriver. But then again actually you do OS kernel development so you probably know better.:-)
I thought Mozilla was going to remove NPAPI plugin functionality. As is Chrome if they did not do it already. Adobe regularly does interesting things like this. Releasing "new" software just when an API is going to get deprecated. (Carbon anyone?).
Stalin allegedly made a speech at a military academy the preceding year where he clearly expected the Germans would invade the year they did. The thing is after all the delays the Germans had in Greece and elsewhere I think the Soviet High Command no longer expected an offensive that year. It was clearly madness to attack in the late summer as this would mean the Germans would not reach Moscow before the winter. The Soviets were also introducing a lot of hardware around that time (e.g. T-34, KV-1). I think they simply did not want to provoke an offensive before the upgrades were completed.
Still even if he did ignore Sorge at the start of Barbarossa the information Sorge provided was crucial to win the Battle of Moscow and Stalingrad later which was the beginning of the Germans defeat.
Russia had their own spies in WW2. Never heard of Richard Sorge? The Russians basically knew Barbarossa was going to happen a long time before it did. They just did not have an exact date.
It's not about tax rates. It's about illegal state aid to specific companies. The EU has plenty of powers regarding these matters including anti-trust law and its in the treaties.
Well if you remember the X-33 debacle, or the Boeing 787 debacle, composites aren't always the best idea. It's a lot easier to repair Al-Li than composites which is important for a reusable vehicle. Also AFAIK you need a large oven to cure the composites or an autoclave. I wouldn't be surprised if it took a lot longer to manufacture a composite structure which has an impact on production capacity and cost per unit. Al-Li might seem trivial now but a couple of decades back when NASA was doing the DC-XA they had to source a Al-Li propellant tank from Russia. Fact is Al-Li did not enter common usage in aerospace that long ago.
Composites are not a requirement for an SSTO. You need lightweight structures but composites are not necessarily the best way to do it. I have seen all sorts of proposals on how to get to the required weight fractions but without actual hardware flying its hard to be sure what's better. I am not even sure an SSTO is that necessary to begin with as long as the second stage is simple and cheap enough to be discarded.
Skylon has too many unknown factors in it. I think the engine itself has interesting ideas but that it's a bad match for an SSTO. The LAPCAT proposal they made for an HST with the SCIMITAR engine makes more sense.
The game content isn't worth $60 USD. The procedural generation and art style kind of reminded me of Spore. That game is like 8 years old now. Spore was another hugely hyped piece of fail.
There were two main reasons. One was that Osama bin Laden was in Afghanistan. The other was that a large amount of the perpetrators of 9/11 had been in Al-Qaeda training camps in Afghanistan.
As for how they found the perpetrators so quickly isn't particularly hard. They checked the passenger lists and Al-Qaeda had bombed the World Trade Center before so it wasn't much of a stretch to assume they were doing it again. So basically all they had to do was check for possible links to Al-Qaeda. Basically anyone who had traveled in the past either to Afghanistan or to a country bordering Afghanistan (where Al-Qaeda had training camps) was suspect.
PS: a manned lunar landing mission could be done with a simultaneous launch of two rockets (can't remember if you need two Falcon 9 Heavies or you can do it with a Falcon 9 Heavy and a regular Falcon 9). This should be possible once SpaceX has their Texas launch site operating in addition to their existing Cape Canaveral launch site. The cost of the rockets for the launch could be quite small if SpaceX reused the lower stages from prior launches.
ULA could also pull it off in theory if they wanted to. The original EELV designs included the Delta 4 Heavy and Atlas 5 Heavy configurations which should have enough payload capacity as well. They would need to modify the launch sites for manned launches and re-certify their launchers for manned missions. But the Atlas 5 is supposed to launch the Boeing capsule for NASA in a couple of years anyway.
AFAIK a Falcon 9 Heavy rocket can do a manned orbital lunar mission in a single launch. The Russians could do it with the Proton rocket (Zond program), with technology available in the 1960s, and the Falcon 9 Heavy has more launch capacity than the Proton. The first Falcon 9 Heavy rocket should launch this year or next.
Developing the lander is easy. SpaceX already designed the SuperDracos for the Dragon V2 capsule. First launch for the capsule supposed to happen next year. The transfer vehicle doesn't sound particularly hard to design either if you start with the Falcon 9 Heavy upper stage and the Dragon V2 capsule.
Flash typically uses different manufacturing processes than the logic processes used for CPUs. So it's not like you can switch a fab from manufacturing CPUs to manufacturing memory like that.
Fact is Intel/Micron and Toshiba were years behind Samsung on 3D-NAND technology. The 3D Xpoint press release smelled a lot like vaporware when I heard about it. Intel and the industry has been working on PCM for decades. Remember Ovonyx? Intel announced a large investment in it around the time the *Pentium 4* came out and it was old even then... The industry has been working on PCM since the 1960s-1970s.
Intel/Micron and Toshiba are manufacturing 3D-NAND this year so there should be a price drop soon as competition heats up. As for 3D Xpoint when (if?) it does get to the market it will have much lower density than NAND. Intel supposedly is aiming for a memory with capacity/speed characteristics between DRAM and NAND Flash. Seems kind of like a niche product to me. Remember the press releases claiming MRAM would replace everything?
Slashdot Mirror
They are supposed to increase the incidence of earthquakes. At least the ones that inject water underground. Google "basel hydraulic rock".
Webkit? You mean KHTML?
Eh the Newton was actually an interesting device for its time. Now if we were talking about Pippin on the other hand...
I still remember when Sony Ericsson used their own special connector (kinda like the old iPhone connector) headphones in the K series phones. Of course could have the privilege of paying an extra $100 or more to get the same exact hardware with an audio jack on a W series phone...
Apple Engineer Talks about the New 2015 Macbook:
https://www.youtube.com/watch?...
They should have luminance limits on indicator LEDs. You just need to know it's on, not to be blinded by the LED.
LEDs have had way too much power since like 2005.
You don't have a clue what you are talking about. POWER has high performance. It's just expensive like heck. But then again their clients are willing to pay the price.
It was the first processor with large embedded DRAM caches that I can think of. It has 4-way SMT, decimal floating point, hardware transactional memory, and gobs of memory bandwidth.
It's not optimized for low-power operation though. It's for people who care about performance and backwards compatibility with legacy IBM hardware.
There are CPUs other than Bulldozer with shared FPU between cores. UltraSPARC T1 (Niagara) is one of them.
Frankly I could care less about FPU performance as all the FP compute bound applications I use are targeted for GPU architectures. I have a Piledriver CPU and for compiling code and running applications it is good enough. It's not like I have to render graphics for Quake 1 with a CPU software rasterizer anymore. On most web servers the FPU will be basically idle. Bulldozer was originally designed for the high end x86 server market and it shows.
AFAIK most OS specific code for Bulldozer is actually workarounds for hardware bugs. I suspected it was going to have bugs (new arch) so waited for Piledriver. But then again actually you do OS kernel development so you probably know better. :-)
The average user cannot exactly ask Pegatron, Compal or Quanta to manufacture 1 million units just because they want a laptop.
I thought Mozilla was going to remove NPAPI plugin functionality. As is Chrome if they did not do it already. Adobe regularly does interesting things like this. Releasing "new" software just when an API is going to get deprecated. (Carbon anyone?).
Stalin allegedly made a speech at a military academy the preceding year where he clearly expected the Germans would invade the year they did. The thing is after all the delays the Germans had in Greece and elsewhere I think the Soviet High Command no longer expected an offensive that year. It was clearly madness to attack in the late summer as this would mean the Germans would not reach Moscow before the winter. The Soviets were also introducing a lot of hardware around that time (e.g. T-34, KV-1). I think they simply did not want to provoke an offensive before the upgrades were completed.
Still even if he did ignore Sorge at the start of Barbarossa the information Sorge provided was crucial to win the Battle of Moscow and Stalingrad later which was the beginning of the Germans defeat.
Russia had their own spies in WW2. Never heard of Richard Sorge? The Russians basically knew Barbarossa was going to happen a long time before it did. They just did not have an exact date.
Yes. They should just continue allowing our better equals to pay %1 tax or less while the rest pay 15x the tax.
I thought we didn't have feudalism anymore.
It's not about tax rates. It's about illegal state aid to specific companies. The EU has plenty of powers regarding these matters including anti-trust law and its in the treaties.
Both. It's called collusion.
Not necessary to expel Ireland. The EU could just stop sending them EU funds (e.g. agricultural and structural funds) for that amount.
Well if you remember the X-33 debacle, or the Boeing 787 debacle, composites aren't always the best idea. It's a lot easier to repair Al-Li than composites which is important for a reusable vehicle. Also AFAIK you need a large oven to cure the composites or an autoclave. I wouldn't be surprised if it took a lot longer to manufacture a composite structure which has an impact on production capacity and cost per unit. Al-Li might seem trivial now but a couple of decades back when NASA was doing the DC-XA they had to source a Al-Li propellant tank from Russia. Fact is Al-Li did not enter common usage in aerospace that long ago.
Composites are not a requirement for an SSTO. You need lightweight structures but composites are not necessarily the best way to do it. I have seen all sorts of proposals on how to get to the required weight fractions but without actual hardware flying its hard to be sure what's better. I am not even sure an SSTO is that necessary to begin with as long as the second stage is simple and cheap enough to be discarded.
Skylon has too many unknown factors in it. I think the engine itself has interesting ideas but that it's a bad match for an SSTO. The LAPCAT proposal they made for an HST with the SCIMITAR engine makes more sense.
The game content isn't worth $60 USD. The procedural generation and art style kind of reminded me of Spore. That game is like 8 years old now. Spore was another hugely hyped piece of fail.
Falcon 9 isn't Al. It's Al-Li with a composite interstage. It also has a lower dry mass than iso-grid designs like the EELVs.
There were two main reasons. One was that Osama bin Laden was in Afghanistan. The other was that a large amount of the perpetrators of 9/11 had been in Al-Qaeda training camps in Afghanistan.
As for how they found the perpetrators so quickly isn't particularly hard. They checked the passenger lists and Al-Qaeda had bombed the World Trade Center before so it wasn't much of a stretch to assume they were doing it again. So basically all they had to do was check for possible links to Al-Qaeda. Basically anyone who had traveled in the past either to Afghanistan or to a country bordering Afghanistan (where Al-Qaeda had training camps) was suspect.
PS: a manned lunar landing mission could be done with a simultaneous launch of two rockets (can't remember if you need two Falcon 9 Heavies or you can do it with a Falcon 9 Heavy and a regular Falcon 9). This should be possible once SpaceX has their Texas launch site operating in addition to their existing Cape Canaveral launch site. The cost of the rockets for the launch could be quite small if SpaceX reused the lower stages from prior launches.
ULA could also pull it off in theory if they wanted to. The original EELV designs included the Delta 4 Heavy and Atlas 5 Heavy configurations which should have enough payload capacity as well. They would need to modify the launch sites for manned launches and re-certify their launchers for manned missions. But the Atlas 5 is supposed to launch the Boeing capsule for NASA in a couple of years anyway.
AFAIK a Falcon 9 Heavy rocket can do a manned orbital lunar mission in a single launch. The Russians could do it with the Proton rocket (Zond program), with technology available in the 1960s, and the Falcon 9 Heavy has more launch capacity than the Proton. The first Falcon 9 Heavy rocket should launch this year or next.
Developing the lander is easy. SpaceX already designed the SuperDracos for the Dragon V2 capsule. First launch for the capsule supposed to happen next year. The transfer vehicle doesn't sound particularly hard to design either if you start with the Falcon 9 Heavy upper stage and the Dragon V2 capsule.
Don't store the disks in a place where they get exposed to light.
I have some disks which have lasted a decade so far. If the media is of decent quality it can last. Other media went bad in a year.
Flash typically uses different manufacturing processes than the logic processes used for CPUs. So it's not like you can switch a fab from manufacturing CPUs to manufacturing memory like that.
Fact is Intel/Micron and Toshiba were years behind Samsung on 3D-NAND technology. The 3D Xpoint press release smelled a lot like vaporware when I heard about it. Intel and the industry has been working on PCM for decades. Remember Ovonyx? Intel announced a large investment in it around the time the *Pentium 4* came out and it was old even then... The industry has been working on PCM since the 1960s-1970s.
Intel/Micron and Toshiba are manufacturing 3D-NAND this year so there should be a price drop soon as competition heats up. As for 3D Xpoint when (if?) it does get to the market it will have much lower density than NAND. Intel supposedly is aiming for a memory with capacity/speed characteristics between DRAM and NAND Flash. Seems kind of like a niche product to me. Remember the press releases claiming MRAM would replace everything?