I wasn't implying that anybody should standardize on battery packs, only that one company had a solution to that problem for their own products. I suspect that eventually, one of two things will happen. Either Tesla will maintain a sufficiently large marketshare to be able to pressure the rest of the industry to move to their own charging standard, or their marketshare will be small enough that a more general industry standard takes off. At this moment, Tesla's superchargers are really the only chargers in their class deployed in the US (CHAdeMO is more widespread, but limited to about half as much wattage), but SAE has a 90 kW version of their J1772 connector that adds some DC connections, and presumably they're working on even higher wattage versions.
I think Tesla deserves some credit for attempting to solve the problem of range anxiety by building out the infrastructure themselves, but ultimately a universal standard would be far more beneficial.
Everyone agreed on it, and Tesla includes an SAE J1772 adapter with each car... but that's a very slow charge, under 20 kW. The SAE is working on higher power standards, but none of them are available or ready yet.
A simple electrical outlet (providing you mean "standard everyday type") can safely provide you at most ~1.5 kilowatts. Charging a long-distance electrical car from such a socket would take days.
A supercharger can do 170 miles of range in a 30 minute charge. A standard 110v outlet would take 40 hours to provide the same range.
It's worth noting that Tesla has adapters for pretty much any charging standard you can shake a stick at, and they include the most common ones with the car (110v, 240v, and J1772). They also sell adapters for the more esoteric sockets, and they even have one for the CHAdeMO fast charge standard, although those aren't cheap ($1000) and CHAdeMO tends to operate in the 50-60 kW range while superchargers are 120 kW (they're apparently looking into doing 150 kW).
There are a rather large number of CHAdeMO stations in the US, but they're very clustered, meant for driving inside a metro area rather than between them. And at less than half the charge rate of a supercharger, I'm not sure it's in the EV industry's best interest to lock themselves into a medium-speed "fast charge" standard like that...
Even the supercharger stations might not be the long-term answer. Improvements in batteries may enable even faster charging in the future, so who knows when all of this stuff will be widespread and standardized.
They came up with a way to get a 200+ mile pack in your Model S in about 90 seconds... an automated station swaps the battery pack. They have some videos showing it off. The plan is to roll them out at supercharger stations. You'll have the choice of supercharging for free (where 15 minutes gets you something like 85 miles of charge), or doing a battery swap where the swap costs about as much as an equivalent tank of gas at local prices (it's a rental, you're expected to stop and do the swap again on the way back to get your original battery).
Seriously, does it really matter anymore? You want to write in JavaScript? Do it. You want to write in something else? Most other languages can be compiled down to Javascript without any major performance penalties. Basically, these days you can do whatever you want in whatever language you want, so what does it matter in the end?
Did you actually read my post? I certainly never said that Apple actively blocking external GPUs was somehow evidence of them taking gaming seriously. That's absurd.
Sorry to double-reply here, but I thought I'd point out that one difference between previous external GPUs and what we're talking about here is the interface. Lots of previous solutions are using USB, which is a very resource intensive and low-speed interface. Thunderbolt is basically just PCI-Express over a cable, so when you route a PCI-Express graphics card over Thunderbolt, the PC and graphics card can communicate in their originally intended form. There is indeed a big drop in bandwidth because you're taking a graphics card meant for a 16x PCIe slot and running it over a Thunderbolt solution that behaves more like a 4x PCIe slot, but that's still fast enough for pretty decent performance. It helps that a graphics card on a 16x slot isn't anywhere close to bandwidth limited.
Summary of the results I posted there: a 2013 11" macbook air gets 69FPS running Bioshock Infinite on max quality at 1366x768 when combined with an external GTX 570. That ain't half bad.
Here's one set of benchmarks from a guy who did it using a rather indirect way involving a thunderbolt-to-expresscard adapter combined with an expresscard-to-pcie adapter:
You can see the benchmarks there for yourself. External monitor benchmarks are higher, probably because of the extra copying that has to go on to use the internal monitor. As an example, the first guy on an 11" 2013 macbook air got 69 FPS running Bioshock Infinite on max settings at 1366x768 (versus 15 FPS on the stock iGPU), and the second guy reported running Battlefield 3 on "Ultra" quality at 40FPS at 1920x1080.
Is there a big performance hit from doing all this, including using a dual-core ultrabook-class CPU? Sure, but it's hard to argue that the results aren't playable. It certainly proves the concept, and a properly supported solution at an affordable price could make one hell of an improvement to a notebook docking solution. Having the portability of an ultrabook, but docking it at home to your home monitor/speakers/mouse/keyboard/storage/network/etc? That'd be pretty nice. For many people, it might obviate the need to have both a desktop for gaming and a notebook for portability.
Apple is responsible for licensing accessories for OS X, regardless of what connection mechanism they use. Considering that such a solution would likely require driver support to work under OS X, that's relevant.
Annoyingly, you'd be able to use discrete graphics cards with any modern Mac if Apple would stop refusing to license thunderbolt PCIe bays. Benchmarks (via enthusiasts hacking together solutions) show that even a Macbook Air can provide good gaming performance (5x or more the framerate of the iGPU) when connected to a high-end graphics card via Thunderbolt (even on the internal display). Since Apple refuses to license them, however, you're restricted to doing it under bootcamp with expensive enterprise-targeted enclosures.
In other words, there is no technical reason why you couldn't simply plug an external discrete GPU into any Mac and instantly get massively improved gaming performance. Apple is actively blocking such things.
I never said that the bluray discs were more economical, only that the GP's 36GB figure didn't make sense. I'd also argue that while tape would beat optical in terms of price, it's questionable if it would win in terms of performance, convenience, or longevity. Let me discuss each:
Performance Tape wins on pure transfer rates for a single reader. LTO-4 does 120MB/s reads, while a 16x BluRay drive will top out at 72MB/s. The performance gap is small at that level, although bigger tapes would widen it (LTO-6 does 160MB/s). What's not clear is what the actual transfer rates would be in practice; Facebook's solution uses six drives in a single rack, while tape libraries also use multiple drives. On the other hand, the gap in random access times is huge. LTO-4 has an average random access time of about a minute, while a bluray disc is about one eighth of a second. Which performance metric (access times versus throughput) is more important depends entirely on the use case.
Convenience If you're in an automated tape or bluray library, convenience is irrelevant. If you're talking about using the media outside of that system, you'll find that a bluray disc can be read by a huge variety of computers equipped with bluray drives, while tape drives are virtually unseen outside of enterprise or the data center. If you're talking about pure density, Facebook's 1PB system claims to fit in a single rack, and looking at some commercial tape libraries it seems to me like they also need around a rack to do 1PB. Of course, if storage density is your goal regardless of cost or power usage, SuperMicro's more insane servers will do up to 4.3 PB in a single 40U rack.
Longevity Manufacturers of LTO tape advertise longevity of up to 30 years. Estimates of the longevity of bluray media is 50-100 years, with special archival discs (like those from Milleniata) advertising well over 1000.
So, while I'm not actually convinced that Facebooks bluray plans make sense, tape isn't as superior as everybody automatically assumes it is. Personally, I'd go for an HDD-based storage cloud. Longevity if properly designed and maintained is indefinite, cost is similar to tape, performance is vastly superior to either. And if you're designing the system with cold storage in mind, power usage could be quite low (shut hardware off, turning on to check health periodically). In other words, build a reliable system out of unreliable hardware by assuming failures will happen.
100GB discs are $40 each if you're buying one of them. But what if you go to the disc manufacture and order 10,000 of them? I think you'd be able to get them a wee bit cheaper.
How do you get 36GB per bluray disc? Commercially available BD-R discs come in a variety of capacities (25, 50, 100, 128) depending on number of layers and density, and 36GB isn't one of them.
10 BDs equals up to 1280GB of storage for quad-layer high density discs, although I can't find any of the 128GB discs for sale, only 100GB discs. Either way, the higher capacity discs are rather expensive, but if you're buying them in big enough bulk, it may not matter as much.
I suspect that the PS4/XB1 8-core processor probably has performance on par with a high-end dual core Haswell i3. If memory serves, AMD's Jaguar cores perform something like half what Intel's do clock-for-clock, and they're running them at half the clockspeed.
The choice to go with AMD likely had more to do with AMD's willingness to design custom solutions rather than any technical superiority on their part. Intel could probably have whipped up a custom solution with more GPU execution units if they wanted to, but they don't seem to have any interest in such things. And the GPU front is definitely one place that AMD has the edge on Intel.
Well, there was the Computer Riot of January 1969, at the time the largest student occupation in Canadian history. Fire did indeed burn down the computer labs.
If there was a giant fire in the maternity ward of the hospital I was born in three months earlier, that would make a pretty big difference in our overall life experiences, yes.
Samsung is Korean, they're the ones with the fabs Apple uses. TSMC (Taiwan) will be getting some of their business in the future (contracts have been signed to split 14nm A9 production between Samsung and TSMC), but they're not doing that yet.
Now, for actual factories, those are in China. But not the semiconductor fabs.
Slashdot Mirror
I wasn't implying that anybody should standardize on battery packs, only that one company had a solution to that problem for their own products. I suspect that eventually, one of two things will happen. Either Tesla will maintain a sufficiently large marketshare to be able to pressure the rest of the industry to move to their own charging standard, or their marketshare will be small enough that a more general industry standard takes off. At this moment, Tesla's superchargers are really the only chargers in their class deployed in the US (CHAdeMO is more widespread, but limited to about half as much wattage), but SAE has a 90 kW version of their J1772 connector that adds some DC connections, and presumably they're working on even higher wattage versions.
I think Tesla deserves some credit for attempting to solve the problem of range anxiety by building out the infrastructure themselves, but ultimately a universal standard would be far more beneficial.
Everyone agreed on it, and Tesla includes an SAE J1772 adapter with each car... but that's a very slow charge, under 20 kW. The SAE is working on higher power standards, but none of them are available or ready yet.
A simple electrical outlet (providing you mean "standard everyday type") can safely provide you at most ~1.5 kilowatts. Charging a long-distance electrical car from such a socket would take days.
A supercharger can do 170 miles of range in a 30 minute charge. A standard 110v outlet would take 40 hours to provide the same range.
CHAdeMO chargers (of which there are many more in the US) are 50+ kW, but they're still less than half the speed of a Tesla Supercharger.
It's worth noting that Tesla has adapters for pretty much any charging standard you can shake a stick at, and they include the most common ones with the car (110v, 240v, and J1772). They also sell adapters for the more esoteric sockets, and they even have one for the CHAdeMO fast charge standard, although those aren't cheap ($1000) and CHAdeMO tends to operate in the 50-60 kW range while superchargers are 120 kW (they're apparently looking into doing 150 kW).
There are a rather large number of CHAdeMO stations in the US, but they're very clustered, meant for driving inside a metro area rather than between them. And at less than half the charge rate of a supercharger, I'm not sure it's in the EV industry's best interest to lock themselves into a medium-speed "fast charge" standard like that...
Even the supercharger stations might not be the long-term answer. Improvements in batteries may enable even faster charging in the future, so who knows when all of this stuff will be widespread and standardized.
They came up with a way to get a 200+ mile pack in your Model S in about 90 seconds... an automated station swaps the battery pack. They have some videos showing it off. The plan is to roll them out at supercharger stations. You'll have the choice of supercharging for free (where 15 minutes gets you something like 85 miles of charge), or doing a battery swap where the swap costs about as much as an equivalent tank of gas at local prices (it's a rental, you're expected to stop and do the swap again on the way back to get your original battery).
So don't use Javascript. Treat it like assembly and write it in something else like C++ that gets compiled down to Javascript.
Seriously, does it really matter anymore? You want to write in JavaScript? Do it. You want to write in something else? Most other languages can be compiled down to Javascript without any major performance penalties. Basically, these days you can do whatever you want in whatever language you want, so what does it matter in the end?
Did you actually read my post? I certainly never said that Apple actively blocking external GPUs was somehow evidence of them taking gaming seriously. That's absurd.
Sorry to double-reply here, but I thought I'd point out that one difference between previous external GPUs and what we're talking about here is the interface. Lots of previous solutions are using USB, which is a very resource intensive and low-speed interface. Thunderbolt is basically just PCI-Express over a cable, so when you route a PCI-Express graphics card over Thunderbolt, the PC and graphics card can communicate in their originally intended form. There is indeed a big drop in bandwidth because you're taking a graphics card meant for a 16x PCIe slot and running it over a Thunderbolt solution that behaves more like a 4x PCIe slot, but that's still fast enough for pretty decent performance. It helps that a graphics card on a 16x slot isn't anywhere close to bandwidth limited.
I included links to two people who did it, including benchmarks, in another reply: http://slashdot.org/comments.p...
Summary of the results I posted there: a 2013 11" macbook air gets 69FPS running Bioshock Infinite on max quality at 1366x768 when combined with an external GTX 570. That ain't half bad.
Here's one set of benchmarks from a guy who did it using a rather indirect way involving a thunderbolt-to-expresscard adapter combined with an expresscard-to-pcie adapter:
http://forum.techinferno.com/d...
And here's a guy who did it more directly using a thunderbolt-to-pcie adapter:
http://forum.techinferno.com/d...
You can see the benchmarks there for yourself. External monitor benchmarks are higher, probably because of the extra copying that has to go on to use the internal monitor. As an example, the first guy on an 11" 2013 macbook air got 69 FPS running Bioshock Infinite on max settings at 1366x768 (versus 15 FPS on the stock iGPU), and the second guy reported running Battlefield 3 on "Ultra" quality at 40FPS at 1920x1080.
Is there a big performance hit from doing all this, including using a dual-core ultrabook-class CPU? Sure, but it's hard to argue that the results aren't playable. It certainly proves the concept, and a properly supported solution at an affordable price could make one hell of an improvement to a notebook docking solution. Having the portability of an ultrabook, but docking it at home to your home monitor/speakers/mouse/keyboard/storage/network/etc? That'd be pretty nice. For many people, it might obviate the need to have both a desktop for gaming and a notebook for portability.
Apple is responsible for licensing accessories for OS X, regardless of what connection mechanism they use. Considering that such a solution would likely require driver support to work under OS X, that's relevant.
Annoyingly, you'd be able to use discrete graphics cards with any modern Mac if Apple would stop refusing to license thunderbolt PCIe bays. Benchmarks (via enthusiasts hacking together solutions) show that even a Macbook Air can provide good gaming performance (5x or more the framerate of the iGPU) when connected to a high-end graphics card via Thunderbolt (even on the internal display). Since Apple refuses to license them, however, you're restricted to doing it under bootcamp with expensive enterprise-targeted enclosures.
In other words, there is no technical reason why you couldn't simply plug an external discrete GPU into any Mac and instantly get massively improved gaming performance. Apple is actively blocking such things.
I never said that the bluray discs were more economical, only that the GP's 36GB figure didn't make sense. I'd also argue that while tape would beat optical in terms of price, it's questionable if it would win in terms of performance, convenience, or longevity. Let me discuss each:
Performance
Tape wins on pure transfer rates for a single reader. LTO-4 does 120MB/s reads, while a 16x BluRay drive will top out at 72MB/s. The performance gap is small at that level, although bigger tapes would widen it (LTO-6 does 160MB/s). What's not clear is what the actual transfer rates would be in practice; Facebook's solution uses six drives in a single rack, while tape libraries also use multiple drives. On the other hand, the gap in random access times is huge. LTO-4 has an average random access time of about a minute, while a bluray disc is about one eighth of a second. Which performance metric (access times versus throughput) is more important depends entirely on the use case.
Convenience
If you're in an automated tape or bluray library, convenience is irrelevant. If you're talking about using the media outside of that system, you'll find that a bluray disc can be read by a huge variety of computers equipped with bluray drives, while tape drives are virtually unseen outside of enterprise or the data center. If you're talking about pure density, Facebook's 1PB system claims to fit in a single rack, and looking at some commercial tape libraries it seems to me like they also need around a rack to do 1PB. Of course, if storage density is your goal regardless of cost or power usage, SuperMicro's more insane servers will do up to 4.3 PB in a single 40U rack.
Longevity
Manufacturers of LTO tape advertise longevity of up to 30 years. Estimates of the longevity of bluray media is 50-100 years, with special archival discs (like those from Milleniata) advertising well over 1000.
So, while I'm not actually convinced that Facebooks bluray plans make sense, tape isn't as superior as everybody automatically assumes it is. Personally, I'd go for an HDD-based storage cloud. Longevity if properly designed and maintained is indefinite, cost is similar to tape, performance is vastly superior to either. And if you're designing the system with cold storage in mind, power usage could be quite low (shut hardware off, turning on to check health periodically). In other words, build a reliable system out of unreliable hardware by assuming failures will happen.
100GB discs are $40 each if you're buying one of them. But what if you go to the disc manufacture and order 10,000 of them? I think you'd be able to get them a wee bit cheaper.
How do you get 36GB per bluray disc? Commercially available BD-R discs come in a variety of capacities (25, 50, 100, 128) depending on number of layers and density, and 36GB isn't one of them.
10 BDs equals up to 1280GB of storage for quad-layer high density discs, although I can't find any of the 128GB discs for sale, only 100GB discs. Either way, the higher capacity discs are rather expensive, but if you're buying them in big enough bulk, it may not matter as much.
I suspect that the PS4/XB1 8-core processor probably has performance on par with a high-end dual core Haswell i3. If memory serves, AMD's Jaguar cores perform something like half what Intel's do clock-for-clock, and they're running them at half the clockspeed.
The choice to go with AMD likely had more to do with AMD's willingness to design custom solutions rather than any technical superiority on their part. Intel could probably have whipped up a custom solution with more GPU execution units if they wanted to, but they don't seem to have any interest in such things. And the GPU front is definitely one place that AMD has the edge on Intel.
Factories are not fabs.
The topic at hand were the great PC fires of spring 1984, pay attention.
Well, there was the Computer Riot of January 1969, at the time the largest student occupation in Canadian history. Fire did indeed burn down the computer labs.
If there was a giant fire in the maternity ward of the hospital I was born in three months earlier, that would make a pretty big difference in our overall life experiences, yes.
Maybe? I'm Canadian.
Apple is currently manufacturing the new Mac Pro in the US, but it's true that they didn't do any manufacturing themselves for a number of years.
Samsung is Korean, they're the ones with the fabs Apple uses. TSMC (Taiwan) will be getting some of their business in the future (contracts have been signed to split 14nm A9 production between Samsung and TSMC), but they're not doing that yet.
Now, for actual factories, those are in China. But not the semiconductor fabs.