And, if you knew how it all was strung together, you'd realise that they aren't always doing that infrastructure in the most efficient manner- or that they're putting the right amounts of effort into the same. In many cases, the wireless data's expensive, because they don't have their infrastructure done up right in the first place. Everyone keeps talking about limited bandwidth over the air- in many cases, they've got more than enough, it's the backhaul that's the problem with "the bandwidth". I know a bit about the nature of this having worked with a company that has traffic monitor hardware in many of the major telecoms. They've not done up their infrastructure to handle that which they're currently offering- they've slapped 3G data on top of a network that was fully optimized for voice traffic done up as SS7 type data (Most of the "improvements" in network infrastructure all equate to SONET and SS7 layered in on top of TCP/IP or expansions of that same thinking...) which isn't designed with TCP/IP type networking in mind.
Caps are something in-between what the GP poster implies and what you imply. Seriously.
$50/GB?? Wow... Verizon's charging $10/GB over the cap (Basically, they're billing at $10/GB and making you pay for 5 or 10 up-front - if you buy the 10GB slot, they give you a $20 discount...)
Still, can't get Sprint out where I am...only Verizon and AT&T serve the area I _need_ wireless Internet in. T-Mo's not an option either. I'm in one of their data and voice gaps.
Unfortunately, they don't offer wireline data to the Horse Farm I operate... The only choices right now are Verizon and AT&T 3G (Both work reasonably well...) or Huges or WildBlue Satellite Fraudband. Seriously. While the "economics" might not be there for one versus the other, unless the wireline players get their damn act together and wire up more than they have- the mobile players are going to have to concede a bit there. And, it should be noted, while it's true, they've got limits (wireline vendors have them too...) they're not hitting them, not even close- they're being cheap on their backhaul more than anything else.
Building your own machines will be a bit of a problem if all the new motherboards do the same thing. Do you honestly think the DIY vendors will not march to that drum unless they're gunning for the Linux user crowd in the first place?
That's less due to Android, and more due to the OEMs producing the devices. It's them that make it a minefield. And it's goofball apps that do things like record phone conversations, interact with specialized hardware on the devices, or side step screwball stuff from the "customized" Androids they're putting on things.
Yes. Until you try and turn it into a multi-core architecture with parallel branches, speculative & out-of-order execution, and all the other things x86 does in its sleep that are virgin territory with ARM. At that point, it's no longer proven and mature -- it's Internet Explorer 6 with lots of band-aids and a few upgrades pigtailed on & held in place with lots of duct tape.
Might be new territory for ARM, but it seems like that they're managing it pretty well within the context of their IA- A9's do REALLY well at it (so much so, that it's able to do 1.5x performance over the Atom at clock-for-clock and consume the same power profile as the earlier devices while doing it. X86's the minefield with all the old legacy calls, screwy memory access methods (unaligned reads/writes allowed, for example...) and the like. It's as much all the old stuff from X86 as what you mention (both combined, really) that makes an X86 a complex processor.
Excuse me... Large amounts of existing software would be easily ported over to the phone with or without X86. Code that couldn't be just simply recompiled is probably NOT something you want roaming loose on an embedded platform.
Actually, what LLVM brings to the table is final code delivery. If you've got a common intermediate code generator that produces fairly to really good native code (LLVM...) you can submit the intermediate code as part of an installer and as part of the process spin out a final binary. In theory, of course. So far, nobody's actually DONE something like that- but it's been more to the lack of someone seriously attempting it than other reasons at this time.
Because that only works in areas that they have coverage. Heh...in the area I would've needed their voice and data services, they've got bupkus and AT&T and Verizon have 3G, with Verizon having the more solid offering in the area.
If their "slower" connection was something realistic...say 768k or so...it's be a winner. You get dropped to 128k max at best when you hit the cap. There needs to be a bit better tiers than what they're offering. I can assure you I can blow through 2GB in about 2 or so weeks of average usage. I should know- I was doing that with Verizon and I was being careful with my usages. 5GB and 10GB should be there with a higher throttling speed if you buy one or both of those higher tiers. If they did that, they'd have a LOT more takers on their plans- just price them similarly to what Verizon's doing with the advantage of not being billed if you roll over.
He tested this with Ubuntu 11.04 under Unity and FreeBSD with KDE. It's a known that the configuration in Ubuntu there is a fairly massive performance sink on top of the already bad one Compiz can introduce with OpenGL stuff. As some pointed out in Phoronix' forums...all they did was benchmark Compiz/Unity versus KDE for all intents and purposes.
All things being equal, the results should be very similar if you're doing apples-to-apples comparisons since the OSes in question are similar in nature in this specific regard.
The big question would be, "Does FreeBSD's environment have Compiz or similar running?"
If you've got that running there's performance concerns, etc. that you have to deal with that make overall performance slower. My guess would be that it's not there- which means you're not making apples to apples comparisons between the OSes. If so, it's not that it's faster...it's that Phoronix didn't do the testing right.
Before anyone jumps to the conclusion that green technology is not profitable and therefore a big scam, or a modern religion if you will, with all of its guilt, shame and asking for money, let me state an opinion that might not be popular here: Maybe, just maybe, the subsidies was too low?
Subsidies too low... You've got to be kidding me.
A half a billion dollars. For that sort of money, I could have given the current industries a means unto which they could cleanly and easily manage their customer's USAGE of what they already are doing and green them up to buy us time for something that would work well (None of the current "green" answers are green or sustainable, in truth...)- and without, in the large, being obtrusive about that control. Seriously. I'm worrying about 1/100th of that and envisioning being able to make a go at employing many and kick-starting the business in question. 500 million is a damn waste, especially on something that they were just using the money to build production plants for- it's not R&D money we're talking here that was "wasted".
Most of the people are buying status and "cool" when they're buying those iPads. Those of us that want more control and ability have already bought Android stuff, where you can actually put on software that doesn't get the blessing from The Man. (As proof of this, the Facepla...er...book... app for Honeycomb doesn't exist- but the phone app itself is "stable" on it. All it took to get it loaded outside of the market was getting the APK for that and just loading it on the tablet, something that you can do with/without adb to help with. With an iPhone/iPad that would only be possible if you jailbroke the thing- iPwnt.
Which is fine, really, considering that all you need for "keyboard and mouse" is just a bluetooth keyboard for the tablets. My Acer Iconia A500 has been surprisingly useful, especially once I added a Bluetooth keyboard to the mix. If I expect to do content creation (documents, etc...) I just tote along two lightweight devices instead of one. Bam! Everything my netbook was supposed to do and didn't quite provide.
Having said this, for many, they don't need more than the 200-300 dollar devices (with 3G/4G access...)- the stuff we're seeing is more pricey toys than much else.
Barnes and Noble seems to not mind. The hackability comes from the fact that the device will boot anything that has the right bootloader and OS information on the microSD in the reader. They've not changed that behavior with subsequent releases of the Color hardware. It's such that you can run the pre-release Honeycomb that was hacked out of the Simulator image and a bit of CyanogenMod 7 kernel and other bits on the SD.
Several of the other vendors are no longer preventing the practice by way of locked down bootloaders, etc. Like HTC, they just tell you that it will likely void your warranty and you're on your own after doing it.
More specifically, their digestive tract is actually geared as a Carnivore's- therefore, even if it is that efficient, they're going to need to eat a damned lot of the stuff just to survive.
I didn't know there was such a thing as corporate suicide, but now we know that there is.
Heh...apparently he wasn't paying attention to the IBM v. TSG lawsuit that's still ongoing. Suing IBM for 3+ Billion over IP infringement on the Linux Kernel is pretty much "corporate suicide" and in the most painful, public manner possible. All HP's doing appears to be doing corporate suicide in the lingering zombie-like tradition.
How is it settled? Got data that consistently proves the theory that doesn't have points that might disprove it? No?
It's far from "settled" if you don't have that crucial item. And Science is never "settled", just so you know. When you use terms like that, you're not talking Science.
Is it above the threshold of reasonable denial? You claim that a theory with NO experimental data that is consistent is "settled". Sorry, not buying it. When you can predict with certainty what the rough behavior of the climate is going to be 3-5 years into the future, I might begin to believe that they've got a reasonable handle on things, science-wise. 10+ years, very much so.
They. Can't. Do. Even. That. Much.
Without it, you have theories and maybe some observations that might match the theory- but they may not. And, as one person pointed out earlier...if you have one solid, verifiable data point that does NOT match the theory, the theory DIES if it's Science. Throw it out. Start over. If you don't, it's nothing more than a religion unto itself and isn't Science. Seriously. Moreover, it does not help the position you are taking by people calling the people that don't agree with the theory because of a lack of data or data that conflicts (which are both present in this case...) "deniers". That's NOT Science.
Not really. The moment money enters into the equation anywhere as even a remote motivation, greed takes over and reputation can pretty much be thrown out the window if there's not something solid to back a given set of claims up.
Slashdot Mirror
And, if you knew how it all was strung together, you'd realise that they aren't always doing that infrastructure in the most efficient manner- or that they're putting the right amounts of effort into the same. In many cases, the wireless data's expensive, because they don't have their infrastructure done up right in the first place. Everyone keeps talking about limited bandwidth over the air- in many cases, they've got more than enough, it's the backhaul that's the problem with "the bandwidth". I know a bit about the nature of this having worked with a company that has traffic monitor hardware in many of the major telecoms. They've not done up their infrastructure to handle that which they're currently offering- they've slapped 3G data on top of a network that was fully optimized for voice traffic done up as SS7 type data (Most of the "improvements" in network infrastructure all equate to SONET and SS7 layered in on top of TCP/IP or expansions of that same thinking...) which isn't designed with TCP/IP type networking in mind.
Caps are something in-between what the GP poster implies and what you imply. Seriously.
$50/GB?? Wow... Verizon's charging $10/GB over the cap (Basically, they're billing at $10/GB and making you pay for 5 or 10 up-front - if you buy the 10GB slot, they give you a $20 discount...)
Still, can't get Sprint out where I am...only Verizon and AT&T serve the area I _need_ wireless Internet in. T-Mo's not an option either. I'm in one of their data and voice gaps.
Unfortunately, they don't offer wireline data to the Horse Farm I operate... The only choices right now are Verizon and AT&T 3G (Both work reasonably well...) or Huges or WildBlue Satellite Fraudband. Seriously. While the "economics" might not be there for one versus the other, unless the wireline players get their damn act together and wire up more than they have- the mobile players are going to have to concede a bit there. And, it should be noted, while it's true, they've got limits (wireline vendors have them too...) they're not hitting them, not even close- they're being cheap on their backhaul more than anything else.
Building your own machines will be a bit of a problem if all the new motherboards do the same thing. Do you honestly think the DIY vendors will not march to that drum unless they're gunning for the Linux user crowd in the first place?
That's less due to Android, and more due to the OEMs producing the devices. It's them that make it a minefield. And it's goofball apps that do things like record phone conversations, interact with specialized hardware on the devices, or side step screwball stuff from the "customized" Androids they're putting on things.
Might be new territory for ARM, but it seems like that they're managing it pretty well within the context of their IA- A9's do REALLY well at it (so much so, that it's able to do 1.5x performance over the Atom at clock-for-clock and consume the same power profile as the earlier devices while doing it. X86's the minefield with all the old legacy calls, screwy memory access methods (unaligned reads/writes allowed, for example...) and the like. It's as much all the old stuff from X86 as what you mention (both combined, really) that makes an X86 a complex processor.
Excuse me... Large amounts of existing software would be easily ported over to the phone with or without X86. Code that couldn't be just simply recompiled is probably NOT something you want roaming loose on an embedded platform.
Actually, what LLVM brings to the table is final code delivery. If you've got a common intermediate code generator that produces fairly to really good native code (LLVM...) you can submit the intermediate code as part of an installer and as part of the process spin out a final binary. In theory, of course. So far, nobody's actually DONE something like that- but it's been more to the lack of someone seriously attempting it than other reasons at this time.
Because that only works in areas that they have coverage. Heh...in the area I would've needed their voice and data services, they've got bupkus and AT&T and Verizon have 3G, with Verizon having the more solid offering in the area.
Arrgh...this is what I get for posting quickly on a tablet... The spelling and grammar on that post was atrocious.
If their "slower" connection was something realistic...say 768k or so...it's be a winner. You get dropped to 128k max at best when you hit the cap. There needs to be a bit better tiers than what they're offering. I can assure you I can blow through 2GB in about 2 or so weeks of average usage. I should know- I was doing that with Verizon and I was being careful with my usages. 5GB and 10GB should be there with a higher throttling speed if you buy one or both of those higher tiers. If they did that, they'd have a LOT more takers on their plans- just price them similarly to what Verizon's doing with the advantage of not being billed if you roll over.
He tested this with Ubuntu 11.04 under Unity and FreeBSD with KDE. It's a known that the configuration in Ubuntu there is a fairly massive performance sink on top of the already bad one Compiz can introduce with OpenGL stuff. As some pointed out in Phoronix' forums...all they did was benchmark Compiz/Unity versus KDE for all intents and purposes.
All things being equal, the results should be very similar if you're doing apples-to-apples comparisons since the OSes in question are similar in nature in this specific regard.
The big question would be, "Does FreeBSD's environment have Compiz or similar running?"
If you've got that running there's performance concerns, etc. that you have to deal with that make overall performance slower. My guess would be that it's not there- which means you're not making apples to apples comparisons between the OSes. If so, it's not that it's faster...it's that Phoronix didn't do the testing right.
They've said it was derived from Perspectives on the website. I'm curious as to what changes they've made.
Subsidies too low... You've got to be kidding me.
A half a billion dollars. For that sort of money, I could have given the current industries a means unto which they could cleanly and easily manage their customer's USAGE of what they already are doing and green them up to buy us time for something that would work well (None of the current "green" answers are green or sustainable, in truth...)- and without, in the large, being obtrusive about that control. Seriously. I'm worrying about 1/100th of that and envisioning being able to make a go at employing many and kick-starting the business in question. 500 million is a damn waste, especially on something that they were just using the money to build production plants for- it's not R&D money we're talking here that was "wasted".
Most of the people are buying status and "cool" when they're buying those iPads. Those of us that want more control and ability have already bought Android stuff, where you can actually put on software that doesn't get the blessing from The Man. (As proof of this, the Facepla...er...book... app for Honeycomb doesn't exist- but the phone app itself is "stable" on it. All it took to get it loaded outside of the market was getting the APK for that and just loading it on the tablet, something that you can do with/without adb to help with. With an iPhone/iPad that would only be possible if you jailbroke the thing- iPwnt.
Which is fine, really, considering that all you need for "keyboard and mouse" is just a bluetooth keyboard for the tablets. My Acer Iconia A500 has been surprisingly useful, especially once I added a Bluetooth keyboard to the mix. If I expect to do content creation (documents, etc...) I just tote along two lightweight devices instead of one. Bam! Everything my netbook was supposed to do and didn't quite provide.
Having said this, for many, they don't need more than the 200-300 dollar devices (with 3G/4G access...)- the stuff we're seeing is more pricey toys than much else.
Barnes and Noble seems to not mind. The hackability comes from the fact that the device will boot anything that has the right bootloader and OS information on the microSD in the reader. They've not changed that behavior with subsequent releases of the Color hardware. It's such that you can run the pre-release Honeycomb that was hacked out of the Simulator image and a bit of CyanogenMod 7 kernel and other bits on the SD.
Several of the other vendors are no longer preventing the practice by way of locked down bootloaders, etc. Like HTC, they just tell you that it will likely void your warranty and you're on your own after doing it.
HTC seems to have been the only one to actually "get" it with the others having pretty locked down phones including e-fuse locked down bootloaders.
Can != Does.
More specifically, their digestive tract is actually geared as a Carnivore's- therefore, even if it is that efficient, they're going to need to eat a damned lot of the stuff just to survive.
Heh...apparently he wasn't paying attention to the IBM v. TSG lawsuit that's still ongoing. Suing IBM for 3+ Billion over IP infringement on the Linux Kernel is pretty much "corporate suicide" and in the most painful, public manner possible. All HP's doing appears to be doing corporate suicide in the lingering zombie-like tradition.
How is it settled? Got data that consistently proves the theory that doesn't have points that might disprove it? No?
It's far from "settled" if you don't have that crucial item. And Science is never "settled", just so you know. When you use terms like that, you're not talking Science.
No kidding.
Is it above the threshold of reasonable denial? You claim that a theory with NO experimental data that is consistent is "settled". Sorry, not buying it. When you can predict with certainty what the rough behavior of the climate is going to be 3-5 years into the future, I might begin to believe that they've got a reasonable handle on things, science-wise. 10+ years, very much so.
They.
Can't.
Do.
Even.
That.
Much.
Without it, you have theories and maybe some observations that might match the theory- but they may not. And, as one person pointed out earlier...if you have one solid, verifiable data point that does NOT match the theory, the theory DIES if it's Science. Throw it out. Start over. If you don't, it's nothing more than a religion unto itself and isn't Science. Seriously. Moreover, it does not help the position you are taking by people calling the people that don't agree with the theory because of a lack of data or data that conflicts (which are both present in this case...) "deniers". That's NOT Science.
Not really. The moment money enters into the equation anywhere as even a remote motivation, greed takes over and reputation can pretty much be thrown out the window if there's not something solid to back a given set of claims up.