Reason, shoddy APIs, protocols which were undocumented for tales for more than a decade, screwing them over with IE for 10 years until the pain become almost unbearable, screwing their own dev houses over by discontinuing popular product lines or making them entirely incompatible etc... I still wonder why there still are people happiliy using their stuff in the dev world, seems sort of like masochism to me (I want to use Microsoft software because I feel happy to be screwed backwards every 5-6 years and I love the enduring pain IE induces)
Yes but the question is is it good enough and cheaper, an arm core can be licensed for a few pennies, and then the third party vendors jump in and trim it to any area you like. Atom is just a measly processor subpar to ARM with more power consumption and Intel only adds an even lousier graphics core to the mix. Intel might be one day good enough, but for now they are not, but as I sad Suns approach to x86 was too late, and to half assed. They first brought out first x86 servers and then it seemed like they dropped them again after one model, they got relatively late seriously into the game, but too late for them, they had somewhat of a success but it did not help them out anymore. I remember there was one server line between 2000 and 2005 and in between the relatively successful Cobalt line was killed. And 2006 or so they became serious about x86 at a time when Dell and Co already swept the floor with x86 based servers. Nevertheless Suns x86 servers sold well, but did not save them anymore, the grave already was to deep and on the high end the constant delays of the next gen Sparc killed them in the end.
Actually a defective series happens to every vendor (I would not call them producers anymore because the brand holders clearly are nothing anymore than the middle men between customer and the producers who sit nowdays mostly in china and taiwan, and to some extent in Korea, speaking of selling your silver ware for a dime, that happened here)
But the problem is if this happens a lot and the vendor tries to offload it by neglecting it or by simply trying to resolve it without official comments or by sending the marketing departement in than the reputation goes down the gutters.
I always get the feeling, if a serious damage happens companies first try to go to their marketing departements to resolve the problem instead of speaking openly and in an honest manner with their customers. And of course if the damge is really serious this always fails, because after a certain point people do not want to hear the corporate double speak anymore. (Not that they want to hear it at all in the first place, but in that kind of situation corporate marketing double speak is like a slap in the face, when they expect honest answers and honest solutions) I personally guess this comes from the fact that most of them are run by MBA type of people who lack any serious skill than getting their bottom line numbers right by pushing persons around and selling things, worst persons in charge of disaster control, where they need a more technology focused mind with a clear language and straight forward problem solving mentality.
Also there is another issue: Generally you can ruin your reputation entirely with the outsourcing of support to call centers just because it is cheaper, angry customers hitting a poor sod who has to go through his list and not being able to help really and cannot get out of the context due to language barriers, and not having any sort of technical expertiese and not really being able to forward the call to a proper (more expensive) technitian does not help to get the reputation back. To the worse those guys are not paid to solve problems but to cover as many calls as possible.
Poor support == a few dimes saved for the next year, but also less customers for the upcoming years, it is a simple as that, but does not get into the corporate minds filled by greed.
They tried with the N1 and failed, seems they have more success now they are selling it over the carriers. But I personally think Google can only win in this game, I am one who was driven towards the N1 by the shoddy update politics of the phone makers, and I love Android but I would never ever buy another phone which does not have official google support.
I would not call computers entirely predictable, there are too many influences which can derail them, for instance an accidental bit flip, hardware design issues etc...
The reason why Intel swept Suns floor was that x86 got competitive speedwise and then became faster and cheaper. Add to that the general Windows craze and Sun had a problem. The biggest problem Sun had in the recent years was that x86 servers were cheaper and faster than Suns offerings at the same price and with Linux, unix became a commodity. Sun recognized that too late. Low end Sparc was competitive until 2000 or so and guess what, what the critical turning point in Suns hardware business was...
The current situation is quit different, x86 compatibility is more a hindrance than a must and currently ARM is sweeping Intels floor and will be for the forseeable future unless Intel drops its x86 instruction set. Windows compatibility is a non argument there.
Entirely different situation. Sun simply was in a reality disortion field and never came out of it. ARM knows its job and definitely is not in one.
Besides that dont forget ARM just licenses the core, it is the entire chipset which counts with multitudes of manufacturers. The processor core is not as important as the rest, the radio chipset for the phone communications and the gpu etc... and that from a load of vendors all. Intel has a serious problem to gain a serious ground there because for every Intel design there are probably 5 faster arm designs which are also cheaper.
Moorestown is about 25% of the performance of CortexA9 (which is about twice as fast as a current Atom gen) and only competitive powerwise in idle mode. Hence Intel always just shows the idle numbers not the non idle ones.
Except that ARM is about 4 times as fast on CortexA9 than Moorestown is, also Moorestown numbers always are given in idle, go figure why. The platform simply is not competitive compared to the latest Cortex processor designs.
It does not say anything about the superiority that ARM dominates but if you look at the raw numbers and general performance numbers also performance per watt numbers than you can see why no one really switches over from ARM to Intel in the non netbook space where Windows compatibility is a non issue.
Arm simply has the better overall architecture. Atoms and generally also is cheaper. As soon as you count out the we need windows factor from any archtecture, Intel has lost the game with X86 everywhere. The architecture is garbage and the Risc core + interpreter on top of it just wastes cycles compared to a lean and clear instruction set ARM provides. Intel tries to cover that ground with shrinking die sizes, but now that ARM has got such a strong foothold, others with good fabs are interested to produce ARM with low die sizes as well, globalfoundries wants to produce ARM in 22nm so Intel has lost the game again.
ATOM still sucks big time in performance/watt compared to a Cortex A8 or Cortex A9, heck the Cortex A9 is overall faster than any Atom offering. It will be interesting that we see by the end of the year beginning of next year phones that are faster than the Atom based Netbooks, both in graphics performance and processor performance. (Most of not all phones currently sold are Cortex A8 based)
Jepp I do not know where Intel wants to go with this, ARM is in the Embedded world where Intel is in the Desktop world and so far their only argument, it runs windows (the other one is we have the worst processor instruction set ever designed in human history) does not hold a candle in the embedded world. The issue I see is, unless Intel drops their instruction set entirely and they cannot beat Arm in the Performance/Watt game.
And there is Microsoft only doing a subset of the spec, and users expecting to get the latest whiz bang stuff backported to IE6 and even worse IE6 mobile...
Well add to that that there is no HTML5 either, because the browser is stuck somewhere between IE6 and IE7... Good luck WinMo users in getting whizbang web applications, you will need it.
You have to add here, that back in Microsofts heydey lots of people were exposed as to PCs as their only alternative they did not even know that superior alternatives did exist (how many people did know macs when Windows 3.0 came out, I bet not too many those who did stuck to their macs)
This time the game is totally different, people have been exposed to the iPhone and Android and other alternative in masses, and there is no real benefit of the Windows monopoly they can leverage that people would flock to something inferior. They have to compete by being better and that wont work out as long as they copy Apple and Android and combine the worst issues iOS1 and Android 1.0 had and adding to that some shoddy stuff of their own (like again some garbage ie implementation a decade behind webkit used by iOS and Android)
Actually the biggest battery killer on android program wise are the task managers, the funny thing is Android really has a sophisticated task management system with various sleep states and programs which can expose what they multitask (sort of what iOS has but much more sophisticated, Apple took a lesson from Android in this regard but implemented only half of what Android provides)
The task killers have a tendency to constantly poll and never sleep you are better off without them.
Jepp seems like it, wait until they figure out how shoddy the browser is they are going to deliver with WinMo7, I just wonder what they are going to print to excuse that. As usual with Microsoft it will again be like that with WinMo8 everything will be better. This is a game Microsoft cannot win, unless they rethink their entire not invented here politics.
Theroetically this has been an issue, practically none copy paste on mobile phones has been around for ages, heck ask people who have nokia phones, or palm phones.
Well they left websurfing out deliberately, because they again are going to ship another shoddy ie mobile version (this time featurewise between ie6 and ie7 desktop) as browser of choice, and this time you wont have a chance to get a decent browser on your machine. But I am glad that Android and iOS combined have such a strong market presence that users who cry for dynamic webapps will be left out in the dust on that shitty browser without any mercy!
Jepp that is one of the biggest gripes I have, the mobile world mostly has settled down to the rather latest version of webkit (and opera) and unlike the desktop world you just could say, forget about IE, this version cannot do anything. But now again Microsoft again comes with another utterly shitty version of IE and people suddenly start to expect to see the whizbang stuff Android and iOS are capable of thanks to webkit. And this time Opera as the savior for WinMo is out of the game thanks to the refusal to expose native APIs. I really hope from a web developers perspective that WinMo7 is dead on arrival, otherwise we will run into the same browser mess the Desktop already has with everyone almost at the same level and Microsoft with a market share which cannot be ignored (and of course also their utterly dumb userbase which refuses to upgrade) dragging everything down and adding 30% to the entire development costs.
It is not that much more pricewise, and for the benefit of being able to daisy chain it I am willing to shell out a few dollars more, there are enough case options if the vendors try to hike the price for the standard boxed hds...
I never missed eSata here, eSata simply is inferior to firewire, ok good the HDs which have a decent firewire 800 port cost a little bit more but not that much more and you can daisy chain them. With eSata one string and then you cannot connect anything anymore. My HDs are all firewire 800 for the workhorses and USB for the el cheapo options.
Actually there is one benefit, it is eye tv, that thing is one of the best DTV solutions in existense, without EyeTV I probably already would be running Linux or Win7 on my mini, but EyeTV os too damned good especially in combination with digital cable or satellite. There is a world outside of eye tunes even on Apple and that one is better than what Apple tries to sell you.
That is my main concern as well, (that and the price). Apple has done faulty thermal designs for the sake of looking good in the past. Apple 2, some Macbook pro revisions, the Macbook Air which still in Revision 3 has a faulty thermal design, the famous cube the time capsule etc...
Reason, shoddy APIs, protocols which were undocumented for tales for more than a decade, screwing them over with IE for 10 years until the pain become almost unbearable, screwing their own dev houses over by discontinuing popular product lines or making them entirely incompatible etc...
I still wonder why there still are people happiliy using their stuff in the dev world, seems sort of like masochism to me (I want to use Microsoft software because I feel happy to be screwed backwards every 5-6 years and I love the enduring pain IE induces)
Yes but the question is is it good enough and cheaper, an arm core can be licensed for a few pennies, and then the third party vendors jump in and trim it to any area you like. Atom is just a measly processor subpar to ARM with more power consumption and Intel only adds an even lousier graphics core to the mix. Intel might be one day good enough, but for now they are not, but as I sad Suns approach to x86 was too late, and to half assed. They first brought out first x86 servers and then it seemed like they dropped them again after one model, they got relatively late seriously into the game, but too late for them, they had somewhat of a success but it did not help them out anymore. I remember there was one server line between 2000 and 2005 and in between the relatively successful Cobalt line was killed.
And 2006 or so they became serious about x86 at a time when Dell and Co already swept the floor with x86 based servers. Nevertheless Suns x86 servers sold well, but did not save them anymore, the grave already was to deep and on the high end the constant delays of the next gen Sparc killed them in the end.
Actually a defective series happens to every vendor (I would not call them producers anymore because the brand holders clearly are nothing anymore than the middle men between customer and the producers who sit nowdays mostly in china and taiwan, and to some extent in Korea, speaking of selling your silver ware for a dime, that happened here)
But the problem is if this happens a lot and the vendor tries to offload it by neglecting it or by simply trying to resolve it without official comments or by sending the marketing departement in than the reputation goes down the gutters.
I always get the feeling, if a serious damage happens companies first try to go to their marketing departements to resolve the problem instead of speaking openly and in an honest manner with their customers. And of course if the damge is really serious this always fails, because after a certain point people do not want to hear the corporate double speak anymore. (Not that they want to hear it at all in the first place, but in that kind of situation corporate marketing double speak is like a slap in the face, when they expect honest answers and honest solutions)
I personally guess this comes from the fact that most of them are run by MBA type of people who lack any serious skill than getting their bottom line numbers right by pushing persons around and selling things, worst persons in charge of disaster control, where they need a more technology focused mind with a clear language and straight forward problem solving mentality.
Also there is another issue:
Generally you can ruin your reputation entirely with the outsourcing of support to call centers just because it is cheaper, angry customers hitting a poor sod who has to go through his list and not being able to help really and cannot get out of the context due to language barriers, and not having any sort of technical expertiese and not really being able to forward the call to a proper (more expensive) technitian does not help to get the reputation back. To the worse those guys are not paid to solve problems but to cover as many calls as possible.
Poor support == a few dimes saved for the next year, but also less customers for the upcoming years, it is a simple as that, but does not get into the corporate minds filled by greed.
They tried with the N1 and failed, seems they have more success now they are selling it over the carriers.
But I personally think Google can only win in this game, I am one who was driven towards the N1 by the shoddy update politics of the phone makers, and I love Android but I would never ever buy another phone which does not have official google support.
I would not call computers entirely predictable, there are too many influences which can derail them, for instance an accidental bit flip, hardware design issues etc...
The reason why Intel swept Suns floor was that x86 got competitive speedwise and then became faster and cheaper. Add to that the general Windows craze and Sun had a problem.
The biggest problem Sun had in the recent years was that x86 servers were cheaper and faster than Suns offerings at the same price and with Linux, unix became a commodity. Sun recognized that too late. Low end Sparc was competitive until 2000 or so and guess what, what the critical turning point in Suns hardware business was...
The current situation is quit different, x86 compatibility is more a hindrance than a must and currently ARM is sweeping Intels floor and will be for the forseeable future unless Intel drops its x86 instruction set. Windows compatibility is a non argument there.
Entirely different situation. Sun simply was in a reality disortion field and never came out of it.
ARM knows its job and definitely is not in one.
Besides that dont forget ARM just licenses the core, it is the entire chipset which counts with multitudes of manufacturers. The processor core is not as important as the rest, the radio chipset for the phone communications and the gpu etc... and that from a load of vendors all. Intel has a serious problem to gain a serious ground there because for every Intel design there are probably 5 faster arm designs which are also cheaper.
Moorestown is about 25% of the performance of CortexA9 (which is about twice as fast as a current Atom gen) and only competitive powerwise in idle mode. Hence Intel always just shows the idle numbers not the non idle ones.
Except that ARM is about 4 times as fast on CortexA9 than Moorestown is, also Moorestown numbers always are given in idle, go figure why. The platform simply is not competitive compared to the latest Cortex processor designs.
It does not say anything about the superiority that ARM dominates but if you look at the raw numbers and general performance numbers also performance per watt numbers than you can see why no one really switches over from ARM to Intel in the non netbook space where Windows compatibility is a non issue.
Arm simply has the better overall architecture. Atoms and generally also is cheaper.
As soon as you count out the we need windows factor from any archtecture, Intel has lost the game with X86 everywhere. The architecture is garbage and the Risc core + interpreter on top of it just wastes cycles compared to a lean and clear instruction set ARM provides.
Intel tries to cover that ground with shrinking die sizes, but now that ARM has got such a strong foothold, others with good fabs are interested to produce ARM with low die sizes as well, globalfoundries wants to produce ARM in 22nm so Intel has lost the game again.
ATOM still sucks big time in performance/watt compared to a Cortex A8 or Cortex A9, heck the Cortex A9 is overall faster than any Atom offering.
It will be interesting that we see by the end of the year beginning of next year phones that are faster than the Atom based Netbooks, both in graphics performance and processor performance.
(Most of not all phones currently sold are Cortex A8 based)
Jepp I do not know where Intel wants to go with this, ARM is in the Embedded world where Intel is in the Desktop world and so far their only argument, it runs windows (the other one is we have the worst processor instruction set ever designed in human history) does not hold a candle in the embedded world.
The issue I see is, unless Intel drops their instruction set entirely and they cannot beat Arm in the Performance/Watt game.
And there is Microsoft only doing a subset of the spec, and users expecting to get the latest whiz bang stuff backported to IE6 and even worse IE6 mobile...
Froyo still has not come out of the beta stage :-)
I guess google was waiting for Flash and Froyo now will be rolled out. Lets hope.
Well that is not directly a security problem it is just added info in the url...
Well add to that that there is no HTML5 either, because the browser is stuck somewhere between IE6 and IE7...
Good luck WinMo users in getting whizbang web applications, you will need it.
You have to add here, that back in Microsofts heydey lots of people were exposed as to PCs as their only alternative they did not even know that superior alternatives did exist (how many people did know macs when Windows 3.0 came out, I bet not too many those who did stuck to their macs)
This time the game is totally different, people have been exposed to the iPhone and Android and other alternative in masses, and there is no real benefit of the Windows monopoly they can leverage that people would flock to something inferior. They have to compete by being better and that wont work out as long as they copy Apple and Android and combine the worst issues iOS1 and Android 1.0 had and adding to that some shoddy stuff of their own (like again some garbage ie implementation a decade behind webkit used by iOS and Android)
Actually the biggest battery killer on android program wise are the task managers, the funny thing is Android really has a sophisticated task management system with various sleep states and programs which can expose what they multitask (sort of what iOS has but much more sophisticated, Apple took a lesson from Android in this regard but implemented only half of what Android provides)
The task killers have a tendency to constantly poll and never sleep you are better off without them.
Jepp seems like it, wait until they figure out how shoddy the browser is they are going to deliver with WinMo7, I just wonder what they are going to print to excuse that.
As usual with Microsoft it will again be like that with WinMo8 everything will be better.
This is a game Microsoft cannot win, unless they rethink their entire not invented here politics.
Theroetically this has been an issue, practically none copy paste on mobile phones has been around for ages, heck ask people who have nokia phones, or palm phones.
Well they left websurfing out deliberately, because they again are going to ship another shoddy ie mobile version (this time featurewise between ie6 and ie7 desktop) as browser of choice, and this time you wont have a chance to get a decent browser on your machine.
But I am glad that Android and iOS combined have such a strong market presence that users who cry for dynamic webapps will be left out in the dust on that shitty browser without any mercy!
Jepp that is one of the biggest gripes I have, the mobile world mostly has settled down to the rather latest version of webkit (and opera) and unlike the desktop world you just could say, forget about IE, this version cannot do anything.
But now again Microsoft again comes with another utterly shitty version of IE and people suddenly start to expect to see the whizbang stuff Android and iOS are capable of thanks to webkit. And this time Opera as the savior for WinMo is out of the game thanks to the refusal to expose native APIs.
I really hope from a web developers perspective that WinMo7 is dead on arrival, otherwise we will run into the same browser mess the Desktop already has with everyone almost at the same level and Microsoft with a market share which cannot be ignored (and of course also their utterly dumb userbase which refuses to upgrade) dragging everything down and adding 30% to the entire development costs.
It is not that much more pricewise, and for the benefit of being able to daisy chain it I am willing to shell out a few dollars more, there are enough case options if the vendors try to hike the price for the standard boxed hds...
I never missed eSata here, eSata simply is inferior to firewire, ok good the HDs which have a decent firewire 800 port cost a little bit more but not that much more and you can daisy chain them. With eSata one string and then you cannot connect anything anymore.
My HDs are all firewire 800 for the workhorses and USB for the el cheapo options.
Actually there is one benefit, it is eye tv, that thing is one of the best DTV solutions in existense, without EyeTV I probably already would be running Linux or Win7 on my mini, but EyeTV os too damned good especially in combination with digital cable or satellite.
There is a world outside of eye tunes even on Apple and that one is better than what Apple tries to sell you.
That is my main concern as well, (that and the price). Apple has done faulty thermal designs for the sake of looking good in the past.
Apple 2, some Macbook pro revisions, the Macbook Air which still in Revision 3 has a faulty thermal design, the famous cube the time capsule etc...