Honestly, branches and forks aside (or not), software revisions should be marked with an integer. I vaguely remember that Linus wanted to start doing the same for the kernel.
Similarly, for those who like to append the date as a revision number - smarten up and start using the YYYYMMDD format - it's an integer that always sorts in the correct order, at least until the year 9999 + 1.
I've been waiting for somebody to start pairing the dual core atom 330 with nvidia's geforce 9400m for a while. In my opinion, this is more than enough horsepower for the average end-user desktop, htpc, or netbook. And so power efficient too!
I'm seriously considering one of these zotac boards for an htpc, while using an even more power efficient arm netbook (e.g. always innovating touchbook) for my portable linux workstation. The zotac board would also likely serve as a good hackintosh, no?
Via could also potentially profit if they paired one of their up-and-coming dual-core chips with the geforce 9400m.
Forgive me for disagreeing with you, which I'm doing based on past experience.
I think you misinterpreted the message. If I'm applying to, say, CompanyX as a test engineer, even if I work there for 10 years, it's very unlikely that I would then be accepted for a job as a VLSI engineer. It's a completely different department, and a completely different team.
Then they would need to go through the hiring process for two people and retrain two people, as opposed to hiring and training 1 person. Many large, especially unionized, companies are legally required to go through an interviewing process for each position, whether they already have someone in mind, are offering the job to an internal person, or external.
Your argument is potentially valid for a promotion within the same team and / or department. That's usually the difference between being a junior engineer and an experienced or senior engineer. I would call that a raise, not necessarily a promotion.
Essentially, I'm saying that it's a bad idea to apply to the test engineering department if one wanted to design silicon.
Incidentally, for a job in VLSI, a master's degree is usually a minimum requirement.
The down-side to this, is that lots of students are doing the same thing (higher education) as a result of the economy. Therefore, in a couple of years, there will likely be a flood of people looking for the same job you are, with the same credentials.
Do the master's but keep in mind, that you will need to improve your professional experience on your own, e.g. with open-source / open-hardware design.
If you feel really brave, then come up with a marketable business concept and look for venture capital through government grants. It's a big risk, but can really pay off if you have the right mind for it.
I can't agree more. I was lucky enough to be able to work 100% from home for my company, right out of my bachelor's degree. I told them I wanted to do a master's degree in a relevant field, and they saw a benefit from that. Now I'm doing my master's degree abroad in Germany, where there is no tuition. This allows me to continuously improve my professional experience, taking a slight reduction in salary, while getting a master's in a highly relevant field. The time-zone difference means that I can work evenings from Germany and be online as if I were back in North America, so there really is no difference from my employer's perspective. The lecture schedule is fairly relaxed, and I even have a couple of days off every week.
Given the current state of the economy, any job you apply for will likely also see applications from professionals who have had 5 years of experience or more. In other words, you have slim chances.
In light of that, my suggestion is to do a master's degree but not in the US. Tuition in the US (and Canada for that matter) is far beyond what its actually worth. Go go a European country where tuition is subsidized. The "worst" part is that you have to learn a new language (unless it's an international program in english), and that's actually a very marketable skill.
On the other hand, if the economy were better, I would say that you should get a job; good work experience is exponentially more valuable than academic "training".
However, when you do begin your career and start looking for employment, and I can't stress this point more - do not apply to a lower-level position in hopes that you will one-day get a promotion. Internal promotions require at least twice the time and money of retraining and rehiring that an external hire requires. In short, don't take the first job that comes your way. Hold out for one that seems challenging and dually rewarding.
If you really want to gain some experience, port Linux / Android to a new device while you're doing your master's degree. You don't necessarily need to be employed to gain some practical, modern, and highly valuable experience. There's a whole world of open-design out there waiting for you - take OpenMoko for example. Why not redesign their next handset and make some major improvements?
They do, of course, have a procedure / appliance for re-flashing using some sort of boundary-scan method (e.g. JTAG, read: slow,inconvenient), but they also have another procedure for reflashing using some sort of bootloader-assisted method (e.g. ethernet, read:fast,convenient). Every second counts.
In any case, the real issue is whether or not it is in Nintendo's best interest to enforce their statuatory warranty. The answer is yes.
I'm an expert, and it's not very hard to see if somebody has modified the firmware. This is what I do for a living, and I do a pretty good job of it according to my clients.
First - you are assuming that Nintendo would prefer to use a boundary-scan method such as JTAG to flash the entire ROM (i.e. bootloader, os, and main filesystem) as opposed to single parts of it (i.e. os, main filesystem) using a faster, more sophisticated method such as ethernet. If the bootloader is not damaged, then it's significantly faster to use ethernet (which requires bootloader software) to re-flash the filesystem image. If the bootloader is damaged, then JTAG is usually the only remaining option but JTAG is dramatically slower (this is a well-known fact). If Nintendo is not legally or financially obligated to repair the device, then someone must pay for the time of the employees who facilitate the repair.
Secondly, you're assuming that Nintendo is uninterested in enforcing their own warranty conditions, which is quite naive. Check out the post titled "Is it so hard to understand?". If a warranty is voided by installing unauthorized firmware, then the warranty is, in fact, void. Whether or not the customer informed Nintendo that they had modified the firmware is irrelevant. An experienced engineer can easily examine firmware and detect if was i) a softwarehack or ii) if it was genuinely a very improbable phenomenon that corrupted some bits in the firmware storage device. Statistically, it's the former reason 99.9% of the time. The only other case is a hardware defect, which falls under under warranty, of course.
On the other hand, if Nintendo was aware that they would be required to charge the customer for a repair, in a fully legal sense, they should have a) obtained confirmation that the client would be willing to pay for the repair, or b) sent the unit back to the customer, unrepaired. I find it unlikely that they would not have confirmed the cost of repair with the customer.
People (including you) have to respect that companies are not obligated to repair items that aren't under warranty. Companies do so for two main reasons 1) to increase sales, and 2) out of respect for the environmental regulations. Personally, I am very happy that Nintendo and this customer negotiated a repair, rather than having the console end up in a landfill (assuming that the customer would otherwise not fix the device).
I do understand German, and this receipt says that the internal software (i.e. firmware) was modified.
Realistically, if someone voids their warranty by modifying the firmware on a device, regardless of whether it was done directly by the user or by the homebrew software, then they should expect to pay to have that repaired.
Why? because it takes time. Usually companies have very specific procedures for quickly re-flashing using their existing boot firmware. However, if that boot firmware is modified (i.e. the device is 'bricked') then that procedure needs to be changed. In this case, the engineer would need to 'hack' their own device to get it up to usable standards again (i.e. 'unbricked').
In the non-wii-world (i.e. reality), people that fix things also need to be paid adequately.
This customer should be happy that their bricked wii was even serviceable and that they weren't forced to cough up an extra 40€ to buy a brand-new Wii for 250€.
I put a highly customized version of Gentoo on my EEE. It was essentially the same thing that I had on my workstation, but tuned for binary packages and never writing anything to disk. It took a while to configure, and it's still not perfect, but it's exactly what I want. I tried Ubuntu as well, but I missed Portage:)
ASUS made a mistake with Xandros directly from the beginning, and that's probably what is showing with the number of returned netbooks. If they had gone with Ubuntu, and even just the regular desktop layout, they definitely would have had more appeal.
The Intel Atom based netbooks are definitely good platforms for OS X. A friend of mine had it on his EEE 1000H and I was quite impressed.
If I wasn't addicted to changeability & source (e.g. the ability to change virtually anything), with Linux, then I would probably also have installed OS X. I still have usability issues with it though. Namely, Mac OS X has shortcut keys that are set in stone, much like Windows.
Based on using the Xandros/KDE install on the original EEE, I can understand people's urge to return them. Most people aren't linux-adept or patient enough to learn how and install something more familiar like Ubuntu. Even then, Ubuntu is very different than, say Windows XP.
This is just another excuse for 'free desktop' people to clone the Windows(R) look-and-feel, and maybe even have a highly tuned and compatible WINE installation out of the box. Now there are actually legal grounds for it, because if MicroSoft absorbs the netbook market too, then there really is no Linux sanctuary on consumer devices. That spells monopoly in the NetBook OS market for MicroSoft. Also, in terms of mobile devices, it's also very hard to say whether or not Android will live up to its expectations.
Maybe the user-interface should be standardized somehow for the GUI on consumer products. At least that way, MS would be forced to code to a certain standard, which would be the same standard that everyone would be forced to code to... e.g. Apple, various Linux's, etc.
ISO, or ISO-like standardization has seemingly helped even the playing field in the past; UNIX to POSIX for example, or MS Word to ODF.
It's true - there needs to be a lot more community involvement with the FreeRunner in order for it to be successful. A good example, is the need for the SMedia software stack to start approaching completion, enabling OpenGL-ES-compatible hardware acceleration.
This device is honestly perfect for any undergraduate course in embedded systems, as the designer has full access to the hardware and it is 'unbrickable'. Today, I believe that computer engineers should really be taking a 2-semester course in low-level embedded systems programming. The course should encompass everything from JTAG debugging to writing a bootloader and launching Linux. IMHO, Google should be sponsoring university programs to provide students with this type of learning material, if they ever want Android to really take off.
The title for this article should read something more along the lines of 'OpenMoko Inc Struggle in
Tough Economic Times'. Just like all other companies, they are feeling the economic crunch as well. The original article says that they had to lay-off a fraction of their staff already - something all-too familiar for most companies today.
At the end of the original article, SMP also states that they will, in the future produce another mobile phone. So the title of this article is clearly wrong.
Marvell's SheevaPlug runs at 1.2 GHz and only consumes 5 watts of power (2 watts actually consumed by the processor).
I apologize for quoting myself... but for the sake of correctness...
The PXA168 (the SoC which powers the SheevaPlug) is, in fact, a system-on-a-chip. So the actual processor draws less power than the 2 watts dawn for the entire SoC.
Has a mainboard for x86 ever, in the entire history of Intel, ever drawn 5 Watts in total? Obviously not.
I would say that your response to this article was nonsense. Do you know anything about computer architecture?
What do you think Intel's new Larabee architecture is? Each core in the Larabee architecture uses in-order execution, like most ARM processors. I.e. Intel realizes that MHz-per-mW == success in today's mobile world.
Superscalar / out-of-order execution, present in all modern Intel chips, requires minimally O(n^2) control logic complexity. In most cases, the chip-area dedicated to SS control logic is proportional to the actual execution logic. By reverting to in-order execution (thereby freeing up much more silicon for multiple cores), Intel will see a huge performance increase with a fraction of the power consumption per-core.
And no, today's ARM chips are not running at 300 MHz. TI's OMAP 3530 (BeagleBoard, OpenPandora), which is already old, was designed to run at 600 MHz. Marvell's SheevaPlug runs at 1.2 GHz and only consumes 5 watts of power (2 watts actually consumed by the processor).
Going back to complexity, it's a simple mathematical relationship, given that transistor count is proportional to power dissipation. Using in-order architecture, 2 cores of execution, requires approximately 2 times the power. Using out-of-order execution 2 cores of execution requires 4-times the power (or likely more) for the same number of threads.
Regardless of the chip-vendor, in-order execution is more power efficient AND more space-efficient. Therefore, in the mobile world, where both power and space are 'costly', in-order architectures dominate (yes, that is true today).
ARM has had about a decade of time to perfect in-order architecture, while Intel has been focusing on out-of-order execution, which does give ARM the advantage in the mobile world.
Some ARM chips these days go far beyond what one would expect from a RISC micro-controller.
Two great examples are the BeagleBoard, which also uses the same hardware as the OpenPandora gaming platform. Both of the aformentioned devices are running a fully open-source Linux-powered stack, complete with a fully-functional Desktop environment. For developers, you can program for these devices using the standar GNU tools (c,c++), and of course several JavaVM's are already ported to the OMAP (JamVM,OpenJDK,Kaffee,etc)
.
The TI OMAP 3530 chip that powers the BeagleBoard and OpenPandora has a 3D graphics acceleration unit, a 2D (video) graphics acceleration unit, and a built in DSP for audio and general-purpose number crunching.
Without question a large majority of users do little more with their mobile computers than browse the internet, write email, and occasionally stream video - all things that are fully achievable with today's ARM chips.
ARM chip manufacturers are also starting to put multi-core chips on the market, and while the transistor complexity is nothing like Intel's out-of-order logic, multi-core, still means serious multi-threading. With clock frequencies approaching and exceeding 1GHz, I would not be surprised to see ARM take over the mobile and netbook markets, especially considering how little power they actually consume.
Furthermore, even high-end x86 laptops could benefit from having an ARM co-processor for instant-on and mobile operation, while the x86 processor could be used for more compute-intensive applications such as CAD or video processing. Imagine battery power extended by a factor of ten for everyone!
I would argue that most of the things MS claims to invent were actually invented by promising technology firms who sold their rights to MS. Many people would probably argue against Visual Studio being the best IDE, in favor of Eclipse, and MS was not the first company to have a large touch interface by a long shot.
Ubuntu is great for beginners in the Linux world, but also for people who can't afford to risk compiling code from scratch (i.e. running into bugs, losing productivity time). I compile all of my applications from scratch, and I fix / file bugs often. My reward is that I'm participating in the greater picture, and quite often others tell me that I've helped them out with similar problems.
I also use a desktop configuration that appears like a cross between Windows XP and Mac OS X, but it's running Gnome. That's all being run on an EEE PC that I've significantly customized both hardware and software-wise.
Today, what I call innovative is mostly what's happening on mobile devices. I'm not an Apple 'snob' by any means... and no longer even own an apple product because my 'retro' flash-modded 4th gen ipod was lost on a plane... but I would really have to say, that Apple has been leading innovation in the mobile world for a while. Android has begun to impact the mobile world, and I feel nice that I was one of the first people to port it to a new device. If Apple's mobile platform was open source, I would probably expand on it, but as far as I know, Android is as close as I can get.
I would gladly opt-in for such a plan.
Although, I would prefer that it also carried the right to share media created by artists in other countries. Of course, I would only share media with others who have also opted in for such a plan.
There is far too much complexity associated with governments and other organizations spying on their citizens, repeatedly coming up with ineffective DRM techniques, etc. For a small monthly fee, I have no problems whatsoever paying for the right to have un-complicated rights to enjoy my favourite media and share it with friends.
Slashdot Mirror
Honestly, branches and forks aside (or not), software revisions should be marked with an integer. I vaguely remember that Linus wanted to start doing the same for the kernel. Similarly, for those who like to append the date as a revision number - smarten up and start using the YYYYMMDD format - it's an integer that always sorts in the correct order, at least until the year 9999 + 1.
I've been waiting for somebody to start pairing the dual core atom 330 with nvidia's geforce 9400m for a while. In my opinion, this is more than enough horsepower for the average end-user desktop, htpc, or netbook. And so power efficient too!
I'm seriously considering one of these zotac boards for an htpc, while using an even more power efficient arm netbook (e.g. always innovating touchbook) for my portable linux workstation. The zotac board would also likely serve as a good hackintosh, no?
Via could also potentially profit if they paired one of their up-and-coming dual-core chips with the geforce 9400m.
Forgive me for disagreeing with you, which I'm doing based on past experience.
I think you misinterpreted the message. If I'm applying to, say, CompanyX as a test engineer, even if I work there for 10 years, it's very unlikely that I would then be accepted for a job as a VLSI engineer. It's a completely different department, and a completely different team.
Then they would need to go through the hiring process for two people and retrain two people, as opposed to hiring and training 1 person. Many large, especially unionized, companies are legally required to go through an interviewing process for each position, whether they already have someone in mind, are offering the job to an internal person, or external.
Your argument is potentially valid for a promotion within the same team and / or department. That's usually the difference between being a junior engineer and an experienced or senior engineer. I would call that a raise, not necessarily a promotion.
Essentially, I'm saying that it's a bad idea to apply to the test engineering department if one wanted to design silicon.
Incidentally, for a job in VLSI, a master's degree is usually a minimum requirement.
The down-side to this, is that lots of students are doing the same thing (higher education) as a result of the economy. Therefore, in a couple of years, there will likely be a flood of people looking for the same job you are, with the same credentials.
Do the master's but keep in mind, that you will need to improve your professional experience on your own, e.g. with open-source / open-hardware design.
If you feel really brave, then come up with a marketable business concept and look for venture capital through government grants. It's a big risk, but can really pay off if you have the right mind for it.
I forgot to mention, that an added bonus is cheap vacationing!! Flights are unbelievably cheap here, to fly to Spain, Italy, Greece, etc.
I can't agree more. I was lucky enough to be able to work 100% from home for my company, right out of my bachelor's degree. I told them I wanted to do a master's degree in a relevant field, and they saw a benefit from that. Now I'm doing my master's degree abroad in Germany, where there is no tuition. This allows me to continuously improve my professional experience, taking a slight reduction in salary, while getting a master's in a highly relevant field. The time-zone difference means that I can work evenings from Germany and be online as if I were back in North America, so there really is no difference from my employer's perspective. The lecture schedule is fairly relaxed, and I even have a couple of days off every week.
Given the current state of the economy, any job you apply for will likely also see applications from professionals who have had 5 years of experience or more. In other words, you have slim chances.
In light of that, my suggestion is to do a master's degree but not in the US. Tuition in the US (and Canada for that matter) is far beyond what its actually worth. Go go a European country where tuition is subsidized. The "worst" part is that you have to learn a new language (unless it's an international program in english), and that's actually a very marketable skill.
On the other hand, if the economy were better, I would say that you should get a job; good work experience is exponentially more valuable than academic "training".
However, when you do begin your career and start looking for employment, and I can't stress this point more - do not apply to a lower-level position in hopes that you will one-day get a promotion. Internal promotions require at least twice the time and money of retraining and rehiring that an external hire requires. In short, don't take the first job that comes your way. Hold out for one that seems challenging and dually rewarding.
If you really want to gain some experience, port Linux / Android to a new device while you're doing your master's degree. You don't necessarily need to be employed to gain some practical, modern, and highly valuable experience. There's a whole world of open-design out there waiting for you - take OpenMoko for example. Why not redesign their next handset and make some major improvements?
They do, of course, have a procedure / appliance for re-flashing using some sort of boundary-scan method (e.g. JTAG, read: slow,inconvenient), but they also have another procedure for reflashing using some sort of bootloader-assisted method (e.g. ethernet, read:fast,convenient). Every second counts.
In any case, the real issue is whether or not it is in Nintendo's best interest to enforce their statuatory warranty. The answer is yes.
A BIOS is something that exists on a PC. The article you referenced has nothing to do with this article.
I'm an expert, and it's not very hard to see if somebody has modified the firmware. This is what I do for a living, and I do a pretty good job of it according to my clients.
Agreed!!
A note to all of those kids at home:
If you can't fix it yourself, then be prepared to pay for a repair service when you void the warranty on something :)
That's where you're pathetically mistaken.
I'm most definitely not mistaken.
First - you are assuming that Nintendo would prefer to use a boundary-scan method such as JTAG to flash the entire ROM (i.e. bootloader, os, and main filesystem) as opposed to single parts of it (i.e. os, main filesystem) using a faster, more sophisticated method such as ethernet. If the bootloader is not damaged, then it's significantly faster to use ethernet (which requires bootloader software) to re-flash the filesystem image. If the bootloader is damaged, then JTAG is usually the only remaining option but JTAG is dramatically slower (this is a well-known fact). If Nintendo is not legally or financially obligated to repair the device, then someone must pay for the time of the employees who facilitate the repair.
Secondly, you're assuming that Nintendo is uninterested in enforcing their own warranty conditions, which is quite naive. Check out the post titled "Is it so hard to understand?". If a warranty is voided by installing unauthorized firmware, then the warranty is, in fact, void. Whether or not the customer informed Nintendo that they had modified the firmware is irrelevant. An experienced engineer can easily examine firmware and detect if was i) a softwarehack or ii) if it was genuinely a very improbable phenomenon that corrupted some bits in the firmware storage device. Statistically, it's the former reason 99.9% of the time. The only other case is a hardware defect, which falls under under warranty, of course.
On the other hand, if Nintendo was aware that they would be required to charge the customer for a repair, in a fully legal sense, they should have a) obtained confirmation that the client would be willing to pay for the repair, or b) sent the unit back to the customer, unrepaired. I find it unlikely that they would not have confirmed the cost of repair with the customer.
People (including you) have to respect that companies are not obligated to repair items that aren't under warranty. Companies do so for two main reasons 1) to increase sales, and 2) out of respect for the environmental regulations. Personally, I am very happy that Nintendo and this customer negotiated a repair, rather than having the console end up in a landfill (assuming that the customer would otherwise not fix the device).
I do understand German, and this receipt says that the internal software (i.e. firmware) was modified.
Realistically, if someone voids their warranty by modifying the firmware on a device, regardless of whether it was done directly by the user or by the homebrew software, then they should expect to pay to have that repaired.
Why? because it takes time. Usually companies have very specific procedures for quickly re-flashing using their existing boot firmware. However, if that boot firmware is modified (i.e. the device is 'bricked') then that procedure needs to be changed. In this case, the engineer would need to 'hack' their own device to get it up to usable standards again (i.e. 'unbricked').
In the non-wii-world (i.e. reality), people that fix things also need to be paid adequately.
This customer should be happy that their bricked wii was even serviceable and that they weren't forced to cough up an extra 40€ to buy a brand-new Wii for 250€.
I put a highly customized version of Gentoo on my EEE. It was essentially the same thing that I had on my workstation, but tuned for binary packages and never writing anything to disk. It took a while to configure, and it's still not perfect, but it's exactly what I want. I tried Ubuntu as well, but I missed Portage :)
ASUS made a mistake with Xandros directly from the beginning, and that's probably what is showing with the number of returned netbooks. If they had gone with Ubuntu, and even just the regular desktop layout, they definitely would have had more appeal.
Hear, Hear!
The Intel Atom based netbooks are definitely good platforms for OS X. A friend of mine had it on his EEE 1000H and I was quite impressed.
If I wasn't addicted to changeability & source (e.g. the ability to change virtually anything), with Linux, then I would probably also have installed OS X. I still have usability issues with it though. Namely, Mac OS X has shortcut keys that are set in stone, much like Windows.
Gnome rocks for that, in my opinion.
Based on using the Xandros/KDE install on the original EEE, I can understand people's urge to return them. Most people aren't linux-adept or patient enough to learn how and install something more familiar like Ubuntu. Even then, Ubuntu is very different than, say Windows XP.
This is just another excuse for 'free desktop' people to clone the Windows(R) look-and-feel, and maybe even have a highly tuned and compatible WINE installation out of the box. Now there are actually legal grounds for it, because if MicroSoft absorbs the netbook market too, then there really is no Linux sanctuary on consumer devices. That spells monopoly in the NetBook OS market for MicroSoft. Also, in terms of mobile devices, it's also very hard to say whether or not Android will live up to its expectations.
Maybe the user-interface should be standardized somehow for the GUI on consumer products. At least that way, MS would be forced to code to a certain standard, which would be the same standard that everyone would be forced to code to... e.g. Apple, various Linux's, etc.
ISO, or ISO-like standardization has seemingly helped even the playing field in the past; UNIX to POSIX for example, or MS Word to ODF.
I agree - I've been using this software for close to a decade. It rocks. It's simple, effective, feature-rich, and easy to configure.
It's true - there needs to be a lot more community involvement with the FreeRunner in order for it to be successful. A good example, is the need for the SMedia software stack to start approaching completion, enabling OpenGL-ES-compatible hardware acceleration.
This device is honestly perfect for any undergraduate course in embedded systems, as the designer has full access to the hardware and it is 'unbrickable'. Today, I believe that computer engineers should really be taking a 2-semester course in low-level embedded systems programming. The course should encompass everything from JTAG debugging to writing a bootloader and launching Linux. IMHO, Google should be sponsoring university programs to provide students with this type of learning material, if they ever want Android to really take off.
Tough Economic Times'. Just like all other companies, they are feeling the economic crunch as well. The original article says that they had to lay-off a fraction of their staff already - something all-too familiar for most companies today.
At the end of the original article, SMP also states that they will, in the future produce another mobile phone. So the title of this article is clearly wrong.
Marvell's SheevaPlug runs at 1.2 GHz and only consumes 5 watts of power (2 watts actually consumed by the processor).
I apologize for quoting myself... but for the sake of correctness ...
The PXA168 (the SoC which powers the SheevaPlug) is, in fact, a system-on-a-chip. So the actual processor draws less power than the 2 watts dawn for the entire SoC.
Has a mainboard for x86 ever, in the entire history of Intel, ever drawn 5 Watts in total? Obviously not.
I would say that your response to this article was nonsense. Do you know anything about computer architecture?
What do you think Intel's new Larabee architecture is? Each core in the Larabee architecture uses in-order execution, like most ARM processors. I.e. Intel realizes that MHz-per-mW == success in today's mobile world.
Superscalar / out-of-order execution, present in all modern Intel chips, requires minimally O(n^2) control logic complexity. In most cases, the chip-area dedicated to SS control logic is proportional to the actual execution logic. By reverting to in-order execution (thereby freeing up much more silicon for multiple cores), Intel will see a huge performance increase with a fraction of the power consumption per-core.
And no, today's ARM chips are not running at 300 MHz. TI's OMAP 3530 (BeagleBoard, OpenPandora), which is already old, was designed to run at 600 MHz. Marvell's SheevaPlug runs at 1.2 GHz and only consumes 5 watts of power (2 watts actually consumed by the processor).
Going back to complexity, it's a simple mathematical relationship, given that transistor count is proportional to power dissipation. Using in-order architecture, 2 cores of execution, requires approximately 2 times the power. Using out-of-order execution 2 cores of execution requires 4-times the power (or likely more) for the same number of threads.
Regardless of the chip-vendor, in-order execution is more power efficient AND more space-efficient. Therefore, in the mobile world, where both power and space are 'costly', in-order architectures dominate (yes, that is true today).
ARM has had about a decade of time to perfect in-order architecture, while Intel has been focusing on out-of-order execution, which does give ARM the advantage in the mobile world.
Some ARM chips these days go far beyond what one would expect from a RISC micro-controller.
Two great examples are the BeagleBoard, which also uses the same hardware as the OpenPandora gaming platform. Both of the aformentioned devices are running a fully open-source Linux-powered stack, complete with a fully-functional Desktop environment. For developers, you can program for these devices using the standar GNU tools (c,c++), and of course several JavaVM's are already ported to the OMAP (JamVM,OpenJDK,Kaffee,etc)
.
The TI OMAP 3530 chip that powers the BeagleBoard and OpenPandora has a 3D graphics acceleration unit, a 2D (video) graphics acceleration unit, and a built in DSP for audio and general-purpose number crunching.
Without question a large majority of users do little more with their mobile computers than browse the internet, write email, and occasionally stream video - all things that are fully achievable with today's ARM chips.
ARM chip manufacturers are also starting to put multi-core chips on the market, and while the transistor complexity is nothing like Intel's out-of-order logic, multi-core, still means serious multi-threading. With clock frequencies approaching and exceeding 1GHz, I would not be surprised to see ARM take over the mobile and netbook markets, especially considering how little power they actually consume.
Furthermore, even high-end x86 laptops could benefit from having an ARM co-processor for instant-on and mobile operation, while the x86 processor could be used for more compute-intensive applications such as CAD or video processing. Imagine battery power extended by a factor of ten for everyone!
YouTube Videos:
BeagleBoard
OpenPandora
AWESOME!!
I would argue that most of the things MS claims to invent were actually invented by promising technology firms who sold their rights to MS. Many people would probably argue against Visual Studio being the best IDE, in favor of Eclipse, and MS was not the first company to have a large touch interface by a long shot.
Ubuntu is great for beginners in the Linux world, but also for people who can't afford to risk compiling code from scratch (i.e. running into bugs, losing productivity time). I compile all of my applications from scratch, and I fix / file bugs often. My reward is that I'm participating in the greater picture, and quite often others tell me that I've helped them out with similar problems.
I also use a desktop configuration that appears like a cross between Windows XP and Mac OS X, but it's running Gnome. That's all being run on an EEE PC that I've significantly customized both hardware and software-wise.
Today, what I call innovative is mostly what's happening on mobile devices. I'm not an Apple 'snob' by any means... and no longer even own an apple product because my 'retro' flash-modded 4th gen ipod was lost on a plane... but I would really have to say, that Apple has been leading innovation in the mobile world for a while. Android has begun to impact the mobile world, and I feel nice that I was one of the first people to port it to a new device. If Apple's mobile platform was open source, I would probably expand on it, but as far as I know, Android is as close as I can get.
I would gladly opt-in for such a plan. Although, I would prefer that it also carried the right to share media created by artists in other countries. Of course, I would only share media with others who have also opted in for such a plan. There is far too much complexity associated with governments and other organizations spying on their citizens, repeatedly coming up with ineffective DRM techniques, etc. For a small monthly fee, I have no problems whatsoever paying for the right to have un-complicated rights to enjoy my favourite media and share it with friends.