OSNews

A poor plan and, if it's an hidden exclusive deal, maybe also an illegal one.

Its more likely it doesn't have a thing to do with MS and has everything to do with IP. Its pretty well known that Intel has been using PowerVR IGPs in many of its mobile processors and PowerVR really couldn't care less about Linux support.

So what is Intel supposed to do? The IGPs they have in the core series will blow the power budget, their other in house IGP suck, so that pretty much leaves the PowerVR team.

If Linux would have gotten more than 1% of the market this wouldn't have happened, but getting the devs on the same page is like herding cats so give it up, Intel had to make a call and take the best they could get that would support the most consumers. After all its not like they could call up Nvidia after they shafted them out of the IGP business could they?

Really, Android and Linux Servers only have a 1% market share. TIL

I would not of disagreed with you if you had said Linux on the desktop but you blanket statement that Linux as a whole has 1% share is incorrect. Feel free to continue your rant below.

Do we REALLY have to spell such things out when we are talking about CONSUMER chips now? Really? Since i got modded down by the groupthinkers i guess so.

But you show me ONE, just one mind you, company that offers Linux servers running Clovertrail and I'll be happy to retract, but since we are all supposed to be logical thinking adults here and the entire article is talking about a tablet chip I honestly didn't believe i had to put out that a consumer chip is not gonna target an OS that has 1.05% of the consumer market.

But YOU know and I know that it was really just moving the goalposts, C'mon, be honest. We see this in every single article that has anything to do with Linux desktops, we'll get "But Linux has server share" which has exactly diddly and squat to do with the topic at hand. This chip isn't aimed at servers, routers, or HPC, this is aimed at consumer devices and on X86 Linux has done VERY poorly, this isn't exactly earth shattering news here.

This chip is aimed at low power consumer devices, like phones and tablets... Linux is actually doing very well in this space in the form of Android.

This is a profoundly false statement. Android is available under a very liberal license with full source and all tools needed to build it. It's development model may be closed, but that doesn't mean that the finished product is proprietary. You could take Android and fork it today into a new OS. That's the very essence of being open-source.

The statement "owned by Google" is untrue as well. Nobody "owns" a piece of software, software isn't tangible goods. Rather, one owns the copyright to a piece of software. And even if take it that you meant that Google owns the copyright to Android, that's potentially untrue as well. Depending on how you define Android, the copyright is owned by a large group of people, starting from the Linux kernel, drivers, through lots of the userspace libraries, none of which were developed solely by Google (and thus they don't "own the copyright" to them, but only potentially to portions). The upper Android software stacks are almost exclusively Google-copyrighted, but as I pointed out above, these are available under a very liberal open-source license.

If only it was that easy to compile desktop figures, but sadly it's not. Windows figures are artificially inflated because it's nearly impossible to buy a laptop or pre-build PC without having Windows and nobody has any idea the extent of Linux installations out there.

The most we can do is form estimates based on user agent strings, Linux downloads and Windows sales, but even all that is error prone:
* user agent strings can be and often are faked
* user agent strings are only representative of the visitor base on that particular site - thus is biased towards the type of clientèle that site attracts (ie OSNews.com will have more Linux users than MSN.com)
* user agent strings don't take into account whether it's a personal computer (which can run anything the user wants) or a work computer (which often tends to be Windows PCs due to MS Office and Exchange compatibility)
* Not all Linux downloads are via distro.com or their mirrors (bit torrent et al are often encouraged)
* One Linux ISO could be used to set up an infinite number of PCs
* and I'd already touched on the Windows sales issues earlier.

So in short, arguing any kind of market figures as "fact" is just proving how little you really know about this subject.

Furthermore, I'd be more inclined to go with Thom's verdict: MS have already blocked Linux on many future ARM devices (secure boot UEFI), so this is a back-handed way of pushing Linux out of the portable x86 space as well as Redmond know full well that a public maneuver -such as the mandatory secure boot UEFI- would see Microsoft in trouble with the EU quicker than a whore drops her knickers.

Really? Better tell Dell, System76, and a dozen other Linux OEMs they don't exist then.

And if you are talking retail? As an actual retailer who USED to sell Linux desktops I'll be happy to tell you why that is, because Linux devs destroy any chance by constantly fiddling with the internals. You see with web sales all sales are final, NO after sale support. this is the exact opposite of retail where if you don't offer at least basic support your sales dry up and blow away, and I can tell you that Linux drivers breaking every update made Linux cost me MORE than Windows by a long shot. At $35 an hour frankly it only takes ONE forum hunt because pulse puked on sound or the WiFi driver was totaled for it to make the $80 cost of Win 7 Home the better deal. This is only gonna get worse with Win 8 as its just $40 for Pro.

But hey, don't take my word for it, please look at this list of over 100 major show stopping bugs with links. If you'd like I'll be happy to post the original page from 3 years ago to let you compare and see how many of them still haven't been fixed after 3 years. Linux works on servers because admins get paid a high 5 figures to do nothing but fix the stuff the devs break, this is the complete opposite of the home market. Even one of the RH devs says the Linux desktop is suckage and the entire system is unsustainable, link provided below.

http://linuxfonts.narod.ru/why.linux.is.not.ready.for.the.desktop.c...

https://plus.google.com/109922199462633401279/posts/HgdeFDfRzNe

I didn't say they don't exist. I said they were rare.


Of course we're talking retail. Did you actually read my post?


I'm not really sure why you're arguing this with me because I was never making a claim that Linux was ready for the desktop. Just about market share figures.

In fact did you actually read my post? Because you're doing this tired old habit that many on here exhibit, where they feel compelled to educate your opponent without even thinking about the post your replying to nor the points it raised.

If you read my post again, you'll see that I deliberately avoided any personal commentary on the state of desktop Linux. It's the very type of argument I actively try and avoid these days because everyone feels so passionately that the other person is wrong and needs to be educated otherwise. Thus, quite frankly, a futile debate that almost always ends with stupid comparisons and trolling.

You hate FOSS / Linux, I get that. But please don't take your personal prejudices out on me.

Let's be clear, this is pure unsupported conjecture. (N.B. not agreeing with you doesn't mean I agree with Thom.)



How about at least putting out a binary blob driver, like lots of vendors do? (nVidia, AMD, Broadcom, etc.) This argument has been shown to be wrong before.



You do realize that the non-Apple tablet market (for which this chip is almost certainly targeted) is about 100% Android (and thus Linux)? Windows 8 has 0% of the market. Thus, from a pure business perspective (even accepting your above arguments about IGPs), it makes next to no sense at all to target a non-existent market with a new product, unless there are other motives behind it as well (such as an exclusivity deal with Microsoft). Just to drive this point home a bit better, let's quote the linked article on the The Inquirer:



Is this absolute 100% proof that Intel and Microsoft have an exclusivity deal? No. But it makes it extremely likely, to the point of being beyond a reasonable doubt.

Edited 2012-09-15 10:22 UTC

An x86 port of Android would be a red headed stepchild...

Best case it would be no worse than the more common ARM devices, it would run the same platform neutral or open source apps, with equivalent performance and battery life.

Worst case...

It won't run a large number of existing Android apps which are closed source and only compiled for ARM.
It will use more power than competing ARM designs, and therefore have inferior battery life.


An x86 windows tablet on the other hand, does offer benefits over an arm based windows tablet (availability of existing programs, albeit program unsuitable for use on a tablet ui)... The windows tablets will be different enough to the android tablets to mask any efficiency differences, and the x86 tablets will be sufficiently better than the crippled arm version of windows that many people will be happy to sacrifice some battery life for a massively superior device.

Whilst I agree with the main point of your post, I do question this assumption.

Android apps are written in "Dalvik", which is modelled after Java (it is not derived from Java, since Dalvik has no Java code). As I understood it, a Java app, and hence a Dalvik app, contains bytecode and not compiled binary.

http://en.wikipedia.org/wiki/Java_bytecode
http://en.wikipedia.org/wiki/Dalvik_%28software%29

"Dalvik is the process virtual machine (VM) in Google's Android operating system. It is the software that runs the apps on Android devices. Dalvik is thus an integral part of Android, which is typically used on mobile devices such as mobile phones and tablet computers as well as more recently on embedded devices such as smart TVs and media streamers. Programs are commonly written in Java and compiled to bytecode. They are then converted from Java Virtual Machine-compatible .class files to Dalvik-compatible .dex (Dalvik Executable) files before installation on a device."

Hence I would assume that Dalvik-compatible .dex (Dalvik Executable) files are architecture independent.

http://source.android.com/tech/dalvik/dalvik-bytecode.html

An x86 Android device could therefore execute the exact same Dalvik executable .dex files as an ARM device, without any re-compilation required.

Edited 2012-09-19 02:30 UTC

Love your analogy of herding cats to try bringing linux developers on the same page. Thing is, with cats, you don't herd, you call them to you. I'm sure this approach could also work with developers.
Just a tip from one who breeds Burmese Cats

I can't speak for all developers, but I work for more than just a saucer or milk - my rates are usually catnip and heavy petting.

Yes, it's an MS deal. It's all about money. Some of the responsible are also Microsoft customers who take part in this ripoff when they pay their MS tax.

Whoa, let's be careful with proprietary software bashing or we could cause society grat trouble!
Now "the world would be much worse" is an assumption you must be very insightful - almost omniscient - to make, as it's quite a demanding task to guess how a world without proprietary software would be. Anyway that hypoetical world has absolutely nothing to do with the topic here. It's so far-fetched you must have felt at least on the edge of bad faith writing your comment.
Nobody would freakin ever complain about proprietary software if there wasn't the suspicion - as in this case - that someone's trying to compete outside the boundaries of what's legitimate this way damaging other entities.
What the heck has this to do with hating proprietary software? All players should just play fair, but maybe that's outside the scope for some of them. The details aren't known yet, but I would get very annoyed by a hypotetical company making it more difficult for an open source OS to support a new processor. It sounds like a steer from the usual Intel policy and as such it does not seem to be happening just by chance.

If you think this shows how bad free software supporters are (the vast majority of which I'm sure don't want proprietary software to disappear but just like having more choice and the possibility to know what the code they're using really does) feel free to voice your personal reality, but be aware it sounds quite crazy.

The internet.

Yes. Do you want more examples?
Many of todays programming languages: Python, PHP, Haskell, Erlang, Ruby, D, Caml etc etc.
Stuff that powers the internet: nginx, apache, lighttpd, varnish, RoR, tomcat, jetty etc etc.
The list goes on.

But hey, lets pretend open and free has never advanced or contributed anything, if that makes you feel better.

Python development got funded by the companies Guido worked for.

PHP development is funded by Zend.

Haskell development is funded by Microsoft Research.

Erlang was developed at Ericson, before it got released as open source.

The list goes on.

Very few successful open source projects, if any, would have succeeded if it wasn't for the dirty money that some companies selling closed source software invested on it.

I used to be a big open source fan until I realized, that without a way to pay for bills, being a fan does not help much.

Your reaction makes it very clear you have no clue what you are talking about regarding open systems.

If it wouldn't be for open systems, you would not have internet the way it is today, and it probably would not even exist. You would have a lot of company-specific networks instead, like a Google network, and a Microsoft network, and so on. If it wouldn't be for open systems, email would not exist, ftp would not exist, ssh would not exist, http would not exist.

A lot of technologies in IT would not exist the way they do today. Even OSX wouldn't exist the way it is today, since they wouldn't be able to build it on top of BSD.

That does not mean that there is no proprietary technology that made big advancements or anything like that. But saying that open systems is irrelevant, just shows you are clueless.

Btw, the big difference between open systems and closed systems is that open systems allow others to build on top of them and takes things even further. Closed systems only allow the creator of the closed system to build further on it.

Edited 2012-09-15 08:29 UTC

Open Systems != Open source.

You can have open systems, where interfaces and protocols are defined, without open source.

Actually Email and other communication mechanisms fall under that situation. They are protocols defined by RFC documents, and they exist way before open source.

Wrong.


After all that rambling, some reasonable, some ridiculous, you end it by claiming I said open systems are irrelevant? Just one problem, ...I said no such thing. I suggested the overall contributions to advancement resulting from the two be compared, not that one is relevant while the other isn't. Apparently you don't know the difference which explains, I guess, why you're making ridiculous claims.


Wrong.

Without the open systems, Google wouldn't even exists. They had no way to be populair or would even be able to index all the now public information that they did.

Technological advances driven by the market are all well and good, but they're not very useful if people who would benefit can't afford them.

Care to come out with an example?
Besides if the source is available linux support in slightly longer term means basically universal support.
Exclusivity is exclusive to closed systems world.

moondevil already did -- refer to his previous post.

Why did you ignore my other questions? I'll repeat one of them again and put it in bold so in case you somehow managed to "over look" it the first time: If Intel developed a chip line to work specifically with linux would you still be so upset?

A misunderstanding: my answer referred exactly to that question.

Sweeping generalization, it's the new fact. This is bullshit. There's good OSS software and bad OSS software, just like how there's good closed-source software and bad closed-source software.



Because Linux is the only OSS software in the world and even if it isn't we can extrapolate from it to everything else. There are many OSS projects with rigorous testing, both manual and automated, and there are many closed-source companies that does neither.



Wow, really. This is just as much nonsense as "all OSs software is crap". It's simply not true. Sometimes bugs get fixed in a timely manner, sometimes it takes forever and sometimes it never happens. This is regardless of if the software is closed or open.



Yes, naturally. Closed-source is bug free and when it isn't all bugs are fixed in a timely manner. Hmm..yeah, maybe in some alternate reality.
I've never had this problem, EVER, on my OSS desktops; be it Linux or BSD.



Wait, are you talking about OSS or closed-source software?

It seems to me that you're missing the core ingredient of lock-in here.

A large part of today's dominant proprietary software is so buggy and painful to use that it surely never goes through serious QA anymore. Why do people keep buying then ? Because they need something to support their proprietary formats, their proprietary protocols, their proprietary hardware...

Software actors are not fairly competing in a free market with flying unicorns, glittery rainbows, and happiness all around. Everyone has its dirty weapons that the others cannot compete with, and will not hesitate to use it. Including the open-source actors.

How do you propose to make money out of free software?

In many European countries a software developer gets to earn at least 1000€ per month.

What is your proposal to have a company developing free software, server side, desktop, embedded, whatever platform, while being able to pay the developers and the usual monthly costs?

For the typical answer of offering the software and asking money for support and trainings, what do you do to keep your company alive, when I take your free software and charge less for support?

All you have to do is look at Red Hat. They have been going through 25% growth quarter after quarter pretty much since they went IPO. They don't seem to have a problem with the Open Source way.

Problem is, how many Red-Hats can there be world wide?

Why do you think companies are moving to CentOS?

This question has been answered many times, and a number of business models exist to subsist on free software.

1) You can offer support by the "gurus" of the software
2) You can dual-license your software, asking money for the commercial license
3) You can sell non-free add-ons to your software
4) You can enter contracts for implementing customer desired features
5) You can lure users of your product into consuming your content/visiting your websites/etc.
6) You can create entirely new markets or overtake existing ones

Red Hat employs a number of kernel developers for the first reason. Also they do #4.
Digium (Asterisk) does at least #1, #2 and #3.
IBM, when they invested $1 bln in Linux certainly had #6 in mind.
Trolltech, before they were bought by Nokia, did #1, #2, #3 and #4, and after being bought stopped doing #3.
Mozilla Corporation, the for-profit subsidy of the Mozilla foundation, does #4 and #5

All those examples only work for enterprise software, where people are willing to pay for support.

When you don't pay for support, like desktop software, it is not possible to keep a company afloat, just by doing PayPal donations.

The more important question is, how is that relevant to goddamned anything, much less the previous two posts?

FOSS software projects don't exist to make money; they exist to create software that their contributors want to use. That's the motivation; if I make a program that I find useful, maybe if I make the source available, other people will find it useful too, and maybe improve it to be useful to them. And if I GPL my software, maybe I'll get them to send me the changes they make.

The "broken toilet" problem half-exists: the problem, to the extent that it's a problem, isn't that people won't do nasty and thankless work, it's that developers tend to focus their efforts on the problems that impact them personally as they use the software.

And for-pay software has the same problem. Many commercial software developers spend at least as much effort making you have to use their software as they do making you want to use their software. Many are the small businesses that deeply despise their accounting system, but they can't migrate because the shitty software won't let them export their data.

Free software in a GNU sense ? I don't know. But for open-source software, I'd suggest granting every licensee access to the software source code for private use, and keeping as many redistribution rights on derivatives as you need for yourself.

As far as I'm concerned, source code should always be available alongside software, but free redistribution of said source and its derivatives is a separate matter.

In fact, I've recently worked on a software license draft that attempts to separate both concerns better than current OSS licenses. I will soon submit it to the OSI in order to get lawyer review, and to discuss other potential issues with their licensing gurus. If all is good, and I get OSI approval, I'll then submit something about it on OSnews.

There is more than one way to make money. If one sells a product or service, there are two main ways of making more money: (1) increase the price of the product or service whilst trying to maintain the volume of sales, or (2) reduce the costs of production of the product or service.

Freedom software is quality software made by its users. It is made by collaboration in a loose organisation which is effectively a consumer's co-operative.

http://en.wikipedia.org/wiki/Consumer_cooperative

"Consumer cooperatives are enterprises owned by consumers and managed democratically which aim at fulfilling the needs and aspirations of their members. They operate within the market system, independently of the state, as a form of mutual aid, oriented toward service rather than pecuniary profit."

"The major difference between consumers' cooperatives and other forms of business is that the purpose of a consumers' cooperative association is to provide quality goods and services at the lowest cost to the consumer/owners rather than to sell goods and services at the highest price above cost that the consumer is willing to pay."

So, in effect, freedom software (aka FOSS) makes money via method 2 above. It makes money by reducing production costs (in this case, the cost of software). This of course only works if the product being sold is itself not software. It only works to make more money for businesses where software is a cost of production, and not the product for sale.

Example:
http://www.linuxinsider.com/story/72867.html

'They Will Save a Fortune'

"Linux makes a fantastic embedded platform," consultant and Slashdot blogger Gerhard Mack opined. "I bet even with the gold membership they will still save a fortune compared to going with closed source competitors."

Indeed, "it's completely expected," agreed Chris Travers, a Slashdot blogger who works on the LedgerSMB project. "Embedded space is going more and more towards Linux."

You do seem very confused about the economics of open source. Hope this helps.

Edited 2012-09-17 10:30 UTC

This is not the way it works, at all. To make money out of open source, one uses open source software to produce something, or provide a service, more cheaply than would be the case if one used closed source instead. One makes money by using the software to make or provide something else, other than software, at reduced costs.

Here is yet another example:
http://www.itnews.com.au/News/315490,study-urges-cios-to-choose-ope...

Study urges CIOs to choose open source first

"Norton cited studies from the London School of Economics which found that investments to deploy open source in-house drives longer-term savings of 20 percent over the alternatives."

" Specific to the travel industry, the report’s financier Amadeus has gradually been shifting its sizeable airline transaction processing business (bookings, reservations, flight management) from proprietary platforms to Linux (SuSE, predominantly).

It also uses the Apache web server and Tomcat application server, among other common tools.

The company’s 4500 developers use the open source Eclipse development tool, and its sales and marketing team use SugarCRM.

Amadeus has also become a contributor to the open source effort. Late last month it donated its user interface framework, ARIA, to the open source community under an Apache license.

It is hoped that travel industry customers might save money at the front-end using these free UI templates for common travel industry functions, and invariably then see the value in outsourcing the remainder (back-end processing) to Amadeus.

“We commissioned this study to highlight to our customers and shareholders our use of open systems and contribution to open systems,”"

The White House, for example, uses Linux and Drupal and it's there working in spite of usual attacks.
http://uptime.netcraft.com/up/graph?site=whitehouse.gov
http://www.theregister.co.uk/2001/07/24/white_house_web_site_moves/
http://www.whitehouse.gov/developers


People can see by themselves how it doesn't happen with good software. Don't take my word for it, if people want to try it by themselves, there's a "Virtual machine of Kubuntu 11.04 i386 Desktop, stable version" on
https://docs.google.com/folder/d/0B2UJmdRmDlL1VmRWZ3dfaGdJTUU/edit
to conduct experiments on it. Just don't update it to an upper version of Kubuntu and, before updating a GUI program, close that program.

I usually employ a copy of it as a virtual machine and it works stably, too.

About it, there is a README.txt on the same web address.

If someone has any doubt, on this thread you can make questions.

The same argumenthave been repeated times an again in the last 20 years, every time OSS software is meant to be drployed in direct competition agains software of some corp.
In fact im niw working in a system that has both closed and open components (both mainyained by the same company) but thanks to Java closed libs are easy to decompile. Guess which one has better quality and which better handles corner cases ans has less bugs.

You are confused.

People being payed to write software does not mean closed software.

What you are talking about is business paying for developing software, that has nothing to do with closed software.

Of couse there are old bugs in Ubuntu, this is the first bug and it still hasn't been fixed:

https://bugs.launchpad.net/bugs/1

Because the same is not true for free software....

Please, name one thing that would disappear if all software was legally required to respect the FSF's four freedoms. Malware? Nagware like mIRC? Ten thousand shitty 99 cent fart buttons? Software patents?

The nineties never left. While the plethora of Apache / BSD style licenses have made life much better for developers, at the level of consumers you really need the GPL to improve the situation, and you need the demand to really come from the consumer. For various reasons, the GPL is not so popular, while BSD style licenses have gained and maintain traction -- because developers demand that sort of thing but users don't.

This basically means that it's business as usual for companies, and despite the fact the Microsoft still run an all-proprietary stack, they have to work in this ecosystem. In the past this would've meant bewildered lashing out, but it is now far more calculated. Google have a far more open source stack, but they still use tricks to get hardware makers to dance to their tune (e.g. the Alibaba Android fork).

In any case, BSD-style software went from being something that blindsided companies to one that is now actively part of their "strategy", whatever that may entail (even if it means precluding use). Linux might be GPLed, but it sits in an ecosystem of BSD-style licenses. The situation will not improve until "customers" demand freedom.

Not always. See

http://en.wikipedia.org/wiki/Tying_(commerce)

Yes, lets look at this from a sensible standpoint. Intel has a monopoly on chips in the x86 market. Microsoft has a monopoly on OS's in the x86 market. All anti trust problems I am aware of have specifically targeted the x86 marketplace. Now the main CPU maker is making a CPU that only works with the main OS. How could anyone possibly have any concerns around that fact?

So what.. This is one product out of many and it doesn't render any of the other products incapable or incompatible. If Intel and Microsoft want to partner together to build a system, fine by me as long as other options exist, which they do. If you want people to panic, give them good reason accompanied with actual evidence in support of it.

How many times has peoples initial reaction been to crap their pants and cry foul, only to look foolish when the sky never fell and armageddon never ensued. Too many to count.. And how many times has it all come crashing down? ....never.

You're right in a sense. I remember the big fiasco over the processor serial number in Pentium IIIs, that ended up being fixed with a simple BIOS patch for the ones shipped before Intel backpedaled. I can see something similar happening here, with the specs eventually released or perhaps reverse engineered and rolled into a future kernel update.

Also, if folks are going to call this out as an egregious offense, they better be ready to call out Sony because their proprietary media buttons on my laptop only work under Windows and in fact cause weird bugs under GNU/Linux and Haiku. When is someone going to stand up for my right to have fully working buttons, with FOSS drivers built by Sony programmers to my own personal satisfaction? (For the obtuse, this is sarcasm on my part. I don't care about those buttons.)

But all that said, I do feel that this is some shady stuff being that it's coming from Microsoft and Intel, the two heavy players in the desktop computing world. I think Intel is trying to secure a solid footing in the tablet market with this move, but they are going about it wrong and are going to piss off FOSS fans who have remained loyal to Intel for years. It may not be a particularly large group of people, but they have a loud voice and the power to sway others towards AMD on the desktop and even mobile devices when it comes to x86.

So my take? Intel sucks for this, but in the grand scheme of things it will only serve to tarnish their reputation and nothing more. The world will go on, zombies will remain in the grave and people who give a shit will avoid this particular run of processors, giving their money to Intel's competitors.

/looks at CPU-Z, Authentic AMD/...LOL Wut?

In this case you have an easy and clear choice, AMD has opened up their specs, Intel has chosen to not support Linux on this chip. Unless you are building a supercomputer (which you certainly wouldn't be using Atom chips for) then all you have to do is walk across the street, AMD has the Bobcat I and soon the Bobcat II quads, tada!

Intel has the right to choose who they want to support, remember that is what we are talking here, no software locks, simply not supporting an OS, and YOU have the right to shop somewhere else.

It's called the free market and voting with your wallet. try it, you might like it. I know I'm quite happy with my AMD E350 netbook and my Thuban X6 desktop.

AMD doesn't have products in this space (arm area of power effiviency) and they clearly stated they don't intend to compete there. This leaves Intel as the sole x86 solution. This is really strange bc they have received great support from Google.

As it turns out, AMD does have a tablet product in its roadmap. You know what, they made it Windows 8 only too

OSX runs on x86, so how does Microsoft have a monopoly in the "x86" OS market? Note that in the DOJ/MS case, Macs still ran on PPCs, so the judge was able to rule out Macs from the "x86" market, but today Macs run on x86. The "monopoly" that the judge declared back then isn't in effect today. Also note that the MS/DOJ settlment expired, so MS is no longer under any anti-trust monopoly regulations, not in the US anyway.

If someone is going to bring some "anti-trust" law suit, they are going to have to prove, all over again, that Windows is a monoply OS TODAY, not yesterday, but TODAY. That means that they're going to have to deal with the fact that OSX runs on x86, and LOTS of people use it (on TV, all you see are Mac Books anymore). The idea that Microsoft has monopoly control on x86 OSes today is laughable. "

I tried to mod you up, but OSNews claims I have already modded you o_o Someone seems to have modded you down just because they don't like you or your message, behaviour that I do not condone of.

Anyways, you are indeed correct: the verdict about Microsoft's monopoly status has expired and therefore going through antitrust lawsuits would require enormours amounts of effort. Microsoft would probably still fulfill the requirements for it to be re-declared a monopoly even with Apple around -- after all, Apple only sells their OS for their OWN hardware, not for general consumption, and their market share is still barely a blip in the radar -- and would possibly face some sanctions, but it doesn't look like there is enough incentive for anyone to sue them on antitrust-grounds. Also, even in theory if Microsoft were re-declared a monopoly and faced some sanctions I doubt they would be sanctioned for what Intel did here: the CPU can and does run Linux just fine, Intel is only saying they do not provide official support for it nor do they plan to add support for the advanced features to the Linux-kernel, and as it is not against the law not to support every possible OS out there... well, you get the drift.

Dont worry. Being modded down makes you stand out of the crowd. Infact it prompts me to actually read the posts in question.

MollyC
Microsoft has over 80% of the desktop market. Microsoft recently forced OEMs to add secure boot to their motherboards to support what is currently only a Windows technology by threatening them with the loss of their advertising budgets. With 80% market share, Microsoft is still able to sway the industry to its benefit. Depending on market conditions, these current actions could be considered anti competitive. If this new Atom processor will only run Windows, it is effective locking Linux out of this portion of the market. Intel is also constantly under scrutiny by the DOJ due to its market share. So maybe it will be Intel that gets investigated, instead of Microsoft.

I wish I hadn't already commented or I would have modded you up as well. As I posted above, this really isn't a big deal. Those of us who actually care enough to get offended should just buy AMD or some ARM based tablet and get their Android/Linux jones that way. At some point Intel will get the message and will play ball again, or else they will continue to appease Microsoft and turn the FOSS world towards AMD. I don't see a world-ending catastrophe either way.

If they do intend to publish the CPU specs (as any self respecting x86 company has done for years) then no problem.
But the article tells about running at all and win8 exclusivity, bringing me to the conclusion no specs are going to hit the streets.
And in the space the clover tail work power management is about as important as running at all.

Is that because Windows 8 is more of a "resource hog" and needs the power-saving features because it's more demanding on the tablet's battery?

Or is it because x86 is just a battery guzzler, and no matter what it's stupid to be using it in situations where power resources are highly limited?

But, well, lets morph this thing into something it was never designed to be, because oh no, we can't let better architectures like ARM and MIPS do their thing and compete where they technically mop the floor with us. No, no, no--we need to start invading their territory, even if in the end the customers lose with shittier battery life due to the x86/Windows combination.

Edited 2012-09-16 01:21 UTC

Well, I for one wouldn't mind seeing some serious competition to ARM in the embedded space. Among other things, that could perhaps make them consider making their architecture more developer-friendly, so that there is no need to work on one separate OS build per supported ARM device.

Competition is good, but only when it makes sense. Don't send a freakin' 18-wheeler on the track to compete against a Camaro SS. There are several architectures that would make worthy competitors to ARM in portable systems. x86 is just not one of them.

Hopefully you're right. But even if you are, why the hell would anyone in their right mind want to use an x86 processor in a cell phone or tablet computer? Or *any* battery-operated portable device, for that matter. It's just downright retarded.

The PowerVR is also used in the ARM Pandaboard, and is fully accelerated on Linux. Both hardware mpeg4 encoding and decoding are available via GStreamer plugins and it has full X support with hardware acceleration. The drivers are not open-source, but that's a totally different issue.

So a Windows 8 constrained tablet is bad, but a constrained Android tablet is good, I see.

When they get that for-profit companies conduct business in a for-profit manner and aren't obligated to give charity and free hand-outs to people who beg.

"However unlike Intel, AMD said there is nothing stopping people from running Linux on its Hondo processor."

Regardless of what Intel says, no one will point a gun to anyone trying to run Linux on their processors, so I fail to see how this is any different.

I too fail to see how it is different... yet.

If Intel builds in a hardware check that cannot be overridden by software - I would say that then it would be quite different. Good or bad may be debatable, but would you agree that at that point it would be different?

AMD is not doing the same, there are working open source drivers for the Brazos chipsets, with 2D, 3D and VDPAU video decode acceleration through Gallium3D (currently MPEG2 only). AMD won't market Hondo for Linux, that is all.

Intel on the other hand refuses to provide any Linux source code or programming documentation regarding acceleration on their PowerVR based Atoms. Now with Clover Trail it seems they won't even produce proprietary drivers. Not that their psb and emgd drivers generated much enthusiasm anyway.

Edited 2012-09-15 20:12 UTC

You know what? I'm going to design and build a processor that won't run Linux, BSD, Haiku or OSX because f*ck all of you, I just want it to run my precioussss Windows 8. I won't provide any Linux or other source, documentation or even spoilers for next week's episode of The Vampire Diaries.

Guess what will happen? NO ONE will give a shit. Who does this guy think he is anyway, right?

Okay, now replace me with Intel. You end up with the same result: It's really not that big a deal, because in the tablet space Intel already has zero presence. So what if one chip maker is being retarded, there are half a dozen other chip makers between AMD and VIA on the x86 side, and the various others on the ARM side.

To put it another way, why aren't we up in arms because Broadcom is refusing to release specs and open source drivers for the chip on the Raspberry Pi? Granted, that's a much smaller market, but come on! We're geeks, we should be all over that shit. Broadcom should have issued an apology and posted full source code along with directions to the pot of gold at the end of the rainbow by now, due to our incessant cries of foul play!

Really folks, this is only an issue because some of us have a need to be butt-hurt on a daily basis so we can blame it on "the man". Yeah, Intel are a bunch of dicks for this, but in the end it will only hurt them.

I think that Linux adapted on more various platforms, CPU, computer architectures than Windows ever done. So this is BS. <- PERIOD

Kochise

I doubt Intel is saying that Linux can't be ported to Clover Trail, I think they are saying that they wont do the work or write the documentation necessary. With the motivation that it is not the way they intend the PC to head in the longer run.

How Microsoft is supposed to code RTM without spec and datasheet ? If Microsoft can put their hands on them, understand them and code something from them, Linux coder would undoubtly also.

Intel is going nVidia-like and while they headed in the Linux area for some times with true openess, they are turning 180°. It's only one CPU, but one of how many ?

Kochise

And do you think Windows or OSX doesn't have to? That the code to support the features manifests itself from the thin air?



There has always been advertising, there always will be. If Google suddenly just dropped dead some other entity would replace it and you'd still be seeing ads. As such Google's domination is totally irrelevant on this point, you are just deluding yourself if you believe otherwise.



If users questioning you is bad then the problem is on your end. A good developer knows how to answer that question.

May I add this : Microsoft bought Skype. Now MSN and Skype are ads riden. And I don't think Google have anything to do about that.

Just awaiting Apple's move to cripple iTunes with ads. After all they prevented any other ads operator on iPhone to have the monopoly of it.

Kochise

Dude you just contradicted one post later:

And do you think Windows or OSX doesn't have to? That the code to support the features manifests itself from the thin air? "



Since forever. When a manufacturer wants to ensure their chips will be adopted quickly they often write the code needed and to also ensure there are as few bugs in the supporting code as possible. "



If users questioning you is bad then the problem is on your end. A good developer knows how to answer that question. "

Yes I deviated a bit from the main subject but that is no reason to do personal attacks. Yes I have managed to convince consumers, yet the fact that they are asking for price-free or price-free-supported software shows my point that Google is making software looks like a cereal box price.

Osnews I think there is troll in the forum.

No, I didn't: the code still has to be written by someone, no matter the OS. You were portraying the situation as if Windows and OSX do not need such, only Linux, something that is obviously not true.



Saying the problem is on your end is a personal attack?

No, I didn't: the code still has to be written by someone, no matter the OS. You were portraying the situation as if Windows and OSX do not need such, only Linux, something that is obviously not true.

"

Yes you did when you putted Windows and Mac OS X relationship with hardware manufacturers to the same level as Linux (which is asymmetric) regarding hardware use and then you said that hardware manufacturers write the code when they want their chips are adopted.

Of course you are pointing the obvious (software is written by people) when a developer talks about support is talking about two things one the hardware features are openly or closely documented by the manufacturer and if the manufacturer itself provides some kind of software implementation (driver). Yes every OS development must accommodate to the underlying hardware but when I say adapt I'm talking about an asymmetric relationship with the manufacturer almost as an species adaptation to nature conditions.

GNU/Linux has always had an asymmetric relationship with Intel at least since Intel and Microsoft has been virtually always strategic partners and now with Apple too. To Intel Linux usage is not that important since it doesn't move that much hardware as Windows. Same thing goes to Nvidia and ATI. For Microsoft of course Linux is not a joke at all, it is a potential enemy( as Apple but they have some kind of arrangement).

Intel and Microsoft interest in some kind of hardware "protection" or vendor lock dates from like 10 years ago or more with the TCP A.K.A Palladium architecture and then with Intel's tryout with UEFI. The intention has been there for quite some time but they haven't been able to do so.

My point is that GNUL has always have a harsh relationship with some hardware vendors while Microsoft and Apple have a symmetric relationship with them, they get hardware ahead of everybody else, they even demand features and get specially optimized compilers while GNUL (developers) has to HACK the hardware and do reverse engineer to find out things that big vendors like Apple and Microsoft already know.

And for those who think that a HACK is bad thing, that is a wrong way of seeing things.

Yes Microsoft Windows was for quite some time a piece of crap but in order to keep that mess working the developers must be real bad asses, but that doesn't make Windows a HACK because all of their features were developed with formal methods and intervention of hardware vendors.

And as a final note, implying that only a good developer can give a satisfying answer to a consumer regarding why software cost money is a bold statement because people some people refuse to pay for software and that doesn't make developers bad.

Edited 2012-09-16 03:01 UTC

In actual fact, there is less and less hardware these days which Linux kernel programmers have to reverse engineer.

Some companies, such as Intel, write open source drivers for Linux for their hardware.

http://www.intel.com/cd/corporate/icsc/apac/eng/teams/331393.htm
http://intellinuxgraphics.org/
http://software.intel.com/sites/oss/

http://www.omgubuntu.co.uk/2010/09/broadcom-releasing-fully-open-so...
http://www.zdnet.com/blog/open-source/broadcom-yes-broadcom-joins-t...
http://www.zdnet.com/blog/open-source/behind-the-open-source-turnar...

Some companies, such as AMD/ATI, provide programming specifications so that open source developers can write drivers for Linux:
http://www.x.org/docs/AMD/
http://www.x.org/wiki/RadeonFeature
http://www.linuxdriverproject.org/mediawiki/index.php/Main_Page#Abo...

"We are a group of Linux kernel developers (over 400 strong) that develop and maintain Linux kernel drivers. We work with the manufacturers of the specific device to specify, develop, submit to the main kernel, and maintain the kernel drivers. We are willing and able to sign NDAs with companies if they wish to keep their specifications closed, as long as we are able to create a proper GPLv2 compliant Linux kernel driver as an end result."

Some companies only go half-way to true GPL-copyleft open source, but developers can still write drivers for Linux:
http://www.malideveloper.com/developer-resources/drivers/open-sourc...

Linux drivers where the developers have to reverse-engineer hardware do still exist:
http://nouveau.freedesktop.org/wiki/FeatureMatrix
... but such drivers are most decidedly in the minority these days. It is a relatively simple matter to avoid hardware, such as nvidia graphics, which still require reverse-engineered drivers.

Edited 2012-09-16 07:55 UTC

Using Nvidia graphics cards in linux does NOT require reverse-engineered drivers. Nvidia provides 32 & 64 bit linux drivers, and communicates well with users to resolve bugs and add features/support.

Using Nvidia graphics cards in linux does NOT require reverse-engineered drivers. Nvidia provides 32 & 64 bit linux drivers, and communicates well with users to resolve bugs and add features/support. "

If one uses the binary blob driver provided by nvidia, one cannot upgrade to Wayland in the near future.

Nvidia's binary blob driver is NOT shipped with Linux distributions, one has to download it separately and actually compile a piece of wrapper code. Any issues the driver causes cannot be addressed by the Linux kernel developers. This is most decidedly NOT recommended for non-expert users.

Every time one updates the kernel, one has to recompile the open source wrapper which sits between the binary blob and the kernel. This requires that one has Linux kernel source code installed. It is an utter pain.

For this and other reasons, it is not recommended to use the nvidia binary blob driver. Since the open source nouveau driver is limited due to the need to reverse engineer, that driver although better does not deliver anywhere near the capabilities of the card.

Since it is a is a relatively simple matter to avoid nvidia graphics and use instead a graphics card which does have a functional, well-performed open source driver which ships as part of the Linux kernel itself, (i.e. Intel or AMD/ATI graphics), then using such hardware for desktop Linux systems is overwhelmingly to be recommended.

Edited 2012-09-17 07:48 UTC

Wow. Just, wow.

"We Linux devs have finally managed to go far enough in market share that hardware vendors will care about us. Now that we are here, we don't care for the younger OSs that are struggling just like we did before anymore, and will gladly sign NDAs on specs so as to keep an edge on them"

Edited 2012-09-17 09:37 UTC

Wow. Just, wow.

"We Linux devs have finally managed to go far enough in market share that hardware vendors will care about us. Now that we are here, we don't care for the younger OSs that are struggling just like we did before anymore, and will gladly sign NDAs on specs so as to keep an edge on them" "

The Linux driver project is over five years old. Why didn't you make this complaint back in April 2007?

http://lwn.net/Articles/276973/

"So the LDP was born. It started out as a single place for hardware manufacturers to contact in order to get drivers written for their devices for free. We allowed the ability for companies to sign an NDA if needed to help get over the hurdle that some companies have in releasing their specifications. The NDA process was put into place through the Linux Foundation, and is a 3-way NDA with all of the proper legal documents needed."

The Linux Driver Project would rather not sign an NDA, they would much prefer that companies published programming specifications. However, if a given company insists that an NDA is necessary or there will be no co-operation, then having to sign an NDA over the programming specifications is a small price to pay.

After all, the end product is source code under the GPL. Everyone is free to study that code and determine how it works. Anyone is able even to re-distribute that code, as long as they agree to re-distribute it to others under the same license as they received it. Where is there a problem?

If there is some problem with keeping the GPL driver code open, other parties may still study that GPL code, determine how it works, write a new driver of their own from the knowledge thereby gained, and redistribute their own code under whatever license they please.

I have to really agree with this - the very language of the article is back-assward... Oh noes, the chip maker doesn't support the OS... and?!?

Since when has it even worked that direction? Operating systems support chips, not the other way around!

Since forever. When a manufacturer wants to ensure their chips will be adopted quickly they often write the code needed and to also ensure there are as few bugs in the supporting code as possible.

Basically bootloader, some drivers, etc... Imagine some graphic specific features like Optimus that are not documented and/or not driver provided, no ways to support them into one or another OS. That could lead to some OS having better support while nothing, technically, prevent them from.

And an OS like Linux being a hack, how can a company be comfortable and confident to an OS that is just a hack ? Do you believe a server could suffer stability issues and just answer "It's just a hack" ?

Kochise

Linux, a hack? Well if Linux is a hack, then Windows is one massive pile kludges, fixing one bad design decision after another--going all the way back to MS-DOS in some cases.

Edited 2012-09-16 01:32 UTC

You seriously must be living under a rock if you're this clueless.

really? Because I dont see Intel announcing a new atom based proc for Linux. If this is "the" processor line to compete against ARM, don't you think it should be able to run Linux too?

ValleyView. Code to support it, especially it's GPU, has been going into the kernel and other components (like the X driver) for months now already.

You could, I don't know, run Linux on ARM? If the multitude of chips Intel makes don't fit your needs, buy an ARM, use a normal Atom, or wait until Linux catches up, then buy one of these new chips.

With Linux, it's only a matter of time.

The vendors are better off switching to ARM. Normal Atom isn't an option for these use cases. Intel has dropped the ball way too many times for their X86 SoCs.

This will just push more vendors to switch to ARM if they were even considering X86 in the first place. Intel's poor track record for X86 SoCs has already pushed enough vendors back to ARM because of the poor support from Intel

Downright crooked. But it does seem like what's happening. I never thought to piece those two bits of lock-out behavior together... good point. I would have modded you up if I didn't already post in this topic.

Edited 2012-09-16 00:58 UTC

Intel is dying? Wow, can you pass me some of what you're smoking?

Slowly, yes, because it is in a dwindling sector: PCs.
Even Intel themselves say so:
http://www.forbes.com/sites/steveschaefer/2012/09/07/intel-slashes-...
ARM on the other hand, is leading in PostPC world.

Intel is heading the same way as Sparc, Power, Alpha, IA64 etc...

Might lead in performance, but is catering to a dwindling market, meanwhile a cheaper competitor is attacking from below...

That's an SoC intended for higher end hardware like workstations and servers. Tegra 2 was a little weak without NEON support but Tegra 3 is more than good enough for anything considering Intel's X86 SoCs.

Not gonna happen, Linux on the desktop is not a moneymaker. The Linux desktop market is simply too small to justify the effort involved; in fact it kind of surprises me that anyone supports gaming GPUs on Linux, period.

(Yes, I use Linux. It Works For Me. But it probably wouldn't if I were into 3D games.)

"Gullible Jones: Not gonna happen, Linux on the desktop is not a moneymaker. The Linux desktop market is simply too small to justify the effort involved; in fact it kind of surprises me that anyone supports gaming GPUs on Linux, period."

Well darn! Quick, somebody go tell Gabe Newell he is wrong and this guy is right!.

Edited 2012-09-16 06:19 UTC

Valve is looking for a way out of Mac OS X and Windows stores, which may diminish sales via Steam.

Well, actually...

http://www.theinquirer.net/inquirer/news/2205758/amds-hondo-will-on...

Umm, LOL?

Even if Windows 8 objectively sucks, vendors are still going to be selling Windows 7 machines for quite a while. Joe User will just fall back to Windows 7. And by the time Windows 7 support ends, Windows NT will probably be well obsolete in favor of Barrelfish or Midori or something.

It's possible that large numbers of people may downgrade or skip over Windows 8, but MS is hardly shooting itself in the foot.

Those wanting to build Android devices will select a different chip anyway, since Clover Trail is x86, nor ARM. ;-)