I don't know why Apple does this. Just to confuse the market and make it seem like Apple has some special sauce whereas the reality is that Apple uses the same ARM designs as everybody else, running at the same silicons and Apples "customizations" are really minor hacks to the peripheral support. To me, this comes across as dishonest, and I wonder why they do it especially considering many people will perceive the next iphone as underpowered because of what seems to be an ARM chip from the previous generation.
Apple's "A4" is really an ARM A8 and it would seem that Apple's "A8" is really an ARM A9. Same processor as everybody else for the next phone generation. And there is nothing special or unique to Apple about the PowerVR core Apple will use either.
I'm sure Microsoft could easily find another ex Procter & Gamble product manager to replace Steve Ballmer. What they need is someone to replace Bill Gates, and that isn't so easy.
The road to ruin for Microsoft was already under construction well before Bill Gates retired. It was Bill Gates' idea to try to build Vista around.NET, which lead to the "reset", the most demoralizing and costly misstep of Microsoft's history. Yes, the reset happened on Ballmer's watch but Bill Gates caused it. And Bill Gates hired Steve Ballmer. Lots of other major mistakes made by Bill Gates before and after retiring.
On the other hand, Microsoft so richly deserves the debilitating disease it now suffers from that I think you are right, they really need someone like just like Bill Gates to continue the fall after Ballmer goes.
I'm not an Apple fan, far from it, but your claim is just wrong. The wheel on the ipod? itunes? The original Mac? Zero config networking? Stepping way out with the ipad?
Apple may be many distasteful things, but "not innovative" is not one of them.
it's got a PE of ~23. Going by Benjamin Graham's rule of thumb it needs to grow at 7.5%
And there is a significant risk that AAPL may fail to do that. It might lose its early lead in the smart phone market, or its quality control or its industrial design might degenerate in the absence of Steve Jobs. The market might lose its taste for products with proprietary lock in. There could be more embarrassing engineering failures like antennagate. There could be further PR fiascos like kicking in the front doors of journalist's homes. AAPL might face increased competition from inexpensive products with similar functionality. AAPL's attempts to lock unencumbered video formats out of the web might fail.
What's the hurry? I would rather we take a few years longer and get it right than slap together something substandard that remains unpleasant to use and work with for all eternity.
But with 20 20 hindsight, the major issue was not so much proprietary drivers and not the strange OEM behavior that almost seemed orchestrated against open source, as it was the counterproductive governance of Xfree86.org. Linux video really took off after the fork and by now is getting damn good. It doesn't take too much imagination to see that in a year or two it will lead the industry, just as a number of other aspects of Linux already do.
Then how do you explain Microsoft, which has never had a very positive image, yet managed to dominate both the desktop operating system and Web browser markets?
"In particular, ZFS’s advanced architecture addresses two of our key performance concerns: random I/O, and small I/O. In a large cluster environment a Lustre I/O server (OSS) can be expected to generate a random I/O workload. There will be 100’s of threads concurrently accessing different files in the back-end file system. For writes ZFS’s copy-on-write transaction model converts this random workload in to a streaming workload which is critical when using SATA disks. For small I/O, Lustre can leverage a ZIL placed on separate SSD devices to maximize performance."
The LLNL ZFS study has been pretty widely publicized in the HPC community. Lustre uses the filesystem API rather than mounting in. Until now Lustre used ext under-the-hood for data storage, so the performance improvement from ZFS is relative to ext. ext3/4 may very well outperform ZFS on a workstation or small server, but that's not the what Lustre is used for (even their test system is ~900TB).
Disclaimer: I used to work for LLNL.
Disclaimer: I used to work on Ext3. I would classify the above as "hype from Sun". There is a hidden cost to making all the writes linear on spinning media: the reads become nonlinear. This is usually the wrong tradeoff.
Note that a traditional journal is another way of linearizing writes in that a transaction write transaction can be considered durably recorded to media as soon as the journal write completes.
Benchmarks tell the true story, not hype, and on good information and belief the benchmarks say ZFS is slow.
Re SDL: I would not recommend that any serious project use it, it many just adds a new layer of bugs. For an OpenGL app you only need two things that SDL provides: a drawable surface and input events. Here is code for this directly on X/Glx, it is short and sweet and on the metal. As an extra benefit, Alt-Tab works in fullscreen whereas it does not in SDL. Doing the same on top of DirectX is left as an exercise for the interested reader, please send me the code when you're done.
OpenGL 3.0 finally comes out and people call it a great disappointment?
Pretty much the sum of their complaints was that they decided to deprecate a load of stuff, rather than removing it outright, so old programs still compiled and worked with the new API. How terrible. Direct3D releases a new interface for each version, so porting from DirectX 9 to DirectX 10 is almost as much effort as porting to OpenGL 3. In contrast, porting from GL2 to GL3 just meant some refactoring.
OpenGL 3.1 then actually did remove things, and came with a core that was roughly equivalent to Direct3D in terms of functionality, plus extensions.
I am so glad Kronos did the reasonable thing and didn't throw out the entire API like some noisy game engine hotheads wanted. Immediate most is just so much faster to prototype new ideas with, this fact is often overlooked by developers who are just trying to optimize an existing code base. It would be really stupid to ever actually drop it from development profiles of OpenGL (and if any hardware vendor momentarily forgets this an angry mob of engineering organizations with be quick to remind them). The right thing to do is exactly what kronos did: make the fixed function pipeline including immediate mode an optional profile so you can do your development work in full comfort and hone your code down to the guaranteed-in-hardware subset at your leisure.
Be honest and admit that the OpenGL game market has been microscopic
You're quite the troll, aren't you? Fact: OpenGL drives the entire console world except for XBox. Fact: OpenGL drives the entire smartphone game market. Fact: OpenGL is cross platform, GL is not. Fact: AAA OpenGL titles continue to be released regularly on Windows (id's technical dominance of each new engine generation ensures this).
Now with a much more efficient community process (Kronos) OpenGL is at least at parity with Direct3D and soon will surpass it, with a far nicer, more consistent, and more widely supported API. The Microsoft monkey is very nearly off the back of the graphics community, except for a few trolls like you.
Modern FPS is all about being brain dead and watching what we want you to watch. It is hardly different from advertising.
Really? Have you played Oblivion?
Ah, I read "first person simulator" where you meant "first person shooter". Agree, used to play a lot of shooters, got tired of it. Very little imagination in the genre, just an addiction to higher frame rates and more polygons. I suppose that is why Bethesda ended up aborbing id and not the other way round.
There have been plenty of benchmarks out there showing ZFS's performance besting EXT3 and EXT4 on identical hardware (with one running OpenSolaris and the others on linux)
Unfortunately, Lustre-on-ZFS [zfsonlinux.org] is substantially faster that lustre on ext3, mainly because ZFS combines the features of an lvm and a filesystem
That's bafflegab and incorrect. Or if you disagree, please explain why.
I don't think a lot of people are going to go into many details on this article, because anyone using [lustre] is liking using it in some way to leverage the idea of the cloud inside their respective [businesses]
There are people who know a lot about Lustre and aren't beholden to anyone. It is a GPLed open source project after all.
Slashdot Mirror
Maybe many people admire him for building such a towering business as Microsoft.
Maybe "respect for the law" is not high on many people's list of admirable qualities.
The ODF standard is equally flawed for pretty much the same reasons (except maybe point 3, which I'll concede).
What kind of crack are you smoking?
I don't know why Apple does this. Just to confuse the market and make it seem like Apple has some special sauce whereas the reality is that Apple uses the same ARM designs as everybody else, running at the same silicons and Apples "customizations" are really minor hacks to the peripheral support. To me, this comes across as dishonest, and I wonder why they do it especially considering many people will perceive the next iphone as underpowered because of what seems to be an ARM chip from the previous generation.
Apple's "A4" is really an ARM A8 and it would seem that Apple's "A8" is really an ARM A9. Same processor as everybody else for the next phone generation. And there is nothing special or unique to Apple about the PowerVR core Apple will use either.
I'm sure Microsoft could easily find another ex Procter & Gamble product manager to replace Steve Ballmer. What they need is someone to replace Bill Gates, and that isn't so easy.
The road to ruin for Microsoft was already under construction well before Bill Gates retired. It was Bill Gates' idea to try to build Vista around .NET, which lead to the "reset", the most demoralizing and costly misstep of Microsoft's history. Yes, the reset happened on Ballmer's watch but Bill Gates caused it. And Bill Gates hired Steve Ballmer. Lots of other major mistakes made by Bill Gates before and after retiring.
On the other hand, Microsoft so richly deserves the debilitating disease it now suffers from that I think you are right, they really need someone like just like Bill Gates to continue the fall after Ballmer goes.
Nothing they've done has been "innovative"
I'm not an Apple fan, far from it, but your claim is just wrong. The wheel on the ipod? itunes? The original Mac? Zero config networking? Stepping way out with the ipad?
Apple may be many distasteful things, but "not innovative" is not one of them.
They do not get involved in races to the bottom
Ehm, the tablet market?
it's got a PE of ~23. Going by Benjamin Graham's rule of thumb it needs to grow at 7.5%
And there is a significant risk that AAPL may fail to do that. It might lose its early lead in the smart phone market, or its quality control or its industrial design might degenerate in the absence of Steve Jobs. The market might lose its taste for products with proprietary lock in. There could be more embarrassing engineering failures like antennagate. There could be further PR fiascos like kicking in the front doors of journalist's homes. AAPL might face increased competition from inexpensive products with similar functionality. AAPL's attempts to lock unencumbered video formats out of the web might fail.
What's the hurry? I would rather we take a few years longer and get it right than slap together something substandard that remains unpleasant to use and work with for all eternity.
But with 20 20 hindsight, the major issue was not so much proprietary drivers and not the strange OEM behavior that almost seemed orchestrated against open source, as it was the counterproductive governance of Xfree86.org. Linux video really took off after the fork and by now is getting damn good. It doesn't take too much imagination to see that in a year or two it will lead the industry, just as a number of other aspects of Linux already do.
Then how do you explain Microsoft, which has never had a very positive image, yet managed to dominate both the desktop operating system and Web browser markets?
Perhaps having no respect for the law helped?
If it wasn't for that meddling Apple it be the year of Linux on the desktop!
It does seem to be the year of Unix on the desktop, not all that far away from Linux really.
It appears to be based on the linked site:
"In particular, ZFS’s advanced architecture addresses two of our key performance concerns: random I/O, and small I/O. In a large cluster environment a Lustre I/O server (OSS) can be expected to generate a random I/O workload. There will be 100’s of threads concurrently accessing different files in the back-end file system. For writes ZFS’s copy-on-write transaction model converts this random workload in to a streaming workload which is critical when using SATA disks. For small I/O, Lustre can leverage a ZIL placed on separate SSD devices to maximize performance."
The LLNL ZFS study has been pretty widely publicized in the HPC community. Lustre uses the filesystem API rather than mounting in. Until now Lustre used ext under-the-hood for data storage, so the performance improvement from ZFS is relative to ext. ext3/4 may very well outperform ZFS on a workstation or small server, but that's not the what Lustre is used for (even their test system is ~900TB).
Disclaimer: I used to work for LLNL.
Disclaimer: I used to work on Ext3. I would classify the above as "hype from Sun". There is a hidden cost to making all the writes linear on spinning media: the reads become nonlinear. This is usually the wrong tradeoff.
Note that a traditional journal is another way of linearizing writes in that a transaction write transaction can be considered durably recorded to media as soon as the journal write completes.
Benchmarks tell the true story, not hype, and on good information and belief the benchmarks say ZFS is slow.
Re SDL: I would not recommend that any serious project use it, it many just adds a new layer of bugs. For an OpenGL app you only need two things that SDL provides: a drawable surface and input events. Here is code for this directly on X/Glx, it is short and sweet and on the metal. As an extra benefit, Alt-Tab works in fullscreen whereas it does not in SDL. Doing the same on top of DirectX is left as an exercise for the interested reader, please send me the code when you're done.
GL is not
Ahem. DX is not.
OpenGL 3.0 finally comes out and people call it a great disappointment?
Pretty much the sum of their complaints was that they decided to deprecate a load of stuff, rather than removing it outright, so old programs still compiled and worked with the new API. How terrible. Direct3D releases a new interface for each version, so porting from DirectX 9 to DirectX 10 is almost as much effort as porting to OpenGL 3. In contrast, porting from GL2 to GL3 just meant some refactoring.
OpenGL 3.1 then actually did remove things, and came with a core that was roughly equivalent to Direct3D in terms of functionality, plus extensions.
I am so glad Kronos did the reasonable thing and didn't throw out the entire API like some noisy game engine hotheads wanted. Immediate most is just so much faster to prototype new ideas with, this fact is often overlooked by developers who are just trying to optimize an existing code base. It would be really stupid to ever actually drop it from development profiles of OpenGL (and if any hardware vendor momentarily forgets this an angry mob of engineering organizations with be quick to remind them). The right thing to do is exactly what kronos did: make the fixed function pipeline including immediate mode an optional profile so you can do your development work in full comfort and hone your code down to the guaranteed-in-hardware subset at your leisure.
Be honest and admit that the OpenGL game market has been microscopic
You're quite the troll, aren't you? Fact: OpenGL drives the entire console world except for XBox. Fact: OpenGL drives the entire smartphone game market. Fact: OpenGL is cross platform, GL is not. Fact: AAA OpenGL titles continue to be released regularly on Windows (id's technical dominance of each new engine generation ensures this).
Now with a much more efficient community process (Kronos) OpenGL is at least at parity with Direct3D and soon will surpass it, with a far nicer, more consistent, and more widely supported API. The Microsoft monkey is very nearly off the back of the graphics community, except for a few trolls like you.
Modern FPS is all about being brain dead and watching what we want you to watch. It is hardly different from advertising.
Really? Have you played Oblivion?
Ah, I read "first person simulator" where you meant "first person shooter". Agree, used to play a lot of shooters, got tired of it. Very little imagination in the genre, just an addiction to higher frame rates and more polygons. I suppose that is why Bethesda ended up aborbing id and not the other way round.
Modern FPS is all about being brain dead and watching what we want you to watch. It is hardly different from advertising.
Really? Have you played Oblivion?
Or to state the obvious, port to OpenGL.
But ZFS is cool and trendy among geeks! Ext3/4 can't beat that!!
Got me :-/
There have been plenty of benchmarks out there showing ZFS's performance besting EXT3 and EXT4 on identical hardware (with one running OpenSolaris and the others on linux)
Link please.
If you are comparing ZFS performance on linux, then, yes, it is slower
No, I am comparing Ext3/4 on linux to ZFS on Solaris.
And by the way, is your opinion based on benchmarks, or on hype from Sun? I strongly suspect the latter.
Unfortunately, Lustre-on-ZFS [zfsonlinux.org] is substantially faster that lustre on ext3, mainly because ZFS combines the features of an lvm and a filesystem
That's bafflegab and incorrect. Or if you disagree, please explain why.
I don't think a lot of people are going to go into many details on this article, because anyone using [lustre] is liking using it in some way to leverage the idea of the cloud inside their respective [businesses]
There are people who know a lot about Lustre and aren't beholden to anyone. It is a GPLed open source project after all.
Lustre on ZFS It, apparently, scales much better than the traditional LVM/RAID/etc. backend methods.
By the way, where did you get that idea?