My biggest beef with Prescott is that Intel rather foolishly lengthened the pipleine and monkeyed with the core design without making the subsequent changes needed to increase clock speed. AMD had it right all along - efficient IPC and low clock speed.
You don't read slashdot do you? This post links to some very recent research that claims lengthening the pipeline is actually the right thing to do.
There might not be any cost to you, however I would argue that there is a cost to many. For me it would be to compromise my values and use a product from a monopoly I do not want to support. I for one make an effort to never support MS. I believe in Capitalism, and feel that a monopoly is devastating to a system based on capitalism. It would cost me very much to have to switch to MSN search.
Yeah, its terrible to support an American company that provides the federal government with billions in tax revenue.
Use Linux? Thats great and good for you for saving money and using something that is better for you. But I can't understand how so many people, particularly teeny boppers and twenty-somethings, really hate Microsoft and feel it violates their values!!.
With very high switching costs, a monopoly is pretty much guaranteed and is a topic in many business strategy classes. If this breaks your values, then I think you are lacking something in life:)
Only in some designs. Architectures like MIPS allow for both a cache-coherent or non-cache-coherent design. In a non-cache-coherent design, the cache is not transparent, and the kernel programmer is responsible for cache management; marking pages as dirty, flushing cache, etc.
There is a big, big difference between caching and keeping caches coherence in a multiprocessor system. You are correct that higher performance can be obtained by doing away with cache coherence. The Cray T3D and T3E machines have no cache coherence and the latter scales to 4096 CPUs I believe.
However not considering MP software with sharing between processors, there are very few uniprocessors that allow software to control caches. The ones that do exist, for example some recent ARM cores, do allow software to pin certain blocks/pages/whatever in cache. This is certainly impractical for user applications and probably too impractical for a kernel as well. It creates a tight coupling between hardware and software.
Not to be a troll, but I sure hope 2.6 makes X more responsive.
Whenever my compile reaches a large link point, interactive response is terrible...my mouse pointer jumps all over the place.
I know that I could improve things by tweaking the kernel and such, but this box is managed by our IT department (it is RH9) and I don't even have root.
By comparison, if I compile the same app in Windows XP, the system is smooth as butter. Nothing jumps, no MP3s skip, etc.
Itanium only kicks ass on SPEC because of their ridiculously large caches and high memory bandwidth.
They have everything to do with Itanium (not IA64). Memory system performance is all that really matters anymore...not instruction sets as you seem to think.
I don't care how fancy your instruction set or superscaler core is. It doesn't mean squat when it has to sit there for a thousand cycles waiting for a word from memory.
No, this does not signal that Itanium is doomed. Have a look at www.spec.org and look at the CPU2000 scores. Itanium is starting to kick some serious tail.
However Itanium is not a desktop chip-- its too big. 64-bit x86 will be a consumer product for desktops.
It's LGPL that is hurting Linux, as Linux is about free software, LGPL is only about proprietary software. Do you really want Linux to have more non free, closed source, proprietary software??
Yes, I do want more non-free, closed source, proprietary software available on Linux. Photoshop, Microsoft Office, FrameMaker, and Call of Duty are some examples.
QT is awesome. Its license isn't. Before you dismiss this as another ill-informed troll message, consider this:
I am involved with a university research project. We started developing our app using QT, but had to quickly abandon it due to the GPL issue. Due to complications with funding and grants, we can't currently open-source our project. And we will not be obtaining QT Developer's licenses for the following reasons:
1. The bureaucratic mess. A proposal would have to be written, alternatives examined, and so forth. This could take too long. This crap is the way many universities work...most software is given to us or obtained very cheaply with academic licenses.
2. Cost. $1550 goes a long ways in funding graduate students. Our research director would rather hire more students with the money saved. Even Microsoft Visual Studio, which includes far more than GUI widgets, is only a couple hundred bucks for the academic edition.
I think this is one example of why this is hurting the Linux movement. GTK is LGPL...QT should be the same. I know that TrollTech needs to pay their developers, but I sure wish someone like IBM would buy them out and LGPL QT. For something intregal to a platform, such as GUI toolkits, having to spend $1500 for a license to develop a close-source application just isn't good. Also consider small-time shareware developers. I know that shareware isn't popular in Linux, but its a way for a developer to get payed for creating that unique, niche application
Of course my arguments are based on my opinion that KDE/QT is superior to Gnome/GTK, and that if it weren't for the license issue, it would be the dominant platform today.
It really comes down to games. However most developers are increasingly writing their games using cross-platform toolkits so that they can penetrate all three markets. And according to the people I know, Xbox performs the best. Gamespot.com corraborates this.
So unless the few games that are console-exclusive are really what matters to you, or some tangible things like form factors, then the only difference is graphics/gameplay quality and smoothness. And XBox wins here for most cross-platform games.
Using your paint analogy, if an artist draws the exact same picture multiple times using different paints, then the superior picture will be the one that uses the superior paints.
XBox is probably the best console according to game developers. Gamespot.com also always claims that the most fluid and best graphics/gameplay comes from XBox followed closely by Gamecube, and then finally the PS2. This is for comparing games written for all three platforms.
It could very well have to due with the cross-platform rendering toolkits that game developers use. They may not be tuned correctly for PS2.
"In an overclocked system where stability and cooling is important"
Ha ha ha...very funny that you mentioned overclocking and stability in the same sentence.
If you really want a stable system, you won't take the risks of overclocking. What do you gain? 10-30%?! No thanks. That performance gain doesn't make or break my computing experience and I'll just wait 18 months for Moore's law to provide me with a 100% speedup.
I think that water-cooled super-overclocking is for those high-school kids who think its cool to put a huge spoiler on their Honda Civic.
Oh, you say that all these researchers say that increasing the pipeline depth past 20 stages increases performance?! How can that be so when so many slashdotters claim that Intel is big, bad, and evil by trying to trick everybody buying an inferior chip?
Hey armchair architects, have a look at the SPEC benchmarks. Intel knows what they are doing...even with Itanium.
Why don't you have a look at the latest SPEC benchmark scores? The numbers speak for themselves. Itanium scores are very impressive. I was once a doubter too.
"As most of us know, a longer pipeline can lead to slowdowns in the form of branch mispredictions and pipeline stalls
Get off your high horse. Intel architects aren't dummies. Itanium benchmarks are starting to whoop some serious ass and the P4 and Athlon have been neck-and-neck for years. I'm sure Prescott will perform very well.
I can get into all kinds of architecture speak as to why your simplistic notions of mispredictions and pipeline stalls might not be so terrible. Who knows? Maybe Intel will execute both paths of a branch? They've already got partial instruction replay to make squashes much less expensive. With deep speculation, a big instruction window, good bypassing capabilities, and effective non-blocking caches, "pipeline stalls" are not an issue due to branch mispredictions. The bigger issue is memory latency/bandwidth and Intel has always done well with that. A branch misprediction can be easily tolerated...an L2 cache miss can't.
Consumers can replace Google immediately if a better search engine comes along. For this reason, I wouldn't buy stock in Google unless they diversify themselves.
On the other hand, Yahoo has subscription-based services and other things to keep customers loyal such as an e-mail address that can't be moved (unlike wireless phone numbers).
Any experienced developer, without an ego, will know that leveraging existing code is usually the best way to get something done faster and better. The Mozilla rewrite was a terrible idea and is what resulted in its delay. Sure, Netscape code might have been aweful, but incremental improvements are usually better.
I've been a user of Linux since 1993. For the last 5 years or so, I've used FVWM as my window manager.
I decided to try RH9, with the Ximian Desktop, on a new Pentium4 2.0GHz 512mb RAM machine.
It makes this machine feel like a 486...very very slow and pages all the time. I think its due to Nautilus...pathetic I think.
Which makes me wonder what IBM will use? Maybe they will rock the Linux world by releasing a desktop environment that works and is free. I hope so...
KDE is nice, but I can't support it. In my academic project, we had to kill development of a program utilizing QT due to licensing. You see, due to some tricky grants, we can't GPL our project. And with the bureaucracy of academics, finding a way to acquire the QT licenses just isn't worth the hassle and resources.
Linux is great for my uses, but it just isn't there as a desktop for the masses.
Slashdot Mirror
My biggest beef with Prescott is that Intel rather foolishly lengthened the pipleine and monkeyed with the core design without making the subsequent changes needed to increase clock speed. AMD had it right all along - efficient IPC and low clock speed.
You don't read slashdot do you? This post links to some very recent research that claims lengthening the pipeline is actually the right thing to do.
Another armchair architect.
There might not be any cost to you, however I would argue that there is a cost to many. For me it would be to compromise my values and use a product from a monopoly I do not want to support. I for one make an effort to never support MS. I believe in Capitalism, and feel that a monopoly is devastating to a system based on capitalism. It would cost me very much to have to switch to MSN search.
:)
Yeah, its terrible to support an American company that provides the federal government with billions in tax revenue.
Use Linux? Thats great and good for you for saving money and using something that is better for you. But I can't understand how so many people, particularly teeny boppers and twenty-somethings, really hate Microsoft and feel it violates their values!!.
With very high switching costs, a monopoly is pretty much guaranteed and is a topic in many business strategy classes. If this breaks your values, then I think you are lacking something in life
The fight is very easy for Microsoft. All they have to do is make a better search engine. There is no cost nor effort for me to switch search engines.
Only in some designs. Architectures like MIPS allow for both a cache-coherent or non-cache-coherent design. In a non-cache-coherent design, the cache is not transparent, and the kernel programmer is responsible for cache management; marking pages as dirty, flushing cache, etc.
There is a big, big difference between caching and keeping caches coherence in a multiprocessor system. You are correct that higher performance can be obtained by doing away with cache coherence. The Cray T3D and T3E machines have no cache coherence and the latter scales to 4096 CPUs I believe.
However not considering MP software with sharing between processors, there are very few uniprocessors that allow software to control caches. The ones that do exist, for example some recent ARM cores, do allow software to pin certain blocks/pages/whatever in cache. This is certainly impractical for user applications and probably too impractical for a kernel as well. It creates a tight coupling between hardware and software.
Not to be a troll, but I sure hope 2.6 makes X more responsive.
Whenever my compile reaches a large link point, interactive response is terrible...my mouse pointer jumps all over the place.
I know that I could improve things by tweaking the kernel and such, but this box is managed by our IT department (it is RH9) and I don't even have root.
By comparison, if I compile the same app in Windows XP, the system is smooth as butter. Nothing jumps, no MP3s skip, etc.
Try Yahoo Games. Seriously. Yahoo Pool is widely popular and is a great way to mindlessly have fun and chat with people.
If your gf isn't the dork you are, it might be the only thing that works.
Swing has made me sick in the past (Java Swing, not a playground swing :)
But a quick search on Google didn't find me any real SWT API documentation. What gives? All I found were some articles on the eclipse.org website...
Itanium only kicks ass on SPEC because of their ridiculously large caches and high memory bandwidth.
They have everything to do with Itanium (not IA64). Memory system performance is all that really matters anymore...not instruction sets as you seem to think.
I don't care how fancy your instruction set or superscaler core is. It doesn't mean squat when it has to sit there for a thousand cycles waiting for a word from memory.
No, this does not signal that Itanium is doomed. Have a look at www.spec.org and look at the CPU2000 scores. Itanium is starting to kick some serious tail.
However Itanium is not a desktop chip-- its too big. 64-bit x86 will be a consumer product for desktops.
It's LGPL that is hurting Linux, as Linux is about free software, LGPL is only about proprietary software. Do you really want Linux to have more non free, closed source, proprietary software??
Yes, I do want more non-free, closed source, proprietary software available on Linux. Photoshop, Microsoft Office, FrameMaker, and Call of Duty are some examples.
QT is awesome. Its license isn't. Before you dismiss this as another ill-informed troll message, consider this:
I am involved with a university research project. We started developing our app using QT, but had to quickly abandon it due to the GPL issue. Due to complications with funding and grants, we can't currently open-source our project. And we will not be obtaining QT Developer's licenses for the following reasons:
1. The bureaucratic mess. A proposal would have to be written, alternatives examined, and so forth. This could take too long. This crap is the way many universities work...most software is given to us or obtained very cheaply with academic licenses.
2. Cost. $1550 goes a long ways in funding graduate students. Our research director would rather hire more students with the money saved. Even Microsoft Visual Studio, which includes far more than GUI widgets, is only a couple hundred bucks for the academic edition.
I think this is one example of why this is hurting the Linux movement. GTK is LGPL...QT should be the same. I know that TrollTech needs to pay their developers, but I sure wish someone like IBM would buy them out and LGPL QT. For something intregal to a platform, such as GUI toolkits, having to spend $1500 for a license to develop a close-source application just isn't good. Also consider small-time shareware developers. I know that shareware isn't popular in Linux, but its a way for a developer to get payed for creating that unique, niche application
Of course my arguments are based on my opinion that KDE/QT is superior to Gnome/GTK, and that if it weren't for the license issue, it would be the dominant platform today.
Hey asshole, I have friends and family who write console games for a living.
Read my other reply in this thread.
It really comes down to games. However most developers are increasingly writing their games using cross-platform toolkits so that they can penetrate all three markets. And according to the people I know, Xbox performs the best. Gamespot.com corraborates this.
So unless the few games that are console-exclusive are really what matters to you, or some tangible things like form factors, then the only difference is graphics/gameplay quality and smoothness. And XBox wins here for most cross-platform games.
Using your paint analogy, if an artist draws the exact same picture multiple times using different paints, then the superior picture will be the one that uses the superior paints.
XBox is probably the best console according to game developers. Gamespot.com also always claims that the most fluid and best graphics/gameplay comes from XBox followed closely by Gamecube, and then finally the PS2. This is for comparing games written for all three platforms.
It could very well have to due with the cross-platform rendering toolkits that game developers use. They may not be tuned correctly for PS2.
"In an overclocked system where stability and cooling is important"
Ha ha ha...very funny that you mentioned overclocking and stability in the same sentence.
If you really want a stable system, you won't take the risks of overclocking. What do you gain? 10-30%?! No thanks. That performance gain doesn't make or break my computing experience and I'll just wait 18 months for Moore's law to provide me with a 100% speedup.
I think that water-cooled super-overclocking is for those high-school kids who think its cool to put a huge spoiler on their Honda Civic.
Oh, you say that all these researchers say that increasing the pipeline depth past 20 stages increases performance?! How can that be so when so many slashdotters claim that Intel is big, bad, and evil by trying to trick everybody buying an inferior chip?
Hey armchair architects, have a look at the SPEC benchmarks. Intel knows what they are doing...even with Itanium.
Why don't you have a look at the latest SPEC benchmark scores? The numbers speak for themselves. Itanium scores are very impressive. I was once a doubter too.
"As most of us know, a longer pipeline can lead to slowdowns in the form of branch mispredictions and pipeline stalls
Get off your high horse. Intel architects aren't dummies. Itanium benchmarks are starting to whoop some serious ass and the P4 and Athlon have been neck-and-neck for years. I'm sure Prescott will perform very well.
I can get into all kinds of architecture speak as to why your simplistic notions of mispredictions and pipeline stalls might not be so terrible. Who knows? Maybe Intel will execute both paths of a branch? They've already got partial instruction replay to make squashes much less expensive. With deep speculation, a big instruction window, good bypassing capabilities, and effective non-blocking caches, "pipeline stalls" are not an issue due to branch mispredictions. The bigger issue is memory latency/bandwidth and Intel has always done well with that. A branch misprediction can be easily tolerated...an L2 cache miss can't.
Consumers can replace Google immediately if a better search engine comes along. For this reason, I wouldn't buy stock in Google unless they diversify themselves.
On the other hand, Yahoo has subscription-based services and other things to keep customers loyal such as an e-mail address that can't be moved (unlike wireless phone numbers).
Any experienced developer, without an ego, will know that leveraging existing code is usually the best way to get something done faster and better. The Mozilla rewrite was a terrible idea and is what resulted in its delay. Sure, Netscape code might have been aweful, but incremental improvements are usually better.
I've been a user of Linux since 1993. For the last 5 years or so, I've used FVWM as my window manager.
I decided to try RH9, with the Ximian Desktop, on a new Pentium4 2.0GHz 512mb RAM machine.
It makes this machine feel like a 486...very very slow and pages all the time. I think its due to Nautilus...pathetic I think.
Which makes me wonder what IBM will use? Maybe they will rock the Linux world by releasing a desktop environment that works and is free. I hope so...
KDE is nice, but I can't support it. In my academic project, we had to kill development of a program utilizing QT due to licensing. You see, due to some tricky grants, we can't GPL our project. And with the bureaucracy of academics, finding a way to acquire the QT licenses just isn't worth the hassle and resources.
Linux is great for my uses, but it just isn't there as a desktop for the masses.
Good luck IBM.
Believing that a software application is to fault for your lousy presentation is even dumber
The Minolta XD-11 is a awesome camera with one fatal flaw-- the hybrid integrated circuit doesn't last.
I got burned on one of these from eBay. I paid $180 for it, and it worked for about 45 days before the meter died.
Get an SRT instead.
The Pentax K1000 is over-rated and over-priced. It lacks depth-of-field preview.
There are plenty of other, better older manual-focus cameras to choose from.
Of course the sources will be available but they won't offer an installable distribution for free.