OpenBSD strives to prevent holes in the first place before releases. If the linux development cycle would strive to make more "stable" stable releases than the ones that come out now this could be a much smaller concern. Linux's bazaar style development of "release early release often" has major weaknesses in the "Are you sure all the holes are plugged? department".
I wasn't saying apps can't be written to use the CPU's as such that they can't achieve the similar performance of 4.4Ghz. I am saying 99% of users won't be doing signal processing. SIMD instructions like SSE or Altivec are great for signal processing too and can be done really well on one CPU.
You don't get a 4.4Ghz machine by running 2 2.2 Ghz CPU's. Find a hard disk/RAM/Bus that can keep up with such a beast. It would be an immense waste of CPU to even build such a thing. You'd be far better off with two different 2.2Ghz boxes in a cluster. I doubt any applications today that a person would want to run on an SMP machine could even push the limits of a single 2.2Ghz CPU.
Of course if your heat sink falls off...:) That's bad.
Linux just isn't tested enough before its branded "stable". How many people have had VM failures in linux or at least really bad performance of swapping. I have heard several horror stories.
by your own argument why does GNOME exist? It was to replace KDE by its original goals.
Mozilla really doesn't do it for me. There were lots of browsers out there before Mozilla..
I think you need a sanity check... Open source is all about duplication of effort. Linux shouldn't even exist by your argument since 386BSD predated it.. Everyone should work on the one that was there first... hmmm
"Companies like Microsoft, with thousands of hard working, dedicated fulltimers are deemed to prevail..."
I don't know about you but I get tired of seeing MS Word not work when I can play the pinball game "easter egg" that comes with it just fine. I think your "dedicated developer" theory is a bunch of bullshit. I mean have you ever even used ME? Its a piece of crap. Caveat Emptor!! [Brady Bunch quote]...
Now it will twice as fast to heat up to the point that it goes to half speed which is 1Ghz. I guess you can call that progress.
They really have to work on cooling those things better. Dell sells em really cheap though. But I am currently in the market for a Mac of some sort. Who knows... Computers are really cheap these days.:)
A complex datatype is becoming one of the reasons many people are considering C# over java these days. C++ doesn't as of right now have such a type but you can write the class. Supporting it natively in the language could probably yield many improvements.
I think the language still needs templates like or better than C++. I have been dabbling in generative programming and template metaprogramming in C++ is one way to get the job done.
Essentially a meta language telling the compiler how to write code for you is really useful in some cases. Loops get unrolled some computation gets done at compile time and it incurs no run-time cost.
If this guy can improve on the templates in C++ and make them better for generative programming practices I'd be sold on using it.
How does Fear Uncertainty and Doubt come into play?
Better things to do with G4 hardware
on
Case Tweaking
·
· Score: 1
I guess its a natural reaction to kill what is alien to you but to throw away PowerPC hardware is a sin. Its just a better CPU than any Intel I have ever seen. I mean if you had that many G4's you could have built a killer cluster out of it and done some Vector processing on each node with the extra sexy AltiVec stuff that comes on those CPU's.
At this point in my life I view the G4 as an equal to the x86 and the only reason x86 is equal is the range of software geared towards it. I personally find x86 to be "on the way out". Its design has been recycled so many times [16 bit - 64bit, new opcodes no one uses anywhere.] All in the name of backwards compatibility that your OS's don't provide.
If I had access to a G4 I certainly wouldn't throw it away. I would just get OS 10.1 when its out.:)
Actually its not just VLIW. Its EPIC.
Explicitly Parallel Instruction Code. Its in its
own class. It may indeed be VLIW like but its called EPIC!
Yes the compiler optimization is the key on such an architecture. You basically don't want to ever write assembler on an Itanium if you can avoid it. [unless you are the compiler writer I suppose]
I don't understand this reasoning. I also think its fishy that Microsoft would help a GPL'd project after all the crap they just said about the GPL. They may want to turn the "cancer" against us somehow.
I suppose they believe that if they can get every Linux/BSD user to write applications with.NET in mind and not use any other system that they may be able to use the servers they control to hold information hostage somehow. True, the tools are open source but will the data pathways be open or private?
If that is Microsoft's goal to wiping the credibility of Open Sourced systems out they forgot that one thing Open Source people are wonderful at is duplication of effort just for the hell of it. It means next to nothing for another group to spend their free time coming up with an alternative with similar or the same functionality. I mean GNOME started because of licensing issues with KDE. Before GNOME there was the Harmony project which was an attempt to replace all the functionality of QT with a drop in GPL'd replacement. Anyone heard of Lesstif? The free Motif clone? Mesa?
If Microsoft plans to use.NET to squash open source its not a very sound plan at all. It will still make.NET more feasible an option if both Unix and Windows can talk to each other through.NET services and I really believe that is the only goal they have with Mono.... to increase credibility of usefulness.
I know its hard to look at this stuff objectively... I think it is especially hard for those who worship a license like the GPL to ever trust a company that is usually targeted as the enemy. Truly I stopped using linux because I got tired of all the zealots persecuting closed source vendors. I use FreeBSD now because I like the environment [and the kernel source:)] a lot better.
There was actually a court case where a bank made a calculation error with a spreadsheet package [Quatro Pro?]. Anyway they said the results weren't guaranteed. The bank lost the court case and lots of money. This is commonplace in software.
I work for a company that writes middleware for clusters and I get to work on them every day of the week.
I just wouldn't use ethernet to connect such a large machine. TCP/IP for message passing is by far not the best choice for making a collective of machines talk to each other. Also PVM is not the only choice anymore. Try MPI.
Slashdot Mirror
OpenBSD strives to prevent holes in the first place before releases. If the linux development cycle would strive to make more "stable" stable releases than the ones that come out now this could be a much smaller concern. Linux's bazaar style development of "release early release often" has major weaknesses in the "Are you sure all the holes are plugged? department".
I wasn't saying apps can't be written to use the CPU's as such that they can't achieve the similar performance of 4.4Ghz. I am saying 99% of users won't be doing signal processing. SIMD instructions like SSE or Altivec are great for signal processing too and can be done really well on one CPU.
You don't get a 4.4Ghz machine by running 2 2.2 Ghz CPU's. Find a hard disk/RAM/Bus that can keep up with such a beast. It would be an immense waste of CPU to even build such a thing. You'd be far better off with two different 2.2Ghz boxes in a cluster. I doubt any applications today that a person would want to run on an SMP machine could even push the limits of a single 2.2Ghz CPU.
:) That's bad.
Of course if your heat sink falls off...
No excuse. If you say you support something and that support is stable... it should be so.
Fix it... don't say its not worth having anyway. Or if its not worth having strip token ring support out and save a few kB.
Too many cooks.. not enough real stability.
Linux just isn't tested enough before its branded "stable". How many people have had VM failures in linux or at least really bad performance of swapping. I have heard several horror stories.
I have heard the same story of a certain linux box bringing down a certain subnet of a certain university.
of this I am certain
I have never read anything more accurate about what I have been going through. Talk about hitting the nail on the head.
Really?
That's interesting... hmmm I will have to think about that and get back to you...
Really?
That's very interesting...
[Begin sarcasm]
... who cares
Yeah... Ximian is really "fighting" Microsoft with alternatives like Mono... way to go...
[End sarcasm]
seriously
by your own argument why does GNOME exist? It was to replace KDE by its original goals.
Mozilla really doesn't do it for me. There were lots of browsers out there before Mozilla..
I think you need a sanity check... Open source is all about duplication of effort. Linux shouldn't even exist by your argument since 386BSD predated it.. Everyone should work on the one that was there first... hmmm
HEHEHEHE
"Companies like Microsoft, with thousands of hard working, dedicated fulltimers are deemed to prevail..."
I don't know about you but I get tired of seeing MS Word not work when I can play the pinball game "easter egg" that comes with it just fine. I think your "dedicated developer" theory is a bunch of bullshit. I mean have you ever even used ME? Its a piece of crap. Caveat Emptor!! [Brady Bunch quote]...
Troll assmunch go home
You can sit and whine about how its being delayed or you can get off your duff and help.
Even taking some time to run what parts of FreeBSD 5 do exist to give some valuable feedback as to how it behaves on your system could be useful.
I am just as dissapointed as anyone else about the news but I can't help but feel motivated to lend a hand in such bad times.
I will probably try FBSD 5 this weekend and see what's what. Too bad I don't have SMP...
Now it will twice as fast to heat up to the point that it goes to half speed which is 1Ghz. I guess you can call that progress.
:)
They really have to work on cooling those things better. Dell sells em really cheap though. But I am currently in the market for a Mac of some sort. Who knows... Computers are really cheap these days.
A complex datatype is becoming one of the reasons many people are considering C# over java these days. C++ doesn't as of right now have such a type but you can write the class. Supporting it natively in the language could probably yield many improvements.
I think the language still needs templates like or better than C++. I have been dabbling in generative programming and template metaprogramming in C++ is one way to get the job done.
Essentially a meta language telling the compiler how to write code for you is really useful in some cases. Loops get unrolled some computation gets done at compile time and it incurs no run-time cost.
If this guy can improve on the templates in C++ and make them better for generative programming practices I'd be sold on using it.
How does Fear Uncertainty and Doubt come into play?
I guess its a natural reaction to kill what is alien to you but to throw away PowerPC hardware is a sin. Its just a better CPU than any Intel I have ever seen. I mean if you had that many G4's you could have built a killer cluster out of it and done some Vector processing on each node with the extra sexy AltiVec stuff that comes on those CPU's.
:)
At this point in my life I view the G4 as an equal to the x86 and the only reason x86 is equal is the range of software geared towards it. I personally find x86 to be "on the way out". Its design has been recycled so many times [16 bit - 64bit, new opcodes no one uses anywhere.] All in the name of backwards compatibility that your OS's don't provide.
If I had access to a G4 I certainly wouldn't throw it away. I would just get OS 10.1 when its out.
But that's just my $.02
I never thought much of that game... oh well. Who cares.... Boo hoo.
Dave
Actually its not just VLIW. Its EPIC.
Explicitly Parallel Instruction Code. Its in its
own class. It may indeed be VLIW like but its called EPIC!
Yes the compiler optimization is the key on such an architecture. You basically don't want to ever write assembler on an Itanium if you can avoid it. [unless you are the compiler writer I suppose]
Its not everything the HURD wanted to be.
Check the HURD's architecture and you will know why.
The HURD has really interesting goals in mind... Multi-server OS is the key design keeping your comment from being correct.
I don't think AtheOS is multiserver.
I don't understand this reasoning. I also think its fishy that Microsoft would help a GPL'd project after all the crap they just said about the GPL. They may want to turn the "cancer" against us somehow.
.NET in mind and not use any other system that they may be able to use the servers they control to hold information hostage somehow. True, the tools are open source but will the data pathways be open or private?
.NET to squash open source its not a very sound plan at all. It will still make .NET more feasible an option if both Unix and Windows can talk to each other through .NET services and I really believe that is the only goal they have with Mono.... to increase credibility of usefulness.
... I think it is especially hard for those who worship a license like the GPL to ever trust a company that is usually targeted as the enemy. Truly I stopped using linux because I got tired of all the zealots persecuting closed source vendors. I use FreeBSD now because I like the environment [and the kernel source :)] a lot better.
I suppose they believe that if they can get every Linux/BSD user to write applications with
If that is Microsoft's goal to wiping the credibility of Open Sourced systems out they forgot that one thing Open Source people are wonderful at is duplication of effort just for the hell of it. It means next to nothing for another group to spend their free time coming up with an alternative with similar or the same functionality. I mean GNOME started because of licensing issues with KDE. Before GNOME there was the Harmony project which was an attempt to replace all the functionality of QT with a drop in GPL'd replacement. Anyone heard of Lesstif? The free Motif clone? Mesa?
If Microsoft plans to use
I know its hard to look at this stuff objectively
Its got a 50% chance of being correct though and
involves coin flipping.
There was actually a court case where a bank made a calculation error with a spreadsheet package [Quatro Pro?]. Anyway they said the results weren't guaranteed. The bank lost the court case and lots of money. This is commonplace in software.
I work for a company that writes middleware for clusters and I get to work on them every day of the week.
I just wouldn't use ethernet to connect such a large machine. TCP/IP for message passing is by far not the best choice for making a collective of machines talk to each other. Also PVM is not the only choice anymore. Try MPI.
http://www-unix.mcs.anl.gov/mpi/