Servers??? I said where I live not where I work. Having a nat server cause problems with online gaming. Msm also got problems sending files due to the nat.
What are you talking about? There are many browsers which does this. Have you newer used lynx, or some of the browsers made for blind people which talk insted of displaying anything?. Infact there don't exists ANY browser which display the site that the server send, because the server does NOT send any site. It send a html document which is a description of the CONTENT of the page not a description of HOW to render it. Just try to visit msn.com internet explore, and then with a wap phone. This you will get an entire different view of the same information.
Re:IPv6 - Chicken and egg ? - no!
on
Slashdot over IPv6
·
· Score: 5, Informative
This is not true, there are real problems getting ip address from Ripe The result is that where I live we got 500 Computers behind a single nat gateway because we can't get an ip to each use. The result is a lousy network.
Well acording to Ars tehhnica it can only do single-precision floating-point which is 32 bit and thus useless for any rendering which is going to be used with movies. They require atleast 64bit.
This program would not cause any memory allocation, it would just kill itself as soon as it started. Not a problem for windows at all. To be really fun you need something that both fork and allocate memory:}
I don't KNOW it but I really think that pixar use 64 or 80 bit FPU calculations to renderer movies and Altivec can't really handle that. Besides Render man is a BIG application, and adding Altivec support for Renderman would cost more then the 1024 cpu clusster they bought to render movies.
>Sun complaining about the JRE support model for internal projects...THEY ARE THEY JRE SUPPORT MODEL. It would be a bit like Ford recommending people don't use Ford parts for internal work because they'd have to go to Ford to get support for them. Eh?
No, it would be like Ford recommending not to drive a ford car to work because they are to unstable to ensure that people come in time.
>Listing off the memory footprint of various "demo" applications. The "Hello World" reference gives this away as totally bogus. Anyone who's used Java knows about its memory consumption.
Yes but their point is that the solaris version of jre is far worse then the windows version. The solaris version of jre require twice the memory of the windows version to run the "Hello world" program. And the fact that I know about the memory usage of java does NOT make it less of a problem.
>Eclipse provides one of the best, cleanest, well designed Java IDEs out there, and starts up into around 25M on my system
Eclipse might take 25MB to startup, but try to use it for a few hours and se you memory usage go to atleast 100MB. But Eclipse is worth the memory so I just bought an other 256MB ram and are a happy Eclipse user:-}
But Eclipse just confirm the point of the memo. After testing awt and swing, they(The developers of Eclipse) came to the same conclusion as this memo: That awt and swing are unuseable for big projects. I really think sun should think about replacing awt and swing with swt.:-}
>Backward compatibility across minor releases. Everyone familiar with Java knows that 1.2, 1.3, 1.4, are as far from "minor releases" as they could possibly be. There's absolutely nothing "minor" about them.
Well the Sun release manager don't know about. Remember a version number is Major.Minor.Revision so if 1.4 is a major to 1.3 it should have been called 2.0 AND there should have been an easy way to ensure that an application could pick between jre 1.3 and 2.0
I am developing an application(Primary target windows and Mac) which currently run fine in jre 1.3 but got problems running with jre 1.4 due to changes in Swing. The best solution would be just to keep the jdk 1.3 alongside 1.4 but that's not really posible with windows, so now I have to change the app to work with 1.4 and then include(and install) jre1.4 and then hope that our customers don't have any applications that only work with 1.3. Oh what I would pay just to be able to include a version of swing with our application allowing us to run with both 1.3 and 1.4 using the same swing library no matter if the user were using 1.3 or 1.4.
It is correct that JSSE has been integrated into the JavaTM 2 SDK, Standard Edition, v 1.4 but with 1.3 and early versions it was an external module, so the author might just not have noticed they included it with 1.4. Which remind me: Does anyone know why jsse use >5 seconds on a 800Mhz pentium just to create an ssl connection to a webserver? This is really a big problem for us -((
I don't know if this memo is valid, but it really does sum up the problems with using java for client side applications.
Why lowpolygon? My ATI Readon 9700 should be able to draw > 10000 polygons widtout slowing down. Remember it don't even have to calculate light source.
I don't think so. Remember they talk about the Itanium II type chip, not a Pentium. And Itanium could really need a speedup in the fight against sparc and mips
Where does this nonsence with only 15 fps/second come from? I for one do NOT have any problem with seeing the difference between 25 and 50 frames. I can't be the only one who think that movies look jerky.
And an other importent thing to remember is that the framerate really SHOULD be equal to(or at least a multiplum of) the refresh rate. For exampel running 35 fps on a 80 Hz display is NOT a good thing, because some frames will be displayed longer then other frames resulting in something really ugly.
As an analogy: There are people who can't tell the difference between a 128bit Mp3 and a cd, but THAT does not mean that Mp3 is cd quality audio. And to make that extream there are people who complain that cd quality is not good enough because they can hear the difference between a cd(16 bit, 44 Khz) and a 24bit 96Khz sample.
How would you address 16GB ram with only 32 bits and no segments? Besides segments always start at 0(virtual address) which allow for some optimization. (Well nobody does that kind of optimizing anymore, and the only os which allow you to controll the segments is OS/2) but still:}
I would not call WinNT/2K/XP micro kernels. They run the entire hd system AND gui system in kernel space. And MacOS X run the hd system in kernel space to, so i am not sure that would count as a micro kernel either. I would say that the best example of a sucessfull microkernel is qnx neutrino(Qnx 6.2) (get.qnx.com if you want to test it)
They do it to get the scrollbar for the text, so the user can click the "I Agree" button widtout having to scroll. Stupid maybe, but atleast they do have a reason.
It is true for floating point, due to the really ugly x86 floating point system, but all the test I have made using integer operations were won by the Athlon.
>Trivial code, that isn't CPU intensive anyway, >however, will run faster on a higher clock rate >CPU, but it isn't doing much work anyway.
???? What kind of non cpu intensive task will run faster on a cpu with a higher clock?? That don't make any sense. Either it is cpu/memory bound in which case the fastes cpu system will finish or it is io bound(Harddisk,modem,network) in which case the clock of the cpu don't matter.
>-- Basically, everyone who thinks Macs are >slower and more expensive than PCs is either in >denial or lying
Or they are using software that don't utilize altivec. (Yes I did some benchmark on software I use, and the 1GHZ G4 lost to my Athlon XP2000). Still looking forward to that 970 however:}
But I (Or more preciesly the dtp users I admin for) want more ram. Give them a motherboard that can take 12GB ram, and they will max it out, and then complain that there is not really enough ram.
The problem is that usb don't have any support of interrupt, so it might be needed by some drivers to a busy wait which will take 100% cpu time nomatter how fast your cpu is. Ofcause if this *#$&*$& PC hardware had some better timer chips, busy waiting would newer be needed by a driver.
Slashdot Mirror
Servers??? I said where I live not where I work. Having a nat server cause problems with online gaming. Msm also got problems sending files due to the nat.
What are you talking about? There are many browsers which does this. Have you newer used lynx, or some of the browsers made for blind people which talk insted of displaying anything?. Infact there don't exists ANY browser which display the site that the server send, because the server does NOT send any site. It send a html document which is a description of the CONTENT of the page not a description of HOW to render it. Just try to visit msn.com internet explore, and then with a wap phone. This you will get an entire different view of the same information.
This is not true, there are real problems getting ip address from Ripe The result is that where I live we got 500 Computers behind a single nat gateway because we can't get an ip to each use. The result is a lousy network.
You should have hold the training in the weekend, unless he specified WHAT you should be traning.
Just remember to tell the company they need to pay for traning facilities
Then you could have a nice: "Free as in beer, a traning session about free beer vs opensource software"
Well acording to Ars tehhnica it can only do single-precision floating-point which is 32 bit and thus useless for any rendering which is going to be used with movies. They require atleast 64bit.
Martin Tilsted
This program would not cause any memory allocation, it would just kill itself as soon as it started. Not a problem for windows at all. :}
To be really fun you need something that both fork and allocate memory
A good idear until they upgrade from Renderman 11.0 to 11.1 and make some minor changes to the render code :}
I don't KNOW it but I really think that pixar use 64 or 80 bit FPU calculations to renderer movies and Altivec can't really handle that. Besides Render man is a BIG application, and adding Altivec support for Renderman would cost more then the 1024 cpu clusster they bought to render movies.
Martin Tilsted
>Sun complaining about the JRE support model for internal projects...THEY ARE THEY JRE SUPPORT MODEL. It would be a bit like Ford recommending people don't use Ford parts for internal work because they'd have to go to Ford to get support for them. Eh?
:-}
:-}
No, it would be like Ford recommending not to drive a ford car to work because they are to unstable to ensure that people come in time.
>Listing off the memory footprint of various "demo" applications. The "Hello World" reference gives this away as totally bogus. Anyone who's used Java knows about its memory consumption.
Yes but their point is that the solaris version of jre is far worse then the windows version. The solaris version of jre require twice the memory of the windows version to run the "Hello world" program. And the fact that I know about the memory usage of java does NOT make it less of a problem.
>Eclipse provides one of the best, cleanest, well designed Java IDEs out there, and starts up into around 25M on my system
Eclipse might take 25MB to startup, but try to use it for a few hours and se you memory usage go to atleast 100MB. But Eclipse is worth the memory so I just bought an other 256MB ram and are a happy Eclipse user
But Eclipse just confirm the point of the memo.
After testing awt and swing, they(The developers of Eclipse) came to the same conclusion as this memo: That awt and swing are unuseable for big projects. I really think sun should think about replacing awt and swing with swt.
>Backward compatibility across minor releases. Everyone familiar with Java knows that 1.2, 1.3, 1.4, are as far from "minor releases" as they could possibly be. There's absolutely nothing "minor" about them.
Well the Sun release manager don't know about.
Remember a version number is Major.Minor.Revision so if
1.4 is a major to 1.3 it should have been called 2.0 AND there should have been an easy way to ensure that an application could pick between jre 1.3 and 2.0
I am developing an application(Primary target windows and Mac) which currently run fine in jre 1.3 but got problems running with jre 1.4 due to changes in Swing. The best solution would be just to keep the jdk 1.3 alongside 1.4 but that's not really posible with windows, so now I have to change the app to work with 1.4 and then include(and install) jre1.4 and then hope that our customers don't have any applications that only work with 1.3. Oh what I would pay just to be able to include a version of swing with our application allowing us to run with both 1.3 and 1.4 using the same swing library no matter if the user were using 1.3 or 1.4.
It is correct that JSSE has been integrated into the JavaTM 2 SDK, Standard Edition, v 1.4 but with 1.3 and early versions it was an external module, so the author might just not have noticed they included it with 1.4. Which remind me: Does anyone know why jsse use >5 seconds on a 800Mhz pentium just to create an ssl connection to a webserver? This is really a big problem for us -((
I don't know if this memo is valid, but it really does sum up the problems with using java for client side applications.
Why lowpolygon? My ATI Readon 9700 should be able to draw > 10000 polygons widtout slowing down. Remember it don't even have to calculate light source.
But I paid for my RedHat CDs
But you don't have any right to use the software for the Radon 9700 if you did not buy one.
I don't think so. Remember they talk about the Itanium II type chip, not a Pentium. And Itanium could really need a speedup in the fight against sparc and mips
Martin
Where does this nonsence with only 15 fps/second come from?
I for one do NOT have any problem with seeing the difference between 25 and 50 frames. I can't be the only one who think that movies look jerky.
And an other importent thing to remember is that the framerate really SHOULD be equal to(or at least a multiplum of) the refresh rate. For exampel running 35 fps on a 80 Hz display is NOT a good thing, because some frames will be displayed longer then other frames resulting in something really ugly.
As an analogy:
There are people who can't tell the difference between a 128bit Mp3 and a cd, but THAT does not mean that Mp3 is cd quality audio. And to make that extream there are people who complain that cd quality is not good enough because they can hear the difference between a cd(16 bit, 44 Khz) and a 24bit 96Khz sample.
Martin Tilsted
How would you address 16GB ram with only 32 bits and no segments? Besides segments always start at 0(virtual address) which allow for some optimization. (Well nobody does that kind of optimizing anymore, and the only os which allow you to controll the segments is OS/2) but still :}
Martin Tilsted
>Laws like this do nothing but raise costs for >consumers
That's the point. That the consumer do not only pay for the goods, but also for the needed safe disposel of the goods. What's wrong with that?
Martin Tilsted
Why not just code in c++ and use the string class insted of making your own in C? It would not be any slower, and it would be just as safe.
Martin
Considering that neither C or C++ have any thread support, it don't require much for java to be better there.
I would not call WinNT/2K/XP micro kernels. They run the entire hd system AND gui system in kernel space. And MacOS X run the hd system in kernel space to, so
i am not sure that would count as a micro kernel either. I would say that the best example of a sucessfull microkernel is qnx neutrino(Qnx 6.2) (get.qnx.com if you want to test it)
They do it to get the scrollbar for the text, so the user can click the "I Agree" button widtout having to scroll. Stupid maybe, but atleast they do have a reason.
Martin Tilsted
It is true for floating point, due to the really ugly x86 floating point system, but all the test I have made using integer operations were won by the Athlon.
>Trivial code, that isn't CPU intensive anyway, >however, will run faster on a higher clock rate
>CPU, but it isn't doing much work anyway.
???? What kind of non cpu intensive task will run faster on a cpu with a higher clock?? That don't make any sense. Either it is cpu/memory bound in which case the fastes cpu system will finish or it is io bound(Harddisk,modem,network) in which case the clock of the cpu don't matter.
>-- Basically, everyone who thinks Macs are >slower and more expensive than PCs is either in >denial or lying
:}
Or they are using software that don't utilize altivec. (Yes I did some benchmark on software I use, and the 1GHZ G4 lost to my Athlon XP2000). Still looking forward to that 970 however
Do you think that it will be THAT much delayed?
Don't look god for amd then
Martin
But I (Or more preciesly the dtp users I admin for) want more ram. Give them a motherboard that can take 12GB ram, and they will max it out, and then complain that there is not really enough ram.
Martin
The problem is that usb don't have any support of interrupt, so it might be needed by some drivers to a busy wait which will take 100% cpu time nomatter how fast your cpu is. Ofcause if this *#$&*$& PC hardware had some better timer chips, busy waiting would newer be needed by a driver.
Martin Tilsted