You know these kids are just going to buy their chips on the street now. Who knows what those are cut with? Very likely inferior conductors.
We will see a lot of kids getting hopped up on modchips and most likely addicted, only to suffer painful side effects like electrocution and intestine blockage due to the poor production standards of street fabs.
I've been developing software that runs on Solaris 64 bit systems for many years. Of course I am glad that 64 bit systems will be more affordable, but 64 bit microprocessors have been around for a decade and I just don't get excited by the Apple and Intel technology hype.
No, I said it does not apply to the G4. I didn't make a generalization. Sorry if it sounded that way. The Power4 is totally different processor than the G4, just to underscore that point.
I couldn't really read all that (attention deficit disorder). Another advantage to a 64 bit architecture is the possibility to run in a single address space, which is more efficient. See Intel's Itanium 2 software developer manual Vol 1 for a brief overview of SAS vs MAS.
Of course Pentium 4s totally smoke G4s. A Pentium 4 2.53 is three times faster than a G4 867 according to specint2000/specfp2000. Amazingly, that three times performance difference is paired with a three times megahertz difference.
System specint2000/specfp2000
AMD Athlon XP 2100+ 1733 720/613
Intel Pentium III 1133 461/320
Intel Pentium 4 2533 882/861
Intel Itanium 800 314/645
Intel Itanium 2 1000 ~700/~1350
IBM POWER4 1300 804/1202
Sun UltraSPARC III 1050 537/701
Apple G4 867 257/154
What were you saying about the megahertz myth? It doesn't apply to the slow G4?
Gentoo sounded interesting so I went and read up a little.
Here's the answer to my question: "Our new kernel is based upon 2.4.19 and includes Robert Love's preemptive kernel and scheduler-hints patches, Ingo Molnar's O(1) scheduler and smptimers patches, and Rik van Riel's reverse mapping patches, for enhanced responsiveness and scalability under high loads and in high-end multiprocessor systems."
Time Sharing environments are fine for servers and multi-user machines, but they're a real waste on a single user desktop machine.
Says you. What about a corporate environment where you don't want to give users full access to mess things up? On NT/Win2k/XP, you don't give them Administrator and on Unix you don't give them root. "Users" allow you to isolate processes, group processes, and give different permissions to those processes. It's a nice system even if there's only one human user on a machine.
When you're working on code, you have to recompile a project repeatedly,
You should not need to recompile the entire project repeatedly during development, only source which has changed and dependencies on that source. And yes I've seen dumb developers that recompile the entire project unnecessarily.
Oh and.. please let me know what languages these are that are removing pointers so that I may refrain from using such a limiting language.
Pointers that can point to arbitrary memory are completely unnecessary (except for certain hardware I/O), dangerous for stability and correctness, and the cause of a lot of security problems. C and C++ let you arbitrarily cast pointers which is dangerous. To me casting is a symptom of a design flaw (Java references included).
Also, it's easier for a compiler to optimize bounded array accesses so
there's no performance benefit using pointers.
You're assuming that the goal is to implement patented ideas without licensing. I see the point of patenting to protect nonobvious intellectual property created by an inventor under the assumption that it will lead towards more inventions.
I'm not a big fan of patents, but I'm arguing from the capitalist point of view.
Many patent ideas are novel and not at all obvious results of a fixed amount of research. Even an expert may never come up with these ideas in a lifetime. In other words, the value of a patent doesn't correlate with the time it took an inventor to develop the idea. You can't assume that you could plug in another expert like a lightbulb and have invented TV or relativity or whatever.
Slashdot Mirror
-Kevin
We will see a lot of kids getting hopped up on modchips and most likely addicted, only to suffer painful side effects like electrocution and intestine blockage due to the poor production standards of street fabs.
-Kevin
the switch "backlash" was old before it started. ads are lies. get over it
Ah, cool. I was wondering if there were any common single address space OSes.
To Mr. Weenie Flamebait Moderator: the truth hurts, doesn't it?
-Kevin
-Kevin
-Kevin
It's not just the processor - that system is far superior to the Mac you'll buy next year.
-Kevin
Yeah, Apple's real cutting edge, circa two years too late.
-Kevin
-Kevin
Of course Pentium 4s totally smoke G4s. A Pentium 4 2.53 is three times faster than a G4 867 according to specint2000/specfp2000. Amazingly, that three times performance difference is paired with a three times megahertz difference.
System specint2000/specfp2000
AMD Athlon XP 2100+ 1733
720/613
Intel Pentium III 1133
461/320
Intel Pentium 4 2533
882/861
Intel Itanium 800
314/645
Intel Itanium 2 1000
~700/~1350
IBM POWER4 1300
804/1202
Sun UltraSPARC III 1050
537/701
Apple G4 867
257/154
What were you saying about the megahertz myth? It doesn't apply to the slow G4?
-Kevin
The G5 vaporchip is supposed to be essentially a 64 bit G4.
-Kevin
I'd rather have a Power4 (which is available now of course) than wait a year for a crappy stripped Power4.
-Kevin
Here's the answer to my question:
"Our new kernel is based upon 2.4.19 and includes Robert Love's preemptive kernel and scheduler-hints patches, Ingo Molnar's O(1) scheduler and smptimers patches, and Rik van Riel's reverse mapping patches, for enhanced responsiveness and scalability under high loads and in high-end multiprocessor systems."
O'Reilly Gentoo article
I may give Gentoo a whirl. Mandrake is getting stale.
-Kevin
-Kevin
Says you. What about a corporate environment where you don't want to give users full access to mess things up? On NT/Win2k/XP, you don't give them Administrator and on Unix you don't give them root. "Users" allow you to isolate processes, group processes, and give different permissions to those processes. It's a nice system even if there's only one human user on a machine.
-Kevin
-Kevin
You should not need to recompile the entire project repeatedly during development, only source which has changed and dependencies on that source. And yes I've seen dumb developers that recompile the entire project unnecessarily.
-Kevin
I've seen makefile Java builds that invoke javac for each .java file. That's very inefficient.
Try using ant if you're not already.
-Kevin
-Kevin
Pointers that can point to arbitrary memory are completely unnecessary (except for certain hardware I/O), dangerous for stability and correctness, and the cause of a lot of security problems. C and C++ let you arbitrarily cast pointers which is dangerous. To me casting is a symptom of a design flaw (Java references included).
Also, it's easier for a compiler to optimize bounded array accesses so there's no performance benefit using pointers.
-Kevin
-Kevin
I'm not a big fan of patents, but I'm arguing from the capitalist point of view.
-Kevin
Many patent ideas are novel and not at all obvious results of a fixed amount of research. Even an expert may never come up with these ideas in a lifetime. In other words, the value of a patent doesn't correlate with the time it took an inventor to develop the idea. You can't assume that you could plug in another expert like a lightbulb and have invented TV or relativity or whatever.
-Kevin