Ah, here is one who has seen the REAL problem with the popular garbage collectors. This is not a problem inherit to garbage collectors: Try looking up "conservative garbage collectors". I have only seen this for C++, though.
A more fundamental problem is that memory is only one of many resources a typical industrial program must manage. GC takes over memory management, but leaves the other scarce resources -- file descriptors, sockets, mutexes, database connections -- to be managed manually, as in C. (Java has this problem, for instance.) "Finalization" simply cannot provide the necessary guarantees.
As I remember the predictions, it was something along temperature rising of about 0.1 to 0.2K at this point. Not exactly killing increases.
In any case, the global warming effect prediction seems to be measurable now. Unfortunately, my source is in Danish, but since the "scientist" (Lomborg?) who starting this "it's all ok, nothing to worry about" fad of today is Danish as well, I'll go ahead and post it.
http://www.dmi.dk/dmi/index/viden/drivhuseffekten. htm
I do not believe that the human race will be killed off by this. But it might be expensive/inconvenient to live with.
The only opensource (GPL) UML modeller I'm aware off is Umbrello, from the KDE suite. But it needs some more work before it is really useful, IMHO. But do check it out; I use it for class diagrams, and have submitted 2 patches so far. If everyone does that... what am I saying?
[...] large, long-term projects, since that's where lack of documentation, etc., will really bite you in the ass [..]
I have been programing for a few years by now. Never have "documentation" figured largely in those projects. And the documentation that does get written is never actually read in my experience, except by the few who could easily do without.
Of course, that was in a non-XP environment, and thus doesn't really count. My methology says: write the documentation in the code. The rest will be obsolete by the time anyone wants to read it. In any case, people will complain that the documentation was missing. (Whether the documentation was actually written and announced is immaterial).
I am, however, a great fan of documented procedures. Those save a lot of pain.
The RAM is like a window into the data on the swap file. So, at some point, the area the RAM can see is so small that swapping will become excessive, and thus make the computer unworkable.
The 2xRAM is mostly an experience based rule for the limit to when adding more swap space no longer helps things. Note that the rule is quite old, and may need updating (though I still use it: 512Mb RAM->1Gb swap partition.).
It is also an old hardcoded requirement for early linux, I believe, though that is of mostly historical interest.
where:1 is whatever window/virtual console you like to have your game on. If you don't want a window manager at all, you should probably use:0. If you run a window manager, use the easy-to-remember-but-hard-to-hit alt-ctrl-F7 and alt-ctrl-f8 to switch between game/desktop.
I usually run my games besides my windows manager this way. One of the things that IS better under Linux... unlike alt-tab it's actually stable.
As I remember the predictions, it was something along temperature rising of about 0.1 to 0.2K at this point. Not exactly killing increases. In any case, the global warming effect prediction seems to be measurable now. Unfortunately, my source is in Danish, but since the "scientist" (Lomborg?) who starting this "it's all ok, nothing to worry about" fad of today is Danish as well, I'll go ahead and post it. http://www.dmi.dk/dmi/index/viden/drivhuseffekten. htm
I do not believe that the human race will be killed off by this. But it might be expensive/inconvenient to live with.
The only opensource (GPL) UML modeller I'm aware off is Umbrello, from the KDE suite. But it needs some more work before it is really useful, IMHO. But do check it out; I use it for class diagrams, and have submitted 2 patches so far. If everyone does that... what am I saying?
I have been programing for a few years by now. Never have "documentation" figured largely in those projects. And the documentation that does get written is never actually read in my experience, except by the few who could easily do without.
Of course, that was in a non-XP environment, and thus doesn't really count. My methology says: write the documentation in the code. The rest will be obsolete by the time anyone wants to read it. In any case, people will complain that the documentation was missing. (Whether the documentation was actually written and announced is immaterial).
I am, however, a great fan of documented procedures. Those save a lot of pain.
The RAM is like a window into the data on the swap file. So, at some point, the area the RAM can see is so small that swapping will become excessive, and thus make the computer unworkable. The 2xRAM is mostly an experience based rule for the limit to when adding more swap space no longer helps things. Note that the rule is quite old, and may need updating (though I still use it: 512Mb RAM->1Gb swap partition.). It is also an old hardcoded requirement for early linux, I believe, though that is of mostly historical interest.
I'd suggest something like
/usr/bin/games/ut2004 -- :1
:1 is whatever window/virtual console you like to have your game on. If you don't want a window manager at all, you should probably use :0. If you run a window manager, use the easy-to-remember-but-hard-to-hit alt-ctrl-F7 and alt-ctrl-f8 to switch between game/desktop.
startx
where
I usually run my games besides my windows manager this way. One of the things that IS better under Linux... unlike alt-tab it's actually stable.