.i = 0x5f3759df - ( t.i >> 1 ); // what the fuck?
This code apparently works for all float numbers (except for the negative ones
of course that cannot be computed) to good accuracy. There's a paper written
about this here:
http://www.lomont.org/Math/Papers
le, as the max. relative
error is 0.175% over all floats.
--
Thilo Schulz
signature.asc
Description: This is a digitally signed message part.
___
ioquake3 mailing list
ioquake3@lists.ioquake.org
http://lists.ioquake.org/listinfo.cgi/ioquake3-ioquake.org
B
On Sunday, 18. July 2010 19:02:10 Michael Menegakis wrote:
> Add: not there on an old binary without sdl.
The sound mixer can only play a given number of sounds in parallel. If this
number is exceeded, sounds will be dropped. This is normal behaviour.
--
Thilo Schulz
signature.
rn quake3 actually has a level with a moving train with you playing on
the train. However, technically the train is static, only the surrounding
canyon is moving and you die as soon as you leave the train.
--
Thilo Schulz
signature.asc
Description: This is a digitally s
ings in multiples of 50ms.
But yes, it definitely is possible.
--
Thilo Schulz
signature.asc
Description: This is a digitally signed message part.
___
ioquake3 mailing list
ioquake3@lists.ioquake.org
http://lists.ioquake.org/listinfo.cgi/ioquake
ke
movie encoding. This here is not the application for it.
--
Thilo Schulz
signature.asc
Description: This is a digitally signed message part.
___
ioquake3 mailing list
ioquake3@lists.ioquake.org
http://lists.ioquake.org/listinfo.cgi/ioquake3-ioquake.org
B
u to learn about it. This is the key where it waits for data on
the socket, and where you have have to set the timeout correctly in order to
send the delayed packets at the correct times.
--
Thilo Schulz
signature.asc
Description: This is a digitally signed message part.
___
h has a too low resolution.
Well, I made it always use the select() syscall instead of NET_Sleep. You can
see the patch here:
http://thilo.tjps.eu/download/netsleep.patch
Please test it and tell me what you think. I did some rough testing and
everything seems to work fine.
--
Thilo Schulz
signature.
On Sunday, 30. January 2011 03:51:02 Thilo Schulz wrote:
> Please test it and tell me what you think. I did some rough testing and
> everything seems to work fine.
Well, I just tested playing on a remote server and there's still a problem
obviously. I'll give an updated version
)
--
Thilo Schulz
signature.asc
Description: This is a digitally signed message part.
___
ioquake3 mailing list
ioquake3@lists.ioquake.org
http://lists.ioquake.org/listinfo.cgi/ioquake3-ioquake.org
By sending this message I agree to love ioquake3 and libsdl.
triangles for the snapshot indicator (upper line, should be blue).
What's new: Compare the load with normal idq3 and the new ioquake3 in the task
manager. I think you'll notice the difference.
--
Thilo Schulz
signature.asc
Description: This is a d
Don't worry, I'll have a look at it and apply it if it doesn't break too much.
After all, zakk wanted it. That VBO stuff is interesting as well.
--
Thilo Schulz
signature.asc
Description: This is a digitally signed message part.
___
i
inux kernels will exhibit this problem. There is an easy
workaround: set cvar com_busywait 1 and it should revert back to the old
behaviour.
Once you've updated your system I expect this workaround will not be necessary
anymore.
--
Thilo Schulz
signature.asc
Description: This is a
y outdated:
Indeed not.
> Note that all that fancy new GEM, KMS and so forth mumbo jumbo is
> not available on FreeBSD. So it might be that the video drivers are
> somewhat dated, since the Xorg project has stopped caring about
> portability.
They do not play a role in this.
--
Hz setting for the scheduler. This allows sub-
millisecond precision in the timeout, which is ideal for what we need.
--
Thilo Schulz
signature.asc
Description: This is a digitally signed message part.
___
ioquake3 mailing list
ioquake3@lists.ioquake.
On Sunday, 20. February 2011 06:57:26 M. Kristall wrote:
> Thilo meant your kernel Hz. You should be able to hit 125fps with
> kern.hz = 1000, which looks like the default since FreeBSD 6.
Yes.
--
Thilo Schulz
signature.asc
Description: This is a digitally signed messag
p://www.www0.org/w/Optimized_executable;_builds_of_ioq3_engine_for_urt
Is your solution portable, and how many change does it require?
--
Thilo Schulz
signature.asc
Description: This is a digitally signed message part.
___
ioquake3 mailing list
ioquake3@list
t any problems to me. This will go into trunk once it's been tested.
I tested with these mods:
baseq3
missionpack
World of Padman
OpenArena
Western Quake3 (old mod)
They all seemed to work fine.
--
Thilo Schulz
signature.asc
Description: This is a digitally signed me
kpro ioquake3-2]$
Can you please test which SVN version introduced the bug?
--
Thilo Schulz
signature.asc
Description: This is a digitally signed message part.
___
ioquake3 mailing list
ioquake3@lists.ioquake.org
http://lists.ioquake.org/listinfo.cg
Basically, most of the x86 assembler code is reusable, only memory
operations need to be handled specially.
This also addresses the problem with the long loading time for the xm_86_64 VM
that we've had thus far.
--
Thilo Schulz
signature.asc
Description: This is
On Wednesday, 22. June 2011 19:51:23 Nerius Landys wrote:
> Any idea why this would happen consistently after a couple of weeks of
> running? What code is responsible?
Use GDB and watchpoints to determine the code that causes this problem.
--
Thilo Schulz
signature.asc
Description: Thi
h appreciate it.
Yes, however, I tried to make the calls into and out of VMs as independent
from calling conventions as possible, at least in vm_x86.c.
vm_x86_64.c only supports the cdecl calling conventions, which makes it more
difficult to port this to other platforms or compilers than gcc.
ve it?
--
Thilo Schulz
signature.asc
Description: This is a digitally signed message part.
___
ioquake3 mailing list
ioquake3@lists.ioquake.org
http://lists.ioquake.org/listinfo.cgi/ioquake3-ioquake.org
By sending this message I agree to love ioquake3 and libsdl.
alues back to memory
> ret
> qsnapvectorsse ENDP
>
> (assuming the global round mode is round-to-nearest ofc).
Hmm, yeah. I guess that would work too. Did you do performance comparisons via
things like profiling?
--
Thilo Schulz
signature.asc
Description: This is a digital
to its
protocol version, that's why we made the protocol version configurable at
startup.
On a sidenote, the cvar naming is settled now, com_protocol and
com_legacyprotocol will be the way to go from now on and I don't think it's
gonna change anymore.
--
Thilo Schulz
signat
orts, and I don't need to clear the patches. Thanks zakk and
congratulations to Zack!
--
Thilo Schulz
signature.asc
Description: This is a digitally signed message part.
___
ioquake3 mailing list
ioquake3@lists.ioquake.org
http://lists.ioquake.
doxygen guidelines
to begin with.
A patch would leave no piece of code untouched, this is a nightmare from the
point of view of someone who regularly merges ioquake3 changes to his project.
I say don't bother.
--
Thilo Schulz
signature.asc
Description
least
shell access with a working debugger environment on it. Can you provide this
for me? I'd be more than happy to try to resolve this.
--
Thilo Schulz
signature.asc
Description: This is a digitally signed message part.
___
ioquake3 mailing
equest these tools be installed though:
- My favorite editor joe
- gdb
- cgdb, an ncurses frontend to gdb which makes my life a bit easier
Thank you very much, Ryan!
--
Thilo Schulz
signature.asc
Description: This is a digitally signed message part.
___
st"
> float to int conversion, but I'd prefer to keep the default rounding mode
> for all the other operations.
That's the problem. There are no guarantees on what is standard, and what is
not. At all.
--
Thilo Schulz
#include
#include
main()
{
float test1
On Wednesday, 21. September 2011 21:12:36 Victor Roemer wrote:
> Anyways, is there any progress in this, anymore info that I can provide.
No. I need ssh shell access with a development and debugging environment.
--
Thilo Schulz
signature.asc
Description: This is a digitally signed mess
On Wednesday, 21. September 2011 21:12:36 Victor Roemer wrote:
> Sorry guys, I jacked up my email rules.
>
> Anyways, is there any progress in this, anymore info that I can provide.
The bug has been fixed in latest SVN
--
Thilo Schulz
signature.asc
Description: This is a digital
27;s
not unfixable though. Try commenting the line SDL_config_minimal.h:38 as a
quick and dirty workaround.
As a note, I haven't really managed to test ioquake3 on ppc64 Linux as that is
quite an exotic architecture. But you should be able to make it work.
--
Thilo Schulz
signa
file named "Makefile.local" and put the following
line into it:
USE_VOIP=0
Or alternatively use this:
USE_INTERNAL_SPEEX=0
but you'll need to install libspeex from your distribution in this case then.
--
Thilo Schulz
signature.asc
Description: This is a dig
2.so.0.7.0 s none AARDsdl
Obviously, /opt/lib is not included in the paths that are searched by your
linker. Solaris must surely have a way to configure this.
--
Thilo Schulz
signature.asc
Description: This is a digitally signed message part.
ver. I'm afraid this
is not nearly enough information that would point us to the actual error, I'm
sorry.
--
Thilo Schulz
signature.asc
Description: This is a digitally signed message part.
___
ioquake3 mailing list
ioquake3@lists.ioquak
e mieux à Thimothee de moi.
--
Thilo Schulz
signature.asc
Description: This is a digitally signed message part.
___
ioquake3 mailing list
ioquake3@lists.ioquake.org
http://lists.ioquake.org/listinfo.cgi/ioquake3-ioquake.org
By sending this message I
the change.
I have to agree with Ludwig. This is bad practice.
--
Thilo Schulz
signature.asc
Description: This is a digitally signed message part.
___
ioquake3 mailing list
ioquake3@lists.ioquake.org
http://lists.ioquake.org/listinfo.cgi/ioquake3-ioquake.o
oquake3 already has very low-level modifications, to the virtual machine, to
the way events are processed compared to id's original vanilla quake3 source
code.
--
Thilo Schulz
signature.asc
Description: This is a digitally signed message part.
_
If it happens during gameplay, the server is simply overtaxed.
--
Thilo Schulz
signature.asc
Description: This is a digitally signed message part.
___
ioquake3 mailing list
ioquake3@lists.ioquake.org
http://lists.ioquake.org/listinfo.cgi/ioquake3-ioquake.
sion to write to this directory.
> > What is the best strategy to correct this behavior ?
>
> The problem is that sudo does not update the HOME environment variable
> by default. Run sudo with the -H flag and you should be fine.
Still, ioq3 shouldn't segfault in
ion.
I would look at the "screen" tool. It provides virtual terminals that you can
detach and reattach at any time.
http://en.wikipedia.org/wiki/GNU_Screen
--
Thilo Schulz
signature.asc
Description: This is a digitally signed message part.
___
42 matches
Mail list logo
