For direct access use https://forums.oldunreal.com
It's been quite a while since oldunreal had an overhaul, but we are moving to another server which require some updates and changes. The biggest change is the migration of our old reliable YaBB forum to phpBB. This system expects you to login with your username and old password known from YaBB.
If you experience any problems there is also the usual "password forgotten" function. Don't forget to clear your browser cache!
If you have any further concerns feel free to contact me: Smirftsch@oldunreal.com
It's been quite a while since oldunreal had an overhaul, but we are moving to another server which require some updates and changes. The biggest change is the migration of our old reliable YaBB forum to phpBB. This system expects you to login with your username and old password known from YaBB.
If you experience any problems there is also the usual "password forgotten" function. Don't forget to clear your browser cache!
If you have any further concerns feel free to contact me: Smirftsch@oldunreal.com
227i found and fixed bugs
- Smirftsch
- Administrator
- Posts: 8999
- Joined: Wed Apr 29, 1998 10:00 pm
- Location: NaPali
- Contact:
227i found and fixed bugs
"Reconnect" didn't work correctly anymore. At least nothing critical although annoying.
Sometimes you have to lose a fight to win the war.
- .:..:
- OldUnreal Member
- Posts: 1635
- Joined: Tue Aug 16, 2005 4:35 am
Re: 227i found and fixed bugs
- Fixed a bug with FluidSurfaceInfo where it had wrong shading on larger resolution.
- Altered many of Engine classes to use TouchingActors iterator instead of the Touching list array for better results and fewer bugs.
- Altered many of Engine classes to use TouchingActors iterator instead of the Touching list array for better results and fewer bugs.
(ಠ_ಠ)1823223D2A33224B0 wrote:...and now im stuck trying to fix everything you broke for the next 227 release xD
- .:..:
- OldUnreal Member
- Posts: 1635
- Joined: Tue Aug 16, 2005 4:35 am
Re: 227i found and fixed bugs
- Reworked all of the code on ProceduralMesh so now it's fully functional and bugfree.
- Made player footsteps cause wave ripplets on FluidSurfaceInfo as you walk in it.
- Added a new native function to PlayerPawn to get current client version:
- Added a new PlayerInterpolating state to PlayerPawn to be used on 227 clients when entering "PlayerPath" on SpecialEvent, it allows for online smooth interpolation movement.
- Changed so InterpolationPoint.GameSpeedModifier does not modify Level.TimeDilation in a multiplayer game environment.
- Made player footsteps cause wave ripplets on FluidSurfaceInfo as you walk in it.
- Added a new native function to PlayerPawn to get current client version:
Code: Select all
native final function int GetClientVersion();
- Changed so InterpolationPoint.GameSpeedModifier does not modify Level.TimeDilation in a multiplayer game environment.
Last edited by .:..: on Sun Dec 09, 2012 8:45 pm, edited 1 time in total.
(ಠ_ಠ)1823223D2A33224B0 wrote:...and now im stuck trying to fix everything you broke for the next 227 release xD
- Smirftsch
- Administrator
- Posts: 8999
- Joined: Wed Apr 29, 1998 10:00 pm
- Location: NaPali
- Contact:
Re: 227i found and fixed bugs
- changed CloudZone to avoid crashes with monsterspawners not avoiding to spawn monsters there.
- changed Pawn a bit to improve performance with blood effects.
- changed Pawn a bit to improve performance with blood effects.
Sometimes you have to lose a fight to win the war.
- Smirftsch
- Administrator
- Posts: 8999
- Joined: Wed Apr 29, 1998 10:00 pm
- Location: NaPali
- Contact:
Re: 227i found and fixed bugs
fixed Linux segmentation fault when using ScriptedTextures in packages like UTWeapons
Sometimes you have to lose a fight to win the war.
- Smirftsch
- Administrator
- Posts: 8999
- Joined: Wed Apr 29, 1998 10:00 pm
- Location: NaPali
- Contact:
Re: 227i found and fixed bugs
fixed AI can see through movers on dedicated servers
fixed Coronas being blocked by BlockAll (ignores bHidden)
fixed Coronas being blocked by BlockAll (ignores bHidden)
Sometimes you have to lose a fight to win the war.
- Smirftsch
- Administrator
- Posts: 8999
- Joined: Wed Apr 29, 1998 10:00 pm
- Location: NaPali
- Contact:
Re: 227i found and fixed bugs
fixed Meshes on StaticMeshes rendering issue (such as Superhealth packs on StaticMesh ground)
reverted ScriptedPawn script change (function timer) to old version because of unwanted side effects
reverted ScriptedPawn script change (function timer) to old version because of unwanted side effects
Sometimes you have to lose a fight to win the war.
- Hellkeeper
- Global Moderator
- Posts: 3260
- Joined: Wed May 21, 2008 8:24 pm
- Location: France
- Contact:
Re: 227i found and fixed bugs
Ah, finally. This is very nice to read, thanks for your work. Finally, DmRiot will no longer look stupid with its seemingly empty superhealth packfixed Meshes on StaticMeshes rendering issue (such as Superhealth packs on StaticMesh ground)
You must construct additional pylons.
- .:..:
- OldUnreal Member
- Posts: 1635
- Joined: Tue Aug 16, 2005 4:35 am
Re: 227i found and fixed bugs
Some fixes for SwFMod (FModEx) audio driver:
- Fixed ambient sounds to use game volume and not play at max volume at all time.
- Fixed so ambient sound effects with unknown sound source (net games) to stop sound once out of hearing distance (instead of keeping the sound until mapchange).
- Fixed consolecommands 'GetMusicOffset', 'SetMusicOffset' and 'GetMusicLen' to use correct music offset values for modules.
(ಠ_ಠ)1823223D2A33224B0 wrote:...and now im stuck trying to fix everything you broke for the next 227 release xD
- Smirftsch
- Administrator
- Posts: 8999
- Joined: Wed Apr 29, 1998 10:00 pm
- Location: NaPali
- Contact:
Re: 227i found and fixed bugs
fixed realcrouch could cause falling through movers
fixed spawnnotify not editable in properties window
fixed spawnnotify not editable in properties window
Sometimes you have to lose a fight to win the war.
- Smirftsch
- Administrator
- Posts: 8999
- Joined: Wed Apr 29, 1998 10:00 pm
- Location: NaPali
- Contact:
Re: 227i found and fixed bugs
fixed Linux: relaunch command
updated codebase to work with GCC4.7.2
fixed some small memory heap problem in Linux build
changed optimization to Core2
this resulted in up to + ~30% FPS gain on my testsystem
updated codebase to work with GCC4.7.2
fixed some small memory heap problem in Linux build
changed optimization to Core2
this resulted in up to + ~30% FPS gain on my testsystem
Last edited by Smirftsch on Fri Mar 29, 2013 4:04 pm, edited 1 time in total.
Sometimes you have to lose a fight to win the war.
- GreatEmerald
- OldUnreal Member
- Posts: 5347
- Joined: Mon May 21, 2007 2:30 pm
Re: 227i found and fixed bugs
Wow. Although it makes sense when most things are run on the CPU anyway. Now I wonder what you'd get with march=native...
- Smirftsch
- Administrator
- Posts: 8999
- Joined: Wed Apr 29, 1998 10:00 pm
- Location: NaPali
- Contact:
Re: 227i found and fixed bugs
well, I guess most gain is due to new compiler, not only the optimization. But I see no point to keep anymore Pentium3, since even Core2 is a really old system already today.
Sometimes you have to lose a fight to win the war.
- Smirftsch
- Administrator
- Posts: 8999
- Joined: Wed Apr 29, 1998 10:00 pm
- Location: NaPali
- Contact:
Re: 227i found and fixed bugs
[split] [link=http://www.oldunreal.com/cgi-bin/yabb2/ ... 389637/0#0][splithere][/link][splithere_end]
Sometimes you have to lose a fight to win the war.
- Hyper
- OldUnreal Member
- Posts: 3515
- Joined: Fri Oct 11, 2002 5:41 pm
- Contact:
Re: 227i found and fixed bugs
Nice to hear about the CPU optimization gains. Are older CPU's still supported now? If not, what are the minimum CPU requirements?well, I guess most gain is due to new compiler, not only the optimization. But I see no point to keep anymore Pentium3, since even Core2 is a really old system already today.
Oh, and my "Really old" Core 2 Quad Extreme QX9650 3.0 GHz still performs very nicely in todays environment, so please do not offend him.
Last edited by Hyper on Sun Apr 21, 2013 3:19 pm, edited 1 time in total.
- Smirftsch
- Administrator
- Posts: 8999
- Joined: Wed Apr 29, 1998 10:00 pm
- Location: NaPali
- Contact:
Re: 227i found and fixed bugs
I guess I should have written this a bit more clearly
These optimizations are mostly for Linux client and server performance. Windows doesn't benefit much from it, although still trying to improve performance a bit yet.
Core2 is the oldest CPU supported then in Linux, but maybe I can go back to a bit older CPU's yet now that the rest of the optimizations seem to work flawlessly.
These optimizations are mostly for Linux client and server performance. Windows doesn't benefit much from it, although still trying to improve performance a bit yet.
Core2 is the oldest CPU supported then in Linux, but maybe I can go back to a bit older CPU's yet now that the rest of the optimizations seem to work flawlessly.
Last edited by Smirftsch on Sun Apr 21, 2013 5:14 pm, edited 1 time in total.
Sometimes you have to lose a fight to win the war.
- Smirftsch
- Administrator
- Posts: 8999
- Joined: Wed Apr 29, 1998 10:00 pm
- Location: NaPali
- Contact:
Re: 227i found and fixed bugs
[split] [link=http://www.oldunreal.com/cgi-bin/yabb2/ ... 124211/0#0][splithere][/link][splithere_end]
Sometimes you have to lose a fight to win the war.
- Smirftsch
- Administrator
- Posts: 8999
- Joined: Wed Apr 29, 1998 10:00 pm
- Location: NaPali
- Contact:
Re: 227i found and fixed bugs
not exactly a bug, but:
cleaned up OpenAL to work better with OpenALSoft in Linux, changed some bad coding, improved EFX error handling. Maybe will fix some of the problems in windows people experienced with non CreativeLabs cards as well.
cleaned up OpenAL to work better with OpenALSoft in Linux, changed some bad coding, improved EFX error handling. Maybe will fix some of the problems in windows people experienced with non CreativeLabs cards as well.
Sometimes you have to lose a fight to win the war.
- Smirftsch
- Administrator
- Posts: 8999
- Joined: Wed Apr 29, 1998 10:00 pm
- Location: NaPali
- Contact:
Re: 227i found and fixed bugs
fixed- or rather updated CPU detection code for Windows.
instead
Init: CPU Speed=2527.041901 MHz
Init: CPU Page size=4096, Processors / Cores=4
Init: CPU Detected: PentiumPro-class processor (GenuineIntel)
Init: CPU Features: CMov FPU RDTSC PAE MMX KNI
it provides now more useful data like this:
Init: CPU Speed=2527.080895 MHz
Init: CPU Page size=4096, Processors / Cores=4
Init: CPU Detected: Intel(R) Core(TM) i5 CPU M 460 @ 2.53GHz
Init: CPU Features: MMX SSE SSE2 SSE3 SSSE3 SSE41 SSE42
or on AMD:
Init: CPU Speed=3221.559491 MHz
Init: CPU Page size=4096, Processors / Cores=4
Init: CPU Detected: AMD Phenom(tm) II X4 955 Processor
Init: CPU Features: MMX SSE SSE2 SSE3 SSE4A A3DNOW MMXPLUS A3DNOWEXT
also updated internals:
[code]
extern CORE_API UBOOL GIsMMX;
extern CORE_API UBOOL GIsSSE;
extern CORE_API UBOOL GIsSSE2;
extern CORE_API UBOOL GIsSSE3;
extern CORE_API UBOOL GIsSSSE3;
extern CORE_API UBOOL GIsSSE41;
extern CORE_API UBOOL GIsSSE42;
extern CORE_API UBOOL GIsSSE5;
extern CORE_API UBOOL GIsSSE4A;
extern CORE_API UBOOL GIsA3DNOW;
extern CORE_API UBOOL GIsMMXPLUS;
extern CORE_API UBOOL GIsA3DNOWEXT;[/code]
for C++ mods and maybe internal updates of something in 227.
Like in previous versions, each CPU feature can be disabled via commandline during startup with "NO....", f.e. "NOMMX".
instead
Init: CPU Speed=2527.041901 MHz
Init: CPU Page size=4096, Processors / Cores=4
Init: CPU Detected: PentiumPro-class processor (GenuineIntel)
Init: CPU Features: CMov FPU RDTSC PAE MMX KNI
it provides now more useful data like this:
Init: CPU Speed=2527.080895 MHz
Init: CPU Page size=4096, Processors / Cores=4
Init: CPU Detected: Intel(R) Core(TM) i5 CPU M 460 @ 2.53GHz
Init: CPU Features: MMX SSE SSE2 SSE3 SSSE3 SSE41 SSE42
or on AMD:
Init: CPU Speed=3221.559491 MHz
Init: CPU Page size=4096, Processors / Cores=4
Init: CPU Detected: AMD Phenom(tm) II X4 955 Processor
Init: CPU Features: MMX SSE SSE2 SSE3 SSE4A A3DNOW MMXPLUS A3DNOWEXT
also updated internals:
[code]
extern CORE_API UBOOL GIsMMX;
extern CORE_API UBOOL GIsSSE;
extern CORE_API UBOOL GIsSSE2;
extern CORE_API UBOOL GIsSSE3;
extern CORE_API UBOOL GIsSSSE3;
extern CORE_API UBOOL GIsSSE41;
extern CORE_API UBOOL GIsSSE42;
extern CORE_API UBOOL GIsSSE5;
extern CORE_API UBOOL GIsSSE4A;
extern CORE_API UBOOL GIsA3DNOW;
extern CORE_API UBOOL GIsMMXPLUS;
extern CORE_API UBOOL GIsA3DNOWEXT;[/code]
for C++ mods and maybe internal updates of something in 227.
Like in previous versions, each CPU feature can be disabled via commandline during startup with "NO....", f.e. "NOMMX".
Last edited by Smirftsch on Thu May 23, 2013 3:00 pm, edited 1 time in total.
Sometimes you have to lose a fight to win the war.
- Smirftsch
- Administrator
- Posts: 8999
- Joined: Wed Apr 29, 1998 10:00 pm
- Location: NaPali
- Contact:
Re: 227i found and fixed bugs
updated Linux code the same way.
Sometimes you have to lose a fight to win the war.
- Hyper
- OldUnreal Member
- Posts: 3515
- Joined: Fri Oct 11, 2002 5:41 pm
- Contact:
Re: 227i found and fixed bugs
Does Unreal 227 actually support every mentioned instruction set?
- Smirftsch
- Administrator
- Posts: 8999
- Joined: Wed Apr 29, 1998 10:00 pm
- Location: NaPali
- Contact:
Re: 227i found and fixed bugs
no
There was never much use of these, indeed there is some A3DNow code spread in here and there, although the implementation looks unfinished, it's not usable. Probably they gave up on it as an instruction set very powerful but not available on any CPU.
Currently MMX is used by Galaxy and SSE/SSE2 is used by OpenGL/D3D9. That's it already pretty much.
Everything else has to follow, if there is someone willing to make use of it.
There was never much use of these, indeed there is some A3DNow code spread in here and there, although the implementation looks unfinished, it's not usable. Probably they gave up on it as an instruction set very powerful but not available on any CPU.
Currently MMX is used by Galaxy and SSE/SSE2 is used by OpenGL/D3D9. That's it already pretty much.
Everything else has to follow, if there is someone willing to make use of it.
Last edited by Smirftsch on Fri May 24, 2013 6:22 am, edited 1 time in total.
Sometimes you have to lose a fight to win the war.
- Smirftsch
- Administrator
- Posts: 8999
- Joined: Wed Apr 29, 1998 10:00 pm
- Location: NaPali
- Contact:
Re: 227i found and fixed bugs
[split] [link=http://www.oldunreal.com/cgi-bin/yabb2/ ... 126288/0#0][splithere][/link][splithere_end]
Sometimes you have to lose a fight to win the war.
- Smirftsch
- Administrator
- Posts: 8999
- Joined: Wed Apr 29, 1998 10:00 pm
- Location: NaPali
- Contact:
Re: 227i found and fixed bugs
Added yet another catch for execPlaySound / execClientHearSound crash issues.
fixed bullet holes dont work on flat surfaces (not sure if mentioned somewhere already)
fixed blood drip rotating wrong when standing still
yet some more small things I can't remember atm. Don't worry, there is still progress.
fixed bullet holes dont work on flat surfaces (not sure if mentioned somewhere already)
fixed blood drip rotating wrong when standing still
yet some more small things I can't remember atm. Don't worry, there is still progress.
Sometimes you have to lose a fight to win the war.
- Smirftsch
- Administrator
- Posts: 8999
- Joined: Wed Apr 29, 1998 10:00 pm
- Location: NaPali
- Contact:
Re: 227i found and fixed bugs
oh yes, disappearing Slith carcass bug fixed (slith carcass in corrosive zones, such as slimezone).
Last edited by Smirftsch on Sun Dec 01, 2013 4:03 pm, edited 1 time in total.
Sometimes you have to lose a fight to win the war.