camsr

I use the onboard realtek/ac97/generic crap, it's fine as long as you use the optical output :music_whistling: But I'm not trying to disagree :)

Just updated to 1.2.8, using fresh configs except for keybinds. Game seems stuck when "Connecting to Server..." Does it take longer now or a problem here? Can someone help?

http://forums.eagle.ru/showthread.php?t=117803 Those are my rough binds for using keyboard only with A10. IF you want to use the stick also, and you are a righty, switch the necessary modifiers to use the left section of the keyboard.

I use my mobo's built on soundchip only for SPDIF optical output to my DAC. It sounds so transparent to my ears :) Then the DAC goes into active studio monitors via a balanced XLR connection. If there is any noise, I can't hear it, but it might have a small effect ;) What's good about optical SPDIF is there is no chance of computer ground noise getting into the monitor. That is basically impossible.

Ambisonics looks cool, never heard it before. But it's just like a glorified surround sound with the same problems. Headphones actually do a better job with postional audio. But the perception of the sound from Ambisonics is more realistic.

Guess what? If the flight plan attempts to tell the AI to land at a high speed, or at an angle, they won't land. You must set the speed for the landing approach very low, like 130, and it must be straight also. They won't land with brains!

The only thing I discovered with my own FPS test "mission" was that Structures lag the game considerably. Turning off structures completely got me to 30fps in most situations.

Where does terminal velocity appear?

When the smoke detector goes off and you launch countermeasures and duck for cover

AMDs are fine for concurrent computing (aka multi-threading), maybe not the fastest, but they put all the cores to work. In that type of super-computing deal, price is a bigger selling point usually. TDP and single thread performance are superior in most Intel chips, two very important factors. My q9450 is OCed to 3.2 with ddr2 800, it's always stable too. This one was a real diamond in the rough of Intel offerings, large L2 cache, no L3 cache (good because less latency that way), and easily clockable to q9650 speed. You didn't even have to buy a q9650 because the q9450 just bumps up so easily... it's kinda funny. Newer Intel offerings don't even double it's performance yet, so stuff it, Moore's law :thumbup:

Bass Shaker, that's just subs under the seat right? You may want to try a bandpass box. I've never built one for sitting on though.

Add Chinook to the list also, maybe someday

Real time calculation of complex boundary conditions is not yet feasible. I gotta admit, maybe I bit off more then I can chew, because I am uncertain it would even be an augment to the gaming experience. Thomas N. gives a good synopsis of the importance of sound reflection in his music studios. http://prorecordingworkshop.lefora.com/reply/25934260/FrontToBack-Acoustic-Technique#reply-25934260

If you do make any more, please do share them :) Youtube will destroy audio mercilessly. It's not only Joint Stereo Compression that does it, but the codec itself. WAV or FLAC would be a better choice. I don't know how to encode videos with lossless audio, so you may have your work cut out for you.

Red mode sounds decently better. What's the difference? Some critisism of your videos.... Disable music in the game shots. Don't turn around constantly like a one-footed drunk. Point and wait a bit.

The problem with reverberation being dynamically generated in a video game is a tradeoff between accuracy and computation. The reverberant characteristics of a 6 sided box are relatively straight forward, but add an interface to another 6 sided box (like a doorway between rooms), or a large profiled surface, and the problem gets deeper. There are also material considerations like what the room is constructed of, and honestly, even level designers don't usually know :lol: they aren't being paid to architect real-world-functional environments, they are paid to make game-world-functioning ones. So what I am trying to get at is positional audio is good, but dry at the same time and in many ways unnatural, being anechoic.

It could be that in-ear type headphones are required for the best experience. Sound bypassing the pinna this way should make the not-so-accurate HRTF more bearable.

Yes I read the links and the OpenAL manual. It's very interesting, the whole idea of positional audio, even if I think it is fundamentally flawed by lack of proper acoustical reverberation. When I listened to the Rapture3d video, it was off by a mile. Not the actual positioning, but the frequency balance was all wrong (to me). I guess that is the pinnae "filters" tuned wrong for my ears. If I wanted to spend an hour or so listening to the pre-compiled set of HRTFs at the OpenAL website, I could probably find something close. The flaw here is their notes on each HRTF test lack the proper data to choose by numbers, and use the power of observation.

From the HRTF wiki: "The perceived accuracy of the result depends on how closely the HRTF data set matches the characteristics of one's own ears." IOW, one size HRTF does not fit all, absolutely. Does OpenAL's program make any considerations for this?

But the reverberation of the in-game world does require more computing time. It's very simple to pan a sound to one of two speakers in stereo, and say the sound is coming from it. Everything that happens in between the two speakers is an illusion (http://en.wikipedia.org/wiki/Phantom_center) You are right to say headphones can provide a "true 3d" soundfield (what is that defined as to you, btw?). Headphones do not have a phantom center, and when a sound is panned to one side, only one ear will hear it, and this is not a normal hearing sensation unless something is very close to the ear. The "true 3d" you are talking about only mimics the HRTF (http://en.wikipedia.org/wiki/Head-related_transfer_function) for an anechoic environment (http://en.wikipedia.org/wiki/Anechoic_chamber), what the ears will hear from a sound some distance with no outside acoustical influences. This does provide directional cues to locate audial events, but that's it. Any type of reverberation added after this should be able to simulate the environment that the viewer is seeing. Reverberation can make locational cues less accurate. Simply simulating the reverberant conditions of one room is not walking around in a world full of sound.

SOI TGP, TMS down, doesn't seem to lose track when I do it ;)

A game itself has to be encoded for multi-channel information. 5.1 or 7.1 is actually pretty good, if the game implements it's advantages fully. But the sad truth of the matter is you will find that it's very far from most developers objectives to have a "true 3d" soundfield, not at the very least because it would require much computing overhead. It's not only angle of incidence that gives audial spacial awareness cues, but the reverberant environment of the "world" itself.

Memory leak?

http://technet.microsoft.com/en-us/sysinternals/cc835722.aspx CoreInfo.exe will tell you all about your CPU. If you have a CPU like mine (Core 2 Quad), you will notice this little goodie: Logical to Physical Processor Map: *--- Physical Processor 0 -*-- Physical Processor 1 --*- Physical Processor 2 ---* Physical Processor 3 Logical Processor to Socket Map: **** Socket 0 Logical Processor to NUMA Node Map: **** NUMA Node 0 Logical Processor to Cache Map: *--- Data Cache 0, Level 1, 32 KB, Assoc 8, LineSize 64 *--- Instruction Cache 0, Level 1, 32 KB, Assoc 8, LineSize 64 -*-- Data Cache 1, Level 1, 32 KB, Assoc 8, LineSize 64 -*-- Instruction Cache 1, Level 1, 32 KB, Assoc 8, LineSize 64 **-- Unified Cache 0, Level 2, 6 MB, Assoc 24, LineSize 64 --*- Data Cache 2, Level 1, 32 KB, Assoc 8, LineSize 64 --*- Instruction Cache 2, Level 1, 32 KB, Assoc 8, LineSize 64 ---* Data Cache 3, Level 1, 32 KB, Assoc 8, LineSize 64 ---* Instruction Cache 3, Level 1, 32 KB, Assoc 8, LineSize 64 --** Unified Cache 1, Level 2, 6 MB, Assoc 24, LineSize 64 Two processors share one top level cache. Four processors total. The Core 2 Quad is actually TWO Core 2 Duo stitched together over the Front Side Bus. In a situation like this, it is best to set affinities for one application to one unified cache. Otherwise the penalty in multi threaded applications is like a cache miss AND a read from RAM every time the called thread is located on the other unified cache. The windows scheduler may or may not know about this, so you can guarantee a shared cache by setting affinities.

Hans, what they are trying to say is ED has auxillary income from other softwares. I do agree that we don't want to see the game become abandonware.