Wave Audio
In the PC, everything is represented and processed digitally as 0s and 1s. Similarly, sound is stored as wave data in a file. DACs (digital to analogue converters) are used to convert digital (sound) data into analog waves which we can hear.
Similarly, to record sound (which is analog in nature) it has to be converted to digital before it can be stored in the computer. ADCs (analogue to digital converters) perform this task.
On the PC, you can store sound as wave files with a WAV extension. There are many other formats like AU and VOC. Compact discs store sound digitally, therefore they will also need a DAC chip in the CD player to process and reproduce the sound.
On the SBLive!, all wave audio is handled by the EMU10K1. It handles sample sizes of 4-bit to 16-bit, and sampling rates of 5kHz to 48kHz. Higher sample sizes and sampling rates will result in higher-quality audio that covers a wider range of frequencies. Consequently, these wave files will take up more disk space.
Multi-Channel Wave Audio
The majority of ISA sound cards, including the older Sound Blaster cards and even the Sound Blaster AWE series, are able to play only one wave sound at a time. If another wave sound needs to be played, Windows will not allow it as the card will only support one channel of wave audio.
Games are worst affected by this limitation, because of the constant need to play many sound effects at the same time. One single wave channel cards, the CPU has to mix all the various sound effects and music that are produced by the game, before sending the sound to the DAC on the sound card. This takes up precious CPU time which may slow down the speed of the game.
Thank goodness those days are gone! With more advanced sound chipsets and the higher-bandwidth of the PCI bus, multi-channel wave audio can now be handled in hardware without wasting CPU time.
The SBLive! supports hardware mixing of wave audio. You can start many programs at the same time, and have all of them playing wave audio through the SBLive!. You can have a wave file, MP3 file, AVI video and other software all playing together. (Although it will sound like a mess. <g>) But don't think that many channels of wave audio is useless - your instant messenger like ICQ can still beep you while you listen to MP3s.
Most of the games in the Windows platform use Microsoft's DirectSound API to play sound effects. DirectSound will automatically use hardware mixing when it is supported by the sound card. Otherwise, it will default to CPU/software mixing.
Creative's specfications state that the SBLive! has 131 hardware channels.
Full-Duplex
For most tasks, you do not need to record sound while playing back a wave file simultaneously. The need for full-duplex arose when Internet phones became popular. Normal fixed-line telephones are full-duplex - you can talk into the phone while listening to the other person on the other side at the same time.
Older sound cards could only either record, or play sound, but not do both at the same time. So with an Internet phone, you can only speak or listen at any one time, making phone calls very unnatural and slow - very different from the normal telephone calls that we are all used to making.
Due to market demand, Creative and other sound card manufacturers did provide full-duplex drivers that emulated full-duplex capability through software, although the sound cards themselves do not support the capability in hardware. This took up some amount of CPU time.
Newer sound cards like the SBLive! supports full duplex in hardware. As a result, it does not make as much use of CPU time compared to a software-emulated full-duplex implementation. In the mixer, you can select a sound source to record from, and still be able to use many wave channels with the built in multi-channel wave audio.
Creative states that the SBLive! is able to do full-duplex at 8 standard sample rates commonly used in wave audio.
Digital Processing
All sound that goes through the SBLive! will be processed digitally. This includes the CD audio, MIDI, wave files, microphone, line-in, digital input and other connections that are used. To do so, analog input sources like the microphone, CD audio and line-in will be converted to digital data with an ADC in the EMU10K1.
Why does it need to be in digital? So that the data can pass through algorithms programmed into the EMU10K1 to perform 3D positioning, mixing, adding of EAX and other stuff. Even the bass, treble and volume controls in the mixer is done digitally for cleaner sound. Chances of inducing hiss, distortions, and loss of sound quality are also reduced.
Locking at 48kHz
The downside of processing digitally is that the EMU10K1 needs to convert everything to 48kHz before it is able to use the sound data for processing. This is partly because of the fact that there are so many different sampling rates (ranging from 4kHz to 48kHz) that is supported.
The cards needs to work on all the multiple channels of sound coming in at the same time, and performing mixing, adding of effects, and a myraid of other tasks - all at the same time. Standardizing the sampling rate will make the chipset and the programming of the drivers less complex.
Musicians, content developers and others who need a direct 1-to-1 transfer of digital data at 16-bit 44.1kHz (CD quality audio) have voiced their disappointment at the loss of resolution and sound detail when converting to 48kHz for internal processing.
The majority of users out there who just want to play games in 3D and some MP3s need not bother about this issue, unless you need to do some sound editing and transferring between digital media.
Is there any way to change it? It seems like the EMU10K1 is hardwired this way, and drivers will not be able to change that.
The specifications state that the card runs at internal 32-bit precision at 48kHz. It means that the EMU10K1 is able to process sound data in 32-bit (4 byte) chunks (for mixing, adding of effects, and other tasks). It does not mean that the card can handle 32-bit wave files.
|
Back To Top