Computer audio is one of the most overlooked aspects of a computer purchase. With little information from the manufacturers, users have a hard time figuring out exactly what it is they are getting. In the first segment of this series of articles on Understanding Computer Audio, we look at the basics of digital audio and the specifications may be listed. In addition, we will look at a couple of the standards that are used to describe the components.
All audio that is recorded or played through a computer system is digital, but all audio that is played out of a speaker system is analog. The difference between these two forms of recording play an important role in determining the ability of sound processors.
Analog audio uses a variable scale of information to try and best reproduce the original sound waves from the source. This can produce a very accurate recording, but these recording degrade between connections and generations of recordings. Digital recording takes samples of the sound waves and records it as a series of bits (ones and zeros) that best approximate the wave pattern. This means that the quality of the digital recording will vary based on the bits and samples used for the recording, but the quality loss is much lower between equipment and recording generations.
Bits and Samples
When looking at sound processors and even digital recordings, the terms of bits and KHz will often come up. These two terms refer to the sample rate and audio definition that a digital recording can have. There are three primary standards used for commercial digital audio: 16-bit 44KHz for CD Audio, 16-bit 96KHz for DVD and 24-bit 192KHz for DVD-Audio and some Blu-ray.
The sample rate refers to the number of bits used in the recording to determine the amplitude of the sound wave at each sample. Thus, a 16-bit bit-rate would allow for a range of 65,536 levels while a 24-bit allows for 16.7 million. The sample rate determines the number of points along the sound wave that are sampled over a period of one second. The greater the number of samples, the closer the digital representation will be to the analog sound wave.
It is important to note here that the sample rate is different than a bitrate. Bitrate refers to the overall amount of data processed in the file per second. This is essentially, the numbers of bits multipled by the sample rate then converted to bytes on a per channel basis. Mathmetically, that is (bits * sample rate * channels) / 8. So, CD-audio which is stereo or two channel would be:
(16 bits * 44000 per second * 2) / 8 = 192000 bps per channel or 192kbps bitrate
With this general understanding, what exactly should one look for when examining the specifications for an audio processor? In general, it is best to look for one capable of at 16-bit 96KHz sample rates. This is the level of audio used for the 5.1 surround sound channels on DVD and Blu-ray movies. For those looking for the best audio definition, the new 24-bit 192KHz solutions offer greater audio quality.
Another aspect of audio components that users will come across is a Signal-to-Noise Ratio (SNR). This is a number represented by decibels (dB) to describe the ratio of an audio signal compared to the noise levels generated by the audio component. The higher the Signal-To-Noise ratio, the better the sound quality is. The average person generally cannot distinguish this noise if the SNR is greater than 90dB.
There are a variety of different standards when it comes to audio. Originally, there was the AC'97 audio standard developed by Intel as a means of standardized support for 16-bit 96KHz audio support for six channels necessary for DVD 5.1 audio sound support. Since then, there have been new advances in audio thanks to the high definition video formats such as Blu-ray. In order to support these, a new Intel HDA standard was developed. This expands the audio support for up to eight channels of 30-bit 192KHz necessary for 7.1 audio support. Now, this is the standard for Intel based hardware but most AMD hardware that is labelled as 7.1 audio support can also achieve these same levels.
Another older standard that might be referred to is 16-bit Sound Blaster compatible. Sound Blaster is a brand of audio cards created by Creative Labs. The Sound Blaster 16 was one of the first major sound cards to support the 16-bit 44KHz sampling rate for CD-Audio quality computer audio. This standard is below that of the newer standard and is rarely referenced anymore.
EAX or Environmental Audio Extensions is another standard that was developed by Creative Labs. Instead of a specific format for audio, it is a set of software extensions that modify audio to replicate the effects of specific environments. For example, the audio being played in a computer could be designed to sound as if it was being played in a cave with lots of echo. Support for this can exist in either software or hardware. If rendered in hardware, it uses fewer cycles from the CPU. There are currently five versions: EAX, EAX 2.0, EAX 3.0, EAX 4.0 and EAX 5.0 (or EAX HD). Creative made up to EAX 3.0 open so many sound cards can support some audio extensions up to that level but teh remaining versions are exclusive to Creative Labs hardware.
The situation with EAX got more complicated with Vista and Windows 7. Essentially, Microsoft shifted much of the audio support from the hardware to the software side in order to have greater level of security on the system. This means that many games that handled EAX audio in hardware was now being handled by the software layers instead. Much of this has been dealt with by software patches to the drivers and games but there are some older games that won't be able to use the EAX effects anymore.
Finally, some products may carry the THX logo. This is essentially a certification that THX Laboratories feels that the product meets or exceeds their minimum specifications. Just remember that a THX certified product will not necessarily have better performance or sound quality than one that does not. The manufacturers have to pay THX labs for the certification process.