From
Www.palmsource.com
Most of the newly emerging next-generation palms have their own portable sound equipment. Among them, the enhanced audio has now become a standard configuration, which is also a potential for the potential purchaser for the handheld computer waiting for a long time. This chapter discusses how to create, use and actually reproduce digital audio, and how to purchase devices based on audio hardware. Sound theory should understand how the sound is now on the handheld computer, it is necessary to explain how the digital sound wave is created and perceived. In the real world of non-numbers, the sound is spread like a pressure wave, which is a real-time simulation. The pressure waves cause our tympanic membrane to vibrate (and transmit energy to them). The energy value transmitted per microsecond is not a fixed certain value, but any one in a continuous value. Moreover, the resulting pressure is constantly changing over time, and the speed of the change is almost unlimited. When the sound in the real world is converted to the digital world, it will be limited, and these restrictions do not exist in the real world. In the computer, the sound wave is converted into a digital stream, which is called sample. The energy of the acoustic wave is photographed with a rapid aliquot of the rule, called "sample", and then stores the obtained energy value in the data stream in the form of a number, and the sampling sound is approximated by the above method to approximate their original sound. When playing back, the number in the saving data stream is used to reproduce the shape of the original sound wave. The reproduction quality depends on the accuracy of the simulation of the original sound waves. This basically depends on two factors: the dynamic range and frequency response of sound sampling. The dynamic range of digital sound waves depend on the range of value of each sample point. For example, an 8-bit number can store 256 values, namely 0 to 255 (or -128 to 127). When sampling acoustic waves, the energy of the sound is mapped into numbers, and the maximum volume of the volume is the maximum. Therefore, the 8-bit sound sampling value is between -128 and 127, and 0 means mute. If the dynamic range is too small, the shape of the original sound wave cannot be accurately reproduced, which can only hear noise when playback. The human ear is difficult to detect the volume change of 1/255, this value is exactly the 8-bit precision. And 16 (two times space, resulting in 65536 values) is the standard of high quality audio. The largest portion of the sound wave is mapped into a full range of sampling values, and the part of the volume is small, and the number of digits of the data cannot be taken out, which is different. For example, if you want to sample 1812 prelude, it may be adjusted to take a larger part of the volume from -128 to 127. These parts will take up all 8 bits and may be very sound quality. However, the small volume is not so satisfied. If their volume is only 1/16 of the maximum volume in the entire music, then their value range is -7 to 8, only 4 digits in the entire 8 bits. However, if the entire waveform is re-sampled in 16 bits, even the most quiet part will reach 12-bit precision, the sound quality of the entire sound is better. The frequency response second factor, the frequency response refers to the speed of the sample, that is, the sampling rate, the unit is sampled per second, that is, Hz (Hz). When the sound wave is quickly transformed, the sampling rate must be improved to reflect the shape of the original sound wave. Fast transform sound waves not only express high-frequency audio symbols like bottles, but also express drums, cymbals and other flamesters. If the sampling rate is not enough to reproduce the original sound wave, these sounds sound like the original sound compared to the original sound waves, or less than the original sound "crisp". Sampling sound can be expressed in the highest frequency of its simple transformation between two different values. Every two sampling, these two values will be repeated once, so a maximum frequency of digital sound waves can be generated, called the Nyquist frequency, which is half of the sample rate.
The smallest sound frequency of the person can be perceived is 20000 Hz, so the smallest sampling rate of high fidelity sound is approximately 40,000 Hz. The minimum sample rate of the ordinary sound quality is approximately 8000 Hz. Digital devices in sound formats use different sound quality and size digital sound format. The audio CD is stored for 16-bit 44,000 Hz stereo, the sound quality is very good, and it does not feel the difference between it and the original sound. However, just like a SUV in the parking garage, the audio of the CD sound quality occupies a large amount of storage space, and 10 megabytes are required per minute. When the sound quality is not important, it is often necessary to reduce the dynamic range and / or sampling rate to reduce the file size. It is generally considered that 8,000 Hz sounds have been enough for voice and non-music applications. Compression technology can also be used to reduce file size. However, just like video, the audio is also difficult to compress. To effectively compress, you always need to modify the original sound data. This compression is called "lossless compression", which can cause sound distortion, which typically generates background noise in the final output. For example, Tealmovie's ADPCM compression algorithm can reduce the size of the sound file, but it will hear obvious noise in sound samples. Advanced compression technology, such as MP3, with powerful computing power and clever algorithms, can make the size of the sound file to one-tenth. Although MP3 is still "lossy" compression, it is difficult to hear any changes to most of the sound. Its side effect is to play back MP3 files require extremely powerful computing power; the results have led to MP3, but in fact, it is actually necessary for dedicated hardware support to do this. So you need to know ... The audio hardware is surprisingly, the first Palm Powered handheld has supported the sampling sound. This feature is from the Dragonball processor of Motorola. Its built-in hardware can play back 8-bit 10,000 Hz sampling sound. However, in order to play the sound outside the horn, the speaker, and the alarm, additional filtering hardware devices is required. Although adding new hardware As long as a small fee, because the audio characteristics are not most important, they don't cost it. Today, all Dragonball-based Palm Powered Pockets still have only basic built-in sound support. Some handheld computers, such as TRGPRO, Handera 330, and Sony Clie, which have dedicated filters, so they can play the security sound. Many SONY Pockets also provide headphone jacks - a good multimedia function application feature in the device. Most ARM-based handheld also supports high quality audio. Both SONY NX70 and PALM TUNGSTEN T offer headphone jacks and support sampling sound playback. However, there is a certain important difference, Tungsten T can play back different formats, and uncompressed sampling sounds of various sample rates to reach the CD level sound quality. Although NX70 can play high-quality MP3 files saved on memory sticks, it can only recognize 8-bit, 8000 Hz compression format, so that the sound quality of the game and application is limited. The sound data of the game and application may be temporary, or the data stored in memory. However, Sony NX70, NR70 and its similar models have other special features, such as supporting sound synthesis. These devices cannot play voice. They use very little hardware, synthetic sounds with a similar clock and a sound wave of the sound. Sound synthesis support can not only provide music alarms to the handheld, but also generate music. It sounds like a feathers such as feather key consoles.
The Beat Plus Springboard module adds the same audio support like Sony, while other SpringBoard modules and Kyocera 7135 Smartphone provide MP3 playback support. In most cases, MP3 playback requires support for dedicated hardware chips. These hardware read data directly from the SD card or memory stick. In essence, this is similar to the independent MP3 player, just being very loosely integrated into the handheld computer. This will maximize battery life. But we will soon discover that this practice limits the software program to use them. Software Support When choosing a new device, the most important consideration is not just its sound performance, and the third-party software can use those characteristics. What is hardware is generally not a problem, it is important for developers, which are mostly available. Developers can easily get the required documentation from Motorola, and add sound support to Dragonball (OS 4 and its previous versions) by documentation. Almost all of this series of equipment supports basic audio of various sound quality. Importantly, the audio buffer for the previous Pilot, Palmpilot and Palm III's Dragonball processor is too small, and other operations cannot be performed while playing the sample sound, such as playing video. Starting from Palmiiix, the later model not only improved the screen that was not clear enough, but also added Dragonball-EZ, -VZ and -Supervz chip. These provide support for audio buffering and more important TealMovie audio. However, even Dragonball, support for software sounds often also depends on documents provided to developers. For example, TRGPRO and Handera 330 are equipped with an amplifier controlled by software, which must be explicitly opened by the program before playing the sample sound. Fortunately, Handera (former TRG) provided this information for developers before released. However, some Sony's audio support often is not so easy. For example, when the NR70 or NX70 plays the sampling sound, at least three soft via controls are manipulated, but only one provides public documentation. Palmos 5 promises to provide new sampling sound playback support to reduce these issues. Improved sound support has been added to Palmos 5.1, but in Palm OS on Tungsten T, Palm has revoked these improvements. Moreover, Sony NX70 does not support this new sound library. In addition, different from SONY NR70 is that NX no longer uses Dragonball hardware. Thus, the playback of the sampling sound is greatly limited. To add support, you need a special document, and these documents are hard to get. This greatly reduces support for third-party software. What is the playback of MP3? The playback of the MP3 file is indeed a good feature that can give the handheld computer users with a better experience. Unfortunately, most MP3 playback supports custom hardware chips, so the functionality of software applications is usually limited to the MP3 files on the start and stop playback of the SD or memory sticks. The general procedure cannot directly access the audio hardware, so high-quality MP3 playback function does not represent games, movies, and other applications to play high-quality sounds, which need to access sound data faster from main memory. Summarizing the purchase device, if the audio is important consideration, then what is the most useful audio application is. game? the film? music? The selected device is equipped with headphone jacks, with a good sampling sound playback function. If you want to play the MP3 file, you must confirm that the battery life indicated by the manufacturer is enough.
