| |
I'll be brief:
ASIC or Application Specific Integrated Circuit is a chip that is custom or semi custom designed for a specific application. Whereas electronic systems can be designed using standard components, usually they include many uneeded functions and work arounds or compromizes that add cost and undesireable "side effects", such as excess power and board space consumption, to the overall design. If a manufacturer has a solid design that will be used in large enough quantity to off-set the non-recurring custom engineering and manufacturing costs, then an ASIC can provide a compact, cost saving alternative to the use of standard off-the shelf components. You find ASICs in things like camcorders, many consumer electronics items, cellular telephones, and many other higher volume products.
A FPGA or Field Programmable Gate Array is a logic or mixed signal (logic and DSP functions) device that, as the name implies, is programmable. This has much of the benefits of an ASIC but usually has some contraints on performance charateristics and "gate counts" or the number of individual logic circuits that can be implemented on the part. A gate is a combination of transistor and other component cells that make up the basic logic elements. Theses are organized into bolean logic expressions such and "AND", "OR" logic functions - if one or two signals coming into a gate is "high or on in an "OR" gate, tehn the output is high as well. You can combine the basic logic into complex computing functions. There is a lot more to know about the way this works, but that's not very important to investing. The disadvantage of a FPGA is that they are relatively high cost and consume more power and have longer signal propagation delays than most ASIC solutions.
A DSP or Digital Signal Processor is a part that combines the ability to do analog to digital or reverse conversions and process the analog and or digital representation and other data. These are used extensively, and increasingly in such things as printers, scanners, communications equipment and voice recognition systems where you take an analog signal, such as a voice signal from a microphone, filter the signal to get rid of background noise, convert it to a digital equivalent and then process and compare the resulting waveform representation against patterns that are stored for recognizable speech. You can do much of the same using simple A/D conversion circutis on a sound card combined with DSP altgorithims running on a PC's uP but that consumes a lot of processor overhead and may not fit into pocket protector sized personal assitant device very well.
A LASIC is a relatively new approach that grows out of the development of very fast and higher accurate laser etching systems that are used to "shoot" the chips on a wafer. The LASIC chips have a standard gate structure or matrix implanted on it and the laser is used to hard "program" the specific circuit needed into the standard matrix. I don't know the details of the techniques that CL uses to program the interconnects using the laser - much of the details were not openly disclossed the last time I looked into it. That isn't very important as long as you understand the basic impacts of the technology. The advantage of the LASIC is that is less expensive than a FPGA and doesn't require the non-recurring engineering or set-up costs of using an ASIC. It also consumes less power and can be faster, (if that is desireable for the specific application). The disadvantage of this approach is that it is limited to a relatively small number of parts at this time and is more expensive than a full custom design if the quantities of production are high - but that's the same with FPGAs with which LASICs compete.
Hope that partially answered your questions. |
|
