Cache in your chips and get a bus!Written by Seamus Dolly
Computer related terminology could sometimes be daunting to newcomers. These are relatively new words or hybrids of words, already in existence. Someone with a degree in English, for example, could not necessarily guess, function of a particular devise, by its name. This is because such a devise never existed in history or in history of English language. Tolerance and patience is required. If you work in field of computers, these "new" words must be learned. However, P.C.'s are designed for use by ordinary people, and gritty details regarding their construction, can be largely ignored. A journalist, typist, builder or bookkeeper doesn't need to know what an EIDE or SCSI is, or indeed isn't. Hard drive description is not relevant to majority, as long as devise is functional. A "BUS", though, could be guessed by some, as it carries something, not unlike a conventional bus. In this case, it carries current/electrons/data, and is simply a conductor. A bus conductor, if you like. For now, buses are made of copper with experiments underway to utilise other materials. I read recently that "prions", which are a type of protein, are been considered as conductors. Will some of our computers' components be organic? We shall have to see! Fibre optics, which are, basically speaking, glass pipes around four thousands of an inch (0.1 mms.), in diameter, employ speed of light, for signal/data transfer. A light (diode), is on or off, at one end, and a photo-resistor at other end registers its state. Now, wasn't that simply put! These glass pipes are coated to reflect light inwards, meaning that light can be "bent" around corners, while within its container/pipe. Heat generation is minimal and electrical resistance is redundant. It is heat generation and resistance that brings your "fan" into play, keeping everything cool, and electron transfer at a controlled level. Low resistance= high current. Your starter motor for your car has really thick cables allowing electrons from twelve volt power source (battery) to do their job in turning over you engine, which is a considerably difficult task. A smaller cable would have a higher electrical resistance, causing it to overheat, and eventually, melt. Keep listening for that "fan". It is your computers first line of defence. In fact, it is principle of electrical resistance that is employed by electrical heaters.
| | In Computer Memory what is CAS Latency?Written by Ron Merts
This is question we are asked more than any other question. So, I figured I'd put together a bulletin containing my $0.02 worth!First of all, what is CAS? "CAS" is short for "Column Address Strobe". A DRAM memory can be thought of as a matrix, kind of like a spreadsheet with memory cells instead of numbers and formulas. Like spreadsheet, each cell has a row address and a column address (like "AA57" or "R23C34" in spreadsheet). As you might have guessed, there is also a RAS signal, which is shorthand for "Row Address Strobe". And, what do you mean by "latency"? Latency refers to time that you are waiting to get what you need. Merriam-Webster dictionary defines it as "the interval between stimulus and response". Now, how does CAS work? To understand this let's walk through a simplified version of how memory controller actually reads memory. First, chip set accesses ROW of memory matrix by putting an address on memory's address pins and activating RAS signal. Then, we have to wait a few clock cycles (known as RAS-to-CAS Delay). Then, column address is put on address pins, and CAS signal is activated, to access correct COLUMN of memory matrix. Then, we wait a few clock cycles -- THIS IS KNOWN AS CAS LATENCY! -- and then data appears on pins of RAM. So, for CAS-2 you wait 2 clock cycles and for CAS-3 you wait 3 clock cycles? Bingo! So, CAS-2 is 33% faster than CAS-3? Whoa, not so fast! There are a LOT of other factors in memory performance. Here are a few of main ones: ·Sometimes you have to move to a different row in memory. This means activating RAS, waiting RAS-to-CAS delay, then doing CAS latency thing. ·Other times, you do a "burst" read, when you pull in a lot of data in one big block. In that case, CAS is only activated ONCE, at beginning of burst. ·Also, don't forget most important thing: processors have big caches! The cache is where processor stores recently accessed instructions and data. The cache "hit rate", i.e., percentage of times processor finds information it needs in its own cache, is typically greater than 95%!
|
