Oh, ok... that makes me feel better. Before I deal specifically with video editing, I'm going to touch on the whole P4 vs. Athlon debate, as it's very relevant to this topic. And I apologize if any of this is too basic, but I want to make it clear not just for you, but also for anyone else reading this that may not know as much as you do.
---------
Here's a quote from Thrax at
www.short-media.com/forum that explains the major difference between the Pentium 4 and the AMD Athlon. I am the cooling forum moderator for Short-Media, and I have a great deal of respect for Thrax; he knows what he's talking about.
Quote:
In a CPU, a pipeline is like a factory assembly line—it executes program instructions one stage at a time. The more stages there are in the pipeline, the less time each stage needs to complete its work, so the faster the CPU can cycle. (note- by cycle, he's reffering to one clockcycle- one Hz. 1Hz = 1/1000 of 1MHz, 1MHz = 1/1000 of 1GHz) While most CPUs have four to seven stages, and the Athlon is considered a superpipelined CPU with 10 stages, the P4 has more than 20 stages.
But superpipelines have drawbacks, which partly explains why a P4 doesn’t always perform as well as the clockspeed would indicate. Clock frequency is not an absolute measure of performance. (my emphasis, not his. I CANNOT emphasize this enough)
One drawback of superpipelines is the penalty they impose when the CPU must branch to another part of the program (Say.. Suddenly going from encoding a movie to loading plugins). Like an assembly line that must stop to change the kind of vehicle it’s manufacturing, a CPU pipeline must often pause to load a different stream of instructions. The deeper (longer) the pipeline, the greater the penalty during this switch. Modern processors try to avoid that penalty by using branch prediction units, and preloading registers that the CPU thinks it will have to use next.
So, by adding more stages to the pipeline, the CPU is allowed a higher speed because each stage needs less time to do the work, and when each stage needs less time to do the work, the clockspeed may rise because the demand on the pipeline isn't as high now. |
Let's compare the Pentium 4 and the Athlon XP to cars. Each pipeline the CPU has equates to more weight, but also more power. However, each extra pound you add to the car adds a fraction of a horsepower less.
In this example, the Pentium 4 would be like, say, the Mercedes 560SEL I drive. It's got 238hp, weighs 4200lbs, blah blah blah... The Athlon, Athlon XP, And Athlon 64 CPUs could be compared to a Caterham 7, which weighs about 1500lbs, but only has about 150hp.
If I were to drag race a Caterham in my 'Benz, I would get my ass kicked. Even though the Benz has a SUBSTANTIAL horsepower advantage over the Caterham, the extra weight negates the added horsepower.
Even though the Athlon XP/Caterham doesn't have as much power as the P4/Mercedes, it's still faster, because it's more efficient.
The end result is that AMD's "Performance Ratings" are more-or-less right on. They're a bit conservative on the lower end CPUs, and a bit optimistic on the higher end CPUs, but overall, An Athlon XP or Athlon 64 xxxx+ is basically equivalent to a P4 @ xxxxMHz
---------
For video editing in particular, AMD CPUs are far superior to the Pentium 4, P4-based Celerons (all 1.7GHz+ Celerons), and Pentium 4 Xeons. The Athlon XP absolutely POUNDS the P4 in Premiere, and the Athlon 64/FX/Opteron is even faster than the Athlon XP. The best video editing system you could get would be a dual opteron system. A dual Athlon system would be cheaper, and still very, very fast. If you're on a tight budget, a Pentium 4 is totally out of the question anyhow, since you can get an athlon XP and a motherboard for the price of a 2.4GHz P4
Moving on, RAID stands for Redundant Arrays of Independent Disks, or Redundant Arrays of Inexpensive Disks, depending on who you talk to. The first one makes the most sense to me, because god knows RAID arrays aren't necessarily cheap.
I suggest you read this site:
http://www.acnc.com/raid.html
That is the single best guide I have ever seen about the various RAID levels.
I'll give a brief description of RAID 0, 1, 0+1 and 1+0 here, though.
RAID 0 is not a true RAID level, as it is not fault tolerant. If one drive in a RAID 0 array fails,
all data on that array will be lost. The way Raid 0 works is that each piece of data below a certain size (called the stripe size) is split evenly between all the drives in that array. So, if you have a 2-drive RAID 0 array, each drive gets half of your data. If you have a 3-drive array, each drive gets 1/3 of your data, and so on. RAID 0's main advantage is that it is very, very fast. It is almost as fast as the combined speed of every drive in the array, up to the limits of the controller's interface (133MB/s for regular 32-bit/33MHz PCI, 266MB/s for 64-bit/33MHz PCI, 533MB/s for 64-bit/66MHz, etc.) but the amount the speed increases diminishes for each drive added.
The capacity of a RAID 0 array is the sum of the formatted capacity for each drive in the array.
RAID 1 is kind of the opposite of RAID 0; it's mirroring. Each piece of data gets written to every drive in the array, so that even if every drive in the array but one fails, all of your data is safe. The drawback is that RAID 1 is not much faster than a single drive, and is slower than a single drive in some cases. The other drawback is that the size of the array is the same as the size of one drive in the array- you can have 4 160GB drives, but the useable space is only 160GB.
RAID 0+1 is a combination of RAID 0 and 1. It requires at least 4 drives, and it takes at least 2 RAID 1 arrays and combines them into one RAID 0 array. As long as one drive in each RAID 1 array is working, no data will be lost. But if all of the drives in a RAID 1 array is lost, all of the data in the entire RAID 0+1 array that RAID 1 array was a part of will be lost. It has some of the speed of RAID 0, and most of the reliability of RAID 1. Also, the useable size of the array is limited to 50% of the total capacity of the drives.
RAID 10 (1+0) is the inverse of RAID 0+1. It is a mirrored array in which each "drive" is actually a RAID 0 array. It has most of the speed of a RAID 0 array, but only some of the reliability of a RAID 1 array, and as with RAID 1 and RAID 0+1, the useable capacity of the array is 50% of the actual capacity. It is not as fault-tolerant as RAID 0+1, because if one drive in each of the RAID 0 arrays in the RAID 1 array fails, all your data is lost.
To summarize that, in a 4-drive array:
RAID 0 can sustain 0 drive failures before total data loss
RAID 1 can sustain 3 drive failures before total data loss
RAID 0+1 can sustain up to 2 drive failures, with a maximum of 1 per RAID 1 array
RAID 10 can sustain up to 2 drive failures
as long as both drives are part of the same array
Personally, I run two separate RAID 0 arrays. I get ALL of the speed of RAID 0, and since I back up my important stuff regularly, I get most of the reliability of RAID 0+1.
Note that if you use any kind of RAID, all of the drives used must be as close to identical as possible. The same manufacturer, model, and capacity is required, and build dates as close as possible to one another is preferable.
---------
What is your budget for the entire system?
