Originally posted by: Rollo
But when you try to overclock something you are taking a chance and a risk and you don't know if you will burn the card or not so you are not really deliberately misusing it and trying to burn it because you don't know your exact chances of damaging the card.
This is not a valid argument.
It's analogous to say, "It wasn't a crime for me to drive my car drunk because I didn't know I'd hit those people".
I emailed ATI a year or two ago to see what they thought of OCing. They said it voids your warranty and that they don't approve. I realize this may have changed with the OCer driver features they now include, but IMO, I've never seen a significant gain in performance on the high end cards from OCing, so why risk your money or being a thief?
Just to jump in here, I hate it when people make analogies that overclocking or another such activity is analogous to high crime. In this case (see bolded line above), you compare theft to murder - vehicular manslaughter.
I don't condone either act, but please don't say that the responsibility for overclocking a card and then getting it replaced when it fries is analogous to driving drunk. You can't take someone's life from lying about pushing a video card past it's spec. As a matter of fact, comparing overclocking to anything where a human life is at risk trivializes the other, much more serious act (in this case, driving drunk, which is a serious issue).
Also, isn't it at least a bit of a stretch to say that people who overclock are therefore morally bankrupt people? Does everyone who downloads mp3's and DIVX movies walk into blockbuster, steal their DVDs and shoot the employees and customers? I don't think so.
Not to sound coy here but I wouldn't buy a toga and join the cult of Rollo just yet, simply because he doesn't overclock his hardware (or if he does and it fails, he pays for it). I respect Rollo for his beliefs, no question, and I try to practise that myself, but I wouldn't consider him my moral compass just because of that. I definately think he is right in that respect, however.
In regards to ATI (and Nvidia for that matter) saying they don't condone overclocking - what the heck else are they going to say? That they strongly recommend overclocking? Of course the official stance on overclocking is that they don't endorse it - overclocking causes hardware failure (mainly by inexperienced users and extreme overclockers who don't know what "safe limits" means), which in turn costs them money. However, when ATI, Nvidia and their board partners include overclocking features in their driver suites, anyone can read between the lines enough to see that they don't exactly deny the fact that many of their customers will be overclocking.
--------------------------------
On an unrelated matter, has a high failure rate of the previous generation(s) of cards lead to the steady increase in prices of new video card hardware (witness the exorbitant $400-500 price tags of the new top of the line cards)? Or is this due to the added complexity of the manufacturing of the cards, the high input costs of R&D, marketing, low yield %, etc? As always, it's a combination of factors, but I wouldn't be surprised if the high cost of cards is mainly influenced by their high(er) manufacturing costs. As a matter of fact, many overclockers/hardware "enthusiasts" keep their cards at lower temperatures and in 'safer' conditions than Joe Public's cramped generic case with one fan.
As for the current cream of the crop, right now, neither the X800 nor the FX6800 appeal to me, simply because they're miles above my $200 (US) price ceiling on a video card. Personally, I'd like a hybrid of the cards, one with the GeForce FX's advanced DX 9.1 specs but with ATI's much lower power consumption and smaller card size.