Maximum PC

If I have my pc doing 5 different intensive tasks for me and all of the sudden it crashes.....how can I tell if my computer is too weak or too strong.

Meaning, did it crash cause I overclocked it too much or my system isn't strong enough to do all the tasks?????

Didn't you say that when you turned off some "overdrive" feature on your video card, the problem went away?

Hm.

Yes but it was the CPU O'Clock that made a difference.

Thought about this question for a while and I came to the conclusion that, if your system is popping around super fast and it crashes once in a while, the assumption should be it's O'C'd to high. If your system is plain Jane, even ass dragging and programs are crashing, you are not O'C'd enough.

Just a hypothesis.

You keep saying "crash", but the problem is that this could be caused by a variety of different things. It may be software, it may be hardware. You have to isolate all the variables that lead up to this point, otherwise you'll be chasing dead ends.

Also programs have long decoupled themselves from clock speed or whatever. Either you're using a crappy program, or whatever's being OC'd doesn't like it when you OC it.

"You keep saying "crash", but the problem is that this could be caused by a variety of different things."


Well I agree with you there.

My pc has been running strong, overclocked for almost 3 years without a problem. Running HD on my tv, surfing high speed net and watch HD on the net all at the same time.

That's why I'm more inclined to believe it's a physical problem first. I mean, my SSD died but it's a relatively new tech (at least it was 3 years ago), but it doesn't mean other physical things can.......if not die at least degrade.

You made a point in another thread that it's all just 1's and 0's, there is no in between. No degrading of SATA cables.

Not but 30 years ago people used to say "that's just blood, nothing more". Then we discover there is more in the form of DNA. A whole new branch of science is created because someone somewhere believed there was more than just blood.

Maybe not in my lifetime but I am sure there will be a way where one will be able to measure the quality of a SATA cable and other devices that carry 1's and 0's. I'm pretty sure my old SSD was not working 100% perfectly one day and the next day just DIED.

My brand new ASUS SATA III cables arrived the other day and tomorrow I will be taking apart my pc and cleaning out every nook and cranny.

Just thought of something else. When I first built my pc 3 years ago and ran the Windows Performance Test with it overclocked, it scored the highest score in every category.

After I installed my new SSD (which means every other part was original from 3 years ago) I ran the test again and scored a full point and a half lower. Machine was still overclocked and NOTHING else was different......except the performance score.

Thanks for your help.

Well, for one, go read http://www.pugetsystems.com/labs/articl ... ference-97

For two, there's already a science to improve the quality of cables... it's called chemistry and physics. Not that it matters in the digital world. As long as the other end gets a 0 or 1 reliably, why do you care? In the analog world, this may matter that if I spit out a 12vpp 100KHz sine wave, that I should get something relatively close to that on the other end.

I recall you suggesting that the amount of data you pump through a cable will degrade it. If this is the case, then how come CPU's don't just die? They transfer a lot more electrons than a SATA cable ever will. And yet, we still have computers from the 80s, 90s, etc. that are running just fine.

And, oh noes, a full point and a half lower. Maybe I should start screamin' "what the fuck EVGA" because my GTX 560 Ti doesn't score the same points as some other review's website did.

EDIT: Here's another thing. Why do you think a lot of websites (particularly Engadget) give Monster Cable a ton of flak for selling overpriced cables? Because they throw in a metric ton of "quality improvement" features that have been proven to... improve nothing. In other words, a $10 HDMI no-name brand cable "performs" the exact same as a $60 HDMI Monster cable. Because, like I said, as long as the other side gets a 0 or 1 reliably, it doesn't matter what you else you do to it.