Originally Posted by kevindd992002;15484374
Oh I didn't noticed you posted this message earlier. Sorry man, I had no intention of mentioning your name for false information or something. I actually did that because I read many people using your name in their posts so I thought it was kind of OK with you. Anyway, I understand and sorry again. Cheers
It's just that there are times where it doesn't matter, and then there are times when it does. I don't quite know how to verbalize why it matters this time, but for some reason it does. It's just one of those things.
Although, I could be misinterpreting your intentions or what you're saying or something (I know myself well enough that I can say that this is quite possible). However, to me it feels like if I hadn't said something, then I could end up having people approach me and be like "What did you mean by saying that the heat produced by IBT/LinX with AVX is useless?"
Or, it could result in someone coming into this thread and going "Well, he's wrong"
, or "Well, he doesn't know what he's talking about"
. Either way, it would either mean that I'd have to say "He misquoted me"
, or it could mean that my reputation gets hurt.
In general, it's best to never name anyone. Actually, it doesn't even feel socially acceptable to say things like "Someone here on OCN once told me this. Is it true?"
because then that will cause people to ask who you're talking about.
So to use the question this thread is asking as an example, it possibly would have been better to just say something like "For stress testing Sandy Bridge, is Prime95 Blend and Blend Custom just as god as IBT/LinX with AVX?"
, and you could even finish the question with like "I mean, is the only difference now that IBT/LinX with AVX simply produces more heat but is no more effective than Prime95?"
Besides, you could ask this question on 10 different message boards and end up with 100 different answers. This is just one of those things. However, my personal findings have been that unlike previous CPUs (including as recently as the first generation Core i3/i5/i7 series), using Linpack for stress testing Sandy Bridge is basically unnecessary which means that all it's really
doing is generating more heat for no reason. Of course, it's still an extremely good stress test for previous generation CPUs, but not so much for Sandy Bridge.
I have not tested with Linpack at all, but yet my system has been rock-solid stable at 4.8 GHz with zero problems for over 3 months (since at least the week of July 10th). All I used was Prime95's Blend and Blend Custom.