Hi all,
My post today involves whether or not Hyperthreading improves Benchmark scores, specifically Sisoft Sandra. The reason for my concern is that my P4 box (2.8Ghz) lags well behind the P4 reference systems in Sandra. I'm currently running Win2k and Hyperthreading is not supported without reinstalling. So before I upgrade to XP (and give Bill Gates any more of my hard-earned money) I'd like to know if the reason my scores are lower is because the reference systems use HT. Also, some have suggested that Sandra may not be the best tool to compare myself to others with. Anybody have opinions about a better benching tool?
Well I have an update on the situation. I installed XP Pro and my scores jumped right up to the level of comparable P4s. I am not sure if the speed increase is because XP is faster than 2K or if it's because XP is using both threads. I suspect it's the thread count, and not that XP is faster than 2K.
hyperthreading technically works in windows 2k but windows 2k actually thinks there are 2 physical cpus. in other words xp is designed to work with hyperthreading specificly and win2k is not. you may see performace decreases when using h/t in 2k.
You are right, if 2k see'd it as dual processors, then you will have the overhead involved in running multiple processors. Second processor adds 50% processing power, third and futher processors add 25% processing power, the 50% and 75% there after are used up in overhead. This was sourced from Microsofts netwoking books.
You are about to answer a thread that has been inactive for more than 6 months. If you still wish to proceed, please ensure that your posting is original and does not duplicate or overlap any prior responses to this thread.