I've got two sticks of 256, so I would have to buy two sticks of 512 anyway.Mr.Monkey said:If you use two different types, they will typically run at the slower of the two speeds with Epox boards.
Hence save money by buying a matching set, or spend more and buy 1Gig of the fastest your MB supports!
If he had 3 slots, then you can use any combination of sizes, as long as you do not exceed the max ram supported by the MB.Clown said:You only have two ram slots? Wouldn't it work if he used two 512 sticks and a 256 one, but all the same speed?
So essentially I need to decide whether I want to save a bit of money and go for the 2100 instead of the 3200? I suppose I better wait and see if I got this job I'm applying for first.Mr.Monkey said:Mank: If you are replacing your ram anyway, and have the money to burn then buy the fastest your MB supports. If you are scrimping, then buy one 512 stick of the same speed.
As to if you will be able to tell the difference in game? Prolly not, but it will improve the score in 3dmark 2003 etc.
Weeeeeeeel, seeing as the amount of mantlelogs is infact an index, it is far more useful than you suggest. It's a hell of alot better than nothing.Scooba da Bass said:That doesn't give you any useful comparison though. It's like me deciding to measure the speed of a PC in mantlelogs and telling other people that their PC is 3 mantlelogs slower than mine, sure, you know yours is slower but how does that translate to anything you'll be able to notice?
3DMark '03 has a recommended spec of 512Mb RAM, so how would more memory make it go faster? Last time i looked the resultsbrowser didn't include scores for machines with 20 copies of MSWord running in the background. Besides which it's pretty clear from the resultsbrowser that it's about 80% GPU dependant, which makes for a pretty poor system memory benchmark.Mr.Monkey said:Exactly.
If you wanted to see what sort of benchmark differences you would get, 3dmark 2003 does offer a comparison search engine. So run the benchmark, and then compare it to someone elses system that is the same as yours, with the exception of the faster memory.
People may argue over the validity of the benchmarks, but the comparison system is the dogs bollocks.
Ermm, I quite simply won't answer something like that. I could explain it to you, but I'm pretty sure you are just looking for a flame war. Not really suitable for the tec forum.Xavier said:3DMark '03 has a recommended spec of 512Mb RAM, so how would more memory make it go faster? Last time i looked the resultsbrowser didn't include scores for machines with 20 copies of MSWord running in the background. Besides which it's pretty clear from the resultsbrowser that it's about 80% GPU dependant, which makes for a pretty poor system memory benchmark.
RAM > 512Mb is generally going to let you run more simultaneously, only apps like Photoshop with hooge images which go beyond half a gig are going to appreciate any immediate difference.
That is, unless, by going from 512Mb->1Gb you're also going from a single to dual channel memory configuration. Which he isn't.