Hi peeps,
I am a little confused.. :)
Should I use more, slower memory, or less, faster memory?
I have 2 x 512mb of pc3200, or I have 2 x 1GB pc2700...
I dont play games that much and I dont overclock at all.
Any help please,
Ta,
Me.
Printable View
Hi peeps,
I am a little confused.. :)
Should I use more, slower memory, or less, faster memory?
I have 2 x 512mb of pc3200, or I have 2 x 1GB pc2700...
I dont play games that much and I dont overclock at all.
Any help please,
Ta,
Me.
what motherboard do you use
what do you use your computer for if just surfing 512/1024 wwould be fine
Its in a shuttle sn85g4 v3, and yup, i do just use it to surf... and do a bit of vid encoding everyso often.
oooh good question :)
Tell you what.....why not try both :) and tell US what works best :)
Use 3d Mark and some other stuff, and report back :)
I'd guess 1 gig of fast stuff is better in most stuff....
but if it was 256 fast stuff and 512 of slow stuff the 512 would whoop ass :)
1 gig is normally enough.
But try man...try...and then spill the beans :)
2 x 1gb seems to be the same price whether pc3200 or pc2700 on scan if your talking about the corsair value select if u are talking about nonbranded memory id definitly go with branded stuff
didnt check other sites(im too lazy)
Lol, u should have said that 2 me b4 cause i had some fast 1GB but now have some slow 2GB lmaoQuote:
Originally Posted by Zak33
more is better in this case I imagine... video editing usually eats up all the memory it can find. Faster memory only helps if the rest of your system (cpu, motherboard, etc) can all work at that speed too (and are set to work at those speeds!) Without more system detail and ram info you'll just have to try it and see as stated above ;)
I have 1x512 pc2700 (333fsb?) and 1x512 pc3200 (400fsb?), but I run them both at pc2100 (266fsb?) speeds and lower CAS to match my 266 fsb cpu. Seemed logical, plus I tried playing games with the ram clocked at pc2700 and it made little or no difference (heck, it might have made it worse!) to performance...
I took as scientific approach as I can, and here are my results for anyone passing through here via Google as I did.
Conclusion: More of slower memory measures faster than less amount of faster memory
Test: 4GB RAM 1600Mhz vs. 6GB RAM 1333Mhz
Measurement: Used utility hardinfo which runs speeds test using zlib, sha1, blowfish, ray tracing, and some other encryption algorithms. I ran 3 tests with 1600Mhz at 4GB and then 3 tests with the 1333Mhz at 6GB. Some sampled results:
zlib: 6GB 0.15% faster than 4GB
ray tracing: 6GB 0.00157055% faster than 4GB
Now, those differences are negligible, but it is noteworthy that every test, 100%, was faster with more rather than speedier. Of course, the difference is not noticeable in the overall experience. But more was proven to be better.
CPU I ran this on: Core Quad 2.8Ghz
OS: Linux Ubuntu 8.10