• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Is it worth upgrading tp pc4200 with this config?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

rayik

Member
Joined
May 2, 2002
Just build a new P4 computer. 3.0C on an abit ic7 mb. Memory is 1 GB of Mushkin pc3200 "2-2-2" special (2 x 512 running dual channel).

It will run stable at 260 fsb with divider 5:4 (at only 1.525 vcore).

Thus ram runs at 208 with 2-2-2-7 settings (at 2.8 v - maximum vdimm mb allows).

The ram is holding back any higher oc. I can actually get into windows at 267 (which is 4007!). However with 5:4 divider memory is running 213 which is just too high. System is very unstable.

Played with memtest86+. Maximum the "2-2-2" special will do is 208. Even with looser timings (2.5-3-3-11) memtest returns all kinds of errors at anything over 208.

I've been looking at pc4200 on newegg. It's expensive for 1 GB. With the ic7 I'd have to buy 2.5 CAS (a number of people have reported that with ic7's won't run with CAS 3).

I'm trying to decide if I should sell the Muschkin "2-2-2" special and replace it with some pc4200.

Question: would I actually see a performance increase with pc4200 at 260 w/ 2.5-3-3-9 (1:1) as compared to pc3200 at 208 2/ 2-2-2-7(5:4). Is there actual better performance or just better benchmarks with no real increase in real world usuage. If there is an increase in performance, how big (or small) of an increase could be expected.

Thanks for any help.
 
Batboy and deathstar13 did some testing on this and found that you really need to hit 265+ 1:1 w/ loose timings before you start getting a performance advantage over 230-265 5:4 w/ tight timings. With a 3.0 chip you'd need to move into extreme cooling to handle 265+ 1:1, so, barring making that jump, you're probably fine with the Mushkin :)

Here's the link to their testing.
 
Thanks Styyn for the link.

So it looks like the cheaper think to do is just stay with what I have. (Maybe play around with 3:2 divider to see the max oc the chip will do.)
 
Back