OCZ vs. Corsair vs. Infineon

T

Toad21

Limp Gawd
Joined
Jun 29, 2004
Messages
445
I have had 3 types of memory:

OCZ 3200 2-3-2-6 2x512

Corsair 3200ll-Pro 2x512

Infineon 3200 cas 2.5 2x512


The lower timing ram(ocz, corsair) showed absolutely no gain compared to the alot cheaper Infineon in sisoft sandra 2004. The infineon can goto 230 mhz stable @ 8-3-3-3, and 200mhz @ 6-3-3-2.5. I stuck with the infineon and got just as high overclock as possible for my [email protected] A7n8x-E Deluxe rig. PC mark also remained the same between brands, WHAT THE HELL, I thought a tighter timing would make more of a difference. Someone explain.
 
Why Would Better Timing Ram Perform The Exact Same As Generic!! Do Timings Mean Nothing?
 
Wouldn't ram running 2-2-2-2 beat ram running 8-3-3-3 at the same mhz? Well it didnt according to my tests. Timings mean absolutly nothing. How interesting.
 
Nope, timings mean a lot. SANDRA isn't the greatest benchmark program, it doesn't even actually run the tests. THATS RIGHT, IT DOESNT RUN THE BENCHMARKS!!!! I think it reads the speeds your computer is at when the program starts and uses a mathmatical forumula to calculate the scores. How do I know? Download the program CLOCKGEN for your computer. It allows overclocking in windows. Boot up at stock. Run sandra. Record the scores. Overclock using clockgen and then run sandra again. No difference? I got the same score with my old 3ghz p4 at 3ghz and 4ghz. Great benchmark program :rolleyes: :rolleyes: :rolleyes:

Anyways, on a64 systems, timings don't mean as much as they do on p4 systems, but on i875p systems timings mean quite a bit. AS much as a few hundred 3dmark 01 points or 5-10 FPS in a game. I've tested this over and over again.
 
I saw absolutley no improvements in 3dmark 2003, and no increase in FPS in Splinter cell pandora tomorrow. Maybe 2001 would have been a better judge of computer power, but overall, I dont think ram that is worth $300 and ram that is worth $150 makes much a difference at all for amd 32 bit architecture.

Sisoft does actually test computer speed in most cases. When I run ram benchmark, it uses 700mg of my ram, and 100% of my processor the whole time. That has to be doing something.
 
Toad21 said:
Sisoft does actually test computer speed in most cases. When I run ram benchmark, it uses 700mg of my ram, and 100% of my processor the whole time. That has to be doing something.
Click to expand...

it is doing something. Its calculating your result, not benchmarking it. And 3dmark 03 is WORTHLESS for anything except your video card. Your video card is also the limiting factor in pandora tomorrow unless you have a 6800 ultra (and if you have an axp, you probably don't.) If you don't believe me, why don't you try the clockgen thing yourself? There was a whole thread about this controversy a while back, try using the search feature. The bottom line is that lower latency ram will have some improvement. It just does. Is it more than 5%? In most cases, probably not. Is it worth a $150 premium? Thats up to you. If you don't have a lot of cash it may be in your best interests to use cheaper stuff. I know for me, every little bit counts (thats why i have an amd a64 fx-53/geforce 6800 ultra system on the way). Whatever you do, still get name brand. Try kingston valuram or corsair value series or anything by crucial. Generic memory is a good recipe for disiaster.
 
As he said, OCing with a P4, you'll see a slight difference with better timings. It's a balancing act between higher FSB and good timings.

*******
Off topic, but....

computerpro3,

No disrespect meant, but as per your sig, it's pretty obvious that you haven't actually used all of those PSs based on what it says. Some of them are junk, yes, but the one you have listed that I know you have no experience with is an Achieve. Throughout 8 various systems running today that I built, all have an Achieve 500w PS. And all have run perfectly since day 1 of me installing them. This includes my main gaming rig for the last 2.5+ years, which, previous was an OC from 1.6 to 2.3, and currently has taken an OC from 2.8 up to 3.7 in a 1:1 ratio.

Here's some facts for you:
http://www.techimo.com/articles/i145.html

Don't base your words of wisdom on a blind fad of expense. Base them on experience.
 
The sisoft results showed like 50 meg more per second with the faster ram.

So it did make a difference, of about 1%.

And I do have a 6800gt, and I not happy with all the hype about ram timings when it means VERY little. All I am saying.
 
Badger_sly said:
As he said, OCing with a P4, you'll see a slight difference with better timings. It's a balancing act between higher FSB and good timings.

*******
Off topic, but....

computerpro3,

No disrespect meant, but as per your sig, it's pretty obvious that you haven't actually used all of those PSs based on what it says. Some of them are junk, yes, but the one you have listed that I know you have no experience with is an Achieve. Throughout 8 various systems running today that I built, all have an Achieve 500w PS. And all have run perfectly since day 1 of me installing them. This includes my main gaming rig for the last 2.5+ years, which, previous was an OC from 1.6 to 2.3, and currently has taken an OC from 2.8 up to 3.7 in a 1:1 ratio.

Here's some facts for you:
http://www.techimo.com/articles/i145.html

Don't base your words of wisdom on a blind fad of expense. Base them on experience.
Click to expand...

And here's some facts for you.

That psu review is absolute, total, pathetic, crap. Absolute crap. They don't even use a multimeter!!!!!! They actually use software testing for a psu review. I find that very funny. Where's the multimeter tests? Where's the ripple testing? Where's the maximum load testing? Where's the electrical noise testing? Where's REAL fluctuation testing? Try reading a real psu review at a real review site, like silentpcreview.com. Then tell me how the achieve 500w psu holds up under those conditions.

Here's what a real review's results page should look like:
http://silentpcreview.com/article177-page4.html

and here's how you get those results:
http://www.silentpcreview.com/article148-page1.html

Impressive, no?

So, about the achieve 500w psu again....where's the APFC? Where's the line conditioning? What brand capacitors are in there? Who oem's them? Where are they manufactured? Whats the efficiency? What temperature is it's 500w rated at? How hot does it get? What's the fan noise like (and I'm not talking about manufacturer specs, they usually lie)? Whats it's real output (and I GUARANTEE YOU it's not 500w under normal conditions)?

When you can answer those questions and they come out to satisfactory answers (which they won't, based on numerous posts in hardforums and ocforums psu forums which you can find using the search feature, or you could just make a post asking about it on xtremesystems.org/forums where the best of the best hang out), I will remove achieve from the list. Hey, it could happen, in which case just pm me with the answers and it will be removed.
 
computerpro3 said:
And here's some facts for you.

That psu review is absolute, total, pathetic, crap. Absolute crap. They don't even use a multimeter!!!!!! They actually use software testing for a psu review. I find that very funny. Where's the multimeter tests? Where's the ripple testing? Where's the maximum load testing? Where's the electrical noise testing? Where's REAL fluctuation testing? Try reading a real psu review at a real review site, like silentpcreview.com. Then tell me how the achieve 500w psu holds up under those conditions.

Here's what a real review's results page should look like:
http://silentpcreview.com/article177-page4.html

and here's how you get those results:
http://www.silentpcreview.com/article148-page1.html

Impressive, no?

So, about the achieve 500w psu again....where's the APFC? Where's the line conditioning? What brand capacitors are in there? Who oem's them? Where are they manufactured? Whats the efficiency? What temperature is it's 500w rated at? How hot does it get? What's the fan noise like (and I'm not talking about manufacturer specs, they usually lie)? Whats it's real output (and I GUARANTEE YOU it's not 500w under normal conditions)?

When you can answer those questions and they come out to satisfactory answers (which they won't, based on numerous posts in hardforums and ocforums psu forums which you can find using the search feature, or you could just make a post asking about it on xtremesystems.org/forums where the best of the best hang out), I will remove achieve from the list. Hey, it could happen, in which case just pm me with the answers and it will be removed.
Click to expand...

LOL, chill man, it's 1 review. It seems more like you're bashing the review because it doesn't agree with your mindset. Oh well, to each his own.....

It's still sad that you post such blind information, without having any actual hands on experience with the Achieve PS.
 
You must log in or register to reply here.
Back
Top