[Onepagebook]

This comparison was done between TCCD running over 300 and BH-5 running over 280, this is just "little dessert"

Nothing tweak, just clean install of fresh win2k, x850xtpe is air-cooled.

TCCD:


cpuz verified:
http://valid.x86-secret.com/show_oc?id=5280

FM site reg:
http://service.futuremark.com/compare?2k1=8461587

************************************************** ********
BH-5:


CPUZ verfied:
http://valid.x86-secret.com/show_oc?id=5293

FM link:

http://service.futuremark.com/compare?2k1=8461820

************************************************** ********

here is the second desert.
I think OPP was running 270x13
I definitely can still go up a bit but not today. I am really really tired.
Man, I hope I got another cascade to cooled my videocard.
X850xtpe is non-moded, all air-cooled.

CPUZ validated:

http://valid.x86-secret.com/show_oc?id=5322
fm reg:
http://service.futuremark.com/compare?2k1=8462254

[matttheniceguy]

How does this compare TCCD and BH-5?

In the first run you are at 60 mhz less CPU speed than the second, and isn't 2001 mainly determined by cpu speed? You also have tigher timmings on the BH-5. I don't know if this was because that was all the TCCD could do, but it doesn't make it a very good comparison.

I wouldn't mind seeing a comparison of what sort of speed and timmings either ram can do at different timmings, or if they actually give the same perfomrance at the same speed and performance, but I just don't know what I am sudposed to gain from what you have shown.

[nightic]

Originally Posted by matttheniceguy

How does this compare TCCD and BH-5?

In the first run you are at 60 mhz less CPU speed than the second, and isn't 2001 mainly determined by cpu speed? You also have tigher timmings on the BH-5. I don't know if this was because that was all the TCCD could do, but it doesn't make it a very good comparison.

I wouldn't mind seeing a comparison of what sort of speed and timmings either ram can do at different timmings, or if they actually give the same perfomrance at the same speed and performance, but I just don't know what I am sudposed to gain from what you have shown.

I think you've missed the point.

OPB's comparison details the performance differences available for 2 of the most commonly used RAM types in a modern NF4 setup.

One of the most frequently asked questions when building such a system is whether to run TCCD knowing full it well that it's capable of extremely high FSB speeds with moderate timings OR BH-5/UTT at lower FSB but with the ultra-tight timings it's known for.

If there's a flaw in the testing it's that 280MHz 2-2-2 is far harder to achieve for most BH-5/UTT sticks than 300MHz 2.5-3-3 is for the highest quality TCCD.
I would like to have seen 2-2-2 results at a more readily-available speed of around 255-260MHz.

(Incidentally, my own casual testing, also with 3DMark2001SE has indicated an approximate 45-50MHz disparity between TCCD at 2.5-3-3 and UTT at 2-2-2 (i.e UTT 255MHz 2-2-2 was around equal to TCCD 300-305MHz 2.5-3-3)

[Sjaak]

So a fair conclusion would be to say that for 3dmark01, timings make a huge difference? or did i read it wrong

[Reefa_Madness]

I think that the only sane conclusion that I can come to, based on that comparison, is that I want some bh-5 that will run 270 plus with 2-2-2 timings.

The rest of you guys can reach your own conclusions.

[d]g[ts]

Thanks for this comparison onepage, your results are always beyond normal to say the least. May i ask why you did not run your TCCD faster? I have seen results from you at XS with TCCD running much faster than 300, and curious why you stop here for this one.
Also IIRC you have done tests like this before with bh-5 at 250 and TCCD at 300 where they came out dead even on an NF3 platform. would it still come up even if the memory speeds were at the same difference in speed. I.E 280 to 330? or would one start to lose out? (yes i understand both of those speeds are sick for these types on memory)

Thanks for posting here OPB always a pleasure to see your results