KT400a or NForce2 Ultra 400

[cephei]

if im gonna use single channel ddr for nForce2 Ultra 400, is it still faster than KT400(a)?

 

[ChipDeath]

I think Nforce2 still has a slight lead, even in single channel. Plus far superior OCing features (lock PCI/AGP for instance).

You can get some very cheap KT400(a) boards though, so it could still be a choice for non-oced 'value' systems. I haven't got any experience with that chipset myself, although I have been less than impressed with KT333.

---
$hit Happens. I just wish it would happen to someone else for a change.

 

[FallOutBoyTonto]

Recently the nForce 400 was released, that doesn't have the ability to do Dual DDR. That is also faster than the dual-channel version in some benchmarks. If you're going to run single channel, that chipset would be the way to go.

<A HREF="http://www.amdmb.com/article-display.php?ArticleID=244" target="_new">Read review of Soltek NV400-64 @ AMDmb.com</A>

<A HREF="http://www.anandtech.com/mysystemrig.html?id=24106" target="_new">My System Rig</A>
<A HREF="http://service.futuremark.com/compare?2k3=535386" target="_new">3DMark03</A>

 

[TheMASK]

If u r overclocking i wud suggest u go with nForce2 Ultra 400.

I dont know about KT400A, but with my KT400, i can only go as high as 185FSB. for me that suxx!

<b>ur comp suxx! face it and live with it!!</b>

 

[Spitfire_x86]

Where you've found that nForce2 400 is faster than nForce2 Ultra 400 (dual ch mode)? It's only a nForce2 Ultra 400 without dual channel memory interface.

----------------
<b><A HREF="http://geocities.com/spitfire_x86" target="_new"> My Website</A></b>

<b><A HREF="http://geocities.com/spitfire_x86/myrig.html" target="_new"> My Rig</A></b>

 

[pIII_Man]

wooooo! time for a cofee break...reread this thread and you'll find the link...

I personally have no clue who's idea it was to start putting dual channel on amd systems...just by making the memory controlor more efficient they could have matched or even beat the performance of a dual channel amd system...

200mhz fsb system single channel

fsb bandwith 3200mb/s
mem bandwith 3200mb/s

200mhz fsb on a dual channel mobo

fsb bandwith 3200mb/s (stay the same)
mem bandwith 6400mb/s (doubles...how can the processor take advantage of this increase if its banwith stays the same?)

Now i am not denying that dual channel has the best performance on most amd mobos...but that is not directly related to the dual channel...just the fact that there is some reduced latency in the memory bus...again though...a tweaked LL memory controlor would squeese out the same performance in single channel arangements...

3 386DX-25's...12 volts...glue some ln2 and a wicked amount of overclocking and you get a willamantee minus 36 pins, 33.75 million transistors and a couple hundred mhz...

 

[Spitfire_x86]

I know these things. My point was, nf2 400 mem controller is not a tweaked version of nf2 ultra400, it's the same one minus dual channel mode. So it should be equal to nf2 ultra400 in single channel mode

----------------
<b><A HREF="http://geocities.com/spitfire_x86" target="_new"> My Website</A></b>

<b><A HREF="http://geocities.com/spitfire_x86/myrig.html" target="_new"> My Rig</A></b>

 

[pIII_Man]

well the benchies don't lie that soltek board performs very well (props to soltek)...


Sorry i went trough that whole thing...i kinda figured you knew but i hate how everyone things that doubling the bandwith of the memory is what causes dual channel to increase the speed of an amd cpu...so that was kinda aimed at the whole thg community...

3 386DX-25's...12 volts...glue some ln2 and a wicked amount of overclocking and you get a willamantee minus 36 pins, 33.75 million transistors and a couple hundred mhz...