• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

DDR400 slower than DDR333 on KT400

A

AbRASiON

Senior member
I read this in a few reviews a while back that most DDR 400 can't do CAS 2.0, only 2.5 @ 400 so it works out actually SLOWER than DDR 333 @ cas 2.

My question is, - is this only on the via chipset or is it all chipsets?
Or is it (was it?) a memory limitation at the time of reviews of 400 boards - and now todays DDR 400 can do cas 2.0 fine on all boards?

Anyone tried the Nforce 2, (A7N8X) with "proper" CAS 2 DDR 400 (which is now available from Corsair etc, using Samsung chips - as low as 80$ US per 256mb stick)

Anyone?

More info below.

Google link here

Another google link

See here also, the original review where ALL the 333 sticks beat the 400 sticks 🙁 🙁 ! ! !
 
Originally posted by: motoamd Read the 1st paragraph where Anand talks about KT400, his review has CAS 2 w/nForce2
Click to expand...
Originally posted in Anand's review "We started out by comparing nForce2 to VIA's KT400, however we quickly realized that motherboards based on VIA's KT400 chipset aren't as mature as some of the better KT333 boards (a good KT333 board will outperform a KT400 board at this point). So instead we used the EPoX 8K3A+ (which ended up outperforming the KT400s we originally tried) and didn't bother with DDR400 tests on the VIA platform since we've already proved that the higher bandwidth memory does nothing for the Athlon XP. "
Click to expand...

What are you reffering to specifically???
I hope it's not
"since we've already proved that the higher bandwidth memory does nothing for the Athlon XP. "
Click to expand...
I specifically intend to put a Barton on this board in 6-12 months, so I NEED DDR400 memory as the barton will be a 400 FSB cpu.
ALSO if I overclock whatever CPU I have on there now to SOMETHING x 200mhz (dual channel or not, shrug) I would like my 200mhz to be faster than 166mhz (333) - I don't want some "silly" bug / problem giving me hassles / slow ram bandwidth! 🙂

I want the best speed I can get from DDR400 ram, since several manufacturers now sell CL2 (cas2) DDR400 memory, surely this SHOULD outperform DDR333 Cas2 (cl2) memory right, if this is the case - then that's what I want - and that's what I intend to research now.
 
The KT 400 motherboards do better than the KT 333 motherboards with the same memory, it is just the memory that doesn't work as well. I would get an ASUS KT400 chipset motherboard (best in the biz), get 333 CAS2 memory for now, and wait for a while.
 
Originally posted by: Markfw900
The KT 400 motherboards do better than the KT 333 motherboards with the same memory, it is just the memory that doesn't work as well. I would get an ASUS KT400 chipset motherboard (best in the biz), get 333 CAS2 memory for now, and wait for a while.
Click to expand...


Look I hate to disregard your post, but it's nothing along the lines of the answers I'm looking for.
I will NEVER use another Via board, I SPECIFICALLY will only use the nforce 2, due to the sound features, the speed increase over the via board, the dual lan, the sata (asus board), and the fact it WON'T have bloody "VIA" written on the chipset / manual anywhere (ick)

Now I don't want an nforce2 vs KT400 thread, what I want to know specifically is - will DDR400 CAS 2 work WELL in an nforce 2, or will DDR333 cas2 outperform it due to some kind of "silly problem" as seen in the links provided in the first few posts?

Also your solution is not cost effective, why would I buy 333 cas2 memory now, then I need to replace it with 400mhz ram for barton - some of us like the cost effective solution - which will last 12-18 months, not 3-6

- thanks anyhow.
- AbRASiON
 
Guess I should have pointed it out a little bit more.

The KT400 is an immature chipset. The memory controller doesn't do a very good job comparing to the KT333 chipset. That's why VIA plans to come out with the KT400A chipset.

The nForce 2 should perfrom with DDR400 CAS2 (w/200 MHz FSB) better than DDR333 CAS2 (w/166 MHz FSB).

Did I get this time? 😉
 
Originally posted by: motoamd
Guess I should have pointed it out a little bit more.

The KT400 is an immature chipset. The memory controller doesn't do a very good job comparing to the KT333 chipset. That's why VIA plans to come out with the KT400A chipset.

The nForce 2 should perfrom with DDR400 CAS2 (w/200 MHz FSB) better than DDR333 CAS2 (w/166 MHz FSB).

Did I get this time? 😉
Click to expand...


This answer is definately what I'm looking for - I'd prefer a link with evidence of this etc but It's good to hear that the main problems I've been hearing is due to some kind of issue with the KT400, the way it was originally worded when I first heard it (god knows where, I think it was HardOCP to be honest) made it sound like a DDR 400 limitation or some such.

Now, what's your comments on this though!?!?


Update :

Some users were asking me why I didn't run it in SYNC mode. Yes running in SYNC DDR266 mode is faster than running ASYNC DDR333 mode. That is to say that nForce2 chipset is just like SIS 745 and ALiMagik1 which can't perform well in Async settings. See the results below :

The test below is done based on the following configuration :

AXP 1900+
2 x 256M Twinmos DDR400
Ram timings : CAS 2, 2-2-5, AGP Aperture 256M

SIsoft memory benchmarks

ASYNC DDR333 Dual (133:166) = 2073/1966
SYNC DDR266 Dual (133:133) = 2119/2021

SuperPI (1M)*

ASYNC DDR333 Dual (133:166) = 1min 6secs
SYNC DDR266 Dual (133:133) = 1min 4secs

* the lower the calculation finishes, the better it is.

Q3

ASYNC DDR333 Dual (133:166) = 227.9fps
SYNC DDR266 Dual (133:133) = 233.2fps

Codecreatures Benchmark Pro

ASYNC DDR333 Dual (133:166) = 23.9fps
SYNC DDR266 Dual (133:133) = 24.0fps
Click to expand...




What's with that, and can someone run "sync" mode DDR400 when the cpu is running @ 400 ??
Anyone know what's down with that, I want this to be the definitive "how to get the nforce 2 REAL FAST" thread damnit 🙂

- AbRASiON
 
Got any information on that whole "sync" thing I mentioned though?

Hang on a minute,.... according to your link that you provided, it's exactly what I originally suspected - the DDR333 is beating the DDR400?!?!

am I misreading what you linked??
 
Originally posted by: AbRASiON
Got any information on that whole "sync" thing I mentioned though?

Hang on a minute,.... according to your link that you provided, it's exactly what I originally suspected - the DDR333 is beating the DDR400?!?!

am I misreading what you linked??
Click to expand...

Err... *doh* I misread it myself, thinking the DDR400 was faster.... crap.

I'm going to read that review to see if they say anything about the difference.
 
Sandra benchmark (ram) ddr400 wins (ocworkbench.com)


Toms h/w UT2003, DDR400 gets spanked by DDR333
Quote from end of this review also (which really weirds things up!)

"As you can see, some 3D applications gained performance considerably when running DDR400. That's why power users should definitely go for DDR400 memory - even though the advantage is not realized with standard games and applications. People that run high-end 3D applications like Design Review or 3D Studio Max will clearly experience more performance with the faster memory. All others can safely run DDR333 - but no matter which type of memory you want to use, do not accept anything other than CL2 mode DIMMs! The performance gap to CL2.5 is as big as if you exchanged your Athlon XP 2200+ for the 2100+ model! "


They fail to mention DDR 400 is slower in games, how weird is that?
Perhaps this has something to do with that "sync" mode (he reviewed the DDR333 in Sync mode) but not the DDR400??

That whole sync mode thing @ ocworkbench (again)
 
Yeah, read that summary, it seems as though only the DDR400 beats DDR333 in SPEC but not really much in anything else.

I don't understand the 'sync' and 'async' myself, as to what is the difference between them besides the numbers listed.
 
I've just pumped off an email to Tomshardware, maybe they can shed some light on this, since they have a tendency of "jumping on issues" if they spot any (personally I have no issues with Tom's / Hardocp or anandtech)

Email I sent listed below


> -----Original Message-----
> From: XXXXXXXXXXXX
> Sent: Tuesday, 3 December 2002 3:42 PM
> To: 'comment@tomshardware.com'
> Cc: 'patrick@tomshardware.com'
> Subject: Problem with Nforce 2 review?!
>
>
> http://www6.tomshardware.com/mainboard/02q4/021111/nforce2-16.html
>
> "As you can see, some 3D applications gained performance
> considerably when running DDR400. That's why power users
> should definitely go for DDR400 memory - even though the
> advantage is not realized with standard games and
> applications. People that run high-end 3D applications like
> Design Review or 3D Studio Max will clearly experience more
> performance with the faster memory. All others can safely run
> DDR333 - but no matter which type of memory you want to use,
> do not accept anything other than CL2 mode DIMMs! The
> performance gap to CL2.5 is as big as if you exchanged your
> Athlon XP 2200+ for the 2100+ model! "
>
>
> I notice the majority of the real world apps which are not
> bandwidth intensive are actually SLOWER on the DDR400 memory.
> I have recently read this small update on ocworkbench.com
>
> "Some users were asking me why I didn't run it in SYNC mode.
> Yes running in SYNC DDR266 mode is faster than running ASYNC
> DDR333 mode. That is to say that nForce2 chipset is just like
> SIS 745 and ALiMagik1 which can't perform well in Async
> settings. See the results below"
> (from http://www.ocworkbench.com/2002/asus/a7n8x/a7n8xp1.htm )
>
>
>
> I'm curious if someone there could advise me if you ran your
> DDR333 in "sync" mode or not in "sync mode" - and same for
> your DDR400?
> Surely on a properly set up computer, DDR400 will beat DDR333
> in everything (if CAS2 is applicable in both situations)
>
> If this is not the case, then perhaps there is an issue to be
> brought up with nvidia?
>
> thanks for your help (if you have time) - I'd love to get to
> the bottom of this anomalie.
>
> - Scott
Click to expand...
 
Originally posted by: motoamd
Yeah, read that summary, it seems as though only the DDR400 beats DDR333 in SPEC but not really much in anything else.

I don't understand the 'sync' and 'async' myself, as to what is the difference between them besides the numbers listed.
Click to expand...

Well that's what I intend to find out, what the jebus is this whole sync / async thing and how can I get more information on this.
Is it an asus only feature?
Is it an nforce2 only feature?
Does it only work on 266 / 333 memory or all memory (400? etc)

I need answers, with answers I can solve problems and create plenty fast PC!
(wheee)
 
I've fired off another email, this time to the webmaster of ocworkbench.com in the hope of shedding more light on the sync / async issue so we can get some solid answers.

(my ultimate goal is to see DDR 400 @ cas2 with this "sync" thing on beating DDR333 in ALL benchmarks)

Email below



> -----Original Message-----
> From: XXXXXXXXXXXXX
> Sent: Tuesday, 3 December 2002 4:03 PM
> To: 'webmaster@ocworkbench.com'
> Subject: Asus Nforce 2 "sync" ?
>
>
> Hi,
>
> I note your review of the Nforce 2 you have updated with the
> information on Sync and non Sync (async?) mode on the nforce 2.
> http://www.ocworkbench.com/2002/asus/a7n8x/a7n8xp1.htm
>
> "Some users were asking me why I didn't run it in SYNC mode.
> Yes running in SYNC DDR266 mode is faster than running ASYNC
> DDR333 mode. That is to say that nForce2 chipset is just like
> SIS 745 and ALiMagik1 which can't perform well in Async
> settings. See the results below"
>
>
>
> I would like to know if this "sync mode" can be turned on for
> DDR400 memory and cpu's (overclocked) to 400mhz (example an
> unlocked T-Bred B cpu)
> 11.5x200 = 2300mhz - Tbred B can do 2200mhz no problem.
>
> Can you tell me a little more about the "sync mode" and
> "async mode" as I am in the process of trying to work out why
> the performance in DDR 400 is slower than 333 in several reviews!
>
> We are discussing this at the moment in a thread on anandtech.com
> http://forums.anandtech.com/messageview.cfm?catid=27&threadid=
> 922891&STARTPAGE=1
>
>
> I'd love it if you could respond with some information on
> this situation, as I'd like to see DDR400 (CAS2) performing
> better than DDR333 (cas2) across the board (this is NOT the
> case in Tomshardware's roundup)
>
>
> Thank you for your time.
>
Click to expand...
 
Response from OCworkbench

Ok.. Sync means that your CPU and RAM is running at the same speed
e.g. CPU at 133mhz fsb, RAM is running at 133 MHz (e.g. DDR266) in SYNC
If you o/c your CPU to 166, RAM will run at 166 (e.g. DDR333) in SYNC
if you o/c your CPU to 200mhz, ram will run at 200(e.g. DDR400) in SYNC mode

Unfortunately, nForce2 supports Async mode, that is to say
you can run
the CPU fixed at 133 and RAM at 166 or 200.

It is proven that when you run in Async mode, performance dips and it's "NOT
RECOMMEND"
by NVIDIA. VIA KT333/KT400 Async mode seems to over come this problem.
Click to expand...

Well that sounds exactly like I suspected, and it seems to indicate it will work for DDR400 mode, now all we need is for someone to test this on the Nforce 2 with Cas2 DDR400 to see just how good this benchmark is (whomever does it, most likely will need a very low multiplier CPU or an unlocked one!!!)
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top