nForce2 delay on purpose?

[pelly]

Could NVIDIA be waiting for the 333FSB launch in October to release its nForce2???

Sounds like perfect timing to me....I hope it's sooner though...

:D

[saturnotaku]

Maybe they're looking to iron out bugs, etc to prevent problems people were having with the first nForce release. I hope that's the case because I'd hate for NVIDIA to have history repeat itself.

[LuckyNV]

nvidia seem liek the company to learn from their mistakes though...

[koneill]

There's no delay. Board manufacturers are just going into full production swing and product should be out very shortly.

[TheOneKEA]

Why wait for the 333MHz Athlons? If nVidia gets their nForce2 out now, bug-free and with features that people want, then all we'll have to do is add the processor to get a 33MHz FSB system. If they wait, people might not want the processor with the chipset and may wait a while.

[netviper13]

Well it makes a lot of sense that nVidia would make absolutely sure their boards are compatible with 333 FSB before releasing them.

[pelly]

Why wait for the 333MHz Athlons? If nVidia gets their nForce2 out now, bug-free and with features that people want, then all we'll have to do is add the processor to get a 33MHz FSB system. If they wait, people might not want the processor with the chipset and may wait a while.

They (NVIDIA) could take advantage of the publicity of the AMD-announcement/launch and ride the marketing wave to the bank.

We all know NVIDIA will be waiving the 333FSB flag....we all know it already supports it...

[Dazz]

Yeah if you saw they video all they could go on about was the Athlon's 333Mhz FSB more so then the built in video & sound lol.

[retsam]

i think nvidia is going to have a huge release at fall comdex this year... or at least im hoping that they will. all these new products nforce 2 nv30 nv28 nv 18 why not release them all at once at the biggest trade show in the u.s. hmmm just my two cents




rets

[TheOneKEA]

Good idea. It would cause marketing overload and force people to maybe buy a combination of all three. Coolness :)

[zoolap]

Originally posted by TheOneKEA
Good idea. It would cause marketing overload and force people to maybe buy a combination of all three. Coolness :)

That said some might just get fed up waiting and buy something else.

[TheOneKEA]

Maybe. Then again, maybe not - and the people who don't buy it will probably research it anyway and will eventually buy it at a later date. My father is like that; he'll research something for a long time before buying it; he does that to make sure he's getting what he wants.

[CainSyris]

I think they, along with the rest of motherboard chipset developers, are having trouble with validating PC3200 modules. VIA and SIS have as much as admitted they cannot guarantee reliability with any certainty. So I'm sure nVidia is not immune to this lack of great PC3200 DDR memory.

Or perhaps they went on vacation after the big paper launch for nForce 2, perhaps to the beach? "Woohoo, guys! We've launched the motherboards! Time to go get some sun!"

"But... Bob... we haven't actually shipped any of the chipsets?"

"So what, Joe? Let the motherboard ...peeps handle that! I'm gone.."

"But wait! We haven't actually made any yet... and ... and... they can't do nothing without some chipsets..."

"I can't hear you! I'm taking the secretary out for lunch... won't be back until November!"

"NOOOOOOOO!!!"

And now, the silence. No nForce 2 boards despite being "...available soon" in July. I suppose last year's launch of nForce some six+ months after the paper launch was "soon," too at this rate. ;)

Is there anyone surprised though? Some of them got rashes from the sand and sun.

nVidia has definitely been dropping balls left and right, but I'd never have thought they'd go this long without releasing nForce 2. That's the only thing they've got going to cover the potential disaster with the nv30 release and lower stockprices. Not that they will die. They won't. But certainly, they're not having the quarter of greatness this time, either. They go from a great GF4Ti success rate to ... this. Hell, they haven't even released the AGP8x versions of the GF4Ti line. That shouldn't have taken them this long.

So why do I think nForce 2 is delayed? Because nVidia is having the same problems with DDR400 that the other motherboard chipset developers have. The memory's just not available that can beat good PC2700 DDR with the highest settings.

Quite frankly, the performance of KT400 is good enough of a change to me, though it is not the level of what the KT266a was over the KT133/a. Not without a new memory interface will we see that again. A new memory interface and a new frontsidebus performance to go with it.

So nForce 2 should just go ahead and launch without DDR400 if it can't get it right. Nobody can. Interestingly, though, they should not have killed that nForce+ version they were originally going to release because *obviously* they couldn't get this one out in any good time at all. Then, they'd have something to push as new today.

Hopefully, they'll sort themselves out sometime very soon now, as their complete lack of anything (excepting Detonator 40's) to sell in the last two months as new or revolutionary or even evolutionary in the face of so many other things being all these things by the competition, well, I think they need to fight back. Because I hate to see any lack of competition. Competition is my friend. Lack of competition is sterile and emptiness. And inflated prices.

[TheOneKEA]

You're right. I most definitely agree with you on the DDR400 issue; it seems literally that nobody can seem to create a bugfree, fast, useable DDR400-capable memory controller. VIA's is a mess that only works properly up to DDR333; nVidia's controller is missing entirely, and AFAIK SiS and Intel aren't doing well on those fronts either. Somehow DDR400 seems to be more trouble than it's worth, and like the GF4 MX 460, I'm convinced it will be forgotten once DDR2 is available (WAM......mebbe that's why VIA and nVidia are waiting...............well?)

[StealthHawk]

what problems are apparent with SiS's DDR400 controller?

that includes performance or general buginess(difficulty working with certain reputable modules)

[TheOneKEA]

What you said - validating all the available DDR400 modules to make sure they work correctly.

[koneill]

everyone's talking about DDR400 memory controller... blah, blah, blah.... the truth of the matter is that there's nothing wrong with NVIDIA's memory controller, and there may or may not be a problem with VIA's. i don't know as i haven't tested a board.

the problem is with DDR400 memory; it's latency is significantly higher than that of DDR333. no one makes a cas2 (or even 2.5) DDR400 module, so you're talking at least 3. add to that other memory timings are far less agressive than DDR333's and you're talking about spinning your wheels, if not putting it in reverse.

DDR333 is still the sweet spot. DDR400 will do you no good for either chipset.

[StealthHawk]

Originally posted by koneill
everyone's talking about DDR400 memory controller... blah, blah, blah.... the truth of the matter is that there's nothing wrong with NVIDIA's memory controller, and there may or may not be a problem with VIA's. i don't know as i haven't tested a board.

the problem is with DDR400 memory; it's latency is significantly higher than that of DDR333. no one makes a cas2 (or even 2.5) DDR400 module, so you're talking at least 3. add to that other memory timings are far less agressive than DDR333's and you're talking about spinning your wheels, if not putting it in reverse.

DDR333 is still the sweet spot. DDR400 will do you no good for either chipset.

quite frankly, until we see Athlons with a 333MHz FSB, DDR400 is nothing but a bullet point for marketing and hype unless you overclock your FSB to speeds that could actually make use of the bandwidth.

[pelly]

no one makes a cas2 (or even 2.5) DDR400 module, so you're talking at least 3

Here's the condensed list ( all available from newegg ):

Geil:

GeIL DDR RAM 256MB PC-3200 400Hz Ultra- CAS2 1T
GeIL DDR RAM 512MB PC-3200 400Hz Ultra- CAS 2, 1T Command
GeIL DDR RAM 256MB PC-3500 433Hz CAS 2.5, 6-3-3 2T Timings


Corsair:

256MB PC3200 DDR RAM - OEM 32x8 184-Pin, CAS2.5
256MB 32MX8 PC-3200C2 CAS Latency 2-3-3-6-T1
512MB 64MX64 CAS 2.5 PC-3200PT With Platinum - Silver Heat Spreader OEM
512MB 32MX8 PC-3200 CAS2 With Heat Spreader

Mushkin:

MUSHKIN DDR 256MB PC-3200 - OEM
AMD, VIA, ALI Approved, PC3200 = 333MHz FSB, Rated 2.5-3-3, 6ns

Kingmax:

KingMax DDR memory 256MB PC3200 MPXB62D-KX OEM
Specifications:
Performance range - 200MHz ( DDR400, CL2.5 )




:D

[koneill]

Pelly... you've been sucked in:

As for the Geil, none of that memory is true DDR400. It's DDR333 that has "been tested" to run at 400MHz. Good luck.

None of Corsair's DDR400 is genuine as of yet. I know. I buy from them regularly. Until they start getting ICs from Winbond, it's the same as the Geil (although probably somewhat more reliable).

As for the Mushkin: PC3200 = 333MHz FSB, Rated 2.5-3-3, 6ns

PC3200 does NOT = 333MHz FSB. PC3200 denotes a 400MHz FSB. That memory is DDR333 (PC2700), again, tested to run at 400MHz (it will only work at 2.5-3-3 (as stated) at 333MHz (also stated).

The Kingmax memory is not yet available at 2.5. It should be available shortly.

So many get sucked in and then wonder why their systems' performance or stability is so poor. The wool is being pulled over their eyes. Just because it says DDR400 doesn't make it DDR400.

[pelly]

Pelly... you've been sucked in:

Ok....let's clear this up asap......

Aside from the lack of an "official" or "genuine" JEDEC specification which would essentially fine-tune a module running at 400MHz ( focus on eliminating noise issues especially ), what constitutes DDR400? In my book, it's a module that runs 400MHz.

This "genuine" memory you mention doesn't exist as the JEDEC standard has yet to be finalized....In other words, there is no official DDR400 available by anyone right now.

To focus on this issue more, let's examine the Corsair XMS modules since you are "familiar" with them. Specifically, let's focus our attention to the CAS2 PC-3200. ( CMX256A-3200C2 ).

This memory has been verified to be operate at 200MHz ( 2-3-3, 1T command rate ) and 166MHz ( 2-2-2, 1T command rate ).

100% tested at 200MHz in high performance motherboards.

The way I read these claims, their modules will run at 400MHz. Since my definition of DDR400 is theability to run at 400MHz, this memory ( increased noise and all ) qualifies as DDR400 in my book....

As for your claim that Mushkin PC3200 is only good for 333MHz.....here is a tiny bit of info from Mushkin's website:

Our PC3200 DDR SDRAM is run through a speed bin sort to insure operation at 200MHz memory bus (400MHz DDR).

Evidently, you think the PR monster is out to steal your money...Fortunately, you're wrong and these companies are not about to risk losing the reputation that took them so long to earn.....

In case you think I'm talking foolishness, take a look at the happy customers from newegg who bought Corsair PC3200 memory....That is, unless they're talking foolishness as well....

http://secure.newegg.com/app/CustratingReview.asp?item=20-145-405

:D

[pelly]

http://www.overclockercafe.com/Articles/DDR400/index.htm

This review might illustrate what you were trying to say....The Xtreme memory is garbage...However, this also gets my point across...The Corsair does what it says it will do...runs at 400MHz...

:D

[koneill]

first, don't tell me what i think. i know what i know. you don't know what i think, so leave it at that.

there is a spec for DDR400 just as there's a spec for DDR333. the spec hasn't been finalized, but that's a different matter.

i'm very "familiar" with corsair's memory; do not patronize me. i don't appreciate it, especially when it comes out of ignorance.

As for your claim that Mushkin PC3200 is only good for 333MHz

i never said that. read my post again. don't look at mushkin's words. comprehend them. the memory is tested at 2.5 at 333MHz - NOT 400MHz.

corsair makes fine memory, but that it 'runs' at 200MHz does not make it DDR400. it's as simple as that.

[netviper13]

People, this same thing happened with the PC2700 (DDR333) specs. Modules were released before the JEDEC spec was finalized, so they were just guaranteed to run at that speed. Then the spec got finalized and we now have official DDR 333 modules. Same will happen with DDR 400.

[pelly]

i never said that. read my post again.

Er...ok...one more time for the girls and boys watching at home...

You said
As for the Mushkin: PC3200 = 333MHz FSB



To top it off....you go on to say....
don't look at mushkin's words. comprehend them. the memory is tested at 2.5 at 333MHz - NOT 400MHz.

That's interesting b/c Mushkin explicitly says on their website ( about this memory )
Our PC3200 DDR SDRAM is run through a speed bin sort to insure operation at 200MHz memory bus (400MHz DDR).

I could go on with this...but it is evident that you are not here to hold a rational conversation...For better or for worse, I stick to the things I say...if someone proves me wrong, I am adult enough to admit it and move on....

In this case, I proved you wrong and now you're back pedaling faster than a mongoose on speed...

BTW, what is this mythical DDR400 if it isn't a low-noise module running at 200MHz???

The answer? It IS a low-noise module running at 200MHz.....only some "official" DDR400 will have more headroom than the Corsair modules...

I've said and seen more than enough here.....you'll have to find someone else to play with...I apologize if this whole thing sounds catty.....Lotsa stress at work as of late...patience is next to nothing as of late...

:eek: