Nvidia: High-End Chipset Problem Solved, End-Users Disagree

Nvidia Corp., a leading developer of core-logic sets and graphics processors, said that the problem with Serial ATA controller on its latest high-end core-logic sets is solved, however, some users are still reporting problems with Serial ATA operation on systems running nForce chipsets from the company. Earlier users in several forums, particularly, in EVGA and Nvidia tech support forums, reported about "lock up" and "disk error" issues with Serial ATA hard disk drives and RAID capabilities of the Nvidia nForce 680i SLI core-logic. Some end users even could not install Windows XP operating system, whereas others could not use their systems flawlessly for long and some even reported data corruption. The majority of customers, however, reported no problems. Later on EVGA and Nvidia issued a BIOS version that should fix the problems and make the systems more stable.

"The problem that some users were experiencing related to signal timings on the motherboard. The new BIOS (P32) corrects this," said Drew Henry, Nvidia's chipset chief, in an interview with [H]ard|OCP web-site. Nvidia now blames its internal quality assurance process for not detecting the issues early and claims that in future there will be no such situations when end-users discover issues not seen by the core-logic developer. "We should have caught this in our internal QA process. Unfortunately, we didn't, and we have adjusted internal processes to make sure that this does not happen again," Mr. Henry said.

View: The full story
News source: Xbit Labs

Report a problem with article
Previous Story

How fat is my DRM?

Next Story

SQL Server 2005 Updates

16 Comments

Commenting is disabled on this article.

They just released a beta 2 version of the P23 bios that addresses another SATA issue (HD that have NCQ). We'll see how successful this new beta is at addressing the problem. I have a 680i and have had no big issues, even with 2 seagate 320 GB sata II HD with NCQ enabled. overclocks like a charm. even nTune is working for me. i guess i am one of the lucky ones.

I agree grimson.
BIOS updates will just mask the issue and hide performance settings in the BIOS.
At the end of the day, this chipset is marketed on overclocking potential.
If you remove this potential with a bios update to 'fix' instability, WHY buy this board?!

A new bios will fix it all ...

My guess: no!

These kind of problems are more likely the result of a flawed design.
My guess is that any bios update that would be released to 'masquerade' this problem does the following;
- Enable more failsafe settings so that eventually performance would be sacrificed.

So no bios update ever would not take away a flawed hardware design.
Away is the magic of the ever auto magically all-fix-bios update. People tend to think that a bios update always does the trick (and companies want us believe it)

Quote - grimson said @ #6
A new bios will fix it all ...

My guess: no!

These kind of problems are more likely the result of a flawed design.
My guess is that any bios update that would be released to 'masquerade' this problem does the following;
- Enable more failsafe settings so that eventually performance would be sacrificed.

So no bios update ever would not take away a flawed hardware design.
Away is the magic of the ever auto magically all-fix-bios update. People tend to think that a bios update always does the trick (and companies want us believe it)

I'd tend to agree with you. considering its a very similar issue to the old Nforce4 i suspect it is infact hardware related, They ignored it then and now its finally caught up with them and there not sure what to do about it. From the looks of it the bios updated hasn't fixed things.

Are we talking nforce4 boards here or 680i people? Read the article heading.
All 680i boards have the underlying issue, the problem is visible for most, but will very likely rear it's ugly head further down the line for those who don't currently experience it. It's a fundamental problem with the chipset.

I wonder now if they can fix the Nforce 4 boards which i bet they still don't acknowledge a problem with :(

My Maxtor Maxline III still doesnt work even with the latest bios on my Nforce 4 system. I cant even get windows installed data corruption issues are so bad (Drive is fine btw been running on an nforce 2 for over a year now. AND I DID go though all the trouble shooting so dont bother going down that route with any replies)


What i dont understand is how this slipped though twice and why Nforce4 isnt being fixed when its clearly their issue

I had problems with the Maxtor Maxline III SATA HDs in a RAID array, but it wasn't related to the RAID controller chipset or the chipset driver.

The actual problem was the firmware on the drive. I contacted Maxtor and they supplied updated firmware, which resolved the issue.

Quote - DeanNapper said @ #3.1
I had problems with the Maxtor Maxline III SATA HDs in a RAID array, but it wasn't related to the RAID controller chipset or the chipset driver.

The actual problem was the firmware on the drive. I contacted Maxtor and they supplied updated firmware, which resolved the issue.

Mine came with the latest firmware, I checked this up on them. Datacorruption is very serious still so its definatly not firmware. The raid controller issue might have been resolved by it but stand alone sure isnt and still isnt to this date.

My Current seagate drive has to run without NCQ, As soon as its enabled all goes to hell and within an hour all the data's been trashed. (this is my 2nd nforce4 mobo so thats not the issue.) Definatly an unresolved chipset issue.

Its actually put me off AMD now since Nforce is really the only option for an SLI board.

My next system will either be ATI/ATI GFx/amd or pure intel and nvidia. meh, Was a real disappointment

I have the EVGA 680i board and I do have these issues and this BIOS revision (P23 not P32 as the article says) does not fix the issue only makes it harder to replicate the issue.

NVIDIA suck they can't make a decent chipset to save their life the past 4 chipsets they made have had some kind of Disk error and I'm sick of it I'll never buy another nForce based board again. Fanboys can say what they like about me but I've used there products for many years and every chipset has been half baked. Their Graphics Cards I like but their motherboard making skills are poor.

I have computers all built on nForce2, 3, and 4 chipset boards, and I have no problems with them. To say they make crappy chipsets is false.

Quote - Fanon said @ #2.1
I have computers all built on nForce2, 3, and 4 chipset boards, and I have no problems with them. To say they make crappy chipsets is false.

Then you're not living in the real world.

Quote - WinMacLin said @ #2.2

Then you're not living in the real world.

So that's it? I give you first hand experience, and all you can come up with is that? No, no, apparently I don't live in the real world because my viewpoint differs from yours.

Nvidia hasn't dominated the AMD chipset market because it sucked; it dominated because it is the best chipset line available for that platform, and it's the second best line on the Intel platform. Are they perfect? No. No technology product is. I can't agree with how this situation is being handled, but one problem like this doesn't suddenly turn a good product line to crap. If that were the case, every technology company would be developing "crappy" products.

Quote - NCalliari said @ #1
Definitely not fixed. I would NOT buy a board with an nVidia chipset.

Both the nforce4 (which has been available for months and months) and 680i still have serious issues which nVidia has not fixed and don't seem capable of doing so either!


Bios release thread - people are not happy!

http://www.evga.com/community/messageboard...?TOPIC_ID=23034

Exactly. My friend got one and it still wasn't fixed.

Oh and BTW for NV fans: "Whoopee?"