Nvidia’s High-End Chipset Under Fire for Problems
Posted by Daniel Fleshbourne on 15 December 2006 - 13:30 · 44 comments & 19594 views
- Advertisement
-
-
#1 Posted by denzilla on 15 Dec 2006 - 13:41
- When I first built my PC, I was getting event ID51 errors all over the place using Nviida's IDE drivers. That was with a base model nforce4 chipset, so it must be a defect in their overall design.
-
#2 Posted by pollaxe on 15 Dec 2006 - 13:49
- It's a very odd problem - people seem to have either 100% rock solid boards or complete flake-outs...
-
(1 reply)
#3 Posted by undomiel on 15 Dec 2006 - 14:07
- nvdisp.dll infinite loop error -- still not fixed these many years later
-
(4 replies)
#4 Posted by ScottishLad on 15 Dec 2006 - 17:26
- I had to buy a new Hard disk for my Nforce 4 system. I had a Maxtor Maxline III 250GB sata drive. Been working fine in my server for the last 6 months. Wont work at all in my nf4 system. Crashes within 5mins of turining on.
That was not using Raid. Just a standard Sata Install. I cant believe they havnt fixed these issues yet. -
#4.2 Posted by ScottishLad on 16 Dec 2006 - 15:48
- I've been though it all
Including turning NCQ off. It only prevents it happening within minutes. Instead it happens after a few hrs 
Trust me, I spent weeks on it and got not very far. Talking in a 400+ post thread on the nvidia forums
I wonder if the firewall is still broken too? -
#4.3 Posted by trparky on 16 Dec 2006 - 17:46
- Probably the firewall is. Rumor has it that future versions of the nVidia chipset won't have the firewall chip.
-
#4.4 Posted by ScottishLad on 17 Dec 2006 - 13:33
- Quote - trparky said @ #4.3Probably the firewall is. Rumor has it that future versions of the nVidia chipset won't have the firewall chip.
Thats a disappointment. It has alot of potential
-
#5 Posted by CheeseCow on 15 Dec 2006 - 17:39
- I think that's a bit vague. Is it confined to the SATA or the RAID controller, or is it affecting whole motherboards? Personally I think the idea of having 2 cards running in paralell seems a bit stupid, there must be quite a bit of duplicate functionality then...
-
(1 reply)
#6 Posted by xMorpheousx416 on 15 Dec 2006 - 17:46
- As ex-Quality Manager for the company I worked for, I can tell you how simple the problem really is.
Speed kills.
Especially in the technology industry. The faster you make your product, the faster you get it thru the assembly line...and the faster you make the quality checkpoints (and in the case of CPU's, the faster the clock cycle)...it's more then inevitable that you will run into problems.
IMHO, it's time for the technology industry to slow down, take a deep breath....and start back up with a steady stream of quality products instead of trying to out-do themselves every
"week" with a new product. -
#6.1 Posted by toadeater on 15 Dec 2006 - 19:20
- Quote - xMorpheousx416 said @ #6IMHO, it's time for the technology industry to slow down, take a deep breath....and start back up with a steady stream of quality products instead of trying to out-do themselves every
"week" with a new product.
It's a shame that Nvidia is charging a full $100-250 extra for their 680i boards compared to the p965 boards and they left this SATA problem. If I bought one of those Asus Striker 680i boards for $400 and had this happen, I would go on a shooting spree inside the Asus HQ.
There's a cheaper 650i chipset out, btw, I wonder if it has this same bug?
Still, all the first revision Core 2 Duo boards seem to have major problems with their BIOSes, the JMicron SATA-IDE controller hack, and DDR2 RAM not working at specified voltages. There's alot of buggy junk out there right now, even from companies that are supposedly considered "reliable."
-
#7 Posted by leebobs on 15 Dec 2006 - 17:46
- This is not the first time... google for nForce 4 and Maxtor Hard disks
-
#8 Posted by Crucify on 15 Dec 2006 - 18:19
- This has nothing to do with nforce 4. It's the new 680i chipset thats got major issues.
I had to rma my Asus striker mobo over this very problem only to have it again and worse with the replacement.
I gave up, boxed it for now and slapped my P5B Deluxe back in.
-
#9 Posted by NXTwoThou on 15 Dec 2006 - 18:34
- I got my new machine a few weeks ago, it has an EVGA 680i with mirrored SATA drives and so far everythings been running just fine with XPSP2.
-
(3 replies)
#10 Posted by WinMacLin on 15 Dec 2006 - 19:15
- I have this motherboard and I have these problems. SATA / IDE Disk errors in the Event Log.
Some people on the EVGA site have said there systems have been stable for 3+ weeks and then they check there event log and see all these Disk Errors which no doubt have caused some kind of data corruption. I hope NVIDIA fix this soon because I'm personally at the end of my rope with this motherboard. -
#10.1 Posted by trparky on 16 Dec 2006 - 02:32
- As I pointed out in an earlier comment, try turning off NCQ in the Device Manager.
-
#10.2 Posted by WinMacLin on 16 Dec 2006 - 05:01
- Quote - trparky said @ #10.1As I pointed out in an earlier comment, try turning off NCQ in the Device Manager.
Please don't give me advice unless you read the EVGA thread. No offence but I've been trouble shooting this for over 2 weeks now and that was done on day 1 not day 14.
-
#11 Posted by hapbt on 15 Dec 2006 - 20:13
- A friend of mine has been having problems with his nForce chipset for years, I was wondering how long it would be until it got more mainstream attention. He had to turn off some of the more advanced features of his SATA controller to get rid of all the errors, so now his SATA performs like a regular ATA -- great.
I love their graphics chipsets but yeah, this stuff sucks.
-
#12 Posted by g0wg on 15 Dec 2006 - 20:47
- oh f...ing great ... i just bought a evga 680i :/
edit: the issue is with just SATA in RAID mode? or even normal SATA operation is affected? Are IDE drives affected? Guess I will have to stick my SATA drives in icyboxes for the time being.
Last edited by g0wg on 15 Dec 2006 - 20:53
-
(3 replies)
#13 Posted by WinMacLin on 15 Dec 2006 - 21:36
- SATA, IDE, RAID, are all effected. You still get the errors even if you use a PCI SATA/IDE card or do/don't install NVIDIA drivers.
Sometimes the computer will even lockup installing Windows. It's a MCP error not the South bridge. -
#13.1 Posted by g0wg on 16 Dec 2006 - 00:42
- damn
is there anything useful in the thread in EVGA's forum? like when they are gonna release a bios? -
#13.2 Posted by WinMacLin on 16 Dec 2006 - 01:54
- Quote - g0wg said @ #13.1damn
is there anything useful in the thread in EVGA's forum? like when they are gonna release a bios?
NVIDIA have said it's a "Driver Issue" perhaps they mean Software/Bios because we already know it's nothing to do with the Drivers. An EVGA support staffer told one consumer (over the phone) that there would be a fix released in the first week of January however in the support Forums a different EVGA staffer refuted this claim as incorrect.
-
#14 Posted by krupini on 15 Dec 2006 - 22:03
- Lol, i had hard drive corruption errors on NF4 and it was widely reported. Several bios and chipset updates later, I can safely install nforce ide/sata drivers and everything would work ok, but not with nvidia firewall enabled. Thats still a no no as it brings back the hard drive corruption problems. To this day its not fully fixed. And now nforce 680i has the same problem. Different chipset, i thought they would have nailed it by now.
-
#15 Posted by obsolete_power on 15 Dec 2006 - 22:51
- Ahh this really blows, I hope they have some sort of recall system going because people paid a lot of money for these boards...
-
#16 Posted by shift4 on 16 Dec 2006 - 02:07
- Works fine for me. 2 Seagate 7200.10 320gb drives worked fine with both raid 0 and raid 1.. Presently running raid 0.
EVGA 680i.
-
(2 replies)
#17 Posted by ANova on 16 Dec 2006 - 06:23
- I don't trust any chipsets other than Intel's and I certainly wouldn't spend anywhere near what they're asking for this crap. I am continually happy without nvidia products in any of my machines.
-
#17.1 Posted by MrCobra on 16 Dec 2006 - 07:20
- A lot of people would also make the claim that Intel's chipsets are crap as well. People that jump on new tech the very second it comes out deserves to have problems with it. No one can ever be patient and wait to see how it's doing before buying.
-
#18 Posted by dhitb on 16 Dec 2006 - 07:15
- Oh just freaking great, they've inherited the m5288 timeout error that still has not been fixed in some ULi M1575 boards (including my CFX3200).
-
#19 Posted by +rm20010 on 16 Dec 2006 - 11:39
- For crap's sake, can't they get their chipsets done right. I've had my run-in with disk corruption/PC lockup problems in April because of this nForce4 chipset.

-
(1 reply)
#20 Posted by just_laze on 16 Dec 2006 - 12:42
- Hello!
The Abit NF7-S was notorious for having S-ATA 'issues'.
Admitedly it was a Silicon Image add-in chipset but it still negatively impacted nVidia's ability to produce a stable S-ATA platform due to association.
laze. -
#20.1 Posted by fluxcapacitor on 16 Dec 2006 - 18:23
- I'm running an NF7-S for almost 2 years on the same format now. But I'm running IDE hard drives in RAID via the SATA adapters. It did take me alot of driver troubleshooting to find one that wouldn't corrupt files though.
-
#21 Posted by PsykX on 16 Dec 2006 - 16:36
- Every part of a PC is starting to boil... I think there's a problem with either heat dissipation, cooling or the loss of heat by composites...
-
#22 Posted by XsBones on 16 Dec 2006 - 17:55
- I have the 680i board and never have had a prob. with anything.
-
(1 reply)
#23 Posted by WinMacLin on 17 Dec 2006 - 08:54
- Just thought I would mentioned that NVIDIA have now released a BETA Bios for owners of the EVGA 680i board. Which they say fixes this problem!! *WOOT*

http://www.evga.com/community/messageboard...mp;whichpage=30
This patch should work on any 680i board that uses the reference design. IT WILL NOT work on the Striker or the DFI or any similar boards that are not using the NVIDIA reference board!! you have been warned!
-
#24 Posted by eilegz on 17 Dec 2006 - 17:17
- Nvidia and ATI now should focus in performance but also in energy, thos new vcard and chipset consume more energy, emits more heat and if it continues like that not even water cooling willl help to cool down those beasts
-
(1 reply)
#25 Posted by +aniv on 17 Dec 2006 - 19:49
- IT HAS BEEN FIXED BY EVGA with a BIOS Update!!! PLEASE USE GOOGLE/SEARCH NEXT TIME BEFORE POSTING NEWS!!!
-
#26 Posted by NCalliari on 18 Dec 2006 - 14:37
- It hasn't fixed it for everyone. Who knows what they actually 'tweaked' in this BIOS. Some people can't overclock anymore, others are still getting corruption. This issue is far from over and I sure as hell would not buy this board at the moment. At least wait until it is re-reviewed by some reputable sites, using the new 'Final fixed BIOS'.
Just remember that nvidia NEVER actually fixed the problem with nforce4 boards, they just botched a workaround.
History seems to be repeating itself with the 680i!
Submit to reddit
Submit to blinklist
Bookmark on del.icio.us
Add to furl
Share on Facebook
Add to Windows Live
Users in several forums, particularly, in EVGA and Nvidia tech support forums, report about “lock up” and “disk error” issues with Serial ATA hard disk drives and RAID capabilities of the Nvidia nForce 680i SLI core-logic that sells for $120 per two chips alone. Some end users even cannot install Windows XP operating system, whereas others could not use their systems flawlessly for long and some even report data corruption.
Even though the majority of users have stable mainboards, the number of those, who purchased mainboards based on Nvidia nForce 680i chipset and now report instabilities seems to be significant and the problem – widespread. The issues do not seem to have relation to overclocking or Serial ATA working modes. In fact, users reported problems with RAID in case of previous-generation Nvidia nForce chipsets as well, but, perhaps, earlier the problems were not faced by a significant number of users.