Just archiving everything related to bumpgate since most stuff has disappeared from Google. Maybe people will finally take the time to read some of it and stop shouting "reflow gpu" at every occasion but rather try to understand what the real issue is…
Interesting parts about the failure analysis are "Why Nvidia's chips are defective" and "Why Nvidia's duff chips are due to shoddy engineering"
May complete this post at a later date with more details.
Basically issue is a combination of thermal dissipation of the chip with hotspot, generating mechanical stress due to thermal expansion, wrong underfill material and wrong bump material.
It's not the solder balls between the chip and the board.
Chips affected are mostly from 2006-2008.
MCP and GPUs are affected, both in laptops and desktops.
Some articles list G86, G86A2, G84, C51, G72, G72M, G73, G72A3, MCP67 and NV42. But these aren't the only ones, some G9x and other MCPs were affected too.
That's some later GeForce 6000, almost all GeForce 7000, almost all GeForce 8000, some early GeForce 9000. Affected MCPs are MCP5x (northbridge only, not southbridge), MCP6x and early MCP7x.
NVidia started producing fixed revision for some affected chips in the summer of 2008. Some defective ones were still produced a bit after that.
Fixed revision of affected chips got a new underfill material that's white rather than dark, so it should be easy to recognize them. They'll typically have a datecode after 0830, and some of them have an odd digit instead of an even digit at the end of the model number (eg. G84-603-A2 instead of G84-602-A2).
Note that some chips were still using the bad bump material with the good underfill material (early 9600M) but they're much less prone to failure than earlier series.
Fixed chips include most G84/G86, most of them are still available at see-ic.
Attached is a copy of the pages if ever links get broken. File is *not* a zip, it's a tar.xz but uploaded with .zip extension, so .zip extension must be removed after downloading.
Interesting parts about the failure analysis are "Why Nvidia's chips are defective" and "Why Nvidia's duff chips are due to shoddy engineering"
May complete this post at a later date with more details.
Basically issue is a combination of thermal dissipation of the chip with hotspot, generating mechanical stress due to thermal expansion, wrong underfill material and wrong bump material.
It's not the solder balls between the chip and the board.
Chips affected are mostly from 2006-2008.
MCP and GPUs are affected, both in laptops and desktops.
Some articles list G86, G86A2, G84, C51, G72, G72M, G73, G72A3, MCP67 and NV42. But these aren't the only ones, some G9x and other MCPs were affected too.
That's some later GeForce 6000, almost all GeForce 7000, almost all GeForce 8000, some early GeForce 9000. Affected MCPs are MCP5x (northbridge only, not southbridge), MCP6x and early MCP7x.
NVidia started producing fixed revision for some affected chips in the summer of 2008. Some defective ones were still produced a bit after that.
Fixed revision of affected chips got a new underfill material that's white rather than dark, so it should be easy to recognize them. They'll typically have a datecode after 0830, and some of them have an odd digit instead of an even digit at the end of the model number (eg. G84-603-A2 instead of G84-602-A2).
Note that some chips were still using the bad bump material with the good underfill material (early 9600M) but they're much less prone to failure than earlier series.
Fixed chips include most G84/G86, most of them are still available at see-ic.
- Class action lawsuit settlement:
http://web.archive.org/web/201106270...ettlement.com/ - 12/04/2007
There are no mobile G84/6 problems
As told from the Nvidia perspective
http://web.archive.org/web/201912192...ere-mobile-g84 - 01/07/2008
HP Pavilion dv2000/dv6000/dv9000 and Compaq Presario v3000/v6000 Series Notebook PCs - HP Limited Warranty Service Enhancement
http://web.archive.org/web/200807010...1087277&dlc=en - 02/07/2008
Nvidia takes charge for bad chips, but who is to blame?
Nvidia takes charge for bad chips, blames TSMC
http://web.archive.org/web/201211170...o-is-to-blame-
Form 8-K for NVIDIA CORP
http://web.archive.org/web/200807100...2/nvda8-k.html - 03/07/2008
Nvidia "opens can of whoop-ass" on itself
Millions of failing parts, income way down
http://web.archive.org/web/201912192...idia-whoop-ass - 04/07/2008
Nvidia contract makers in Taiwan low-key over defective chip reports
http://web.archive.org/web/201912202...0704PD210.html
Rumours abound about real reasons for Nvidia's failing chips
Taiwanese sources chip in with their thoughts
http://web.archive.org/web/201912192...-failing-chips - 05/07/2008
Nvidia 8500M series is the defective chipset just announced
https://forums.macrumors.com/threads...ounced.511433/ - 07/07/2008
Nvidia plays the meltdown blame game
Comment Story doesn't mesh with reality
http://web.archive.org/web/201912192...own-blame-game - 09/07/2008
All Nvidia G84 and G86s are bad
Comment No word on MCPs yet
http://web.archive.org/web/201912192...a-g84-g86s-bad - 25/07/2008
NVIDIA GPU Update for Dell Laptop Owners
http://web.archive.org/web/200808300...op-owners.aspx - 28/07/2008
Ten Dell models have defective Nvidia GPUs
Analysis NV is still stonewalling customers and analysts
http://web.archive.org/web/201912192...ve-nvidia-gpus - 31/07/2008
HP pays half for Nvidia's graphic problems
Analysis The numbers game
http://web.archive.org/web/200905290...vidia-problems - 12/08/2008
Nvidia G92s and G94 reportedly failing
Desktop boards this time
http://web.archive.org/web/201912192...rtedly-failing - 18/08/2008
NVIDIA GPU Update: Dell to Offer Limited Warranty Enhancement to All Affected Customers Worldwide
http://web.archive.org/web/200812191...worldwide.aspx - 25/08/2008
Nvidia changes desktop G86 for no reason
Deny, spin, duck, dodge
http://web.archive.org/web/201912192...top-g86-reason - 28/08/2008
Nvidia 55nm parts are bad too
NVGate Changed for 'no reason' once again
http://web.archive.org/web/201912192...vidia-55nm-bad - 01/09/2008
Why Nvidia's chips are defective
Part One A long and complex story
http://web.archive.org/web/200905250...hips-defective
Bumpgate: Part One – A long and complex story
http://web.archive.org/web/201011280...are-defective/
Why Nvidia's duff chips are due to shoddy engineering
Part Two The underfill
http://web.archive.org/web/200905250...efective-chips - 02/09/2008
What Nvidia should do now
Part Three The cock-up
http://web.archive.org/web/200905260...6374/nv-should
- 10/09/2008
Nvidia sued for violations of federal securities law
All may be revealed
http://web.archive.org/web/201912192...securities-law - 22/09/2008
Nvidia chipsets are defective too
BumpGate Only 630, 7050, 7100 and 7150 changed for no reason
http://web.archive.org/web/200908211...sets-defective
- 10/10/2008
Apple notebooks have defective Nvidia chips
Nvidia still won't come clean
http://web.archive.org/web/201912192...e-nvidia-chips - 12/10/2008
HP admits 38 desktops have defective Nvidia chips
Time to update the 8-K filing
http://web.archive.org/web/201912192...e-nvidia-chips - 14/10/2008
Apple Knowledgebase says G92s are defective
Bumpgate You wanted proof...
http://web.archive.org/web/201912192...g92s-defective - 15/10/2008
AMD offers its take on GPU packages, failures
https://techreport.com/news/15707/am...ages-failures/ - 29/11/2008
Nvidia heat causing Macbooks to fail
We are shocked... shocked I tell you!
http://web.archive.org/web/200903260...using-macbooks
- 09/12/2008:
INQUIRER confirms Apple Macbook Pros have Nvidia bad bump material
We break out the electron microscope
http://web.archive.org/web/201912192...-bump-material - 17/12/2008
Nvidia chips show underfill problems
Bumpgate: Electron microscopes reveal a lot
http://web.archive.org/web/201011280...fill-problems/
https://www.semiaccurate.com/2010/07...fill-problems/
http://web.archive.org/web/200908221...hips-underfill - 26/12/2008
Nvidia's Solution To Defective GPUs : Buy Our New Chips
http://web.archive.org/web/200905042....html?doc=6351
Nvidia recommends not buying its defective chips
Told ya so
http://web.archive.org/web/201912192...efective-chips - 13/03/2009
Nvidia 10-K filing goes under the microscope
Analysis No admissions, only arrogance
http://web.archive.org/web/200905211...oes-microscope - 28/04/2009
Pressure mounts on HP over failing Nvidia graphics cards
Fried from the moment when the chip went 'pop'!
http://web.archive.org/web/201109032...graphics-cards - 12/05/2009
Apple, Dell and HP owners sue Nvidia
Want failing graphics chips replaced
http://web.archive.org/web/201311200...ers-sue-nvidia - 15/05/2009
Nvidia's insurance company falls out over faulty chips payments
http://web.archive.org/web/201207020...chips-payments - 19/05/2009
Nvidia's bad bump misery deepens
Bumpgate Insurers bite back
http://web.archive.org/web/201210081...misery-deepens - 26/05/2009
Nvidia is out at Dell
Bumpgate Bumps and payback
http://web.archive.org/web/200905290...63/nvidia-dell - 26/07/2009
Apple to Nvidia: Don't let the door hit your *ss on the way out
http://web.archive.org/web/200912112...ur-ss-way-out/ - 21/08/2009
Nvidia finally understands bumpgate
Two director posts open in Santa Clara
https://semiaccurate.com/2009/08/21/...ands-bumpgate/ - 01/10/2010
Nvidia's bumpgate settlement details revealed
https://www.itproportal.com/2010/10/...ails-revealed/
Attached is a copy of the pages if ever links get broken. File is *not* a zip, it's a tar.xz but uploaded with .zip extension, so .zip extension must be removed after downloading.
Comment