• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Larger-than-GA102 (non-HBM) dies spotted?

Mar 31, 2014
1,320 (0.50/day)
System Name Indis the Fair
Processor R5 3600 (PBO 150/130/130, 73c temp limit, FCLK/UCLK 1866)
Motherboard Asus Prime X470 Pro
Cooling Heatkiller VI Pro, VPP755 V.3, XT45 240mm, 2xA12x25, Arctic P14 case fans
Memory G.Skill Ripjaws V 2x16GB 4000 16-19-19 (b-die@3733 14-15/9-13-26-36 1.45v)
Video Card(s) EVGA 2080 Super Hybrid (A12x25 fan)
Storage 860EVO 500GB, 660p 1TB, WD Blue 3D 1TB, Sandisk Ultra 3D 2TB
Display(s) BenQ XL2430T, Dell P2417H
Case Phanteks Enthoo Pro M
Audio Device(s) DT770 Pro, Topping A50, Focusrite Scarlett 2i2, Røde VXLR+, Modmic 5
Power Supply Seasonic 860w Platinum
Mouse Razer Viper Mini, Razer Gigantus
Keyboard GMMK Fullsize v2 (Gateron Browns)
Software Win10 x64/Win7 x64/Ubuntu
Guy I know has been snooping around on some Chinese forums and came across this (post was quickly deleted later)...

The pics are of a System Function Testing machine which tests the assembled packages for defects in the BGA, substrate and likely some die function testing as well. These dies appear to be larger than GA102s (check the die shot from the reviews), keep in mind that the packages get a metal bumper around the edge added after. They are not TU102s, the peripheral SMTs don't match the layout of TU102 and the dimensions are different...

Note that these are also conventionally packaged, which leads me to believe they must be a (non-HBM) GDDR die larger than GA102...

Nvidia is cooking some thing up.