NVDA ...Fishy things Smell when you dig deeperThink about this for a second...To make things like engines, there are tons of people cramming ideas onto boards and CAD systems then employing CNC's to construct the first engine of your PROTOTYPE.
Even though every Formula one to Toyota Camry has an engine...the mounts, the fuel system, the weight ratios, etc are so drastically different- you could say, the car is built around it.
Now...relating ^ that to chips. You have to do tons of research to fit chips onto things. We arent talking a simple- pop the AMD out of Mr. Gamer's Tower and switch with the NVidia one....Thats as simplistic as it gets. You have data centers that need to have parameters as to heat dissipation, electrical usage, maximum safe tolerance for chip load, etc.
Yet you release something new that is like, hey guys- just pop out the old and slip this bad boy in. I find that hard to believe. How could you have newer power hungry projects that the utilities would have to be informed of to calculate load and transfer line/sub station tolerances, given these things arent located in the boonies. The engineers would have to come up with ways, in how short of a time, to dissipate the higher heat...while calculating what the operating temps of the whole system they are installed in can handle before things go a little burn-outy.
Again....I may just be completely oblivious to it...but like the neighbor who says, "You sure thats a good idea not roping that ladder to the tree" to the 20 year tree pro- its both an inquiry for thought, but also a possible reminder if everyone gets to far up their own whatever to come back to reality.
