The new stuff is supposed to be better, but they likely didn't do enough QC to see if it was really fit for purpose.
Better in what?
It is better for saving cost of manufacturing, for sure,
It may not have better heat transfer performance, it may impact long term realiability of the cards, it may make repairs more difficult.
BTW if this putty really needs to be thinned significantly for automated dispensing, why dont they use some
evaporative thinning agent? They would just squirt the runny putty on the PCBs, let them sit for a while to allow the thinner to evaporate away and firm up the putty and then go on with the assembly. It is a complication (delay, storage of the waiting PCBs, dealing with thinner vapors), but overall even with this complication the manufacturing in mass scale may get cheaper than manually placing the pads on the cards and mounting the coolers.
I believe this still has a potencial for making the manufacturing cheaper (and possibly making them cheaper fo consumers to buy too?!) and it could be possible to make this work.
This is for sure sure an interesting discussion, but the question stands what is GB going to do with the existing cards which have the slippery stuff in them.
The
problem is not only the orientation of the cards when they run. You may a have normally mounted card, with the oily and degraded gel, which still somehow managed to stay in place and perform adequately and the card works, but as soon you remove the card and for example ship it somewhere, it may sit for days in wrong orientation and the gel will just slip away.