br0dee

i  im sure the aftermarket waterblocks are in production already..

metal_god69

Sapphire must be pissed that NVIDIA stole their Vapor-X design lol.

kram36

metal_god69

Sapphire must be pissed that NVIDIA stole their Vapor-X design lol.

Sapphire did not invent the vapor chamber. nVidia no more stole the idea then Sapphire did.

metal_god69

Sapphire must be pissed that NVIDIA stole their Vapor-X design lol.

lehpron

Dear EVGA, I have an innovation idea that you must consider trying (and don't think for a second that I haven't completely thought this through just because some of you reading this may not have thought about it yet):

Make a waterblock using a vapor chamber as opposed to a copper plate.

The current waterblock technology is dependant on the relatively slow convection process of spreading heat from the source into the groved fins where the water is.  The block still absorbs some of that heat, thereby raising its temperature and preventing all the heat from the source to transfer to water.  More surface area that the water has contact with helps, as present and future waterclock designs will show, but ultimately the block is limited by the heat it absorbs during the transfer.

For now this technology is 'good enough' in terms of the return on investment.

If this process were sped up with more evenly distributed heat transfer with little to any absorbsion, the advantaged would be having a thermal resistence of near zero (source load and water temperature would be the same, instead of a few degrees of separation where the block and source would be warmer).   One disadvantage is the potentially high manufacturing and thus unit cost for such "vapor block" could be twice that of current regular nickel-plated copper waterblocks.

Another disadvantage is how far below freezing vapor chambers and heatpipes can go, but its not like they can't go below zero.  This link goes to a company names Thermacore, they make among other things, heatpipes (as well as flexible ones) and vapor chamber -based heatsinks.  Their own specs show that their chambers are 20 times more efficient at heat transfer than a regular plate of copper, as well as being able to operate as low as -40C.

If I could design it, it would use a wapor chamber base 3mm think and 40x40mm.  The top and bottom would be slightly thicker to support the grooved water channels on top and some lapping underneth.  Then an O-ring and your choice of plastic housing to enable tube attachements.  Or the waterblock plate itself can be manufactured separately from the vapor chamber, so the grooved process doesn't puncture the chamber, seeing how they can't be lapped...

Really EVGA, you have to think about this, I foresee it being a necessary step to get ahead, not the cheapest route, but it can be an end all for a while beating out all other companies because no one else has hinted trying such a thing (that doesn't meant they aren't doing it now, so you better get on the ball).