Originally Posted by
rge
I had Dtek V1 with washer and 5.5 nozzle on my E8600. When I upgraded to i7, no way was I taking my loop down twice and no 1366 bracket available, so made my own bracket and left loop as is. First couple weeks, I used DTEK and nozzle on my i7.
The first 1366 that came available was GTZ so I switched to that. With same ambients, I got ~2C better on load at same overclock looking at 4, 4.1 and 4.2 by switching to GTZ with backplate vs DTEK as described above. I had probably 15 runs prime/linpack with DTEK and more than that now with GTZ, any way I look at it short runs or long runs, realtemp usually records about 2C less max temps and looking at avg temps with GTZ.
Though that is not exactly what you are asking, and unfortunately the same laziness that prevented me from taking my loop down more than 1x until 1366 was available, also prevented me from knowing what difference between nozzle on vs off, but given it was 2C from GTZ with it on, my only comment was it was not too bad with 5.5 nozzle on. But cant tell you if would be same or little better with it off, or for that matter a little worse (theory doesnt always predict results). Not to mention I made no effort to control any other variables, though my ambients are usually very controlled.