I was curious myself about which was better push or pull so I did a little experiment with one of my slims.
I suspected that pulling would be better for southbridge temps as cool air would be drawn in from the sides of the case running over the board rather than bathing the components in hot air from the CPU/GPU. But I thought that the CPU/GPU may be slightly cooler with the fans pushing as cool air is being blown directly on the the heatsink. I really wasn`t expecting it to make much difference either way.
I attached a temp sensor to the southbridge with kapton tape and the slim I used is RGH`d so we can monitor temps in FSD.
With this slim I had already installed a 120mm fan in a pulling config.
I placed the xbox next to my TV and left it for a about 10 min to let the southbridge temp settle.
Then I powered up the xbox and started a new Mass Effect 3 game and played it through the opening about 10-15 min, untill you see all the ships burning up in Earth`s atmosphere and the Mass Effect 3 logo appears.
This is the maximum temp the southbridge reached.
And FSD temps.
Now about 10 min and an entire gluestick latter, pushing.http://a.imageshack....2/img1668mq.jpg
Southbridge temp at idle.
Same procedure with Mass Effect 3, max southbridge temp, as I suspected runs a little hotter.
FSD temps on the other hand were a little surprising!
So it would seem certainly in my case that pulling outperforms pushing, so time to break out the glue gun again.
This time I cranked up the radiator, ran through the opening sequence of Mass Effect 3 and then played Need for Speed for about an hour and I still couldn`t replicate the poor temps I got with the fan pushing. http://a.imageshack....1/img1682jv.jpg
I went into this completely un biased and the results were surpising to me, I can`t think why the CPU/GPU temps were so high with the fan pushing, it couldn`t be room temp because even with the radiator turn up uncomfortably high I still couldn`t reproduce those temps. Maybe some fans perform better at pulling and some are better at pushing.
P.S. I`ve found that the stock fan is actually pretty good, in fact at 75% it outperforms the 120mm fan I`ve been using for these tests, but it`s louder. When you set it to 60% it`s about the same dB and runs a couple of degrees hotter than the 120. Oh and removing the fan shroud improves temps, though probably not reccomended if your leaving the fans on Auto as you`ll get less airflow over the southbridge. http://img515.images...63/img1638p.jpg
Does look good though!
Edited by RDC, 08 February 2013 - 02:43 PM.