Originally Posted by doyll
I think "wisdom" has little if anything to do with it. It has more to do with conventional design, optical bay placement, conventional motherboard design with CPU near top, old low power demand GPUs, PSU being mounted in back top pulling air from inside, etc. It's what customer was buying and case builders have no desire to rock the boat too much.. until recently. Oh sure, they "designed" fancy gamer cases but that's more about the look that sells it than performance.
Now we have GPUs making 2 to 3 times as much heat as CPU, often 2 or 3 in a take making 6 to 9 times the heat CPU does. we don't what that heat going into CPU cooler.. at least I don't.
Obviously an intake in the fornt of case feeding air directly to CPU cooler would be ideal, but that isn't practical without modifications.
My suggestions above were for air cooled use, not H2O. For H20 maybe top and bottom in with front and back out might be best. )
I agreed the first time that all are valid points but again, there are competing factors..... which set of factors wins will depend on case specifics..... my thinking is however, if your postulation were an absolute given, that at least one case manufacturer would have tried to distinguish itself with such a design by now. Look at all the independent thought that Phanteks put into this design, seems to me this thought would have occurred to them, and if borne out, would have been another way to distinguish themselves.
I was not suggesting all of top, side, front and bottom be filled with intake fans.
I missing something .... the list says:
1 or 2 top
1 or 2 bottom
Ok so there's intakes in the top, front and bottom..... if ya not gonna short circuit, what's left ? The side panel bottom is not going to be very effective for exhausting hot air .... it's presence even possibly detracts from case cooling as with6 fans blowing in, the path of least resistance is fresh air from the front intakes going right out 3" away.
The combination of 5.25 bay in top and front fans tends to turn the airflow toward the back. Only area of possible "dead zone" is at middle of case / front of motherboard above GPU.. which should not be a problem.
That's where I think Phantek's missed one..... at least as per the Silverstone article.... the front 5.25 area provides a nice location for an exhaust fan tho, many people won't want the sound an heat being exhausted in their faces. A grille in that front door still wuda been welcome. But again, mush shorter path from front to side panel, especially with those HD bays in there deflecting air to the sides.
Again, all your points are valid points ..... but there's valid points in both directions and only detailed testing will show definitively which set "wins" in the end and even that will be case, configuration and installation specific. A floor installation under a desk has different restraints than a desk cubby which is in turn different from an open desktop....what works well in one might not carry over to the others.
But all in all, being geeks, I feel we give a lot more credence to these various mantras we go buy then we if called upon we can adequately justify....
-$2300 in Water Cooling parts netted me 0.1 GHz on my CPU, .... not exactly a winner in the ROI department
-I have extremely low GPU temps but get the same "stable" OCs as under air
-I have spent hours on tweaking fan curves for no real observable gain......fans are inaudible at 850 rpm so why bother turning them down ?
But I had loads of fun and personal satisfaction doing it so was the expense worth it, hell yeah ..... A 0.1 second difference in a race means thousands of dollars in winnings and endorsements to a competitive race driver.....but to do car enthusiasts benefit in any way from tweaking their cars ? No, not at all, .... no one gets a bonus for getting to work 4 seconds earlier. ..... but like them I did get a lot of pleasure and personal satisfaction of figuring things out.
..... and we also tend to long hold onto "wisdom" from past years and hold it as gospel long after the data behind it is relevant.
-Temps won't vary more than 0.5C in a loop....they do, I have managed 3+C
-SP is extremely important on rads .... can be, but not with 9-12 fpi and low rpm fans
-Positive pressure is essential to keep dust out, it's very helpful but having bust positive pressure without inlet fan filters is meaningless.
-The presence of aluminum in a loop will corrode copper.... scientifically impossible.
So like you said, long held "wisdom" should always be questioned and revisited over time.....At this point, from a theoretical standpoint, I'd say it's hardly a difference between back versus back + top in air cooling on large cases like the Enthoo..... On small cases I gotta go with back and top, on large cases, I'm not so sure but I'm still leaning towards back and top but if I'm right or wrong, I doubt it's by any margin of significance . I base my a lot of my thought process on building and confined space ventilation studies and practices where ya can get in and conduct detailed measurements .... bit harder to do when the confined space is 1/20th the size but the theory should hold.
And the result can easily flip flop via influences we don't immediately recognize. Your air restriction at the case bottom is a perfect example, it greatly affected your 480 installation whereas in my build it had little effect. While we were both similarly affected by the small inlet area, my inflow was augmented by having air intake thru the case floor thru the large open vents between the 280 and the front fans. The disadvantage was offset and rendered to a "no impact" because conditions changed.
If someone looked at my test and said .... the impact of the reduced inlet area is negligible at worse, they'd be correct.
If someone looked at your test and said .... the impact of the reduced inlet area is huge and the only remedy is ..... , they'd be correct.
The problem only arises when the reader looks at the conclusion and tries to apply the conclusion across the board, without recognizing the conditions that you are bound to run into some surprises. A perfect example of why there are no absolutes and why we have to look at the reasoning behind the mantra, not take it as gospel and make sure that it is relevant to all situations we apply it to.