There seems to be a lot of comments regarding the cpu temps for the 8700k and the tests on kickstarter. People will always be skeptical of the test results.
I really wish it could fit an L12s and I know other people who want to air cool would too. It shouldn't be the notion that you'd have to increase the case size every time a company like Noctua releases a new sff cpu cooler. But if it's one of the best cpu cooling companies and they revise a longstanding sff cooler, it should at least be considered more. I mean the L12s was made 4+ years after the L12.
Instead you should think of like: what air cooler will accommodate newer and possibly upcoming cpus? Hence a lot of concern with the 8700k and an L12. The 8700k is what a many people currently want to put in this thing, only to have to settle for an sff cooler released in 2012/13 when it has since been redesigned. So people will be very hesitant to put their new $400+ cpu and rely on an older generation cooler.
Don't get me wrong, I have an unopened L12 sitting in the corner for this. But that's because it's my best option for air cooling that I'm limited to and it's discontinued. Am I concerned about having to downclock my 8700k? Yes. If the case were 4mm wider to accommodate what could be the modern version of the best sff air cooler, would I get an L12s? Of course I would.
What concerns me is the logic that we'd be getting the best modern parts for this new case, but we're going to have to settle for the best cooler from 4+ years ago. Don't think of it as increasing the case by 4mm to fit the newest Noctua cooler. Think of it like increasing the case by 4mm to allow more people to be able to put newer i7's in it and the next 1-2 year's upcoming cpus which the L12s might be better suited for.
We have not tested the L12S so we don't know if it potentially would be a better fit with the 4mm increase in width of the case. Actually no one can know for sure before it has been tested in reality. We did a new tests an published it on the KS campaign as you say. This time using more aggressive fan settings that allow the fans to spool up more earlier and potentially causing more noise.
We understand these results are hard to believe. When we first saw them on the first prototype we had to make reruns only to believe it ourselves and then it hit us - this will be hard to communicate without people questioning the results. So we have urged the reviewers to run these tests and share their results because this is one of the biggest USPs of the Ghost, but so far they haven't really done this - still hoping Dave will do it! If you have any ideas on how we could communicate this more convincingly and impartially without putting our hopes in reviewers' hands we would love to do it. Maybe someone here lives close by and could swing by to confirm!? We could do a live 3Dmark run during the live stream today, probably at the end of it not to waste time.
The point is that The Ghost performs well with any of these coolers. When you experience this yourself we believe you'll be happy the case was not made any bigger. There is no need for down volting or delidding with any of these coolers! The key to this performance is the "Back-to-back architecture, only the fact that it gives superior vertical heat-pipe orientation is important.
The L12S compared to the L12: you call it a 4 year old design, but what is really different and why is the S better other than the fact that it is available? I'm not sure I'd prefer the S even with the 4mm case width increase only because it is newer, i'd base my decision on facts available and then test the candidates myself to understand the implications these facts have in my specific application:
- The S is lifted give it sufficient ram clearance when using a 120 mm slim fan - but the clearance is still lower than that of the L12, no?
- You get one fan only with the S
- It is taller
- Same recommended TDP?
- The S has a copper base, potentially an improvement but I'm not sure... heat traveling between different mediums is never optimal, if it would have been all copper I'd understand
In our view the L12 is the superior alternative, it has given us a solid baseline, a starting-point from where our expectations should keep rising. The question is now: Where do we go from here and how can it be improved further? Yesterday we reached out to Noctua (located not far from Stockholm) to initiate a dialogue with them to see if we can answer this question together. We don't know where that will lead at this point but we see several outcomes that would truly be better than increasing the size of the Ghost S1.
Link to test