specifically, "Heat PIPE tech"
 |
They exist, the technology is there,
But do YOU have one? |
https://hardforum.com/threads/do-heat-pipes-contain-liquid-inside.1649396/
I think they're full of it (the forum, not the pipes)
Cutting one open and finding nothing inside would trigger the response that the gas has evaporated (lol) gas?
More professionally-sounding sites claim to use a small amount of water
(which is much more believable if true)
But if you discover that your tubes have nothing, then what, cheap counterfeit pipes?
No wick nor spongy material, just..pipe?
I'm in the camp of the heatpipe being empty (having no clothes, prolly a story you've never heard so forget it)
But maybe more reputable manufacturers have sprung to life since my youth (I doubt it.)
Wouldn't they spend extra time crowing about what's in their pipes?
But they don't.
Maybe really expensive ones do, but *They* know you really want a AIO (water pump with many fans) so I doubt they bother either.
Heat-conducting metal cylinders, that's all.
Now if my comments-section worked you could disagree,
Or say that the pipe you just destroyed has "condensation" you might not have noticed (drip)
I really need to find the product-page for the picture-fragment on top.
WE don't see any heatpipe-competition among heatsink makers,
It's usually taken for granted. More pipes, more cooling.
But is there (in theory) better pipes?
If there were pipe making companies, they're keeping mighty quiet about them,
promoting the rumors that elves at corsair and Noctua make them by hand.
Sintered Powder FTW (whut?)
Let's also say (in plain language) that heatsinks, cheapo ones, rely on gravity to work, so if your motherboard is stuck on a wall (most are) then gravity is nullified to that enormous gargantuan (huge) fin/heatpipe assembly.
So I say, it depends a LOT on the pipe.
Reviewers in general use test-benches...the motherboard is laid flat (horizontal) not on the case-wall, and cheap heatsinks work excellently, they get their freebie, win-win.
Moral, lay the heatsink (buy a heatsink) that sets *higher UP* than the CPU.
Heat rises. But this assumes liquidy-pipe cheap AIR-COOLED heatsinks exist, and I posit that they do not. (but maybe some do)
 |
Yes, I saw that biggie (wrong) answer on Google too.
Water-cooled, water PUMP, so gravity is not a factor. |
I'm looking for an air-cooled-heat-pipe answer, this is proving difficult to google.
Heatsink fins and water-cooled systems are WAY different subjects.
Grass-roots opinions:
- horizontal fans blowing down onto a heatsink is wasteful
- laying a case on its side is ill-advised.
- Following all the bad advice I just said, you really need an AIO instead.
Well shit.
My motherboard IS horizontal, the heat transfers *vertically*, and if moisture-filled heatpipes exist there, they use gravity to recycle the moisture, the moisture inside the heatpipes,
So I must be googling it wrong (don't trust reddit/quora)
@84F, 36C
(Nyah)
Speaking of not trusting reddit/Quora,
TDC: what is it, what can it do?
and everyone wants it *higher* than mine, which is a piddly 95 (actually, I dared "96")
Not trusting butmunch freeloaders who apparently have gobs of $$ to buy new stuff,
I wanna try "TDC=100" which is (to my mind) the captain of a submarine taking her down to incredible depths to check stuff out (on a slow day)
800-feet? Bah, take her to 1200 (*gasps and meaningful looks*).....
Well, yeah, it made a difference, a just-enough difference.
Not a "stuff a cake in your mouth and swallow it without choking" difference (so it's still sensible.)
Running Prime95 now (passing 750 feet (creeeak)
..
Passing 800 feet (*hissss-creak*)
Moving on, I've gotten to 89c and once, 90c with prime95 running 4 cores.
But what is irksome is that the time-sensitive workings of Timespy demands complete and utter quiet in the PC, no nervousness nor multitasking, because decent settings look real bad if you run timespy at the wrong time.
I've been thinking about "affinity" and "priority" and I wonder if that's cheating (if it even works.) So I give up and go back to tweaking TDC.
At 98 it was very slow, at 102 it was fast, which is illogical because the test does not tax the CPU all that much.
"CPU Profile" would be more appropriate but I know Timespy better.
The results for CPU profile are hard to find.... I'm very close to a 5900x on one Japanese site.
Plus it's past my naptime, will I make an irrecoverable error?
TDC@ 98, Timespy-CPU-15152 (best run)
TDC@100, Timespy-CPU-15162 (meh)
TDC@102, Timespy-CPU-15717 (was I lying? Huh?)
I think I'll go fritter some more with the settings.
Vddg-CCD 0.930
Vddg-IOD 0.970
TDC@102, Timespy-CPU-15800
Listening to "some guy said" pays off maybe 40% of the time....
Unfortunately the scores never last,
and I'm back to old and slow.
It was nice while it lasted though.
 |
| The day After |
No comments:
Post a Comment