Jump to content
Search In
  • More options...
Find results that contain...
Find results in...

unholy_walrus

Member
  • Content Count

    63
  • Joined

  • Last visited

Awards

This user doesn't have any awards

About unholy_walrus

  • Title
    Member
  • Birthday May 19, 1987

Profile Information

  • Location
    Finland
  • Gender
    Male
  • Interests
    water cooling, hunting, hockey, diving, music, movies,

System

  • CPU
    i7 4770k @ 4.5GHz 1.41v
  • Motherboard
    ASUS Z97-AR
  • RAM
    2x8GB Corsair Vengeance Pro 2133MHz
  • GPU
    EVGA GTX 1080 SC2
  • Case
    Fractal Design Define S
  • Storage
    Samsung 850 EVO 250GB, WD Green 1TB x2
  • PSU
    EVGA Supernova 750 G2
  • Display(s)
    180hz 1080p G-Sync
  • Cooling
    Custom H2O
  • Keyboard
    G810
  • Mouse
    G502
  • Sound
    Hifimaan HEXv2
  • Operating System
    W10

Recent Profile Visitors

1,056 profile views
  1. 1) Define "hot". Sure, the radiator heats the air to an extent. That's how radiators work. The point is that (at least I claim) the effect on internal component temps is minimal. 2) What factors did I not take into account? Saying something isn't being taken into consideration and then not mentioning what that something is - that's not great. Note that while I would never claim that my testing is anywhere near even adequate, I do think that if running rads at intakes, dumping all that "hot" air into the case, was really that bad temp-wise, I should've seen at least some indication of it d
  2. There is no "hot air". I wish I could find where this whole concept of "having rads as intakes is bad" was investigated and debunked, because I know I've seen that just a while back. I'd be willing to bet that even in a fairly extreme case such as mine the effect to case temps is mininmal, even negligible. What I can say personally is that when I tested 240 rad front intake-360 rad top exhaust-120 fan back exhaust vs 240 rad front intake-360 rad top intake, 120 fan back exhaust, even when stressing both the cpu and gpu the difference in MB temps & a case temp probe was negligible. Hardware
  3. I won't pretend to know the science of it, but I do know that even with a 240 & 360 rads set as intakes and a single 120mm fan as an exhaust, the internal case temp won't reach even 40c during load at 25c ambient temp (Define S). Setting all rads as intakes has had only positive effects when running both cpu & gpu under water, based on my experience in the past few years. I don't mind having a very slightly higher case temp if as a result the liquid temp is lower.
  4. If you don't want to spend a whole lot of money, Noctua's NF-P12 redux fans are shockingly good for rad use considering the price, at 500-1k rpm. Their performance tends to drop off at higher rpm's compared to NF-F12 and especially NF-A12x25, however I don't think you should ever run fans much faster than that or it sort of defeats the whole idea of watercooling. I could never afford the NF-A12x25 or NF-F12 for my rads, however even if I could, I'd still go with the NF-P12 redux for that use case. If I have to run fans on a custom loop at 1k+ rpm, I've done something wrong or I have a very sma
  5. I would suggest getting some other fans for your rads. The NF-S12A is a fantastic fan, but it doesn't do great on rad use. I wouldn't use it even on low fpi and/or slim rads. The performance really isn't great in my experience. The NF-P12 / NF-P12 redux would be a significant upgrade already, and you can't go wrong with the NF-F12 (though I prefer the NF-P12's sound profile at 500-1000rpm). Then there's the new NF-A12 fans that are supposed to be great, but shockingly expensive. The NF-P12 redux is my new go-to for radiator use at 1000 rpm or less for a reasonable price. If you don
  6. It's very common to have some of the fins get damaged/bent when handling rads. They're made of aluminum or copper, thus making them fairly soft. All of my rads have bents fins from dropping fittings on them etc. Just like @Semper said, you'll lose some cooling effectiveness, but very little - unless the damaged are is large and prevents airflow significantly. In your case, you shouldn't even notice it. Do make sure the liquid channel is indeed intact, but if there is no visible damage on it at all, you should be completely fine. Luckily the channels themselves aren't quite as frag
  7. Loop leaked when one of my EK 45° adapters completely failed, so I re-did the whole thing. Basic loop, basic system. But it works, and it's all I can do with no budget.
  8. I had a leak due to a 45° EK-AF fitting that pretty much completely failed. It's a miracle I didn't end up with serious damage. I figured I might aswell move the rad from the roof to the front and generally re-do the loop. It's basic as fuck, but I'm working on a non-existent budget. The camera on my Honor 6 really doesn't do it justice either. I'll still need to fix the LED strips and sleeve those ugly ugly wires, but meh. I'm a lazy bastard.
  9. If you plan on watercooling your card, go for EVGA. Otherwise the Gainward card is the obvious choice because of the fantastic cooler it has. Also, didn't EVGA fuck up the Pascal cards pretty badly, missing thermal pads and all that? I'm not sure which cards were effected or if they've finally fixed the issue.
  10. Well, here's my super fucking ghetto setup for now. Notice the Noctua boxes that work as my monitor stand lol. Plan is to finish the liquid cooling project and get a new G-Sync monitor, but until I can scrape up the money for all that I'm stuck with this mess. The monitor has basically no adjustment options, but at least it's 1080p 144hz. As for the PC, a complete overhaul is coming once I can get an EK 280 rad and the fittings I require (WINK WINK EK). For now this works just fine. Note that all this looks about a billion times better IRL thanks to the horrible camera on my Honor
  11. Those temps are perfectly normal if you don't have a delidded chip. My liquid cooled 1.32v 4770k hits 65-70c in Witcher 3 with sub 40c liquid temp, which is why I'll be delidding this chip like every Haswell I've ever owned. So far I've just been too lazy to do so, but soon I'll re-do my entire loop so I won't have any excuses not to do it any more. The TIM between the IHS and the CPU die isn't good but it isn't exactly horrible either. The problem is usually just bad contact. Delid the chip, get rid of the epoxy glue or whatever the gunk is they use to glue the IHS to the chip and
  12. What kind of heatsinks did you use for the VRM if I may ask? Seeing this has made me want to try something on a reference 780 I'll get soon. I have a bunch of small copper heatsinks I used on one of my R9 290s somewhere around here but I suspect they may be way too high to fit inside.
  13. I was referring to the power delivery though, not the vram chips.
  14. I still find most full cover blocks to be better looking, but I have to admit this is a very nice mod. It's definitely the best one of these mods I've ever seen - most of them look amateurish and sloppy, unlike yours. The idea is nice and the end result looks sharp. My only real concern when comparing this to a full cover block would be VRM cooling, but it seems that the reference design provides no VRM heatsinks to begin with (correct me if I'm wrong, I've never seen a reference 980Ti up close) so I suppose it should be okay with just airlow from the blower fan.
  15. I'd suggest going for whichever decent quality custom card is the cheapest where you live. The GTX 1060 should give you a slight performance edge in gaming right now, but I wouldn't be surprised if the tables were turned a bit in a while. If the cards are within like 10-15€ of eachother, I'd probably advice you go with the GTX 1060, but both cards will do well. If I had to choose a card, I'd probably go for a Sapphire RX 480 Nitro. Sapphire generally makes fantastic cards and I've yet to hear anything bad about the 480 Nitro. The 4GB Nitro is by far the best deal when it comes to b
×