Overclock.net banner

741 - 760 of 800 Posts

·
Registered
Joined
·
686 Posts

·
Registered
Joined
·
2,012 Posts
That mount looks great. And less the 10C spread is what you want to shoot for and looks like you have nailed it. Stick with what your doing there. With out delid thats most likely the best your gunna get. Good job.
Ok thanks :thumb:
 

·
Registered
Joined
·
1 Posts
Hey All,

Picked up a Magnitude am4 full nickel a couple weeks ago for my 3950x and can say I'm having some middling results... Cinebench r20, 15 min loop: Full Nickel Magnitude 62c | Plexi Velocity 60c.

Made sure the inlet for the water is focused towards the chaplets (Side of CPU closest to the PCIE slots).I remounted probably a dozen times now, with good even past spread. Tried the .6p vs .5p plates. Tried completely different mounting hardware (ditched included screw/springs/mounting plate standoffs for screw/wash going directly into rear mounting plate). All resulting in the same 2c hotter than EK Velocity.

Doing a little digging through this thread I see on page 43, where user "R0CK3T" is seeing a drop from 83.6 C to 77~75c making a similar switch on a 3900x. Cant really find much else for AMD/am4 Magnitude user results.

Looking at his ("R0CK3T") loop config, the only guess I would have to why he could be seeing better performance than on the Magnitude would be the use of a DDC pump vs the D5 pumps i'm using in my system?

Is block really that hungry for high pressure? Or am I missing something else?

Just feels like I'm falling more inline with intel results ...

Loop in serial configuration:
Start -> D5 pump/res (running @ 100%/18W) -> CPU water block -> 280x30mm rad -> D5 pump/res (running @ 100%/18W) -> ek 2080ti block -> 420x30mm rad -> 420x30mm rad -> \r back to start
 

·
Registered
Joined
·
2,553 Posts
TLDR how much cooler would a magnitude be over my EK Supremacy EVO? CPU is a 10700K.
 

·
Vermin Supreme 2020
Joined
·
25,778 Posts
probably within margin of error.


while it fits mainstream, that's not really the intended use case of these new megalithic blocks.


I <3 my Optimus SigV2, and they're actually in-stock unlike their AM4 block.
 

·
Registered
Joined
·
6 Posts
Still running the .5P
Hey Matt,

Thanks for all your posts. Just on the waterblock and delid approach you took, did you need anything other than washers at the rear to get it to sit flush on the die, or do you need different screws? I have one coming with a flatplate.

Edit - I just found this: "before closing the loop i tried the footprint to be sure the block make contact with the 9900KS die and it worked, so i can confirm standoffs are short enough to use the Magnitude without an IHS" - I think the 10900k is lower (with thicker IHS) so I guess the question is, do you think I will be able to keep screwing the waterblock down and meet the die surface with enough pressure to be effective?

BTW I sanded down the die (10900K) a little bit and it was convex which I noted by the middle being worn down first (wasn't totally flat).

Thanks
 

·
Registered
Joined
·
326 Posts
Hey Matt,

Thanks for all your posts. Just on the waterblock and delid approach you took, did you need anything other than washers at the rear to get it to sit flush on the die, or do you need different screws? I have one coming with a flatplate.

Edit - I just found this: "before closing the loop i tried the footprint to be sure the block make contact with the 9900KS die and it worked, so i can confirm standoffs are short enough to use the Magnitude without an IHS" - I think the 10900k is lower (with thicker IHS) so I guess the question is, do you think I will be able to keep screwing the waterblock down and meet the die surface with enough pressure to be effective?

BTW I sanded down the die (10900K) a little bit and it was convex which I noted by the middle being worn down first (wasn't totally flat).

Thanks
hi, you might try with a thin layer of thermal paste to "test" the footprint.
It seems to have thinner pcb (and die also if i remember good), compared to 9900k cpu.

---> Mine 8086k don't work with Magnitude direct die, this makes me assume that standoffs are barely enough to make contact with 9900k thin, your 10900k should be a half way size between 8086k and 9900ks. it might works as well as not

 

·
Registered
Joined
·
6 Posts
hi, you might try with a thin layer of thermal paste to "test" the footprint.
It seems to have thinner pcb (and die also if i remember good), compared to 9900k cpu.

---> Mine 8086k don't work with Magnitude direct die, this makes me assume that standoffs are barely enough to make contact with 9900k thin, your 10900k should be a half way size between 8086k and 9900ks. it might works as well as not

Hey, thanks for the reply.

Yeah, good idea to try that. I don’t have the water block yet so I’m trying to figure out if the thread on the stand offs keep going or they stop before the die, and if washers are needed then can they be placed behind the board to solve that or is there a limit to how much they can do.

I really do want to go direct if I can. I suppose I could use m3 threaded screws (assuming that is the right thread type) and then it’s just a matter of getting the right amount of pressure.

I have a frame for the 10900k that I’ll use from rockitcool on the way, but the only info I’ve found searching for this ekwb mag waterblock and direct die is literally only your posts - so thanks :thumb:
 

·
Registered
Joined
·
686 Posts
Hey, thanks for the reply.

Yeah, good idea to try that. I don’t have the water block yet so I’m trying to figure out if the thread on the stand offs keep going or they stop before the die, and if washers are needed then can they be placed behind the board to solve that or is there a limit to how much they can do.

I really do want to go direct if I can. I suppose I could use m3 threaded screws (assuming that is the right thread type) and then it’s just a matter of getting the right amount of pressure.

I have a frame for the 10900k that I’ll use from rockitcool on the way, but the only info I’ve found searching for this ekwb mag waterblock and direct die is literally only your posts - so thanks :thumb:
What motherboard are you using with your 10900k?
 

·
Registered
Joined
·
326 Posts
Hey, thanks for the reply.

Yeah, good idea to try that. I don’t have the water block yet so I’m trying to figure out if the thread on the stand offs keep going or they stop before the die, and if washers are needed then can they be placed behind the board to solve that or is there a limit to how much they can do.

I really do want to go direct if I can. I suppose I could use m3 threaded screws (assuming that is the right thread type) and then it’s just a matter of getting the right amount of pressure.

I have a frame for the 10900k that I’ll use from rockitcool on the way, but the only info I’ve found searching for this ekwb mag waterblock and direct die is literally only your posts - so thanks :thumb:
another thing you might try is to screw standoffs without using the plastic washer. doing this don't tighten too much or you may damage the motherboard pcb
 

·
Registered
Joined
·
6 Posts
Ah yes, just looked at the instructions and that makes sense. The washers go on the top side and so raise the height a little bit. I might consider grinding the standoffs downs bit if I need to so that I can reduce the height a little and keep the washers.

Cheers buddy!
 

·
Registered
Joined
·
686 Posts
@sibzilla and yes, pay attention at capcitors that are round the cpu socket, cause if they are too close to socket, your Magnitude may collide with them when using a direct die solution :)
thanks cris for that reminder
Hope he doesn’t have any of the Asus boards or he’s gunna be super disappointed when he try’s to mount the mag. Lol.
 

·
Registered
Joined
·
6 Posts
What motherboard are you using with your 10900k?
Crikey, totally missed your post. That is forum browsing on a phone at midnight for you!

I think I should be OK with the caps from taking a look. The board is a gigabyte Aorus ultra - seems fairly spacious.
 

·
Registered
Joined
·
6 Posts
I decided to ask EKWB today and they said that it won’t meet the die and that they can’t advise I attempt direct die cooling but that if I were to I’d need shorter standoffs so I’ve got some M4 screws, bolts and varying size brass couplers - once I figure out what’s required I’ll report back assuming I don’t bugger the whole job up. Looks like I’ll be waiting another week or so before EKWB send me the block and flat plate anyway.
 

·
2 Easy
Joined
·
3,493 Posts
My temps have been really good with this block using a 10900K, definitely impressed.
 

·
Registered
Joined
·
6 Posts
My temps have been really good with this block using a 10900K, definitely impressed.
i finally got all the gear and can confirm no modifications required for direct die. I used a rockitcool kit to hold the cpu in place and a flat plate from EK, but there’s plenty of clamping pressure installing as normal. Temps are excellent. Great bit of kit.
 

·
Overclocker in training
Joined
·
14,315 Posts
Hi,
Never checked but I doubt they are interchangeable.

Just did and no way
Screw holes are different and depth of cold plate is very different.
Mags cold plate is very shallow so you'd need a 1/4" gasket at least and would still leak lol
 
741 - 760 of 800 Posts
Top