Threadripper BUILDERS thread

[Markfw]

Preordered EK water block. Estimated delivery date says August 18th with DHL Express. Hope that's correct. More than likely I'll be building on the 21st, I suspect.

I hope you get it then. If not, by then I can tell you how my $120 AIO 360 is working@ 4ghz

 

[FiLeZz]

EK-Supremacy EVO Threadripper Edition ordered

 

[ub4ty]

There's something comfy about having stressed out and gotten your parts locked in and pre-ordered before the review mania.

I feel like I just quickly glanced at reviews today and said to myself 'yep, as expected. This is why I ordered it'. I'm primarily looking forward to proper mobo reviews, 3200 vs 3600 ram reviews and cooler reviews at this point to lock in those parts.

Feels comfy man !

Noctua, if you're selling the HSF combo this Saturday, get some review samples to the e-celebs !!!
and I still got a bone to pick w/ you about your poo color scheme

 

[mattiasnyc]

Do we need pepe on this forum?

 

[IEC]

Hmm. Well looks like MSI sent some X399 boards to Newegg via air earlier than the 8/16 ETA, because my board is preparing to ship as well.

Unfortunately that means I may need to order an AIO for temporary usage, as I will have CPU + motherboard in hand by Tuesday.

Edit: Solved that problem. Ordered an Arctic Cooling Liquid Freezer 240 which I can use to cool my mATX R5 1600X system once I get my full blown waterblock from EK.

 

[kaporkle]

Hmm. Well looks like MSI sent some X399 boards to Newegg via air earlier than the 8/16 ETA, because my board is preparing to ship as well.

Unfortunately that means I may need to order an AIO for temporary usage, as I will have CPU + motherboard in hand by Tuesday.

Edit: Solved that problem. Ordered an Arctic Cooling Liquid Freezer 240 which I can use to cool my mATX R5 1600X system once I get my full blown waterblock from EK.

I got order confirmation for my MSI board as well. I do re-call it was originally set to ship 8/10 on newegg before they changed it to 8/16.

 

[AdamK47]

Went to the local Micro Center today to see the latest goings on. They of course had the 1950X and 1920X CPUs. They also had the Asus X399 Zenith Extreme motherboards along with some others that escape memory.

They were just sitting there on the shelves. Mere inches from my grasp. A purchase would have instantly netted me a new Threadripper CPU and motherboard without the need of online ordering and the checking of shipping status.

 

[Markfw]

Went to the local Micro Center today to see the latest goings on. They of course had the 1950X and 1920X CPUs. They also had the Asus X399 Zenith Extreme motherboards along with some others that escape memory.

They were just sitting there on the shelves. Mere inches from my grasp. A purchase would have instantly netted me a new Threadripper CPU and motherboard without the need of online ordering and the checking of shipping status.

You lucky dog... most of us don't have that luxury.

 

[CHADBOGA]

Do we need pepe on this forum?

Yes. Pepe is Godly.

 

[kaporkle]

For those of you getting the MSI board, which RAM kit are you ordering? I'm thiiis close to pulling the trigger on some HyperX (HX432C16PB3K4/32). It's Hynix, but also the only 32GB 4x kit rated at 3200 on the QLV. Are Hynix chips still an issue?

 

[ub4ty]

Do we need pepe on this forum?

Relax and go with it. You'll be seeing alot more of him in the timeline ahead...

 

[ub4ty]

Went to the local Micro Center today to see the latest goings on. They of course had the 1950X and 1920X CPUs. They also had the Asus X399 Zenith Extreme motherboards along with some others that escape memory.

They were just sitting there on the shelves. Mere inches from my grasp. A purchase would have instantly netted me a new Threadripper CPU and motherboard without the need of online ordering and the checking of shipping status.

Everyone going to Microcenter ... Pics please !!!!! I need the satisfaction of seeing threadripper in a retail setting along w/ x399 mobos

 

[mattiasnyc]

Relax and go with it. You'll be seeing alot more of him in the timeline ahead...

You know he's been appropriated by a certain movement, yes?

 

[ub4ty]

You know he's been appropriated by a certain movement, yes?

It's going to be alright lad. Just give it time.
Pepe is colored for a reason.

 

[MarkPost]

Noctua TR coolers already available:

http://stores.ebay.de/NOCTUA-Designed-in-Austria

 

[ub4ty]

Noctua TR coolers already available:

http://stores.ebay.de/NOCTUA-Designed-in-Austria

US link : http://stores.ebay.com/NOCTUA-Designed-in-Austria
Pricing was advertised at launch to be :

• NH-U14S TR4-SP3: EUR 79.90 / USD 79.90
• NH-U12S TR4-SP3: EUR 69.90 / USD 69.90
• NH-U9 TR4-SP3: EUR 69.90 / USD 69.90

Source : http://noctua.at/en/noctua-presents-cpu-coolers-for-amd-s-ryzen-threadripper-x399-and-epyc-platforms

The Euro pricing is correct on Ebay. However, the USD pricing is not correct and seems to be a direct currency conversion. As such, there is an ~$12 markup + $10 shipping from germany with an arrival date of Aug 18th-22nd. Hmmmm...

Also, I was quite suspect about the huge gap between the heatpipes and the expanded contact surface they engineered for threadripper :

There is zero gap present on the non-threadripper NH-U14s, NH-U12S, NH-U9 models :

This indicates that they likely only extended the contact plate w/o extending the heatpipe mating. The issue with this is, the only thing that transfers the heat up into the fins is the heatpipes. Thus, if they aren't mated with the contact plate extension that covers the dies, you have a point of thermal inefficiency, the effective contact surface is really no bigger than the above (non-threadripper) and there's potentially partial effective die coverage. I was 100% correct in my detailing of AMD's compatible listed AIO cooler's contact surface and I am likely correct here.

I'd love for a Noctua rep to reply to this. I am going to send them an email shorty for clarification w.r.t to THE GAP.

* Inquiry E-mail was sent to Noctua. Purchase is on hold until they clarify this to sufficient detail.

In fairness, I also have no clue what's going on with Artic's Threadripper solution either :

From a width standpoint, I doesn't appear to be a full coverage and the width sure doesn't look like it covers two diagonally arranged gapped dies. What looks good is that the heapipes don't appear to have a yuge exit gape on the long side. However, it doesn't seem to be wide enough.

For socket reference :

 

[mattiasnyc]

Pepe is colored for a reason.

Are you?

Being coy doesn't help your case. Spell it out.

 

[krumme]

US link : http://stores.ebay.com/NOCTUA-Designed-in-Austria
Pricing was advertised at launch to be :

• NH-U14S TR4-SP3: EUR 79.90 / USD 79.90
• NH-U12S TR4-SP3: EUR 69.90 / USD 69.90
• NH-U9 TR4-SP3: EUR 69.90 / USD 69.90

Source : http://noctua.at/en/noctua-presents-cpu-coolers-for-amd-s-ryzen-threadripper-x399-and-epyc-platforms

The Euro pricing is correct on Ebay. However, the USD pricing is not correct and seems to be a direct currency conversion. As such, there is an ~$12 markup + $10 shipping from germany with an arrival date of Aug 18th-22nd. Hmmmm...

Also, I was quite suspect about the huge gap between the heatpipes and the expanded contact surface they engineered for threadripper :

There is zero gap present on the non-threadripper NH-U14s, NH-U12S, NH-U9 models :

This indicates that they likely only extended the contact plate w/o extending the heatpipe mating. The issue with this is, the only thing that transfers the heat up into the fins is the heatpipes. Thus, if they aren't mated with the contact plate extension that covers the dies, you have a point of thermal inefficiency, the effective contact surface is really no bigger than the above (non-threadripper) and there's potentially partial effective die coverage. I was 100% correct in my detailing of AMD's compatible listed AIO cooler's contact surface and I am likely correct here.

I'd love for a Noctua rep to reply to this. I am going to send them an email shorty for clarification w.r.t to THE GAP.

Lay this to rest.
The physical area is a non issue
http://www.guru3d.com/articles_pages/amd_ryzen_threadripper_1950x_review,5.html

"

The physical contact surface area is not an issue.

Lukewarm at best is what it is. The location of the two dies sits under the cooling block, these are soldered to the heatspreader, ergo heat doesn't leak away towards the visible sides.

"
Guru 3d review page 5

No problem what so ever
Nada. Zip.
The gab on the noctua is irrelevant before it even started.

 

[ub4ty]

Are you?

Being coy doesn't help your case. Spell it out.

 

[Reinvented]

Everyone going to Microcenter ... Pics please !!!!! I need the satisfaction of seeing threadripper in a retail setting along w/ x399 mobos

This feels creepy...and I feel a bit violated. But, I guess we can.

 

[ub4ty]

Lay this to rest.
The physical area is a non issue
http://www.guru3d.com/articles_pages/amd_ryzen_threadripper_1950x_review,5.html

"

The physical contact surface area is not an issue.

Lukewarm at best is what it is. The location of the two dies sits under the cooling block, these are soldered to the heatspreader, ergo heat doesn't leak away towards the visible sides.

"
Guru 3d review page 5

No problem what so ever
Nada. Zip.
The gab on the noctua is irrelevant before it even started.

It is an issue my friend which is why custom full coverage water blocks are being made and why people are going out of their way to make contact plates that cover the entire heatspreader. A review site that actually does a good job of getting into such details highlight this here :
http://www.gamersnexus.net/news-pc/3008-threadripper-cooler-and-thermalpaste-coverage-vs-die-ihs

So no.. I wont put it to rest as it quite clearly is a thermal inefficiency.
DUH :

This is exactly what I highlighted in previous commentary and it is quite clearly 100% accurate.

So, as for Gur3d's coverage

Threadripper is a big chip, the heatspreader will be bigger than the contact surface area your cooper cooling block offers.

FALSE. The heatspreader wont be bigger than the contact surface area of correctly engineered contact surfaces. Straight from their own contradictory website :
http://www.guru3d.com/news-story/ek-offers-amd-threadripper-supremacy-evo-water-blocks.html

As for this crap :

This does not show the thermal inefficiency issue as there is a huge area that's not being shown underneath the mounting bracket and even further in to the copper mating surface and subsequent outter surface that has screw holes that don't contact the heatspeader directly above the die.
Per Guru3d's own words :

The thermal image above is a little arbitrary,

That being said, there is clearly thermal leakage if you look closely that is between a valuation of M2 and M4 7*degrees C. I'm quite confident its alot higher than that if you got thermal information under the bracket and then closer to the copper contact plate and above one of those screws. I'm quite sure its about 10-15*C difference which is a big issue w.r.t to potential warping and a friggin die that has some cores running pronounceably hotter than others. Just because Guru3d didn't do a sufficient job (arbitrary by their own words) doesn't make the case any stronger that these unspec'd AIO cooler contact surfaces are thermally inefficient and don't cover the dies. That my friend is factual as proven by the correct coverage by Gamer's Nexus in actual pictures under the brackets and under the copper contact plate. Again for your viewing pleasure :

See fin surface.. See screw holes directly above the die surface?

 

[ub4ty]

This feels creepy...and I feel a bit violated. But, I guess we can.

 

[krumme]

It is an issue my friend which is why custom full coverage water blocks are being made and why people are going out of their way to make contact plates that cover the entire heatspreader. A review site that actually does a good job of getting into such details highlight this here :
http://www.gamersnexus.net/news-pc/3008-threadripper-cooler-and-thermalpaste-coverage-vs-die-ihs

So no.. I wont put it to rest as it quite clearly is a thermal inefficiency.
DUH :

This is exactly what I highlighted in previous commentary and it is quite clearly 100% accurate.

So, as for Gur3d's coverage

FALSE. The heatspreader wont be bigger than the contact surface area of correctly engineered contact surfaces. Straight from their own contradictory website :
http://www.guru3d.com/news-story/ek-offers-amd-threadripper-supremacy-evo-water-blocks.html

As for this crap :

This does not show the thermal inefficiency issue as there is a huge area that's not being shown underneath the mounting bracket and even further in to the copper mating surface and subsequent outter surface that has screw holes that don't contact the heatspeader directly above the die.
Per Guru3d's own words :

That being said, there is clearly thermal leakage if you look closely that is between a valuation of M2 and M4 7*degrees C. I'm quite confident its alot higher than that if you got thermal information under the bracket and then closer to the copper contact plate and above one of those screws. I'm quite sure its about 10-15*C difference which is a big issue w.r.t to potential warping and a friggin die that has some cores running pronounceably hotter than others. Just because Guru3d didn't do a sufficient job (arbitrary by their own words) doesn't make the case any stronger that these unspec'd AIO cooler contact surfaces are thermally inefficient and don't cover the dies. That my friend is factual as proven by the correct coverage by Gamer's Nexus in actual pictures under the brackets and under the copper contact plate. Again for your viewing pleasure :

See fin surface.. See screw holes directly above the die surface?

My friend.
You dont know the thickness of the heat spreader.
So its all irrelevant anyway.
Guru 3d took the effort to meassure it. And touch.
Nobody else did. They all made drawings.
Temp was never an issue.
It just wasnt even if its hard to grasp. But the heatspreader obviously does its job well and better than i anticipated. Just stop the crusade.
I will personally prefer a plate that covers all and even 5 degree is well going for imo. We agree.
But this noctua gab is surely 100% a non issue. Even 10-12 c is easily taken care of.

 

[IEC]

Are you?

Being coy doesn't help your case. Spell it out.

Take this garbage to PM, no one cares about your being coy about being insulted by a meme, allegedly appropriated or not. Memes don't belong to or represent a single group, no matter what media tells you to think. We're on the internet. Deal with it.™

As far as the discussion about IHS coverage, heatpipe contact, etc:
1) The gains from optimizing either or both of these things will be marginal at best
2) Marginal gains can certainly help when you are pushing the limits
3) For the vast majority of users outside of enthusiasts/hardcore OCers, I doubt it makes a difference

That said, I should have an opportunity to compare a 240mm AIO (AC Liquid Freezer 240) versus an EK Supremacy EVO Threadripper edition water block with full custom loop, 360mm rad, etc. We will see if it makes any significant difference or not. And yes I will be pushing the limits and voiding warranties, as usual.

 

[mattiasnyc]

Take this garbage to PM, no one cares about your being coy about being insulted by a meme, allegedly appropriated or not. Memes don't belong to or represent a single group, no matter what media tells you to think. We're on the internet. Deal with it.™.

you deal with it. Put me on ignore if you don't like it. Quit whining about whining.