EK Supremacy EVO Threadripper TR4 Waterblock Review

EK is one of the most heralded names in the enthusiast computer cooling industry. It has been building waterblocks for custom CPU cooling since 2003. It has now thrown its hat into the AMD Threadripper waterblock ring. Does the Supremacy EVO waterblock have the performance to back up its reputation against tough competition?


EK Supremacy EVO Threadripper Temperature Results

For our temperature testing we put our EK EVO up against our previous champion from XSPC, the RayStorm NEO. In that testing we showed the RayStorm to come out ahead of the Bykski, but by less than 1 degree celsius.

Our cooling loop is comprised of all XSPC components: D5 Photon Reservoir/Pump Combo V2; RX480 Radiator V3; Raystorm Neo Waterblock.

We used our ASUS ROG Zenith motherboard and AMD Threadripper 1950X overclocked to 4GHz. 4GHz seems to be the highest 100% stable with our Threadripper under any load. We have also seen the AIO coolers from Enermax come close to supporting 4GHz, but not quite. Seeing that outdoor temperatures have cooled off here a bit in Texas lately, we were working with a very static ambient temperature of 72F/22C. That said, the temperature data here is NOT directly comparable to previous reviews we have done.

The Threadripper die temperature noted below was our peak temperature over a 1 hour run of Prime95 Small using FFT instructions.

Article Image

Article Image Article Image

First and foremost, I was quite a bit surprised at our temperature delta considering EK's reputation in the industry. On the good side of things, the EK block did actually support our 4GHz clock speeds without issue which our previous AIO systems would not handle with a big caveat covered below.

We had done several test matings and we felt good that our Prolimatech Pk-1 Nano Aluminum Thermal Compound was doing well. We used the same application techniques that we have found worked best in the past and we got consistent repeatable temperature results.

Article Image Article Image

(Sideways Mounted Block / Stock Mounted Block)

After deciding that our TIM and application was not the issue, I went back in broke the EK block down again to make sure I had put it reassembled it correctly. Like I mentioned on the previous page it is easy to put back together wrong and I wanted to double check myself. The EK block assembly was not correct! I had mounted it back into the bracket "sideways." I had put the block assembly in rotated 90 degrees. So I reassembled and retested with the proper stock configuration. In this configuration, you have more cold plate in contact with the surface of the Threadripper's rectangular IHS. What is odd, is that I got a better result with the cold plate mounted "sideways." With the EK block in "stock" configuration I got a peak temperature of 84C then I would start dropping Prime 95 threads, and this of course decreased our wattage load, negating the run as stable and giving us non-comparable temperature data. Going back and forth between these configurations proved it was not a fluke. The flow of the EK block produced better temperature results (barely) in our sideways misassembled configuration. In stock configuration it would not support our 4GHz Threadripper overclock.

Article Image

(Sideways mounting.)

After going back over everything, while I had previously doubted it, it seemed that nycdarkness in our forums was solid with his information that he shared, even though the EK representative on Facebook accused him of "bad guerilla marketing." We did not see the deltas as large as he did, and obviously we are using a completely different cooling loop, but the 6.6C delta that we recorded between the EK and the RayStorm is a world apart when we are talking about overclocking CPUs. With the EK block in stock config it would not support 4GHz overclocking, just like nycdarkness has seen in his testing.

Another thing to keep in mind here is that the AMD Threadripper starts throttling at 85C. At our ambient testing temperature of 72C/22C, we are seeing good overclocking temperatures that do not throttle with our "trick," but we are only 4.6 degrees C away from throttling. If you are dealing with higher ambients it seems fully possible that the EK block would not support non-throttling under full usage loads as nycdarkness was experiencing.

Here is exactly what EK states in writing about its Threadripper waterblock.

With AMD releasing their new X399 chipset based HEDT processors, came the need for a water block with a larger cold plate contact surface. The primary goal in designing the new EK-Supremacy EVO Threadripper Edition water block was to cover the entire IHS of the newly released HEDT AMD Ryzen Threadripper processor. With a dense micro-fin structure that counts 52 grooves with a spacing of 0.25mm apart, the EK-Supremacy EVO Threadripper Edition water block offers the best possible cooling performance! The water block uses award-winning EK-Supremacy EVO cooling engine with specialized jet insert and jet plate combination to ensure best possible cooling for X399 chipset based CPUs.

EK states that its Threadripper waterblock offers the "best possible cooling," not only once, but twice on its PR page. We obviously know that is not the case, so what is the issue? I fully think it is due to the much smaller microfin footprint on the EK block. We put together another video to specifically address this.

The Bottom LIne

The EK Supremacy EVO Threaripper Edition seems to be a very well built waterblock that was not truly designed for Threadripper CPUs as it applies to high end enthusiast usage. The size of of EK's microfin footprint seems to be crippling its performance. The orientation of the coldplate also seems to impact cooling as well as we found out by accident, which I guess is a good thing, but surely only further points to EK's lacking in testing and design. Maybe they have bigger fish to fry and are resting on their laurels?

If we look at the EK block from a value proposition compared to the XSPC RayStorm NEO, the picture is a bit different, as the XSPC is now going for $110, where the EK Threadripper block is selling for $78 in is basic black trim. That said, the XSPC design looks a lot better and has Frag-Harder lights, but that may or may not hold value for you. Where this all goes off the rails for EK is when you remember that we have the Bykski Threadripper waterblock in the mix as well at $65. The Bykski underperforms the RayStorm but only by a degree for a lot less cost.

In the Threadripper enthusiast waterblock market, EK fails at performance and EK fails at value. The EVO is a capable waterblock that is far from "bad," but if you are looking to overclock that Threadripper to 4GHz, we would have to suggest you look elsewhere. On the upside, there are so few products on the market right now, you don't have to worry about getting confused and spending your hard earned money unwisely.

Article Image

If you like our content, please consider supporting us on Patreon for as little as $1 a month.

UPDATE: Worth mentioning is that since our review, and pretty much the only real review online, EK has issued an apology for this product to its customers.