Since 2018, the PC processor market has seen a very good variety of generations come and go. And with good purpose: for the reason that final replace of our cooling take a look at protocol, we’ve had three generations of Ryzen CPUs and 4 on the Intel facet. In reality, on the time of our final replace, Intel was nearly to launch its Core i 9000, whereas AMD was solely at Ryzen 1000.
So why the change of protocol? Quite just because, on the one hand, the variety of cores has elevated but once more. On the opposite, CPU design has turn into radically completely different. While Intel retains a monolithic design on its thirteenth era, AMD has switched to MCM for the reason that Ryzen 3000. All this adjustments the best way processors warmth up… Not to say their energy consumption, which has risen sharply. These are the explanations which have prompted us to rethink our methodology.
So, with a view to measure the conduct of AIOs and followers on Intel and AMD platforms, we’ve opted for the next two configurations:
Configurations for our new cooling protocol :
- AMD :
- AMD Ryzen 7 7900X
- MSI X670E Carbon WiFi
- Kingston Fury Beast DDR5 2×16 GB (KF560C40BBAK2-16)
- GTX 690 for show
- NZXT C1000 Gold V2
- Thermaltake Commander F6 rheobus
- Intel :
- Intel Core i5 13600K + Thermalright Contact Frame
- MSI Z790 Carbon WiFi
- Kingston Fury Beast DDR5 2×16 GB (KF560C40BBAK2-16)
- GTX 690 for show
- NZXT C1000 Gold V2
- Thermaltake Commander F6 rheobus
Between the 2 platforms, we word the addition of a Contact Frame positioned on the Intel processor. The intention is to enhance the contact floor with the heatsinks we take a look at. All followers and AIOs are in contrast with this machine.
Room temperature:
Please word that each one our CPU cooling assessments are carried out exterior the case, in a room maintained at round 20°C (±2°C). So, in winter, we flip up the warmth, and in summer time, because the assessments are carried out in Brittany, it’s not an issue as a result of it rains on a regular basis and there’s by no means any solar. Joking apart, in summer time we take our readings when temperatures are most favorable, within the morning for instance.
Ventilation administration:
For our new protocol, fan administration continues to be entrusted to our Thermaltake F6 rehobus. For all-in-one watercooling, the followers are linked to the rheobus, as is the pump.
We set the cooling between three settings:
- High pace: 12V followers and 12V pump
- Medium pace: 8V followers and 12V pump.
- Low pace: 5V followers and 12V pump.
Compared with our earlier protocol, we go away the pump operating at full pace, whereas the fan runs at low pace. The energy consumption of the 13600K and 7900X is nowhere close to that of our outdated 1700X.
Nevertheless, the intention stays to have three clearly outlined noise ranges.
As our rheobus permits us to regulate the air flow to the closest tenth of a volt, we’ll be operating the followers based on three profiles: 5V, 8V and 12V. So why 8V? To be, roughly, in the course of the followers’ working vary.
However, some followers might not be appropriate for 5V. In which case they both don’t begin up, or they do begin up, however run at a ridiculously low pace, exterior their specs. The fan is then working in a clearly unsuitable electrical atmosphere. If such a case arises, we’ll enhance the voltage barely, in order that we are able to as soon as once more adjust to the fan’s specs. We set a minimal of 500 RPM, until the information sheet signifies a decrease minimal.
The intention behind this reflection and adaptation is to acquire outcomes similar to actual makes use of and doable cooling settings.
CPU profiles :
Next, we wish to take readings with increased and decrease energy consumption, to raised present the conduct of CPU cooling. We have due to this fact outlined 4 secure profiles in our X670E Carbon WiFi and Z790 Carbon WiFi, every corresponding to 3 completely different energy consumption profiles:
- AMD :
- 65W: Multiplier: x24.5 and VCore at 1,000V
- 100W: Multiplier issue: x33 and VCore at 1.132V
- 150W: Multiplier issue: x39 and VCore at 1.246V
- 200W: Multiplier issue: x40 and VCore at 1.360V
- Intel :
- 65W: E-Core/P-Core multiplier: x30/x30 and VCore at 1.105V
- 95W: E-Core/P-Core multiplier: x35/x35 and VCore at 1.210V
- 130W: E-Core/P-Core multiplier: x35/x45 and VCore at 1.270V
- 170W: E-Core/P-Core multiplier: x40/x50 and VCore at 1.350V
Of course, there are methods of optimizing profiles when it comes to frequency/consumption/heating, however that’s not the target right here. Our intention is that, on the time of our readings, energy consumption was all the time round 100W/150/200W on our AMD platform. The similar goes for Intel, with 95W, 130W and 170W, all of them secure.
This provides us an thought of the efficiency of cooling programs at completely different energy consumption ranges. For instance, 65W will solely be used for aircooling heatsinks and 120 mm AIOs. This corresponds to the consumption of low-power CPUs such because the Core i3 13400F.
Between 95W and 100W, we discover the identical energy consumption because the Ryzen 5 7600X and Ryzen 7 7800X3D. Between 130W and 150W, we’re now across the energy consumption of a Ryzen 9 7950X3D or a Core i9 10900K. Finally, at 170W-200W, we’re on a par with the @inventory consumption of a Ryzen 9 7900X or a Core i7 12700K.
Unfortunately, we might have preferred to have gone additional when it comes to energy consumption on our Core i5 13600K. Unfortunately, we’re compelled to notice that past 170W, it’s tough for us to acquire constant outcomes with our AIOs, because the temperatures reached are so excessive.
Our strategy:
Before becoming a brand new heatsink, we clear our CPUs and apply Arctic MX-4 as thermal paste. We join the followers to the rheobus, similar to the pump on an AIO.
At start-up, we choose the 100W/95W profile for the primary sequence of readings, 150W/130W for the second and 200W/170W for the final sequence. Heating instances will final round 10 minutes per sequence, and will likely be damaged down as follows:
6 minutes of heating, throughout which the temperature is allowed to stabilize.
At the top of those 6 minutes, we’ll have two sequence of averaged readings of two minutes every.
We’ll repeat this era of 2-minute readings till the temperature has stabilized (stopped rising).
To stress our Ryzen 9 7900X and Core i5 13600K, we’ll run them underneath CPUID’s PowerMax. We’ll run an SSE-type take a look at, which we’ll go away operating for a vast time. The intention is to generate a linear load utilizing 100% of our processors to see how the cooling programs behave.
For the temperature readings, we use the HWiNFO64 software program and report the Tdie temperature. There are 9 temperature readings: 3 fan modes with 3 completely different CPU settings.
Noise measurements are taken through the heating interval. The sound degree meter is positioned 40 cm from the sound supply within the route of the fan suction. The benefit is twofold: we are able to measure as many variations as doable from one cooling cycle to the subsequent, and we’re not disturbed by air actions.
In order to measure solely cooling noise, we block the graphics card fan and make sure that the semi-passive energy provide shouldn’t be triggered at the moment.
These measurements will allow us to give you temperature graphs, noise graphs and graphs displaying the evolution of the temperature/noise ratio based on fan mode. In this fashion, we’ll go into extra element concerning the conduct of 1 cpu cooler or AiO versus one other.
And off we go!
You’ve already had an opportunity to see what it’s like, albeit partially, in our assessments of the Asus AIO Ryuo III 240 and 360, and the NZXT Kraken Elite 360 and Kraken 240 RGB.
We’ll quickly be publishing an article that includes the readings of those kits on the Intel platform, to see how they carry out on our Core i5 13600K.
