Review Thermalright Royal Pretor 130 Review: The Crown Jewel of Air Cooling

[btmedic04]

Honestly neither one of those coolers are particularly relevant today. They both came out over a decade ago and the original Peerless Assassin was already performing around the same. There may only be two charts with the D15, but it should be fairly obvious it's not something anyone should consider using today.

It would certainly be nice to see a retest of the D15 G2 should that be possible though.

If you're worried about performance it's probably not an issue, if it's aesthetics I totally understand that one. Any air cooler that isn't offset your only options tend to be lifting the fan, removing it entirely, or moving to the back for pull (this can be very problematic noise wise depending on fan).

My main concern is with my x670e gene's dimm.2 slot that resides where typically slots 3 and 4 for ram would be. I'm currently using both m.2 connections on it and can't give that up unfortunately

 

[bit_user]

Maybe later Tom's will get a CMS update that can give us more content width on desktop browsers.

I've heard a few other Toms authors complain that they're made to use the same CMS as all of Future's publications. We can hope for improvements, but you'd think they would've taken care of some of these issues long ago, if they had any interest in doing so.

 

[Darkbreeze]

Yeah, I increased to 120-130% zoom-in for better sizing. But usually articles here are pretty good for giving you click-to-zoom options; it's just not here for whatever reason.

Dynamic web pages are meant to give different display options for mobile vs desktop browsers, but it's pretty clear desktop screen width is wasted on the articles. The forums have the left/right rails with the main content in the middle, so it looks like their system just dumps the left/right rails for articles. Maybe later Tom's will get a CMS update that can give us more content width on desktop browsers.

This is definitely not the first time. And it's a weird thing too. When the "Who's who in power supplies" article first came out years ago the charts worked fine. This was the article that informed on who the OEMs were for most the known power supply manufacturers out there. Then, at some point, they didn't. Then we changed to Xenforo and there were apparently some changes on the editorial side of things as well at that point and then they definitely didn't. I haven't even bothered to look for a long time because for one it hasn't been updated in years and years and two, it never got fixed.

 

[thestryker]

My main concern is with my x670e gene's dimm.2 slot that resides where typically slots 3 and 4 for ram would be. I'm currently using both m.2 connections on it and can't give that up unfortunately

I don't recall seeing any height measurements on the unit Asus uses for that, but the two SSDs don't overlap much. It's probably around 50mm high so the fan would definitely need to be raised. On my PA140 (heatsink height is identical to this cooler) I was able to install ~49mm high DIMMs, but definitely had to move the fan quite a bit which I imagine could be tight on a lot of smaller cases.

 

[btmedic04]

I don't recall seeing any height measurements on the unit Asus uses for that, but the two SSDs don't overlap much. It's probably around 50mm high so the fan would definitely need to be raised. On my PA140 (heatsink height is identical to this cooler) I was able to install ~49mm high DIMMs, but definitely had to move the fan quite a bit which I imagine could be tight on a lot of smaller cases.

Yeah, I'll have to do some more research and see what i can come up with. I can always stick with an aio, but I'd like to move back to air cooling for simplicity sake

 

[AB382B]

How does Thermalright Royal Pretor 130 compare to Phantom Spirit 120 EVO?

 

[bit_user]

BTW, did the Intel setup involve a contact frame, or did you stick with the motherboard's standard CPU hold-down mechanism?

Answering my own question...

Source: https://www.tomshardware.com/pc-components/air-cooling/cooler-master-hyper-612-apex-review
​

@Albert.Thomas , I think you really owe it to readers to highlight the fact that you're using a contact frame on LGA1700. Otherwise, it's a disservice to readers, as people will not be able to reproduce your results.

Even if you add a test case where you use the standard CPU hold-down mechanism, Noctua found that prolonged use of the standard mechanism eventually warps the CPU's IHS. So, to do it properly, you'd really want to swap CPUs when you swap hold-down mechanisms.

Here's Noctua's data on LGA1700 IHS concavity, depending on the hold-down used. They didn't explicitly test contact frames, but those are designed to approximate the "washer mod".

https://noctua.at/en/nh-d15-g2-versions-explained
​

I wondered how it was possible a single cooler could get such good performance on both LGA1700 and AM5, when Noctua had to rely on at least two different bases to achieve this.

@PaulAlcorn what do you think about even adding a note about this, in the Verdict section? Given the potential of contact frames to void motherboard warranties, I think we can't just assume everybody is going to use one. If they're not, then this might not be the best cooler for them.

 

[Albert.Thomas]

Answering my own question...

​

​

Source: https://www.tomshardware.com/pc-components/air-cooling/cooler-master-hyper-612-apex-review​

​

@Albert.Thomas , I think you really owe it to readers to highlight the fact that you're using a contact frame on LGA1700. Otherwise, it's a disservice to readers, as people will not be able to reproduce your results.

Even if you add a test case where you use the standard CPU hold-down mechanism, Noctua found that prolonged use of the standard mechanism eventually warps the CPU's IHS. So, to do it properly, you'd really want to swap CPUs when you swap hold-down mechanisms.

Here's Noctua's data on LGA1700 IHS concavity, depending on the hold-down used. They didn't explicitly test contact frames, but those are designed to approximate the "washer mod".

​

​

https://noctua.at/en/nh-d15-g2-versions-explained​

​

I wondered how it was possible a single cooler could get such good performance on both LGA1700 and AM5, when Noctua had to rely on at least two different bases to achieve this.

@PaulAlcorn what do you think about even adding a note about this, in the Verdict section? Given the potential of contact frames to void motherboard warranties, I think we can't just assume everybody is going to use one. If they're not, then this might not be the best cooler for them.

My bad, I must have somehow snipped the section about it on testing methodology when I updated that part when I incorporated AMD's Ryzen 9 9950X3D. All of my testing on Intel's i7-14700K is performed with a contact frame to insure bending does not impact results.

 

[bit_user]

My bad, I must have somehow snipped the section about it on testing methodology when I updated that part when I incorporated AMD's Ryzen 9 9950X3D. All of my testing on Intel's i7-14700K is performed with a contact frame to insure bending does not impact results.

Thanks for the reply.

I am concerned that too many people would miss it, if you only mention it in the section about Test Setup. It seems to me that the entire reason the Verdict is right at the top of the article is an admission that many people will not read the entire thing. Therefore, due to the significance of a Contact Frame on the cooler's Intel performance, I think it's something that deserves to be highlighted.

I would not be at all surprised if the convex base version of Noctua's NH-D15 G2 outperformed this, when both are used with the standard ILM (i.e. no contact frame).

 

[Albert.Thomas]

Thanks for the reply.

I am concerned that too many people would miss it, if you only mention it in the section about Test Setup. It seems to me that the entire reason the Verdict is right at the top of the article is an admission that many people will not read the entire thing. Therefore, due to the significance of a Contact Frame on the cooler's Intel performance, I think it's something that deserves to be highlighted.

I would not be at all surprised if the convex base version of Noctua's NH-D15 G2 outperformed this, when both are used with the standard ILM (i.e. no contact frame).

I've been trying to find the words to respond to this correctly, I sometimes have issues finding the right words due to the traumas that I have survived.

I prefer to use a contact frame because it insures consistency in results.

I recognize that bending can impact results, and for those who have CPUs which have suffered bending a convex solution may be more ideal.

But that begs a question: Wouldn't the efficiency of a convex plate vary depend on the extent a CPU has bent? I would imagine that would create a huge variation in testing results. With that in mind, is it really better to test without a contact frame? Wouldn't I need multiple CPUs with differing levels of bending to account for this properly?

 

[thestryker]

I have lots of thoughts about using a contact frame, but I'll start by saying that for a reviewer I don't think there's any other option than to use one.

The amount of impact a contact frame will have up front on performance is largely related to the design of the motherboard it's being used with and the cooler cold plate design. A $100 board is going to bend more than a $500 reinforced one. Likewise the flatter the cold plate is the worse it will perform without one. While I wouldn't recommend contact frames for every builder personally I wouldn't use any of the high wattage LGA1700 parts without one.

I haven't heard of air coolers doing this in general (the exception of course being Noctua having 3 versions of the D15 G2), but Asus and Thermalright have used a highly convex center (comparatively speaking) on their AIOs. While this will not typically be a negative from an end user point of view it can skew review results (depending on reviewer hardware setup) which doesn't happen with a contact frame. Obviously testing a D15 G2 HBC with a contact frame would be a disservice, but doing a general review on a cooler with a cold plate specifically designed to counter bending is also of nebulous value unless testing it on a brand new CPU/motherboard.

The other big issue for reviewers is bending over time. Even on a reinforced board due to the poor design from Intel it will get worse over time. While that won't matter for someone using a system, because the cooler will deform with it, for reviewers who are constantly testing new coolers it's a problem.

I don't think there's a good way to explain all the variables and nuance behind using one or not in a succinct enough way for cooler reviews. I do think the paragraph that has been used is about as good as it can be. I don't really think there's any particular reason to change how it has been noted in the past, just that it needs to be noted in any reviews where it's used. If anything were to be changed I think a note in the charts themselves next to the CPU saying something along the lines of "with contact frame" would be the way to go.

 

[bit_user]

I've been trying to find the words to respond to this correctly, I sometimes have issues finding the right words due to the traumas that I have survived.

Thanks for mentioning this. I can sometimes be rather blunt, but I hope I didn't cause offense or alarm. I'm just trying to look at this from a perspective of arming readers with all the information to make the best decision for their needs.

I prefer to use a contact frame because it insures consistency in results.

I'm not trying to argue against your use of a contact frame. I have one, on my LGA 1700 board, as I think many of us do. However, we should also recognize that not everyone will, and many are still unaware of the benefits. That's why I think it deserves to be mentioned in a place that's more prominent.

But that begs a question: Wouldn't the efficiency of a convex plate vary depend on the extent a CPU has bent?

This seems to be why Noctua created 3 different bases for their NH-D15 G2. In their testing, they include the case of a CPU that was initially used with the standard ILM, but then the user switched to using that (or the "washer mod').

Otherwise, I think they only would've needed 2 different bases for the G2 - a flat and convex one.

would imagine that would create a huge variation in testing results. With that in mind, is it really better to test without a contact frame? Wouldn't I need multiple CPUs with differing levels of bending to account for this properly?

In the ideal world, one could test everything. I can see an argument for/against contact frames. Given the enthusiast focus of this site and limited testing resources, I agree with your decision to use a contact frame. I just think it should be noted that the verdict was based on using one.
: )

 

[EvilScotsman]

Can you guys please make a point of including the previous revisions of a company's cooler in your benchmarks. You have the royal knight and royal pretor which are both 2025 revisions, but the people considering buying these will often be wanting to compare them to prev. revisions. For example, the phantom spirit. Including the peerless assassin 120 and phantom spirit would've been good for this since the PA is so popular. Could easily drop a couple of the AIO's to put more air coolers in. Doesn't make sense to have quite so many AIO's in an air cooler comparison benchmark. Your benchmarks are a great resource, but the selection of included coolers is always a bit weird.

 

[Albert.Thomas]

Can you guys please make a point of including the previous revisions of a company's cooler in your benchmarks. You have the royal knight and royal pretor which are both 2025 revisions, but the people considering buying these will often be wanting to compare them to prev. revisions. For example, the phantom spirit. Including the peerless assassin 120 and phantom spirit would've been good for this since the PA is so popular. Could easily drop a couple of the AIO's to put more air coolers in. Doesn't make sense to have quite so many AIO's in an air cooler comparison benchmark. Your benchmarks are a great resource, but the selection of included coolers is always a bit weird.

Thanks for the feedback. I agree that we need more air cooling comparison results. I don't plan on changing the test setup again for another year or so. If I continue to send my reviews to Tom's Hardware, I'll focus on getting more air coolers tested!

 

[Jamminator]

The reason ThermalRight is so cheap is a combination of, A) they own their own factories that do OEM work, B) most of their products are surplus from the pandemic bump that caused shortages and factories to ramp up production, now they have excess inventory they can't move.

I know some of their AIOs sold dirt cheap on Amazon are actually a couple years old. You also see reports of people complaining their included TX7 thermal paste is dried out (because it's old).

 

[bit_user]

The reason ThermalRight is so cheap is a combination of, ... B) most of their products are surplus from the pandemic bump that caused shortages and factories to ramp up production, now they have excess inventory they can't move.

Well, that might apply to some of their older models and some of their sub-components, but even their newest model coolers are quite inexpensive.

I know some of their AIOs sold dirt cheap on Amazon are actually a couple years old. You also see reports of people complaining their included TX7 thermal paste is dried out (because it's old).

Interesting data point. I never use the thermal compound included with coolers, so I wouldn't even know!

 

[liq7]

Was just about to buy PS120 Evo but looks like it’ll have to be this instead! It’s a shame there is no RGB to break up the plain black square. The K12 fans included with the 120 Evo was tastefully done.

Edit I've just realised another known reviewer has already compared this to the PS120 Evo and the Royal Pretor actually got outperformed by it so now I'm really confused. The Evo has one extra heatpipe but not as chunky fans. At this point clearly both of them are perfectly adequate but they are actually identically priced right now for me so would rather not miss out on a free 5% performance. Please please test the PS120 Evo on your current setup since it was last tested on the previous gen and is probably the most used air cooler for enthusiasts/gamers

 

[8086]

They are not compatible. Noctua's mounting system is 100% proprietary, as are most all aftermarket air coolers. Not sure why you'd need to use the Noctua mount anyhow since basically all aftermarket coolers other than very cheap top down style models generally come with their own backplate, bracket and hardware.

I currently have Noctua installed in a few of my machines and I have tons of surplus mounting hardware lying around. Being able to "upgrade" by dropping in a new cooler just makes life easier.

 

[Jamminator]

Well, that might apply to some of their older models and some of their sub-components, but even their newest model coolers are quite inexpensive.

Well even newer models still have a lot of crossover parts...fittings, radiator block, etc. But ya, TR is basically the factory selling out their backdoor direct to consumers. No distributors or middlemen markup.

 

[tristanx]

Constant price hike of Noctua made it disappear from my radar. Might get Royal Pretor and screw around with it.