is the nh-d15 overrated?

[Guest]

8 minutes ago, WoodenMarker said:

Did you read the post I linked? If not, I'll just paste it here:

The LGMRT matches or falls behind the NH-D15S half the time and only just matches the NH-D15 in a few cases. 

Beating the NH-D15 in a test result or two out of many doesn't mean it's just as good.

Here's another example of how lower loads don't differentiate coolers well: https://www.tweaktown.com/reviews/6958/cryorig-h7-cpu-cooler-review/index6.html

The H7 beats the NH-D14 at stock. The performance flips and the gap is wider once oc'ed. 

Yes, these are true, real-life results but they won't reflect the norm. 

All of those sources (besides nikkitech, but I very much doubt the reliability of that source (https://old.reddit.com/r/sffpc/comments/9gx7tl/this_nikktech_review_shows_that_argon_ar05_is_9/)) show the LGM being 1-2+- within the NH-D15, which can be attributed to sensor error (Besides, even if the sensors were 100% perfect, it wouldn't matter, because whether an OC is stable or not isn't going to be decided by 1-2 measly degrees). 1-2 degrees which is most likely sensor error shouldn't decide the tier a cooler goes in.

[corrado33]

2 minutes ago, ears_ears said:

All of those sources (besides nikkitech, but I very much doubt the reliability of that source (https://old.reddit.com/r/sffpc/comments/9gx7tl/this_nikktech_review_shows_that_argon_ar05_is_9/)) show the LGM being 1-2+- within the NH-D15, which can be attributed to sensor error (Besides, even if the sensors were 100% perfect, it wouldn't matter, because whether an OC is stable or not isn't going to be decided by 1-2 measly degrees). 1-2 degrees which is most likely sensor error shouldn't decide the tier a cooler goes in.

Weren't you arguing in your OP over a 0.25 degree difference? Saying how the LGM was "better" than the D15 because of it?

[Guest]

1 minute ago, corrado33 said:

Weren't you arguing in your OP over a 0.25 degree difference? Saying how the LGM was "better" than the D15 because of it?

No, I was arguing that the LGM doesn't deserve to be 1 tier below the NH-D15 just because the brand isn't as well known. They have identical performance for all intents and purposes; 1-2 degrees isn't going to change a thing.

[corrado33]

Just now, ears_ears said:

No, I was arguing that the LGM doesn't deserve to be 1 tier below the NH-D15 just because the brand isn't as well known. They have identical performance for all intents and purposes; 1-2 degrees isn't going to change a thing.

And the fact that the noctua comes with 2 high quality fans vs one for the LGM? Allowing for redundancy, push pull efficiency gains, and likely longer life? 

 

If you've forgotten, the D15S (aka the single fan version) is in the same tier as the LGM. 

[Guest]

9 minutes ago, WoodenMarker said:

I do believe there's a case to be had. Otherwise, it wouldn't make sense for me to have been thinking about this for the past week. 

'Common sense' with how the LGMRT is supposed to be better than the Macho Rev. B along with benchmarks showing it performing similarly to the NH-D15 had me suspecting if it should be moved up. This tends to happen to many coolers on the tier list but I tend to find the result I originally settled on as correct after double and triple-checking old reviews as well as trying to find new ones for a better comparison. 

I've noticed that mid-priced coolers (windale 6, macho direct, H60 2018, H5, fuma rev.b) tend to be much lower on the tier list than their performance demonstrates.

[corrado33]

1 minute ago, ears_ears said:

I've noticed that mid-priced coolers (windale 6, macho direct, H60 2018, H5, fuma rev.b) tend to be much lower on the tier list than their performance demonstrates.

If you would read the text surrounding the rankings, you'd know that the rankings are based on fan quality/bearing quality/mounting hardware quality/ease of mounting as well. 

[WoodenMarker]

8 minutes ago, ears_ears said:

All of those sources (besides nikkitech, but I very much doubt the reliability of that source (https://old.reddit.com/r/sffpc/comments/9gx7tl/this_nikktech_review_shows_that_argon_ar05_is_9/)) show the LGM being 1-2+- within the NH-D15, which can be attributed to sensor error (Besides, even if the sensors were 100% perfect, it wouldn't matter, because whether an OC is stable or not isn't going to be decided by 1-2 measly degrees). 1-2 degrees which is most likely sensor error shouldn't decide the tier a cooler goes in.

You linked a reddit post for SFF coolers.

Results will vary depending on the system airflow and the performance changes can be drastic in SFF cases. It's nearly impossible to find two reviewers with results that don't 'disagree' in many places if you assume performance is absolutely static and independent on other factors. 

3 minutes ago, ears_ears said:

1-2 degrees isn't going to change a thing.

That's the difference between the tiers most of the time. I could combine tiers but what's the point in losing granularity?

2 minutes ago, ears_ears said:

I've noticed that mid-priced coolers (windale 6, macho direct, H60 2018, H5, fuma rev.b) tend to be much lower on the tier list than their performance demonstrates.

A good rule of thumb is that coolers in a tier will match but not usually beat the performance of coolers in the tiers directly above and below it. What you won't find for example is a Tier 5 cooler consistently beating Tier 4 coolers. 

[WoodenMarker]

1 minute ago, corrado33 said:

If you would read the text surrounding the rankings, you'd know that the rankings are based on fan quality/bearing quality/mounting hardware quality/ease of mounting as well. 

Uh, are we looking at the same list or did I make a typo? My list specifically states that it focuses on performance and nothing else.

[corrado33]

Just now, WoodenMarker said:

Uh, are we looking at the same list or did I make a typo? My list specifically states that it focuses on performance and nothing else.

I remember reading about fan bearing quality as well? Maybe different lists

[Guest]

On 12/7/2018 at 7:38 PM, corrado33 said:

And the fact that the noctua comes with 2 high quality fans vs one for the LGM? Allowing for redundancy, push pull efficiency gains, and likely longer life? 

 

If you've forgotten, the D15S (aka the single fan version) is in the same tier as the LGM. 

Can we please get rid of the adjectives? 

 

A better way to describe the NH-D15 vs the LGM is, "And the fact that the noctua comes with 2 fans vs one for the LGM?". Getting rid of the highly subjective tone, and therefore making your viewpoint seem more objective. 

 

Fans last a very long time. You'd be hard pressed to find a fan that fails, even after 5+ years of usage. Also, they're very easy to replace. The heatsink is what does 90% of the job anyway.

 

On 12/7/2018 at 7:43 PM, WoodenMarker said:

You linked a reddit post for SFF coolers.

Results will vary depending on the system airflow and the performance changes can be drastic in SFF cases. It's nearly impossible to find two reviewers with results that don't 'disagree' in many places if you assume performance is absolutely static and independent on other factors. 

That's the difference between the tiers most of the time. I could combine tiers but what's the point in losing granularity?

A good rule of thumb is that coolers in a tier will match but not usually beat the performance of coolers in the tiers directly above and below it. What you won't find for example is a Tier 5 cooler consistently beating Tier 4 coolers. 

If the average temp difference between tiers is 1-2 degrees, there's a problem. 

 

IMO the tiers should be lowered to 5. 

 

1 - 360mm, 280mm, and high end 240mm AIOs like corsair h110i, h100i, thermaltake floe riing 360mm, CLC 280

2 - High end Air coolers and 240mm AIOs like nh-d15, LGM, dark rock pro 4, kraken x62, EVGA CLC 240, R1 ultimate

3 - Mid-high end air coolers (typically 3-5 degrees or less within tier 2): FSP windale 6, thermalright macho direct, macho rev.b, scythe FUMA, mugen, CRYORIG H5, dark rock 4

4 - Better 212 evo's: H7, freezer 33 esports, windale 4

5 - 212 evo level: Gammaxx 400, Pure rock slim (if you change the fan curve in the BIOS), m9a/m9i, 212 evo, Freezer 33 non-esports edition

[WoodenMarker]

Just now, ears_ears said:

IMO the tiers should be lowered to 5. 

For what benefit?

[Guest]

2 minutes ago, WoodenMarker said:

For what benefit?

Having 9 tiers with only a 1-2 degree difference between each just serves to overcomplicate things, especially for people who aren't as experienced with coolers. I know I'm getting overwhelmed with the sheer amount of tiers and coolers on the list.

[WoodenMarker]

4 minutes ago, ears_ears said:

Having 9 tiers with only a 1-2 degree difference between tiers just serves to overcomplicate things, especially for people who aren't as experienced with coolers. I know I'm getting overwhelmed with the sheer amount of tiers and coolers on the list.

What's complicated about it? A cooler in a higher tier performs better. That's all there is to it. The purpose of the list is to separate coolers when performance blurs and overlaps in many reviews. I'd add more tiers if possible but that takes a lot more data and time. 

There are 550+ coolers tiered and it's only getting longer. *shakes fist at obscure coolers I can't find good reviews on*

Feel free to ask questions if something isn't clear. 

[corrado33]

25 minutes ago, ears_ears said:

Can we please get rid of the adjectives? 

 

A better way to describe the NH-D15 vs the LGM is, "And the fact that the noctua comes with 2 fans vs one for the LGM?". Getting rid of the highly subjective tone, and therefore making your viewpoint seem more objective. 

 

Fans last a very long time. You'd be hard pressed to find a fan that fails, even after 5+ years of usage. Also, they're very easy to replace. The heatsink is what does 90% of the job anyway.

 

So are you arguing that there are no push pull benefits, and redundancy isn't a good thing? 

 

Color me confused as those two FACTS are well known to be TRUE. Also, noctua DOES make some of the highest quality fans you will find. There's a reason they're well known. And no, fans don't USUALLY outright fail, but they do get louder over time. 

 

YOU should get rid of your argumentative tone. Every thread you've been in you've tried to argue. 

 

The tier list is based off of the best data that the author could find. Just because YOU don't like it doesn't mean it needs to change. Many more people (who aren't confused by the number of tiers) like it, therefore it will stay. 

[Guest]

I personally think having less tiers but higher temp difference would clear up a lot of things, but I respect your opinion. Thank you for taking the time to explain your side of the argument to me. 

[Guest]

Sorry for bumping, but I found absolutely NOTHING that suggests the D15 is in any way better than the LGM even with its dual fan setup

 

https://www.tweaktown.com/reviews/7986/thermalright-le-grand-macho-rt-cpu-cooler-review/index6.html

https://www.techpowerup.com/reviews/Thermalright/Le_Grand_Macho/6.html

 

Also quieter as well. 

 

And to Mr. Corrado33, there may be a dual fan setup on the NHD15 and push pull, but if there's no real life effect, it's basically mute. The Thermalright uses more metal on the heatsink to achieve its cooling effect. Also, fans have a limited effect on the effectiveness of a cooler; it's the heatsink that really makes things work. 

 

and as to WoodenMarker's sources:

 

http://hardwareoverclock.com/Thermalright-Le-Grand-Macho-RT-5.htm (This could be explained by the fan curve, as the temps aren't even hitting 60C, but unless the RPM of both the D15 and the LGM are shown we have no way of knowing.) 

https://www.nikktech.com/main/articles/pc-hardware/cpu-cooling/cpu-air-coolers/7107-thermalright-le-grand-macho-rt-cpu-cooler-review?showall=&start=5 (I searched up Nikkitech, almost nothing came up. I don't trust this source, no one on reddit even talks about it nor on overclocker forums. They probably didn't even use probes. Also, fan curve.)

https://www.hardwareluxx.de/index.php/artikel/hardware/kuehlung/38997-thermalright-le-grand-macho-mit-ty-147-b-im-nachtest.html?start=4 (Again, fan curve. The only reliable way to test the effectiveness of a cooler is max load max OC.)

https://overclockers.ru/lab/show/73449_4/obzor-i-testirovanie-passivnoj-sistemy-ohlazhdeniya-thermalright-le-grand-macho (Wow, it's the Russian site that shows RPM. You can clearly see they have different fan curves: single fan NHD15 and LGM have the same temps at 1500RPM and 1200RPM respectively. The NHD15S brought down to the RPM of the LGM goes 2 degrees higher.)

https://www.techpowerup.com/reviews/Thermalright/Le_Grand_Macho/6.html (Funnily enough, it's the same source I'm using. I'm assuming you meant the very last benchmark with the FPU load. This is an anomaly, as every other benchmark on that site tested them to be the exact same. This could be explained by: probe inaccuracy, the uncertainty of electronics, slight differences in the ambient temperature, differences in paste applicance, etc.) 

http://pcfoster.pl/artykul/thermalright-macho-le-grand-rt-–-test-chlodzenia/1655-4.html (They set the fan curve, but possible external factors make the differences mute.) 

https://whatnext.pl/test-chlodzenia-thermalright-le-grand-macho-rt/ (Where is the NH-D15 on this?)

 

Overall the main problem with all of these sources is that none of them were tested with 70+C temps, making fan curve a big factor into the temp differences here. (Except techpowerup and the Russian site, but neither of those sources helps prove the point that the NH-D15 is better than the LGM.)

[aezakmi]

Tl;dr but yes, it is.

[WoodenMarker]

3 hours ago, ears_ears said:

http://hardwareoverclock.com/Thermalright-Le-Grand-Macho-RT-5.htm (This could be explained by the fan curve, as the temps aren't even hitting 60C, but unless the RPM of both the D15 and the LGM are shown we have no way of knowing.) 

https://www.nikktech.com/main/articles/pc-hardware/cpu-cooling/cpu-air-coolers/7107-thermalright-le-grand-macho-rt-cpu-cooler-review?showall=&start=5 (I searched up Nikkitech, almost nothing came up. I don't trust this source, no one on reddit even talks about it nor on overclocker forums. They probably didn't even use probes. Also, fan curve.)

https://www.hardwareluxx.de/index.php/artikel/hardware/kuehlung/38997-thermalright-le-grand-macho-mit-ty-147-b-im-nachtest.html?start=4 (Again, fan curve. The only reliable way to test the effectiveness of a cooler is max load max OC.)

https://overclockers.ru/lab/show/73449_4/obzor-i-testirovanie-passivnoj-sistemy-ohlazhdeniya-thermalright-le-grand-macho (Wow, it's the Russian site that shows RPM. You can clearly see they have different fan curves: single fan NHD15 and LGM have the same temps at 1500RPM and 1200RPM respectively. The NHD15S brought down to the RPM of the LGM goes 2 degrees higher.)

https://www.techpowerup.com/reviews/Thermalright/Le_Grand_Macho/6.html (Funnily enough, it's the same source I'm using. I'm assuming you meant the very last benchmark with the FPU load. This is an anomaly, as every other benchmark on that site tested them to be the exact same. This could be explained by: probe inaccuracy, the uncertainty of electronics, slight differences in the ambient temperature, differences in paste applicance, etc.) 

http://pcfoster.pl/artykul/thermalright-macho-le-grand-rt-–-test-chlodzenia/1655-4.html (They set the fan curve, but possible external factors make the differences mute.) 

https://whatnext.pl/test-chlodzenia-thermalright-le-grand-macho-rt/ (Where is the NH-D15 on this?)

 

Overall the main problem with all of these sources is that none of them were tested with 70+C temps, making fan curve a big factor into the temp differences here. (Except techpowerup and the Russian site, but neither of those sources helps prove the point that the NH-D15 is better than the LGM.)

The reviews have a testing methodology section.

Hardware overclock 

Quote

2nd fan speed:

The maximum speed of the respective fan is used because PWM controllers distort the result.

NikkiTech: 

Quote

Every CPU cooler in this database is tested with the bundled 140mm/120mm/92mm/80mm fans while working at both idle speed and 100% of their speeds for all the temperature tests.

Hardwareluxx:

Quote

Temperature in degrees Celsius
Series fan, maximum speed

TechPowerUp:

Quote

During all these tests, fans are set to run at 100% in the BIOS, with temperatures being recorded by AIDA64.

For Techpowerup's review, 'every other benchmark on that site tested them to be the exact same' also shows the NH-D15S performing the same or even better than the NH-D15. There's some interpretation here but I see that meaning the test isn't heavy enough to differentiate similarly performing coolers like the NH-D15 and the NH-D15S. The last test is actually heavy enough to show a 2c difference. The 1c error of margin they state doesn't account for this.

 

PCfoster:

Quote

st. c [mniej=lepiej], went.fabr. MAX

means st. c [= better less] fan. Factory. MAX

 

As for whatnext.pl, the Ninja 4 and Fuma are all coolers on the list. There aren't always direct comparisons so comparing it to similar coolers means more data and hopefully a more accurate representation of performance. Are you going to say that the Ninja 4 is better than the NH-D15?

 

70c+ is an arbitrary point that will favor some coolers more than others and exclude many potential points of data -- so will looking only at direct comparisons. Lack of data will mean a much higher margin for error.

What's taken into mind when creating the tier list is whether a cooler on average is better than the cooler below it and worse than the ones above it. This means comparing the cooler to as many others as possible and not one specific listed in a tier.

Probe inaccuracy, the uncertainty of electronics, slight differences in the ambient temperature, and differences in paste applicance are factors that exist but aren't good reason to ignore a result. The point of having so many reviews to look at is that enough data points will create a more accurate average despite these factors. They are real life results that are varied like actual results from users of the same cooler in different systems. 

[vjizzle2384]

On 12/6/2018 at 8:15 PM, PopsicleHustler said:

I have Dark Rock 4 cooling R7 2700x, and from performance charts I found on the web, its almost identical performance-wise. And it looks miles better. So yea, its a little bit overrated.

I want to get one for my 2600x you think it's better than a 240mm Liquid AIO cooler ?

[PopsicleHustler]

3 hours ago, vjizzle2384 said:

I want to get one for my 2600x you think it's better than a 240mm Liquid AIO cooler ?

240mm who is an overkill for 2600x. A decent air cooler will be enough to hit 4.2Ghz.

[_Hustler_One_]

I will always simply recommend TLGMRT to anyone who asked for an air cooler for their high end CPU to OC, just because it is big and have serious surface area, despite its size it's only weighing 1060 grams with the fan included (not too much stress onto the motherboard) and only 159mm height, also it offers full compatibility, no conflict to RAM, VRM heatsink, and Pci-E slot, compatible with most of the case on the market because of the 159mm height, also it's quiet and top tier performance, it has black top plate heatsink fin to give a little flavor of aesthetics, and relatively cheaper, what more can you expect?

 

Regarding warranty, it doesn't matter too much as air cooler by general has a little point of failure and very durable, so for any 2 or 6 years warranty, for me it doesn't count so much. TLGMRT is relatively cheaper than NH-D15 because of its warranty while performs at similar level. I think the extra price for 6 years warranty on NH-D15 is indeed looks reasonable but not necessarily for us to pay more to get 6 years of something that we most likely won't get the use of it.

 

Why while I say this I use NH-D15 instead? Doesn't it contradict to what I stated? It's because I wanted to have a top tier performance and top tier aesthetics at the same time. I feel there is no other air cooler which offers both performance and aesthetics at the highest level except NH-D15 fully black Chromaxed, it's very elegantly sleek and classy, even I must pay the extra only for the looks, but for me it's totally worth it! The total price I paid to get the NH-D15 to be fully black chromaxed is still much better than the awful 'Cooler Master Air Maker 8'.

Other than that, I still choose TLGMRT as my top choice of an air cooler.

[AngryBeaver]

This whole Argument is pointless. All that matters is according to the list Water > Air. So stop arguing over which Air cooler is the best of the second rate options. Just go out and get a nice AIO and be happy you are now a member of the water-cooled master race.

 

I mean you can spend $90 on a NH-D15 or you can spend 60 on an ML240 lite with RGB. The ML will perform similar if not better in some situations, it is still quiet enough to be mostly inaudible unless running full out. Plus you get a massive bump in Aesthetics.

 

If you don't want a cheap AIO then you can toss in an extra 20 bucks to get something more high end.

[_Hustler_One_]

20 minutes ago, AngryBeaver said:

This whole Argument is pointless. All that matters is according to the list Water > Air. So stop arguing over which Air cooler is the best of the second rate options. Just go out and get a nice AIO and be happy you are now a member of the water-cooled master race.

I will never get the convenience to go water cooling as their risk and point of failure is higher than the air coolers. So this discussion only matters to someone like me or others who never or will never ever go to liquid cooling.

 

Just can't imagine if things like this happen to me..

 

[aisle9]

13 hours ago, ears_ears said:

Sorry for bumping, but I found absolutely NOTHING that suggests the D15 is in any way better than the LGM even with its dual fan setup

NH-D15: 6-year warranty

Le Grand Macho: 2-year warranty

 

If I'm looking at a cooler to run passively in a low-TDP system (and the LGM is very capable of that), I'm buying the Thermalright. If I'm looking for something that's going to require active cooling and the case has room for it, I'm going for a cooler with a longer warranty. Good fans don't often fail entirely, but they will start clicking over time. Replacing them with equivalent fans would cost a significant percentage of the overall price of the cooler, and warranty support that ends in two years is a red flag.

[AngryBeaver]

11 minutes ago, _Hustler_One_ said:

I will never get the convenience to go water cooling as their risk and point of failure is higher than the air coolers. So this discussion only matters for someone like me or others who never or will never ever go to liquid cooling.

 

Just can't imagine if things like this happen to me..

 

So on this point. This is just the typical run around you can sometimes get in other countries when trying to get RMA suppport. They always recommend you contact the seller, but they are also able to handle the warranty and are required to if the seller does not hold up their end.

 

When buying an AIO it is important to look at the warranty of it. Most of the big names out there cover the item and any parts it damages in the event of a failure. I am pretty sure CM does too, but their website is showing down for me atm so I cannot confirm.

 

So some of the problems with the post you link are that the OP hasn't taken the fight to them yet. They have a legal obligation to honor their warranty and any parts that their product damage due to being faulty. So once the OP does what is needed he will walk away from this once again whole.

 

Now if we were talking about custom water cooling then you would have a valid argument, but when dealing with an AIO your butt is normally covered. So while there might be more points of failure with an AIO it doesn't mean they happen often. The biggest concern is the leak and those are very very rare. The other items do fail time to time, but that is the same with an air cooler and the fan dieing. The only major downside to an AIO is that they have a maximum lifespan which is when the fluid inside evaporates enough it no longer functions, but this is normally outside the warranty period and I would replace these when the warranty is up just to stay covered if it does fail.

 

So in the end I have no problem recommending an AIO to someone. They are perfectly safe imo and leaks are about as rare as someone killing their board with a giant air cooler (Yes it happens). The major difference is the people that kill a board via the heavy cooler causing damage either overtime or when moved don't normally know why it won't boot and don't normally have an obvious sign. Where as if a AIO leaks you can see it pretty easily.