The Megahalem vs H50 Battle Royale Results

[kyleax1]

The Heavy Weight Mega Vs The Light Weight H50

vs

I have owned a mega since I built my system last month, but wanted to test it against the H50. I have finally finished all the tests and the graphs have been updated with the new info. Based on my results I got better results by pulling air from inside my case and exhausting it out the rear even though Corsair suggests you to do it the other way around.

IBT(Intel Burn Test also known as Linpack) has been run at High to ensure the same amount of ram is used in every test and this is the test I used to find the max load temps.

I tested 3 settings:Stock, 3.9ghz with HT on, 4.1ghz with HT off
I ran 4 variations to these 3 tests: Idle, Window Panel, Fan Panel, Gaming
I even threw in the GPU temps so people could see how the panels and coolers affected the temps of the GPU during normal game play. I don't think the gaming/GPU results were very accurate though, I feel they varied some based on the maps played. I should have just ran vantage, but o well.
UT3 was the game used in all the gaming tests


This was not a stability test, the testing was only done to see temps.


Test System:
i7 860
CM Storm Sniper
Asus Maximus III
Geforce GTX 295
Windows 7
IC7 thermal paste
Ambient temp 20c
2x Ultra Kaze 3k @ 100% Push/Pull used on mega and H50

Pics of the system:

What I did was averaged the 4 cores temps I got to give a straight average for each test, however the detailed results are in the 2nd post. I took screenshots of every test ran showing: temps, cpu verification, IBT verification, and GPU temp verification and the link to them is also in the second post.



I put together graphs for easy reading, but as I said before if you want detailed results just read the 2nd post.

Quick Brake Down:


Stock
IBT- Mega wins
Gaming- Mega wins
Idle- H50 exhaust wins


3.9ghz Hyperthreading On
IBT- H50 exhaust wins
Gaming- Mega wins
Idle- H50 intake wins


4.1ghz Hyperthreading Off
IBT- H50 exhaust wins
Gaming- Mega wins
Idle- H50 exhaust wins


As you can see they both trade blows, the mega does better while playing games when only part of the cpu is used, the H50 exhaust does better during the Intel Burn Tests and they swap blows on Idle(but Idle results are very inaccurate no matter how you look at it).



Pro's and Con's

H50 Pro's
*Small- Good for smaller cases and ram with high heat spreaders
*Good for window cases- Temps are the same in IBT with side fan & window
*Easy to Install
*Good price/performance ratio

H50 Con's
*Push/Pull on 38mm fans difficult to mount- had to use zip ties


Mega Pro's
*Better cooling if running your cpu on stock settings
*Had lower temps while gaming
*Easy to Install
*Good price/performance ratio

Mega Con's
*Big- May not fit in all cases
*Need for good airflow- If your case has bad airflow it might not cool as well
*May have to pay extra for mounting brackets for your cpu
*Have to pay extra for 38mm fan clips or make your own if using 38mm fans



Conclusion:
Both coolers appear to do great for the money. I think it just comes down to case airflow, case size, ram size and personal taste in looks. I do not see either of these 2 coolers achieving a higher overclock then the other.





Discuss freely and enjoy looking over my results

Keep in mind these results only speak for the i7 860 and results may vary based on case airflow and cpu type.

 

[kyleax1]

Picture Album of every screenshot taken of every test ran


Detailed test results:


Megahalems Push/Pull

Stock
Megahalems side fan on Idle: 34-34-32-28 =32avg
Megahalems side fan on Load IBT 5 loops: 50-50-50-49 =50avg
Megahalems side fan on Load games 10min: 39-39-40-37 =39avg
GPU: 62
Megahalems side window Idle: 32-33-38-29 =33avg
Megahalems side window Load IBT 5 loops: 50-50-51-50 =50avg
Megahalems side window Load games 10min: 43-42-43-40 =42avg
GPU: 66
3.9 ghz HT on 1.375v
Megahalems side fan on Idle: 48-55-54-44 =50avg
Megahalems side fan on Load IBT 5 loops: 88-92-84-93 =89avg
Megahalems side fan on Load games 10min: 63-62-57-57 =60avg
GPU: 63
Megahalems side window Idle: 51-55-46-47 =50avg
Megahalems side window Load IBT 2 loops: 88-90-82-91 =88avg (attempted 5 times, failed each time)
Megahalems side window Load games 10min: 65-66-59-60 =62.5avg
GPU: 66
4.1 ghz HT off 1.375v
Megahalems side fan on Idle:41-53-46-49 =47avg
Megahalems side fan on Load IBT 5 loops: 86-87-80-88 =85.5avg
Megahalems side fan on Load games 10min:59-60-59-60 =59.5avg
GPU: 62
Megahalems side window Idle: 39-44-55-42 =45avg
Megahalems side window Load IBT 5 loops:86-89-81-88 =86avg
Megahalems side window Load games 10min: 61-63-60-61 =61avg
GPU: 66



Corsair H50 Push/Pull (pulling cool air from outside of case)

Stock
Corsair H50 side fan on Idle: 35-31-38-29 =33avg
Corsair H50 side fan on Load IBT 5 loops: 53-51-53-51 =52avg
Corsair H50 side fan on Load games 10min: 43-42-45-40 =42.5avg
GPU:63
Corsair H50 side window Idle: 34-35-39-33 =35avg
Corsair H50 side window Load IBT 5 loops: 53-52-54-50 =52avg
Corsair H50 side window Load games 10min: 43-42-45-40 =42.5avg
GPU:64
3.9 ghz HT on 1.375v
Corsair H50 side fan on Idle: 41-40-49-39 =42avg
Corsair H50 side fan on Load IBT 5 loops: 88-90-87-90 =89avg
Corsair H50 side fan on Load games 10min: 69-66-61-59 =64avg
GPU:62
Corsair H50 side window Idle: 42-44-44-42 =avg43
Corsair H50 side window Load IBT 5 loops: 90-89-89-90 =89.5avg
Corsair H50 side window Load games 10min: 66-68-66-60 =65avg
GPU:63
4.1 ghz Ht off 1.375v
Corsair H50 Caseless Load IBT 5 loops: 85-84-83-84 =84avg
Corsair H50 side fan on Idle: 43-48-44-45 =45avg
Corsair H50 side fan on Load IBT 5 loops: 85-85-83-85 =85avg
Corsair H50 side fan on Load games 10min: 65-64-65-63 =64avg
GPU:62
Corsair H50 side window Idle: 41-46-43-38 =42avg
Corsair H50 side window Load IBT 5 loops: 86-85-83-86 = 85avg
Corsair H50 side window Load games 10min: 63-63-65-62 = 63avg
GPU:61



Corsair H50 Pull/Push (blowing hot air out rear of case)

Stock
Corsair H50 side fan on Idle: 35-33-35-31 =33.5avg
Corsair H50 side fan on Load IBT 5 loops: 50-51-52-50 =51avg
Corsair H50 side fan on Load games 10min: 46-44-47-44 =45avg
GPU:63
Corsair H50 side window Idle: 31-30-33-28 =30.5avg
Corsair H50 side window Load IBT 5 loops: 54-53-55-52 =53.5avg
Corsair H50 side window Load games 10min: 44-43-45-41 =43avg
GPU:65
3.9 ghz HT on 1.375v
Corsair H50 side fan on Idle: 45-40-48-41 =43.5avg
Corsair H50 side fan on Load IBT 5 loops: 90-89-84-89 =88avg
Corsair H50 side fan on Load games 10min: 67-66-65-60 =64.5
GPU:63
Corsair H50 side window Idle: 40-43-47-39 =42avg
Corsair H50 side window Load IBT 5 loops: 86-89-85-91 =88avg
Corsair H50 side window Load games 10min: 67-60-61-63 =63avg
GPU:64
4.1 ghz Ht off 1.375v
Corsair H50 side fan on Idle: 40-38-42-43 =41avg
Corsair H50 side fan on Load IBT 5 loops: 85-85-82-85 =84.5avg
Corsair H50 side fan on Load games 10min: 66-63-67-63 =65avg
GPU:62
Corsair H50 side window Idle: 47-43-46-44 =45avg
Corsair H50 side window Load IBT 5 loops: 85-85-83-85= 84.5avg
Corsair H50 side window Load games 10min: 61-60-61-59 =60avg
GPU:63

 

[3t3P]

wow very comprehensive testing well done, you've probably answered a lot of real-world questions
shame H50 won't fit in my case!

 

[the_electrician]

Fantastic effort.
I can see that you have put a lot of time and effort into mega vs h50.
Both are fantastic coolers.

I think the mega vs h50 should be on Harry Hill next week to which is best.

Fight!!!

 

[gary996]

Very nicely written well done

 

[Tolmekian]

Great first post, well done keep it up.
Now test the nh-d14 just as thoroughly when you can get your hands on one, it's looking to be a real winner.

 

[JonJ678]

Your first post here is results based on 38mm 3000rpm fans, and extra credit for being the first person I've seen use a shroud on a heatsink. Excellent.

Thank you for joining the forum.

 

[kyleax1]

Thanks

My home is overclock.net, but since I've seen a couple different forums with the same "should I get mega or H50" questions I decided to post my comparison on those sites as well.

I hope this information is useful to everyone.

A side note..adding a shard to the mega made no temp difference, but adding fans in the bay drive slots dropped the mega's temps by 2-3c on LinX load.

 

[Geek.]

good work many thanks for doing this!

 

[manbearpig]

thanks for doing this, i think i will go for the megahalems although they both seem about as good as each other!

 

[sorn21]

Excellent post mate, I was an early adopter of the H50 and I'm glad to see my wee cooler is still a rival to "high-end" air cooling.

 

[JonJ678]

You're probably sick to death of the testing process by now, but further to the shroud not helping (disappointed, I hoped it would do):

A line of pvc tape between shroud and fan to reduce air lost between them, and a piece of card taped on top of the megalahems to stop air escaping out of the top will likely improve matters. If the set up is as close to a sealed wind tunnel as possible then all the air goes over the fins rather than blowing out the sides. Air leaking reduces the effective pressure over the fins and so reduces the airflow.

It's more than reasonable if you never want to run comparisons again though, it's quite tedious

A con for the H50 is that it's a sealed unit, when too much of the coolant evaporates it's game over. It also has a pump, which inevitably makes noise. Also a risk of it springing a leak. I basically don't trust it over a block of metal.

I've read a fair few threads on overclock.net but never signed up, some very knowledgeable people there.

Cheers

 

[kyleax1]

Ya I have no interest in taking my mobo out again to switch coolers and taping up the mega. I am yet to see anyone go through that kind of trouble to see if it lowers temps.

However there has been a lot of fan volt/speed testing done on the mega and it has bwwn shown that 2x San Ace 38mm h101 fans in push pull produced the same temps when operating at 100% and at 60%.

 

[JonJ678]

In that case it's not airflow which is limiting and taping up the top will achieve nothing. Good to know, cheers.

 

[krooton]

A con for the H50 is that it's a sealed unit, when too much of the coolant evaporates it's game over. It also has a pump, which inevitably makes noise. Also a risk of it springing a leak. I basically don't trust it over a block of metal.

I'd have to agree, the H50 will eventually have diminishing returns, whilst the mega won't. If you plan to keep the heatsink for a few years, the Mega seems a safer bet, plus easier to sell on later.

 

[dtokez]

A con for the H50 is that it's a sealed unit, when too much of the coolant evaporates it's game over. It also has a pump, which inevitably makes noise. Also a risk of it springing a leak. I basically don't trust it over a block of metal.

Cheers

how can evaporation take place in a sealed unit?

 

[blister]

wow comprehensive going for h50 anyway but great to see results of this before i buy thanks

 

[FilterFish]

how can evaporation take place in a sealed unit?

Nothing at this price is ever completely sealed

A quote from Corsairs product blurb..

the H50 is a closed-loop system that is supplied pre‑filled and fully connected using low permeability tubing to virtually eliminate evaporation and leaking.

I've underlined the important word

 

[dtokez]

haha i think that underlined word alone has put me off the h50!

looks like its got the performance but its not for me if it will hit the bin in a year

 

[3t3P]

I assume also corsair have glued the tubes in place for better sealing and leak reliability