Will Ninja 4 lose performance over time?

Cooling Processors quietly

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Post Reply
lb_felipe
Posts: 270
Joined: Wed Dec 24, 2008 12:21 pm
Location: São Paulo, SP, Brasil

Will Ninja 4 lose performance over time?

Post by lb_felipe » Mon Mar 07, 2016 5:03 pm

This site says yes.

http://aphnetworks.com/reviews/scythe-ninja-4
Something to take into consideration is the heatpipes are not soldered to the fins of the heatsink. At first, this is not an issue, but should be taken into account in the long term, because you will experience a drop in performance after many cycles of expansion and contraction.
However, the heatpipes are not soldered to the fins as aforementioned, which means over time the fins and heatpipes will lose contact, greatly reducing performance in the long term. The heatpipes will lose contact because of expansion during times when the heatsink will warm up, and contraction when it cools down.
-- Final APH Numeric Rating is 5.9/10
The Scythe Ninja 4 performed well in all of our tests, but how long will it last?

xan_user
*Lifetime Patron*
Posts: 2269
Joined: Sun May 21, 2006 9:09 am
Location: Northern California.

Re: Will Ninja 4 lose performance over time?

Post by xan_user » Mon Mar 07, 2016 8:08 pm

FUD IMO.
none of my heatsinks have soldered fins. never any performance drop here.

quest_for_silence
Posts: 5275
Joined: Wed Jun 13, 2007 10:12 am
Location: ITALY

Re: Will Ninja 4 lose performance over time?

Post by quest_for_silence » Mon Mar 07, 2016 11:08 pm

lb_felipe wrote:This site says yes.

http://aphnetworks.com/reviews/scythe-ninja-4

Literally a shitload of bullshits from a bunch of illitterate asian teens: forget them (the review was written by... a guest writer! Unbelieable!).

EDIT: I failed to understand what "illiterate" means (mistakenly used for "not graduated yet"), "asian teens" was not meant with denigratory sense (but with mere descriptive connotation).
Last edited by quest_for_silence on Tue Mar 15, 2016 11:17 am, edited 1 time in total.

edh
Posts: 1621
Joined: Mon Mar 17, 2008 1:49 pm
Location: UK

Re: Will Ninja 4 lose performance over time?

Post by edh » Tue Mar 08, 2016 12:30 am

What a load of nonsense. Most heatsinks have non-soldered joints against heatpipes. Thermal expansion will happen, yes but with aluminium you're looking at less than 0.1% linearly over a 40K rise above ambient and there is no reason to assume this will have such a detrimental effect. I will not believe this unless they can prove it.

quest_for_silence
Posts: 5275
Joined: Wed Jun 13, 2007 10:12 am
Location: ITALY

Re: Will Ninja 4 lose performance over time?

Post by quest_for_silence » Tue Mar 08, 2016 1:52 am

edh wrote:What a load of nonsense. Most heatsinks have non-soldered joints against heatpipes.
And there, soldered joints are meant to improve thermal transfer (so to squeeze the last drop of cooling prowess), never ever for that accounted "reason" (bullshit)...

frenchie
Friend of SPCR
Posts: 1346
Joined: Tue Jul 08, 2008 4:53 am
Location: CT

Re: Will Ninja 4 lose performance over time?

Post by frenchie » Tue Mar 08, 2016 1:16 pm

bull poop !!!!!

lb_felipe
Posts: 270
Joined: Wed Dec 24, 2008 12:21 pm
Location: São Paulo, SP, Brasil

Re: Will Ninja 4 lose performance over time?

Post by lb_felipe » Tue Mar 08, 2016 2:23 pm

Thank you guys for prevent that misinformation is taken as rule. I feel fine by knowing that, and because I can keep recommending Ninja 4 to friends.

MikeC
Site Admin
Posts: 12285
Joined: Sun Aug 11, 2002 3:26 pm
Location: Vancouver, BC, Canada
Contact:

Re: Will Ninja 4 lose performance over time?

Post by MikeC » Fri Mar 11, 2016 7:41 am

I've been testing heatsinks, using them, monitoring them, etc, since before this site was created... probably 15+ years. I have never noticed any cooling performance drop over time with a heatsink that has friction fit heatpipes running through thin fin stacks. In fact, except for one or 2 heatsinks with broken or cracked heatpipes or with a fan whose bearing was seizing up, I've never actually seen any HS perform worse with age. This is not to say it couldn't happen, but it seems highly likely. EDIT: UNlikely, I mean.

CA_Steve
Moderator
Posts: 7650
Joined: Thu Oct 06, 2005 4:36 am
Location: St. Louis, MO

Re: Will Ninja 4 lose performance over time?

Post by CA_Steve » Fri Mar 11, 2016 9:26 am

MikeC wrote:This is not to say it couldn't happen, but it seems highly unlikely.
Fixed! :)

lb_felipe
Posts: 270
Joined: Wed Dec 24, 2008 12:21 pm
Location: São Paulo, SP, Brasil

Re: Will Ninja 4 lose performance over time?

Post by lb_felipe » Fri Mar 11, 2016 1:27 pm

I was thinking... If the fins get contracted, dilated, etc, the heatpipes too, then the pressure between both is the same in every condition, right?

quest_for_silence
Posts: 5275
Joined: Wed Jun 13, 2007 10:12 am
Location: ITALY

Re: Will Ninja 4 lose performance over time?

Post by quest_for_silence » Fri Mar 11, 2016 2:11 pm

lb_felipe wrote:I was thinking... If the fins get contracted, dilated, etc, the heatpipes too, then the pressure between both is the same in every condition, right?
Right (more or less, pipes balance fins).

quest_for_silence
Posts: 5275
Joined: Wed Jun 13, 2007 10:12 am
Location: ITALY

Re: Will Ninja 4 lose performance over time?

Post by quest_for_silence » Sun Mar 13, 2016 3:50 am

MikeC wrote:I've never actually seen any HS perform worse with age. This is not to say it couldn't happen, but it seems highly unlikely

I agree, Mike: and, as a matter of fact, that's what I should have written when commenting some days ago.

On a second thought I realized that in case unlikely doesn't mandatorily (literally) mean impossible.
Given that, a further investigation should have been advisable for the sake of knowledge.

That's why some days ago I contacted the authors, and particularly mr. Kwan from Calgary in order to verify how such statements are backed.

While my interest towards the specific argument were somewhat appreciated, at least to some degree, mr. Kwan noted that my "snarky" remarks (like "Literally a shitload of..." you may see above) was not respecteful at all, and he was right: definitely when someone used cinism or sarcasm against me I would be unlikely "happy"... long story short, I begged mr. Kwan/APH pardon for being inpolite, and I hope my apologies may be accepted, so I'm doing here.

Set aside my own personal slope, I've asked him the reasons of their convincement: long story short (I don't want to duplicate here the long exchange with mr. Kwan), their main points seem to be two logical arguments.
First of all that a reduced contact due to incongruent material expansion/contraction through many cycles repeated daily for about a year, will cause a decrease in cooling performance more probably than any other logical alternative. Secondly no involved manufacturer (Cooler Master, Scythe, and so on) had ever made any claim against their reviews, even if informed about. According also to mr. Kwan the first argument was somehow (I didn't well understood atm) endorsed by http://schulich.ucalgary.ca/profiles/markus-r-dann
Positively mr. Kwan also his willingness to perform a one year later redux of that Ninja 4.

What do you think about, Mike?

Well, I think that's all.

MikeC
Site Admin
Posts: 12285
Joined: Sun Aug 11, 2002 3:26 pm
Location: Vancouver, BC, Canada
Contact:

Re: Will Ninja 4 lose performance over time?

Post by MikeC » Sun Mar 13, 2016 7:33 am

What do I think about that? Hmmmm.

Good that you made amends with Mr. Kwan and continue a dialog.

Sure a retest is interesting... But perhaps not as definitive as one might think. Why?

1. Natural accuracy of all review site cpu heatsink testing platforms is no better than 1 degree C. IE, repeat the same test, uninstalling and reinstalling the HS each time, and the results will vary by at least 1 degree. This is virtually guaranteed. That 1 degree variance might swamp any changes in performance due to metal fatigue from thermal cycling.

2. Assuming the test platform is the same, there is also a year's wear, tear and potential change in all the components, from motherboard and CPU to PSU. How much effect will this have? Possibly a little
Enough to cause a 1 degree change? Possibly.

However, if Kwan's thesis holds, then perhaps the drop in performance will be significant enough for the above to be trivial. I await results with curiosity but not bated breath.

baii
Posts: 69
Joined: Fri Nov 20, 2015 3:32 pm

Re: Will Ninja 4 lose performance over time?

Post by baii » Tue Mar 15, 2016 5:23 pm

All (i think) of the thermalright have migrated to non-solder, I think they call it fin penetration or w.e. Don't see anyone complaining ~ and I think thermalright know what they are doing~.

There were actually some debate (hr-02 change from solder to non-solder(hr02 macho) in 2012?) in the asian world few years ago when I search. People at that time pretty much conclude they perform the same.

So that's 3 years old, I think some people still have one in their machine.

fjf
Posts: 192
Joined: Wed Apr 06, 2005 9:44 am
Location: Europe

Re: Will Ninja 4 lose performance over time?

Post by fjf » Fri Apr 15, 2016 9:09 am

I have an original ninja on a core2duo 8 years old working fine...

kater
Posts: 891
Joined: Thu Sep 07, 2006 11:20 pm
Location: Poland

Re: Will Ninja 4 lose performance over time?

Post by kater » Fri Apr 14, 2017 3:20 am

At SPCR's recommendation, shortly after I joined the site, I bought SCNJ-1000, aka the one & only original, the iconic, the holy, errr, excuse me, I get carried away at times...

Anyway, my '06 Ninja is now cooling a 2500K @ 4.4 GHz. Clocked at 4.6 GHz, the CPU proves to be a little too much for the cooler, paired with a 1400 RPM Glidestream - at least under synthetic Linx torture. Gaming - all cool & fine, stays under 70 C. But some synthetic loads push it beyond 85 C. Also, given how much V I need for 4.6, I keep it at 4.4 tops.

Sadly, no way to say whether there's been any change in its performance over the years. For obvious reasons, duh.

Still, just for sh.... & giggles, I will be trying to obtain a modern hi-endish (single?) tower cooler, to see how my Ninja compares to it. As much as I like the Ninja and the fact that I can show off a +10 y.o. cooler on a still relatively powerful CPU, I fear it won't hold - if only due to the imperfect clamping system - bolted systems really shine in the pressure department.

Olle P
Posts: 711
Joined: Tue Nov 04, 2008 6:03 am
Location: Sweden

Re: Will Ninja 4 lose performance over time?

Post by Olle P » Wed Apr 19, 2017 12:24 am

Press fit is not an issue.
Dust build-up and fan ageing are more pressing issues for a cooler in long-term service.

kater
Posts: 891
Joined: Thu Sep 07, 2006 11:20 pm
Location: Poland

Re: Will Ninja 4 lose performance over time?

Post by kater » Wed Apr 19, 2017 12:37 am

In general - yes, quite likely.

But I don't think it's the issue here - the heatsink is squeaky clean, it's regularly dusted; the fan is a fresh Glidestream - the original fan has been replaced long time ago, and then the replacement fan got replaced by the one I have now - again, a squeky clean, very smooth fan reaching 1400 RPM.

The heatsink itself does have some "circular" play, if you know what I mean - it doesn't tip, obviously, but it can be rotated just a little bit.

Just this weekend my brother gave me his ancient 775 system, a P43 board with E8400 cooled by Mugen 2 - the first Scythe to feature a proper, bolted clamping system - and this one won't budge even 1 mm, it's rock fast. Sadly, the mounting system can't be transplanted between these 2 coolers.

Post Reply