Scythe Ninja

Danger Phoenix · 4 Apr 2008 at 02:12

I have just changed from an Arctic Cooling 64 Pro to a Scythe Ninja hoping that my temperatures would be lower, but in fact they are the same

I thought the Ninja was better?

Chronictank · 4 Apr 2008 at 02:28

Depends on the airflow in the case, it is good for what it is and has the added bonus that you can put a silent fan on it

Danger Phoenix · 4 Apr 2008 at 03:02

Well I just got a new case (Lian-Li PC-A71) and I used to have a 120mm fan strapped to my A64. I have just realised that the temperatures are actually higher at the moment

FrenchTart · 4 Apr 2008 at 11:10

Perhaps you need to give the artic silver a couple of days to settle in. Usually makes a few degrees diff in temp

GhostlyPea · 4 Apr 2008 at 11:12

As Aekeron said ^ Also have the CPU and Heatsink bases been lapped? Sometimes they don't give very good contact.

- Pea0n

Danger Phoenix · 4 Apr 2008 at 11:37

Nope... neither have been lapped. Would this make a big difference? How would I go about doing it?

GhostlyPea · 4 Apr 2008 at 12:50

Theres plenty of posts in this forum part about it, its reasonably easy and I havn't heard of anybody killing a CPU doing it providing your careful, although it will void the warranty. As for the temps, well some people report not a lot of difference, maybe 1 or 2 degrees but thats when both parts are allready reasonably flat, some report massive drops of up to 10 degrees, but ofcourse I don't think thisis as common. Mainly the point that

"doesnt matter how good a cooler is, if it doesn't have good contact it wont do much good"

stands up well. If your up to it give it a go and GL

- Pea0n

gt_junkie · 4 Apr 2008 at 15:15

Are you talking load or idle temps Danger?

I've used the Freezer Pro, Scythe Ninja, Big Typhoon, Ultra 120 Extreme and Titan Amanda on an Opty 165.

The ninja returned equal 2nd best temperatures. The Freezer pro returned the worst so i'd say something's wrong unless you're looking at idle temps (which don't tend to change as much in my experience).

gt

warren_1979 · 4 Apr 2008 at 16:47

Like Aekeron said give it a week for the AC5 to settle. Probably see a 5 degree drop in temps.

Danger Phoenix · 4 Apr 2008 at 17:08

Idle temperatures are around the same, but the load temperatures are around 4 degrees higher

I don't think I will bother with lapping, but I will wait a week or so for the AC5 to settle (I had not even mentioned AC5 in this thread, so how did you guys guess I was using that

)

Big.Wayne · 4 Apr 2008 at 18:04

Danger Phoenix said:
Idle temperatures are around the same, but the load temperatures are around 4 degrees higher

Take it off, examine the paste to see if its made good contact, also check to see nothing is stopping the heatsink sitting squarely on the CPU (i.e its not resting against a capacitor etc).

Perhaps even use less paste? You will know once you take off the heatsink, the clues should be there!

A Scythe Ninja should eat an Arctic Freezer 7 Pro alive in terms of temps!

Danger Phoenix · 4 Apr 2008 at 19:24

Funnily enough I just reseated the heatsink before I read the post, lol!

The first time round I got my housemate to sort it out because he is better at this sort of thing than I am. There was enough thermal paste there to probably sort out 2-3 CPUs

Anyways, I just found out that his AS5 is like 3 years old and when I just applied it again, it feels thicker than usual. I am going to buy some more immediately!

The Arctic Freezer 64 kept my load temperatures below 56 but this one is going above that

Ace Modder · 4 Apr 2008 at 20:08

56 is quite high too...

Danger Phoenix · 4 Apr 2008 at 20:22

Sorry, I should have mentioned that I am running an Opteron 165 @ 2.7Ghz (no increase in vcore) - so the overclock explains the high temperatures.

Danger Phoenix · 8 Apr 2008 at 16:00

Right, I got some new AS5 and re-did the heatsink. Now my temperatures are even worse than before, with one of the cores hitting 61 on full load

My Arctic Freezer 64 Pro was never this bad...

Chronictank · 8 Apr 2008 at 17:09

how are you applying the paste?
also pics of the base of the ninja may yield some info

Danger Phoenix · 8 Apr 2008 at 18:04

I am spreading it across the whole IHS as I have found that the blob method does not work too well. I can't take a photo of the base at the moment because its fitted on my CPU, but maybe you could tell me what I should be looking for?