Sl-120 vs cnps9500

[alejandroAT]

hi all,

just recently i changed my Sl120 cpu hs with the zalman cnps9500 in order to achieve better temperatures.

With the Sl120 i was running my x2 4400 at 40C idle and 47C under EXTREME load (2 instances of prime95).

I wasnt happy with 47C so i decided to try out the 9500 everyone is raving about.

Both hs were mounted using AS5.

All i can say is that my cpu now runs at 46C idle and can reach 50C under small load (eg guildwars running).

At first i applied the paste with a plastic card spreading an even and very thin layer on the cpu. Temps were high so after 5 days i cleaned it up using isopropyl alcohol and applied the paste in the form of a rice grain in the middle. Didnt change a thing! still 46C at idle.

The 9500 is supposed to be pushing all the hot air towards the back exhaust fan thus quickly getting rid of the hot air. With the Sl120 the air was pushed through the HS and then spread 360degrees all around .

My question is this: how can a copper hs with such an efficient design and 6 heatpipes have 6C more at idle than an aluminium hs with 5 heatpipes and a no so efficient design?

and dont tell me i have to wait for it to cure cause that will only make a difference of 2-3 degrees.

 

[Ika]

SI120 is bigger?

 

[alejandroAT]

the fan is bigger but the overall mass of the zalman is more. Not to mention that the zalman is copper whereas the thermalright is aluminium. copper is supposed to be better at thermal conductivity!!

Another thing is that the zalman is mirror finished at the surface of contact whereas the sl120 is poorly machined and full of deep lines.

In general the zalman has all the attributes to make it much better than the sl120.

this is why i dont understand it running my cpu 7C hotter

 

[SparkyJJO]

I have the 9500 mounted on my 3700+ OC'd to 2.86GHz @ 1.5v, and I get load temps in the low 40s and idle temps somewhere in the 30's with the fan speed set at 1650 RPM. I'm not totally sure about idle temps cuz my computer is running F@H all the time so I pretty much have load temps constantly. It heats up more under heavy load, like gaming or something like that, but only by a couple degrees.

When the room temp was somewhere in the 80's F - I'd say between 82 and 85*F - My CPU under heavy load would top out around 48*C. I assume your CPU produces more heat than mine does when both are at stock, but I'd guess that mine would at least be equal to yours when mine is OC'd.... strange.

Maybe you just got a defective one, one of the heatpipes is bad? I dunno....

EDIT by the way, the zalman 9500 has 3 heatpipes, bent in a figure-8 shape to give it the performance of 6 supposedly. I'd guess not completely equivalent to 6 pipes but probably close.

 

[alejandroAT]

i see we have the same room temps? over here in athens its 26C (80F).
what other fans are u using?
i have the antec p180 case and my hdds are mounted inside "cooldrive light" boxes and mounted at the top just in front of the 9500 in order to supply it with air directly.....my 7800gt has a Arctic Cooling NV5 which sucks air through the heatsink and vents out the back so it doesnt affect case temps. basically the whole front of the case is intake while the back is outake.

(the x2 4400 is a very hot cpu - imagine that its 2 cpus running at 2.2 gh)

 

[Operandi]

47c is a great load temp and much, much lower then it has to be. Why anyone would want lower temps is beyond me.

 

[alejandroAT]

because i payed 550 euros for my cpu and i want it to last the longest possible....lower temps mean longer lifetime.

 

[Operandi]

Originally posted by: alejandroAT
because i payed 550 euros for my cpu and i want it to last the longest possible....lower temps mean longer lifetime.

So you want it to last what 5000 years? Seriously 47c is ice cold for a CPU like that on air.

 

[RallyMaster]

Tip: sacrifice 3 of your 5.25" bays and mount a 120mm in there. It will provide air to the CNPS9500 and will make almost a wind tunnel effect at the upper part of your case.

 

[alejandroAT]

not 5000 years mate but 3-4 years at least at top performance....will start overclocking as well when the cpu becomes old for new apps and thats why i m concerned with the best cooling solution...planning ahead!!

as for sacrificing the bays....i cant afford that...the very top is occupied by the cd drive....
2nd and 3rd slots (in front of the 9500) are occupied by a pair of Cooldrives (coolermaster's hdd cooling box) which are basically hdd heatsinks with fans which suck air from the front , blow it through the hdd hs, over the RAM and into the 9500 which in turn vents out the back through the help of a 120mm papst. The wind tunel is there but its not working as well as just having the si120 blowing violently all over the place! : (

 

[alejandroAT]

just a quick note here..

i turned on the cool'n'quiet feature and the cpu temp dropped 10C to a nice and comfortable 40C. If more power is used then the core switches back up to normal for the duration. checked it with cpuz

 

[scrawnypaleguy]

Are you reading your temps from a probe or from a software monitoring program? One thing that I might say might be the problem is that if you're monitoring using a program, it's using a sensor on the board. Some of the air from the downward blowing si-120 may be hitting that probe, thereby giving you lower temps than the sideways-blowing 9500. Just stick with the si-120, I love mine.

 

[Operandi]

alejandroAT I really think your making too big of a deal out of nothing. Even a CPU running at 70c 24/7/365 will last far longer then you'll ever want to use it.

Originally posted by: scrawnypaleguy
Are you reading your temps from a probe or from a software monitoring program? One thing that I might say might be the problem is that if you're monitoring using a program, it's using a sensor on the board. Some of the air from the downward blowing si-120 may be hitting that probe, thereby giving you lower temps than the sideways-blowing 9500. Just stick with the si-120, I love mine.

Actually the temps hasn't been read that way since the early socket A days.

Still different boards seem to read the temp differently. I think newer CPUs are starting to put the thermal diode closer in the CPU die so maybe eventually we?ll have temps that can be universal between boards.

 

[JEDIYoda]

Originally posted by: Operandi
alejandroAT I really think your making too big of a deal out of nothing. Even a CPU running at 70c 24/7/365 will last far longer then you'll ever want to use it.

Originally posted by: scrawnypaleguy
Are you reading your temps from a probe or from a software monitoring program? One thing that I might say might be the problem is that if you're monitoring using a program, it's using a sensor on the board. Some of the air from the downward blowing si-120 may be hitting that probe, thereby giving you lower temps than the sideways-blowing 9500. Just stick with the si-120, I love mine.

Actually the temps hasn't been read that way since the early socket A days.

Still different boards seem to read the temp differently. I think newer CPUs are starting to put the thermal diode closer in the CPU die so maybe eventually we?ll have temps that can be universal between boards.

I have the SI 120 on this rigg and I have the zalman 9500 on my gaming rigg.
There literally is no difference in temps. maybe at the most in my case 1c-2c.
You cannot go wrong with iether!

 

[Henny]

Originally posted by: alejandroAT
hi all,

just recently i changed my Sl120 cpu hs with the zalman cnps9500 in order to achieve better temperatures.

With the Sl120 i was running my x2 4400 at 40C idle and 47C under EXTREME load (2 instances of prime95).

I wasnt happy with 47C so i decided to try out the 9500 everyone is raving about.

Both hs were mounted using AS5.

All i can say is that my cpu now runs at 46C idle and can reach 50C under small load (eg guildwars running).

At first i applied the paste with a plastic card spreading an even and very thin layer on the cpu. Temps were high so after 5 days i cleaned it up using isopropyl alcohol and applied the paste in the form of a rice grain in the middle. Didnt change a thing! still 46C at idle.

The 9500 is supposed to be pushing all the hot air towards the back exhaust fan thus quickly getting rid of the hot air. With the Sl120 the air was pushed through the HS and then spread 360degrees all around .

My question is this: how can a copper hs with such an efficient design and 6 heatpipes have 6C more at idle than an aluminium hs with 5 heatpipes and a no so efficient design?

and dont tell me i have to wait for it to cure cause that will only make a difference of 2-3 degrees.

The 9500 only performs well at full fan speed. For a given noise level it's not best of class by any means.

 

[m1ldslide1]

I made a big mistake when I installed my 9500, and that was swapping my motherboard at the same time. With the stock AMD HSF on my 3700+ (asus motherboard) I was getting around 29C idle and about 40C load. After swapping mobos and installing the 9500 (with AS5), I was getting... 29C idle and about 40C load (abit mobo). Best $90 I ever spent! Anyhow, I can only hope that the fact that there is NO difference is attributable to the difference in mobos and the software that monitors the temps.

I have a 120mm intake blowing directly over the NB and the 9500, and then an uber-powerful 120mm case exhaust on the back of my case. (plus the 120 on my PSU and another 90mm lower intake). Anyhow I have the whole windtunnel effect, which I thought would be perfect for the 9500. Unfortunately the 9500 fan faces the top of my case, so I'm thinking about perforating the top of my case and installing another 120.

 

[JEDIYoda]

Originally posted by: Henny

Originally posted by: alejandroAT
hi all,

just recently i changed my Sl120 cpu hs with the zalman cnps9500 in order to achieve better temperatures.

With the Sl120 i was running my x2 4400 at 40C idle and 47C under EXTREME load (2 instances of prime95).

I wasnt happy with 47C so i decided to try out the 9500 everyone is raving about.

Both hs were mounted using AS5.

All i can say is that my cpu now runs at 46C idle and can reach 50C under small load (eg guildwars running).

At first i applied the paste with a plastic card spreading an even and very thin layer on the cpu. Temps were high so after 5 days i cleaned it up using isopropyl alcohol and applied the paste in the form of a rice grain in the middle. Didnt change a thing! still 46C at idle.

The 9500 is supposed to be pushing all the hot air towards the back exhaust fan thus quickly getting rid of the hot air. With the Sl120 the air was pushed through the HS and then spread 360degrees all around .

My question is this: how can a copper hs with such an efficient design and 6 heatpipes have 6C more at idle than an aluminium hs with 5 heatpipes and a no so efficient design?

and dont tell me i have to wait for it to cure cause that will only make a difference of 2-3 degrees.

The 9500 only performs well at full fan speed. For a given noise level it's not best of class by any means.

Henny....you are wrong.
In all the reports I have read and in my own personnal experience the 9500 works best with the fan throttled down.
Just do an internet search of reviews for the 9500 and you will see thats the case.

 

[Operandi]

Originally posted by: JEDIYoda

Originally posted by: Henny

Originally posted by: alejandroAT
hi all,

just recently i changed my Sl120 cpu hs with the zalman cnps9500 in order to achieve better temperatures.

With the Sl120 i was running my x2 4400 at 40C idle and 47C under EXTREME load (2 instances of prime95).

I wasnt happy with 47C so i decided to try out the 9500 everyone is raving about.

Both hs were mounted using AS5.

All i can say is that my cpu now runs at 46C idle and can reach 50C under small load (eg guildwars running).

At first i applied the paste with a plastic card spreading an even and very thin layer on the cpu. Temps were high so after 5 days i cleaned it up using isopropyl alcohol and applied the paste in the form of a rice grain in the middle. Didnt change a thing! still 46C at idle.

The 9500 is supposed to be pushing all the hot air towards the back exhaust fan thus quickly getting rid of the hot air. With the Sl120 the air was pushed through the HS and then spread 360degrees all around .

My question is this: how can a copper hs with such an efficient design and 6 heatpipes have 6C more at idle than an aluminium hs with 5 heatpipes and a no so efficient design?

and dont tell me i have to wait for it to cure cause that will only make a difference of 2-3 degrees.

The 9500 only performs well at full fan speed. For a given noise level it's not best of class by any means.

Henny....you are wrong.
In all the reports I have read and in my own personnal experience the 9500 works best with the fan throttled down.
Just do an internet search of reviews for the 9500 and you will see thats the case.

Right, the 9500 dose very well at low speeds.

SPCR Review: "Excellent low-airflow performance"

FrostyTech Review: "As a heatsink, the Zalman CNPS9500 LED functions best as a low noise cooler. With that translucent 92mm fan glowing blue and spinning at 1350RPM, the heatsink maintains good temperatures and excellent noise reduction."

 

[alejandroAT]

Originally posted by: Operandi
alejandroAT I really think your making too big of a deal out of nothing. Even a CPU running at 70c 24/7/365 will last far longer then you'll ever want to use it.

Operandi please dont say that again to an X2 owner cause he might not know and follow your advice.

In part you are right...most if not all intels can go up to 90C.

But i know the X2 isnt supposed to go to 70C....it shuts down to prevent damage.....

As for m1ldslide1....mate.....dont do that......just turn the 9500 so it points towards the rear exhaust,,

if you really want to open a hole on the top of the case then make sure its an exhaust and not an intake as it will surely mess up your airflow....you see hot air travels up : )

 

[Operandi]

Originally posted by: alejandroAT

Originally posted by: Operandi
alejandroAT I really think your making too big of a deal out of nothing. Even a CPU running at 70c 24/7/365 will last far longer then you'll ever want to use it.

Operandi please dont say that again to an X2 owner cause he might not know and follow your advice.

In part you are right...most if not all intels can go up to 90C.

But i know the X2 isnt supposed to go to 70C....it shuts down to prevent damage.....

As for m1ldslide1....mate.....dont do that......just turn the 9500 so it points towards the rear exhaust,,

if you really want to open a hole on the top of the case then make sure its an exhaust and not an intake as it will surely mess up your airflow....you see hot air travels up : )

The X2 dose not shut down at 70c, stability may or may not be compromised at such temps, but it's nowhere near hot enough to cause physical damage.

It's definitely higher then I would recommend you run your CPU at but it's still in the safe range.

 

[m1ldslide1]

As for m1ldslide1....mate.....dont do that......just turn the 9500 so it points towards the rear exhaust,,

if you really want to open a hole on the top of the case then make sure its an exhaust and not an intake as it will surely mess up your airflow....you see hot air travels up : )

The problem with pointing the fan towards the rear exhaust is that the bracket attached to the motherboard will only allow the 9500 to face up or down. I was really bummed when I found this out, and I have no idea how to fix it??? IS there some way to reorient your HSF bracket?

If I get a case fan on the top it'd definately be an exhaust. I'm just skeptical that any of it's going to make a huge difference from this point on.

 

[skinnypotbelly]

You can twist the retention clip so that the HSF can point towards the back. Seems like you have the retention clip fiting perpendicular to the heat pipes. You can actually fit the clips between the heatpipes, parallel to them.

 

[alejandroAT]

Check out these links Operandi

http://www.gen-x-pc.com/cputemps.htm
http://www.digital-daily.com/cpu/athlon-64-x2-4800/

the first states the max operating temp as 65C and the second as 71C.

Also read this below (the 4800 is also a Toledo core)
-------------------------------------------
So, what does the ADA4800DAA6CD stand for? With ADA all is clear - it is AMD Desktop Athlon 64. Then there goes the processor's performance index - 4800+. The letter D after the index says that the processors installs on Socket 939. The first letter A indicated the supply voltage 1.3-1.4V. But the second A may be confusing after reading the specifications since it points to the maximum core temperature 71°C (seems like AMD have been overcautious). The number 6 means that the overall L2 cache size is 2048 K, the CD is already familiar to you.
--------------------------------------------

Could you explain plz where u found this information that the X24400 does not shut down at 70C and that damage is not a possibility?
Also, these cpus do not have an internal temp sensor.....the reading relies on the mb sensor that reads the outter casing temp of the cpu...so the inner temp is always higher.

 

[alejandroAT]

 

[Henny]

Originally posted by: JEDIYoda

Originally posted by: Henny

Originally posted by: alejandroAT
hi all,

just recently i changed my Sl120 cpu hs with the zalman cnps9500 in order to achieve better temperatures.

With the Sl120 i was running my x2 4400 at 40C idle and 47C under EXTREME load (2 instances of prime95).

I wasnt happy with 47C so i decided to try out the 9500 everyone is raving about.

Both hs were mounted using AS5.

All i can say is that my cpu now runs at 46C idle and can reach 50C under small load (eg guildwars running).

At first i applied the paste with a plastic card spreading an even and very thin layer on the cpu. Temps were high so after 5 days i cleaned it up using isopropyl alcohol and applied the paste in the form of a rice grain in the middle. Didnt change a thing! still 46C at idle.

The 9500 is supposed to be pushing all the hot air towards the back exhaust fan thus quickly getting rid of the hot air. With the Sl120 the air was pushed through the HS and then spread 360degrees all around .

My question is this: how can a copper hs with such an efficient design and 6 heatpipes have 6C more at idle than an aluminium hs with 5 heatpipes and a no so efficient design?

and dont tell me i have to wait for it to cure cause that will only make a difference of 2-3 degrees.

The 9500 only performs well at full fan speed. For a given noise level it's not best of class by any means.

Henny....you are wrong.
In all the reports I have read and in my own personnal experience the 9500 works best with the fan throttled down.
Just do an internet search of reviews for the 9500 and you will see thats the case.

If this review is accurate, the 9500 is among the worst performers (even worse then Intel Stock HSF) on low but best on high:

Here