Scythe SCNJ-1000 Ninja heatsink
-
- Posts: 1424
- Joined: Sun Dec 12, 2004 5:00 am
- Location: New York, NY
-
- *Lifetime Patron*
- Posts: 368
- Joined: Fri Jul 23, 2004 11:33 am
- Location: King of Prussia, PA
I'm curious to how motherboard temps, memory temps, and especially northbridge temperatures varied with the use of this cooler in passive mode and versus a cooler that has a fan blowing down onto the motherboard.
If I'm not mistaken, the passive NB HS usually are designed to take the airflow off of the CPU HSF right?
Does the sideways blowing fan of this cooler "skip" over the northbridge? And in passive mode (like set up in the P180) does the airflow over the NB decrease in a drastic amount, or is the airflow adequate enough to cool these hot chipsets?
Thank you.
If I'm not mistaken, the passive NB HS usually are designed to take the airflow off of the CPU HSF right?
Does the sideways blowing fan of this cooler "skip" over the northbridge? And in passive mode (like set up in the P180) does the airflow over the NB decrease in a drastic amount, or is the airflow adequate enough to cool these hot chipsets?
Thank you.
-
- Posts: 1424
- Joined: Sun Dec 12, 2004 5:00 am
- Location: New York, NY
-
- Site Admin
- Posts: 12285
- Joined: Sun Aug 11, 2002 3:26 pm
- Location: Vancouver, BC, Canada
- Contact:
Very good question. This is the topic of an article that I am working on right now. The short answer is that as long as there is a small amount of airflow across the motherboard being created by the HS fan, PSU fan and/or the case exhaust fan, all is fine with all the components you mention. There are exceptions when you combine very low airflow and certain very high power CPUs / chipsets -- the efficiency of voltage regulators, in particular, can be dramatically affected by inadequate cooling.thetoad30 wrote:I'm curious to how motherboard temps, memory temps, and especially northbridge temperatures varied with the use of this cooler in passive mode and versus a cooler that has a fan blowing down onto the motherboard.
If I'm not mistaken, the passive NB HS usually are designed to take the airflow off of the CPU HSF right?
Does the sideways blowing fan of this cooler "skip" over the northbridge? And in passive mode (like set up in the P180) does the airflow over the NB decrease in a drastic amount, or is the airflow adequate enough to cool these hot chipsets?
Thank you.
-
- *Lifetime Patron*
- Posts: 368
- Joined: Fri Jul 23, 2004 11:33 am
- Location: King of Prussia, PA
Ok cool I can't wait for that article.
I currently have a very poorly designed case with very little allowable airflow (I have the TJ05, I could cut the case, but I need to return it due to it being broke, so I'm waiting until the P180 gets reviewed before sending this one in).
I have an Asus P5AD2-E Premium board, and an XP-120. All 120mm fans are Antec, with all on high.
Chipset temps as measured by my thermal probe routinely hit 45-50C, and in high-use, such as high disk I/O and CPU usage, it can get as high as 55-60C. Overclocking with these temperatures is a no go: The chipset is just too unstable at these temps.
I am using a 3.4GHz P4 Prescott (I know, I know...) and this by itself puts out a lot of heat.
So, what I'm curious about is how cool the airflow needs to be in order to be "ok". In otherwords, what should the case temp be in conjunction with the airflow?
Mine right now is around 38-45C in idle/load. So, my chipset gets a lot of airflow with the fans, but I am thinking (and common sense tells me) that my case temp is just too high for any "airflow" to have any kind of cooling effect on component temperatures.
Thank you.
I currently have a very poorly designed case with very little allowable airflow (I have the TJ05, I could cut the case, but I need to return it due to it being broke, so I'm waiting until the P180 gets reviewed before sending this one in).
I have an Asus P5AD2-E Premium board, and an XP-120. All 120mm fans are Antec, with all on high.
Chipset temps as measured by my thermal probe routinely hit 45-50C, and in high-use, such as high disk I/O and CPU usage, it can get as high as 55-60C. Overclocking with these temperatures is a no go: The chipset is just too unstable at these temps.
I am using a 3.4GHz P4 Prescott (I know, I know...) and this by itself puts out a lot of heat.
So, what I'm curious about is how cool the airflow needs to be in order to be "ok". In otherwords, what should the case temp be in conjunction with the airflow?
Mine right now is around 38-45C in idle/load. So, my chipset gets a lot of airflow with the fans, but I am thinking (and common sense tells me) that my case temp is just too high for any "airflow" to have any kind of cooling effect on component temperatures.
Thank you.
with this product, esp in the p180 picture, it is really clear how choosing/designing the cpu heatsink and the case together is a good idea. particularly for low noise/low airflow systems. when you put heat source inside a box, like a cpu inside a case, it doesn't make sense to consider them as two separate systems (the cpu cooling and the case airflow). from the point of view of airflow management, they're coupled so much that they're basically the one system. and when aiming for a configuration with best airflow (i.e. most efficient cooling for low flows) it starts to look suboptimal to have something like xp-120 with a fan sucking or blowing along the width axis of the case.
one fix was intel's duct in the case side over the cpu (chassis air guide intake). the other option is this ninja kind of idea. i suppose it's up to careful measurements and experiment to determine which, if either, is a better design direction.
this does look promising. am also interested to read the article concerning the effects on the mobo's VR etc
but seriously, i guess the ninja would be better in the p180 setup because it has the correct axes for airflow with respect to the case fans (see IsaacKuo earlier post). looks like the 2005 thing encourages only one of those airflow directions.
actually with all the previous tower heatsinks i can remember there has always this question of which direction it would do best mounted in - i like the way this design prevents you from worrying about that
one fix was intel's duct in the case side over the cpu (chassis air guide intake). the other option is this ninja kind of idea. i suppose it's up to careful measurements and experiment to determine which, if either, is a better design direction.
this does look promising. am also interested to read the article concerning the effects on the mobo's VR etc
i suppose because spcr was sent a ninja and not a ncu-2005?thetoad30 wrote:WHY would the NCU-2005 not be used instead in this setup that the review has (With the SNCJ-1000 in passive mode with 2 fans of the P180 near it)?
but seriously, i guess the ninja would be better in the p180 setup because it has the correct axes for airflow with respect to the case fans (see IsaacKuo earlier post). looks like the 2005 thing encourages only one of those airflow directions.
actually with all the previous tower heatsinks i can remember there has always this question of which direction it would do best mounted in - i like the way this design prevents you from worrying about that
-
- *Lifetime Patron*
- Posts: 368
- Joined: Fri Jul 23, 2004 11:33 am
- Location: King of Prussia, PA
I suppose that's true.
However, if I know what most people are going to do if they are silent enthusiasts... they are going to close up that top blowhole and let the rear fan do all the work.
Seems to me, if you can mount the NCU-2005 the correct way, it would be the best to use, according to that graph in the other thread.
Not doubting this review, as I am seriously thinking about getting the Ninja for my P180 case... since I don't think I'm going to be closing up the top blowhole. But we'll see.
However, if I know what most people are going to do if they are silent enthusiasts... they are going to close up that top blowhole and let the rear fan do all the work.
Seems to me, if you can mount the NCU-2005 the correct way, it would be the best to use, according to that graph in the other thread.
Not doubting this review, as I am seriously thinking about getting the Ninja for my P180 case... since I don't think I'm going to be closing up the top blowhole. But we'll see.
-
- Posts: 419
- Joined: Sun Sep 19, 2004 1:05 pm
- Location: Palo Alto, CA
just for those who are curious; I went from a NCU-2005 to a Ninja (actually now using my second system with a Ninja). I found much better near-passive cooling wih the Ninja, even when the case airflow was biased in a single axis.
And for those concerned about weight:
-Do not try this at home-
My motherboard is currently sitting on a shelf in the mobo tray of a P160. While it was running I picked it up, turned it on it's side, and shock it quite hard. Considering I am still typing this post, the resuls were favorable.
And for those concerned about weight:
-Do not try this at home-
My motherboard is currently sitting on a shelf in the mobo tray of a P160. While it was running I picked it up, turned it on it's side, and shock it quite hard. Considering I am still typing this post, the resuls were favorable.
-
- Posts: 1283
- Joined: Thu Aug 07, 2003 2:43 pm
- Location: Charlotte, NC, USA
If one is interested into a comparism of the Ninja and the Shogun (and NCU-2005), just check this german page: http://www.dirkvader.de/page/Scythe_Towers/index.html.
Conclusion: Ninja is best at 660rpm and 1500rpm, NCU is best at passive. Fan used was the one supplied with the Shogun.
Conclusion: Ninja is best at 660rpm and 1500rpm, NCU is best at passive. Fan used was the one supplied with the Shogun.
I guess some of you haven't seen ddrueding1's experience with a Scythe Ninja and an A64E.
First I thought that he had a very good CPU, but now I realize it's a very good HSF as well!
I'm just waiting to se some serious reviews of Asetek Vapochill Micro and Zalman CNPS 9500, they both look interesting.
MikeC: You didn't happen to get any of them from Taiwan?
First I thought that he had a very good CPU, but now I realize it's a very good HSF as well!
I'm just waiting to se some serious reviews of Asetek Vapochill Micro and Zalman CNPS 9500, they both look interesting.
MikeC: You didn't happen to get any of them from Taiwan?
http://www.vr-zone.com/?i=2316&s=24
looks like there is a copper one in the works.
And a VGA heatsink.
looks like there is a copper one in the works.
And a VGA heatsink.
-
- SPCR Reviewer
- Posts: 8636
- Joined: Sat Nov 23, 2002 6:33 am
- Location: Sunny SoCal
-
- Site Admin
- Posts: 12285
- Joined: Sun Aug 11, 2002 3:26 pm
- Location: Vancouver, BC, Canada
- Contact:
IIRC, the copper version is not necessarily planned for release. The Scythe folks told me at Computex that there are many iterations of products that they play with but never actually put into production -- which makes sense for a company that shows as much ingenuity as Scythe. They do show such samples for attention and discussion value at shows. This item, I am sure, is one that will never see a SKU.
-
- Posts: 20
- Joined: Thu Jun 16, 2005 3:06 pm
-
- Posts: 1283
- Joined: Thu Aug 07, 2003 2:43 pm
- Location: Charlotte, NC, USA
-
- Posts: 20
- Joined: Thu Jun 16, 2005 3:06 pm
Thanks MikeC and Shadowknight, that helps. I was planning to get an XP-120 before I saw this review and that was my only worry about switching.
The Ninja seems like it'll be much better just from an airflow perspective--with a 120mm on the right side of the Ninja pushing air into it and the 120mm exhaust fan two inches away pulling air out of it. Not to mention what looks like an easier installation.
The Ninja seems like it'll be much better just from an airflow perspective--with a 120mm on the right side of the Ninja pushing air into it and the 120mm exhaust fan two inches away pulling air out of it. Not to mention what looks like an easier installation.
Just following up on the Ninja vs. NCU-2005 discussion.
I emailed Scythe Co., Ltd. (Japan) regarding this subject a couple of months ago. Below are the questions I asked and the answers I received:
<me> I would like to know how the "SCNJ-1000 Ninja Fanless Heatpipe CPU Cooler" cooler compares to the NCU-2005. Is is better, worse, etc?
<scythe> As for the SCNJ-1000 Ninja fanless heatsink, it would be rated lower than the NCU-2005 as NCU-2005 is the high-end fanless heatsink with heatlane technology (heatlane technology can transfer the heat much faster and efficient than the normal heatpipe system). This also reflects to the price between NCU-2005 and SCNJ-1000.
<me> So is the NCU-2005 the very best fanless CPU cooler you provide, or is there something even better available?
<scythe> At this moment, NCU-2005 is the very best fanless heatsink we provide. SCNJ-1000 Ninja fanless heatsink is the 2nd best which we wanted to introduce a more affordable fanless cooler into the market.
So based on that advice I was planning to go with the NCU-2005. But after reading this thread the general consensus appears to be that the Ninja is the better of the two.
All in all, im confused
I emailed Scythe Co., Ltd. (Japan) regarding this subject a couple of months ago. Below are the questions I asked and the answers I received:
<me> I would like to know how the "SCNJ-1000 Ninja Fanless Heatpipe CPU Cooler" cooler compares to the NCU-2005. Is is better, worse, etc?
<scythe> As for the SCNJ-1000 Ninja fanless heatsink, it would be rated lower than the NCU-2005 as NCU-2005 is the high-end fanless heatsink with heatlane technology (heatlane technology can transfer the heat much faster and efficient than the normal heatpipe system). This also reflects to the price between NCU-2005 and SCNJ-1000.
<me> So is the NCU-2005 the very best fanless CPU cooler you provide, or is there something even better available?
<scythe> At this moment, NCU-2005 is the very best fanless heatsink we provide. SCNJ-1000 Ninja fanless heatsink is the 2nd best which we wanted to introduce a more affordable fanless cooler into the market.
So based on that advice I was planning to go with the NCU-2005. But after reading this thread the general consensus appears to be that the Ninja is the better of the two.
All in all, im confused
I just had a thought:
Is the SCNJ-1000 Ninja REALLY orientation independent?
It looks like there are 3 heatpipes in one orientation next to the CPU in one layer, and then 3 more heatpipes in the perpendicular orientation in a second layer.
Logically, you'd think that the first layer of heatpipes will cool the CPU more effectively than the second layer of heatpipes. Right?
Thus, depending on how the heatsink is installed, it might be slightly more effective with airflow toward the rear, or with airflow toward the top of the case.
Am I making sense, here?
Is the SCNJ-1000 Ninja REALLY orientation independent?
It looks like there are 3 heatpipes in one orientation next to the CPU in one layer, and then 3 more heatpipes in the perpendicular orientation in a second layer.
Logically, you'd think that the first layer of heatpipes will cool the CPU more effectively than the second layer of heatpipes. Right?
Thus, depending on how the heatsink is installed, it might be slightly more effective with airflow toward the rear, or with airflow toward the top of the case.
Am I making sense, here?
-
- Site Admin
- Posts: 12285
- Joined: Sun Aug 11, 2002 3:26 pm
- Location: Vancouver, BC, Canada
- Contact:
Maybe, maybe not, but my guess is no. The connection between all the heatpipes and the base looks very good, it's hard to imagine there'd be much difference, if any. And even if there was, the fins themselves spread the heat out anyway, so...IsaacKuo wrote:I just had a thought:
Is the SCNJ-1000 Ninja REALLY orientation independent?
It looks like there are 3 heatpipes in one orientation next to the CPU in one layer, and then 3 more heatpipes in the perpendicular orientation in a second layer.
Logically, you'd think that the first layer of heatpipes will cool the CPU more effectively than the second layer of heatpipes. Right?
Thus, depending on how the heatsink is installed, it might be slightly more effective with airflow toward the rear, or with airflow toward the top of the case.
Am I making sense, here?
You know I was wondering about a variation of this question. Generally, my question applies to some of the vertical tower designs. One of them is the Thermaltake Sonic Tower, of which I was wondering, if you have a fan running on it (either on the side or sandwiched) the downwind pipes are going to be fed warm air from the upwind side. Would this translate into tempeature gradient at the cpu surface, and thus a constant stress in the silicon? Does anything like that, if it exists at all, matter in the sceme of things , perhaps in relation to processor lifespan?MikeC wrote:Maybe, maybe not, but my guess is no. The connection between all the heatpipes and the base looks very good, it's hard to imagine there'd be much difference, if any. And even if there was, the fins themselves spread the heat out anyway, so...IsaacKuo wrote:I just had a thought:
Is the SCNJ-1000 Ninja REALLY orientation independent?
It looks like there are 3 heatpipes in one orientation next to the CPU in one layer, and then 3 more heatpipes in the perpendicular orientation in a second layer.
Logically, you'd think that the first layer of heatpipes will cool the CPU more effectively than the second layer of heatpipes. Right?
Thus, depending on how the heatsink is installed, it might be slightly more effective with airflow toward the rear, or with airflow toward the top of the case.
Am I making sense, here?
ON another note, beside the weight issue, I think the passive northbridge temperature is the most critical issue of not having stray downdraft air. I base this on a test that someone did online of a tower design a while back. If I find it I will post the link of that showed the temperatures.
PS. What is the designated maximum heatsink weight from AMD, and also from Intel. Will my board warp over time with this thing on. By the way, I appreciate the assurance that most all the weight is in the base.
-
- Posts: 1283
- Joined: Thu Aug 07, 2003 2:43 pm
- Location: Charlotte, NC, USA
The maximum weights from Intel and AMD are for shipping, not stationary mounting. Unless you're moving the thing really roughly, it shouldn't damage your motherboard in any way... but I wouldn't take bets on how safe the Sonic Tower is, though. There;s a picture on the forum showing the centre of gravity to be about 1/3 of the length of the heatsink.
-
- Posts: 103
- Joined: Mon Jun 20, 2005 11:30 am
I am building a new computer from scratch and this heatsink looks tempting. I'll be getting a Athlon 64 3500+ and *probably* a Gigabyte K8N Ultra-9 motherboard. It will be going in an Antec SLK3000-B case.
My question is, will this heatsink fit without blocking anything? And as I'm not overclocking anything, could I operate in fanless mode or should I just get a Nexus fan @ 5V?
EDIT: ddrueding1 apparently has the same CPU and motherboard as I, but his signature gives no clue as to his case.
My question is, will this heatsink fit without blocking anything? And as I'm not overclocking anything, could I operate in fanless mode or should I just get a Nexus fan @ 5V?
EDIT: ddrueding1 apparently has the same CPU and motherboard as I, but his signature gives no clue as to his case.