Nagamaki temps

Every time I play Far Cry 2 the game will eventually crash. Haven't really timed it, but about an hour to an hour and a half of time, so, I started tracking temps. I thought that this might be the reason. It isn't (9.1 drivers fixed, thanks E-Wrecked), but it led me to experiment and I thought you'd all be interested in what I've found so far. It's also interesting to note that most people comment my observed temps are higher than other Nagamaki's, but here's what I've got. Each of these temp shots is taken after the computer crashes out of Far Cry 2 for unknown reasons. All were subjectively about the same amount of time with similar room temp.

Temps with the computer sitting, as delivered, on the desk-



Temps with a cheap device that raises the computer up to allow better airflow underneath-



Temps with generic USB powered 2 fan notebook cooler-


As you can see, each iteration decreases temps a couple of degrees C.

Finally, I removed the bottom to see what that would do-


Quite a difference! When looking at the bottom of the computer, you can see that there is not a lot of space open to allow air to flow in. Some of the incoming 'cool' airflow is intended to come in from the sides. Mark talked about that in his initial posts on the Nagamaki. His first prototype had many more holes on the bottom than what I have-



Apparently he decided not to drill as many for the production Nagamaki, though I think it's more for professional appearance. Obviously, opening up the bottom makes a difference! While I had the bottom off, I took pictures of the modifications to the bottom-


Notice that all the vent holes, except for the areas modified by Mark over the CPU cooler, have a thin filter cloth. The filter definitely works, but also will limit airflow even further. I will probably remove the cloth over the GPU cooler as well, though I'd like to get some feedback from Mark first as I think he probably had a reason for not removing it. The CPU fan will require more cleaning as a result. As a matter of fact, here's what mine looked like. Didn't come out too well in the pictures, but you can just make out the baby dust bunnies. A picture of my GPU fan shows virtually no dust in comparison. (I talked to Mark after the original post, he was worried about removing all of the filter material as some people do NOT regularly clean their fans. However, removing it will help airflow, but you will need to clean. I'll post some kind of guide for it if someone doesn't beat me to it.)


CPU fan distant-



CPU fan up close-



GPU fan close-


So, the CPU fan will require a little cleaning, and well worth it to keep the temps down more. Most notebooks don't have the filter covering over vents, and for good reason as any avoidable restriction to airflow in a package that already has airflow limitations is not acceptable. Most people clean their notebook vents regularly anyway, it's just what you have to do to keep the performance up. I cleaned the CPU cooler while I was in there.

Lastly, I took a picture of the whole interior for everyone. A clean install, no apparent plastic between the heatsinks and the video cards like others are having trouble with, no paste spilling out the sides either. I probably just have GPU's that run slightly hotter, bad luck there. Either that, or a mistake in the application of t-paste or something. Can't get it right every time!

Interior Nagamaki-


A close look at the image shows why one of my cards runs about 5C higher than the other. As you can see , the heat pipe from one card attaches to the heatsink after the first one. So, the air flowing through cools the first one down, which heats the air some. Then it flows through the section of the cooler that is cooling the second card. As the air is already hotter, it can't cool the second card as much. The temperature difference indicates that the air temp difference is likely in the neighborhood of 5C, which is very close to the ideal heat rise over a heat exchanger. Usually 10-15 degrees F is considered ideal from a design standpoint, I think 5C is close to 8-9 deg C, so close.

Comments? Thoughts?

 

I thought K|N modified their Nagamaki's by adding extra surface areas to the heatsinks.

I don't see anything extra so maybe that wasn't done to yours and why you're running hotter than other Nagamakis?

Something like what they did to this Executioner: http://www.killernotebooks.com/executioner_sr/executioner_sr_heatsink_mod.html

 

This rig doesn't offer much room inside to mod the sinks. You could essentially add some small vertical fins, but it would be costly. And, not very effective. I think that your temps are in check. Mine idles @ 40c with this new qx9300 running on "Balanced" mode in Vista. All I can recommend is that you apply new AS5 to the heatsink. Takes about 15 minutes.. otherwise, I don't think you have anything to worry about.

 

dbl pust.. Falken, download HWMonitor's newest version..it is free and won't expire. I saw your current program is expiring. You don't need advanced version to monitor GPU temps any longer.

 

The insides of this thing are a work of art technologically. The build quality exudes confidence in a product.

 

I call it, sexy.

 

Word up, E-diddy...lol

 

I remember from this locked thread a while back that it was said that the K|N heat mods to the Nagamaki were more than just holes drilled into the bottom.

mhooper claimed there was a bolt on mod for the GPU that I assumed was like those mods done to that K|N Executioner in the link.

But then again I've never actually seen any pictures of the heatsink mods themselves so I can't be sure. The closest thing to a pic would be the one RunningGunner posted in that thread which shows what looks like a heatsink through the extra drilled holes.

 

E-Wrecked- Thanks, I'll do that. Saves me some money!

everyone else- Mark specifically said that he didn't have any specific heatsink mod for the Nagamaki, he has a new design that he's working on for the 4870's coming out. That's the one Hooper is talking about. The cooling solution is specific to each type of computer, and this one didn't need much initially.

I will probably take off the filter over the area for the GPU cooler. I'm inclined to think that at least one of my cards is just running hot, but I don't know. It's not completely out of whack like some people's I've seen that are hitting 90's. All that has to happen is the card that's being cooled first is running higher than normal, the second card will have to be hotter based on the temperature of the air hitting the second stage of the heat exchanger.

Yes, I was impressed with the interior build quality of the machine, it's very strong and well put together. The chassis feels pretty indestructible, though if you read my review, I point out some of the design feature of the machine, like the built-in card reader, that could use improvement.

Anyway, the main point of this was to show how different solutions affect cooling performance. Once I get a high end notebook cooler I'll add that data too. I don't expect that that will add much, the biggest issue with these computers and cooling is getting more cooling air into the case, as can clearly be seen from the results of removing the bottom.

I would like to stress again that I'm not worried about the temps, I started tracking them when I thought that was the problem, but with the new drivers I don't have the problem anymore (at least, it takes 3 hours to crash out now). It's definitely not heat related. Hope you all found this interesting.

 

what cooler are you planning to get?
The nzxt cryo? i suggest that, otherwise talk to hooper about that special cooler that was supposed to be released for the whitebook. dont know if there is any actual release planned.

 

not sure, hadn't done much research on that yet. Zalman 2000 was one I heard was good. If the custom Whitebook cooler gets done, then yes, that's what I'd prefer.

 

This is the cooler:
http://www.fudzilla.com/index.php?option=com_content&task=view&id=10066&Itemid=1

It's made SPECIFICALLY for this rig. I'm gonna byu it fo' sho'

Intel used it for their public testing of the QX9300 OC'd

EDIT: found another link http://www.coolitsystems.com/index....-intel-laptops&catid=2:press-release&Itemid=3

 

That is friggin' awesome!

 

Isn't it though? johnksss dropped that link on us when the OCZ first came out. Been holding onto it since. I should call that guy in Miami to see about getting ahold of one.

 

The temps under load in the review unit seem much higher than the temps experienced here wth a Quad Core.

Did you get an OEM or ES QX9300 E-wrecked? How's the OC going so far?

Seriously fighting the urge to pick one of these up and go to town. Saw on eBay they have some M17 WXGA+ LCD assemblies now. I have a spare QX9300, 4gb DDR3, 5300 and 320gb sitting on the table behind me.

 

Yeah that cooler looks like it would really do its job.

 

I went with a QS with E0 stepping.. QGNG model - It's literally the retail release, ES. Same stepping and all. I stole this thing for $525 w/ 3 day shipping. Couldn't pass it up for that! I am having some minor issues though, not from the CPU - but my Crossfire keeps dropping in the middle of 3dmark runs. Not sure why. It never did it with the x9100 - and I assume this is why most folks benchmarks with the qx are lower than they should be. I think it's a power distributrion issue.. the x9100 ROCKED in synthetic benches. I'm still testing.. the CPU is stable, so it's definately not the problem. Anyhow, I suggest the OCZ and you can always flash your M17 BIOS

 

did you apply the patch?

 

$525? Yeah, you can't pass that up! You could turn around and resell your X9100 and break even or even get a little coin back in your pocket at that pricing.

That's odd. The external PSU looks pretty beefy. What is its rating?

I'd snag an Xotic or similar barebones CF and replace the LCD assembly or if a dirt cheap M17 came along @ 1440x900 w/ CF, that would work too.

 

http://www.slashgear.com/coolit-lau...r-intel-core-2-extreme-based-laptops-2014318/