Initial Impressions on Sager NP8170

Correct me if I am wrong but the whole point on FN+1 is to keep your things cool under load is it not? its not something you turn on and off as you please. To achieve decent temps during applications that are very stressful you just keep the fans running on full speed.

 

HT was right when he said the fan at max was like a hair dryer. The thing is LOUD. I don't think I could sit and game for any substantial amount of time and listen to that fan noise.
Honestly I think the max fan should kick in only during extreme situations. Not under load while doing normal gaming as the computer was intended.

HT I think your temps do seem a little high. Here is what I got running Prime95:

Spoiler

This was after running Prime95 for 10 minutes Using the "In-place large FFTs" Torture test on 8 threads.

 

Fn+1 turns fans on max speed, it's a toggle.

Yeah, I just played Battlefield Bad Company 2 for an hour, and here's my temps:

CPU: 85-89-85-81
GPU: 82

That being said, I have to say I love gaming with this power. Played 1080p @ max settings (default max) and fps remained above 50, and sat comfortably around 60 at all times. This larger screen is beautiful as well.

 

o no...it makes me want to stay away for IC diamond upgrade (not due to material, but people who do not apply it correctly).

noise is anothr concern to me...hopefully they can improve it to the level of asus G73. (I heard it's quiet, but never heard it myself)

 

Oh one more thing. While this screen is considered "glossy" it certainly isn't bad. By glossy I was half-expecting an M11x glossy which might as well be called "mirror". The reflection lights are somewhat diffuse and not horribly annoying.

 

Well noise is reasonable if you don't run the fans at max speed. If you're wearing headphones (not earbuds) it's not so bad either. The max fan sound it emits is LOUD but not really annoying. Kinda like your cars air vents at max.

 

You know...I hate to say it, but from your temps, it's almost as if you got STOCK compound b/c IC7 should give you the lowest possible temps. Could someone have really applied IC7 so haphazardly and carelessly that it had the opposite effect? That is hard to believe. As reflected in my monaker; given two differing hypotheses, the simpler one is likely more correct, and your rig having STOCK compound could be that simple answer. Something to chew on perhaps?

 

I know that, I am talking about the whole reason behind the toggle is for you to "toggle" it when playing games to keep your stuff cool. Not just to enable it for a few seconds and disable it. The reason behind those bad temps I would think is because your not keeping the fans at full blast the entire time.

 

Well, when comparing temps between two machines, both should be at the same base line.

So if a person gets 80'c without toggling the switch on, another should theoretically be getting around the same temps.

The fact that htwingnut has to toggle his fn+1 on to achieve such temps means there is something wrong with the ICD application.

 

My point is that the fan algorithm should work to keep temps low enough. I shouldn't have to manually enable them. If that means the fans go to full blast on their own then so be it. Unfortunately there's zero configuration in the BIOS for the fans either. I don't like using software fan control either, that's just another app that I shouldn't have to run. Not even sure if something like speedfan would work with it either.

Based on my CPU temps vs. Marvin H Muckley III his are a good 8C cooler running the same program (Prime95). Unless his room is like 10C (mine's about 21C) I think there's some issue.

Even the best thermal paste can cause issues if you use too much or it's unevenly distributed. Either way if I repaste with ICD and my temps drop significantly I'm asking for my $40 back.

 

I'm not too technical when it comes to these things, but does the temperature difference have anything to do with the fact one is a 470m and the other is a 485m? I would think that the 485m would be a bit hotter, but I could be wrong.

 

Yeah I see what you mean now. But realistically speaking since he has a 470m while you have a 485m. Your gpu creates more heat around it, sure it has its own heatsync but it increases the ambient temperature inside the casing aswell so its not unlikely that your CPU temps are higher than his.

Edit: Highspeed4567 you beat me to it XD

 

, but ya these last few posts seemed a bit odd that we were comparing a 470m vs 485m and expected the temps to be close to each other when running GPU benches.

 

Oh no thats not the issue, the issue is the CPU temps. But seeing as the 485m runs hotter than the 470m I would assume the ambient temp inside the chasis would be higher with the 485m causing the increase in CPU temps.

 

You guys are forgetting about the fact that we ran Prime95 which stress tests the CPU. Look at the GPU temps during the test. Both are very close where his is 40 degrees and mine is 39 degrees. However, his CPU temps range from 81-85 where mine are max at 78.

This leads me to believe that something isn't right.

 

I did not forget that, still many things could increase ambient temperatures inside the chassis, be it GPU, Ram, etc. Even if those components aren't being stressed their still running which cause ambient temps to increase.

Also whats your room temp? That could make a world of difference aswell.

I am just trying to help him look at all the alternatives before he decided to open it up and reapply the thermal paste. (Ofcourse that could be it aswell, I am not saying that it isn't).

 

Two computers are not always going to run at the exact same temperatures...

 

Room temp will only make a significant difference at idle. At full throttle the room temp would have to be significantly lower to make a 10C difference (it's not 1:1). There should be little variation between computers, at least +/- 2C at the same ambient temperature.

The GPU won't make a difference here either. Everything else is identical between the machines that would have any influence.

Marvin Muckley - can you run CPU-Z and watch your CPU speed while you run prime95. Mine steadies at 2.4GHz although it seems it should run much higher than that with Turbo Boost.

 

Well then I guess you might aswell re-apply the thermal paste as soon as can to not waste anymore time. The sooner you do it the sooner you know for a fact on whether it was the thermal paste or not.

That would suck if it turns out you paid the extra 40 usd for nothing

 

I just reapplied thermal paste and temps are 75-76-74-73 compared with previous 83-87-84-81. That's 8-11-10-8 deg C difference, pretty significant. Now on to furmark.

BTW, congrats to Green Bay!

 

Hope it is nothing major wrong with your new laptop

 

Annnnnd there is why you only pay yourself $40 to apply thermal compund.

 

boooo! Steelers should've won! lol

 

Right. But I thought what the heck, I don't want to have to go through all that. I know it's not that hard, but really didn't want to have to do that with a new laptop.

Furmark is about 93C, 3C hotter than before, but hopefully with a little curing time it should drop 1 or 2C, I'm not as concerned about that. I also hear the fan kick in over 90C and it drops down to 90C right away.

 

Posted by BabyHemi:



You can see how well segregated the CPU and GPU cooling systems are, so little temperature influence from one to the other.

 

'ht' - what state did you find the original paste in? Was it misapplied? Just curious before I, too, possibly choose to pay $40 as well (though I think K_J is right) for the same thing.

 

I'm also curious what paste you applied

 

Oh I applied IC Diamond, like what I paid $40 for.

After I pulled off the heat sink though, it looked like it was applied decently. Maybe a litte thick, and one small portion of the corner wasn't covered. But overall not bad. However, the paste seemed quite fluid compared to IC Diamond, which is much thicker. It seemed like it was either factory goop or Arctic Silver.

Spoiler

 

and you paid for icd7? makes you wonder if they're actually applying icd7 or if they're using something else.. i too would rather pay $40 to have it be done, then have to take apart the new rig i just paid $2700 for. at prices that high, you'd assume they'd do it right lol

 

Thanks! Well, that makes up my mind to just do it myself.

 

I don't want to scare people off by any means. I don't know what the cause was, but it's clear changing the paste myself, it dropped temps by 8-10C. I am going to request a refund of my $40 tomorrow, we'll see how that goes.

 

Yeah, you should definitely get a refund for that one.

Let us know how it went.

 

From 'Goldfish' stating the original paste did not seem as thick as ICD7, I am of the mind that it wasn't ICD7, and perhaps his chosen-order option didn't make it to the factory floor. He should definitely get his money back! Either way, based on his help here, I am not going to take the chance when I know I can do it and trust what is in my system from day one (for less money).

 

Did it take some time and effort to change the paste yourself? After reading about your experience I figured I should save myself $40 and also change the paste myself.

But I've never repasted before--is it fairly easy to access the cpu and gpu? Thanks.

 

CPU and GPU are easy to access on this laptop.

But you will need something more than just screwdrivers and cloth..

First of all you will need articlean, or some sort of alcohol to wipe the goop.

You will need a coffee filter, or microfiber cloth to wipe the alcohol off with the goop.

Then you will need to buy the paste you wish to apply, especially IC diamond or Arctic Cooling MX-4.

The internals can be seen in the XoticPC reviews, so you shouldn't face too much of a problem.

 

I agree, an 8-10c drop for just applying it yourself means they didn't do a decent job. Good to know the problem was solved though!

 

It isn't that hard to do. I've changed thermal paste dozens and dozens of times, on desktops and laptops over the years. The Sagers have real easy access on these machines. The heatsink for the GPU and CPU each only require four screws removed and it pulls right off.

But as meraki1990 stated, you need the proper tools to do the actual removal and reapplication correctly. I also use ArctiClean kit, coffee filters, soft paper towel, 99.999% alcohol, lint free q-tips, toothpicks, all to use in the cleaning process. I'm pretty meticulous, but it's a messy job to clean up the existing goop.

Then application of the new paste is pretty straight forward. I usually use a credit card (edge cleaned thoroughly with alcohol) to "tint" the surface of the CPU and GPU, basically scrape the card and goop across the surface to fill the tiny microscopic voids. Then apply a straight line bead along the length of the CPU or GPU, and squish the heatsink down on the goop and fasten the screws. Some people don't think you need to tint, but I've always had good success with it, mainly on the heatsink because it's not quite as lapped as the CPU heat spreader.

 

All I use is alcohol, a paper towel, and my finger....

 

lol. That'll work too I guess. Cleaning is the hard part. Application is easy.

What other benches, or info would you like on this machine before I try my hand at a full review. Although there's probably not a lot to say other than what's already been said, but probably good to consolidate and formalize.

By the way my idle temps are really good too:

 

Very nice. Amazing what thermal paste can do when applied correctly.

 

I did a comparison a while ago between the P9700 Core 2 Duo and Q9000 Core 2 Quad.

http://forum.notebookreview.com/gam...er-np8662-comparision-core-2-duo-vs-quad.html

I will probably run several of the same benchmarks to compare. Marvin Muckley also stated he would run benches if he had the game and/or access to the benchmark so we can compare GTX 470m vs GTX 485m.

I think it would be good to know for Core 2 users what kind of performance to expect updating to a new Core i7.

 

Free benchmark demos:

Lost Planet 2
Resident Evil 5
Street Fighter 4
Dirt 2
Just Cause 2

 

Thanks. I have RE5, DiRT 2, SF4 demos, have to get the other two. I suppose I should do Crysis just for fun too. I think you can run the benchmark in the demo, right? I will also be doing the Flight Simulator X benchmark. Curious how well this machine will handle it. That is one of those holy grail of performance to seek. It can consume your entire machine, very CPU biased too.

 

After discussion with Justin @ XoticPC I'm going to wait until after they replace the mobo from the SATA controller recall and see how the paste fares then. Sounds fair to me. To others, don't let one person's issue dissuay you from making the purchase. I dont know what the exact reason was for the higher than normal CPU temps, but the paste spread looked good. I dunno, in either case it's fixed, and hopefully will be after the mobo is replaced.

 

Well ICD7 does not have curing time, so it theoretically should be the same the moment you apply it.

I still can't understand how high the temps were after application of ICD7 by the Sager factory, and also with you paying $40 for the application.

Sager should take more care into applying thermal paste, as this isn't some OEM compound, but a customer's trust in paying $40 for a less than $5 paste, which means the $35 should go into reassurance of quality..

Very disappointing.. to be honest.

 

I'll be very interested to see FSX benches. That game is the death of me when I haul a three hour flight only to have system malfunctions which cause me to crash on landing due to stuttering controls. But that's what happens when you try to play something on a 4-year old machine.

I agree that one person's bad instance shouldn't cause a purchase-option rebellion, but for me, I was already leaning toward not paying $40 for someone else to do what I can easily take care of later on. No worries

 

HT, have you played Total War or any other RTS on the 8170 yet? I am curious as this is one of the main reasons for purchasing this machine. I tried playing Napolean on my Inspiron 9300 and things didnt work out to well.. Needless to say it ended in a hard reboot of the machine..lol. Def time for a new Laptop.

Do you think this machine is overkill for Total War and similar games?Do you think the Sager 5170 would be a better choice? It has the 1024MB DDR3 nVIDIA GT 540M w/ Optimus.

Thanks for all the good info you are providing.

 

I think with Total War you're better off with a higher CPU than anything else, and the i7-2630qm would be great. GT 540m is probably adequate as well.

 

In regards to total war, especially the new ones IE empire and the upcoming shogun. You'd need a combination of BOTH a strong CPU and a GPU. I include the GPU as well because of how detailed the game is in terms of graphics. This is ofcourse assuming you like your games running on high and above, nothing less.

 

Do you mind benchmarking Battlefield BC2 online and let us know how well it runs? I get the feeling that benchmarks from notebookcheck were done on single player... which means nothing to me. I want to feel confident that this thing will plow through BF3 in the future. Thanks