*** Official Clevo P770ZM / Sager NP9772 and P770ZM-G / Sager NP9773 Owner's Lounge ***

I've hit my peak on my GPU Won't be able to break 10k.

http://www.3dmark.com/fs/4130943

 

What is the main difference between NP9772 and NP9377? I see CPU type difference and NP9772 seems slimmer and I think it uses IPS screen. Does it use the same IPS screen as ASUS ROG G751 with G-Sync? Any other difference?

It would be my number 1 choice IF it could support GTX 980M SLi.., but I guess that is not possible if it is to preserve its slim-factor.

Why would Sager want desktop 4790K in the SLIM laptop and 4810MQ in the thick laptop? Seems like it should be the other way around...

 

Will call UPS to see if I can pick.up the laptop a day early on Monday. Hope I can since I've been working a lot only have Monday and Tuesday to setup and play with the machine.

 

10h and I haven't hit 80* on cpu or 65* on gpu. and that is cpu utilization over an hour graph. nice results from repaste. forgot to take pictures when i took my heatsink off but both sides looked good.

 

What did you use to over clock? What did you have to change to get it to work?

 

Clevo just didnt update their SLI line yet...

 

Ordered the batman today, too. Just a small question to the owners:
Did someone already installed Win 7 on it and does the Win 8.1 drivers on the Clevo site also work for Win 7? If not, then a link collection of or to the needed drivers would be helpful.

 

For whatever reason my vbios was locked. After installing Prema's vBios all was well. NVinspector/afterburner work without a hitch.

 

Check your resellers site, sager, xoticpc, prostar,etc. It should come with a driver dvd to copy onto a thumb drive. Or you can download directly from component manufacturer for reference drivers.

I did that for win 7 pro

 

MySN has drivers for 7 and 8.1. I got all of them for win 8.1 from their site and everything works fine

 

Xoticpc came with both.

 

This is the link to the sager collection of drivers it has both win 7 and 8 drivers. http://www.sagernotebook.com/drivers.php?cat=608

 

I picked up the laptop today and a rattle happened. Guess what a screw was loose in the laptop. I opened the cover for the cpu and gave a couple of gentile shakes and out came a screws. I removed the extra storage screws that were not being used at this time.

I have had the laptop for less than 2 days, I am assuming it was a build mistake.

Anyone else get this?

 

lol, why the heck was it so difficult? i thought its rather easy to clean up

locked vbios? huh? where did u get ur machine from again? did u make a vbios dump? maybe prema could take a look at that... hope thats not a new and upcoming trend...

@blightcp: lucky that the screw didnt nick the mobo and fry it, that wouldve been.....nice. not.

 

THX! This should be in the FAQ of the batmans, too.

 

It was thick. Almost like thermal pads but rubbery. Around the edges of both dies i had to let articlean soak a few times and the edges around the GPU die I had use a floss pick to pick out the goo and there was still a little bit left. I got a good paste though as my temps have shown.

 

aaah so u mean the paste was hard to clean up it sounded more like the heatsinks themselves made cleaning difficult

@Phönixfire: was actually thinking the same thing linking those in the batcave OP shortly...

 

I'm so bad when cleaning my own heatsinks, I just use kitchen towls it gets the job as good as it needs to be if using the same paste again however

 

actually, scratch that, the drivers they provide are actually way behind those of MySn, and those are already linked in the OP.... better not to cause any confusion

 

So did you find a key or did the bios have one?

 

Yea I created a dump of my stock vbios and already emailed it to Prema. It has been received, don't know what the conclusion is.

I got my P770ZM from RJtech.com. On a sidenote, I found stable OC on my GPU (980m) at stock voltage +144 on core and I found my stable undervolt on my 4790k. Temps for my CPU max out at 75c and my GPU temps never go over 63c.

Oh also max settings on Shadows of Mordor bench and I average ~68FPS with everything on Ultra. So nice to have 8GB of Vram.

Last edit, anyone know how I would go about getting 346.87 drivers on my laptop? laptopvideo2go.com doesn't have the inf for me when I put in the info. I've done a diff on the inf in the P75xzm thread and I see where all of the additions are made, so do I just need to put my hardware IDs in the correct string?

 

346.87 is linked in the batcave op already modded with the appropriate .inf file foe 970M and 980M on win8.1, so just go ahead, download it and install it

just make sure to use ddu to uninstall ur current drivers in safe mode, then deactivate driver signature enforcement before installing the new ones.

Sent from my Nexus 5 using Tapatalk

 

Two updates. First @jaybee83 I created a stock dump of the 980m vbios. Let me know if you need it. Also I repasted today and am a little concerned I did not do it correctly. Stock voltage I hit temps of 98 degrees and had thermal throttling pretty bad. After repaste using IC thermal I hit 96 under same conditions with no throttle. I expected to drop temps more than this. I took it apart and verified that everything was covered like it should be with the paste and cleaned and did it again. Second go at it and same temps. Is this to be expected? What temps are you guys getting using XTU and stock everything on the 4790K?

I agree there was a ton of stock paste. Mainly on the CPU but it wasn't terrible to get it clean. I just wish my temps dropped more. I'll be interested to know everyone else's temps. Maybe I'll look into getting that past that pukemon got

 

please post some pics after a paste job, would be interesting to know what exactly the coverage on the heatsinks and chips look s like.

thx for the offer on the stock vbios, much appreciated metacarpus already helped me out in that department though.

cant say ive run xtu stress test on stock before since i started with prime95 right away. it is indeed the case with the 4790k that without undervolting itll shoot up into thermal throttling range quite easily, no matter which paste u use...

but lets first check n make sure ur paste job is fine and the heatsink is not warped first things first!

Sent from my Nexus 5 using Tapatalk

 

To check for warping and contact,do a super thin layer spread/business card style. Set the heatsink directly down and screwscrew in the cpu side. Pull it back off and look for good contact all over. If you have spots with little to no contact left by the thermal, you could have a milling or warping problem. If it has nice contact, then it is a matter of finding the right amount of paste. I haven't checked my stock thermals in awhile, so I can't verify (should be earlier in this thread). I'll try to check tomorrow morning for you.

 

Has anyone tried using 2133 ram?

 

Thanks for the reply. After pasting I have not hit any thermal throttle using XTU. Again this is using stock voltages. I did undervolt but until I verify the paste job I'll leave those results on the side for now. Unfortunately I did not take any pictures of the heat sink paste and used the last bit on my second past job. Upon removal of the first paste job it did seem to cover the entire CPU without much excess at all. Looking at the heat sink it seemed that the cover was fine. I just expected larger drops in temp upgrading from stock paste. I do want others temps to compare but tomorrow I will order better paste and take pics etc. when that comes in. I used the pea method. Using a smaller pea on the GPU and larger PEA on the CPu since it's fairly large. Maybe I should have used the three dot method but I'm not convinced coverage is the issue. It may in fact be normal so I don't want to raise concern but I do want to verify that it is normal if you us see what I'm saying.

 

I've installed 32GB of Kingston 2233 ram and it's running perfectly without any issues.

 

Have you verified it is operating at 2133 and not 1866?

Are you able to post a screenshot?

 

I do Pea method myself. The spread is to check well for any protrusions or areas of no contact that pea and line (x) method may not catch (as they fill in all extra space). But happy to hear it worked!

 

Stock everything I was hitting 91* on only one core. The others were about 89*. After some undervolting, 85* tops. I repasted last night. Peak temp was 79* initial run with about 24*C ambient temp. All my runs after are 78*. Todayy ambient temps is 25*C. Note that what I just mentioned is with turbo multipliers off. I did a run awhile ago with turbo back on and max was 85* one one core and 83* on the others. All auto fans. You can kock 5-10*, closer to 10* off with full fans on all my xtu runs.

My prime95 runs peak at 98* auto fans. With full fan somewhere in 80's. Forgot about where. Don't really like running prime but I'm still unstable.

As for repasting I noticed I had good contact from whatever hard as hell to get off stock precut stuff they used before. Was like square wafer putty. As jay noted you shouldvr noticed what kind of contact your heatsinks were making with die first pull. Mine definitely looked good and my initial paste that I was worried about turned out great. Gelid doesn't really have a curing or break in process but I noticed after the first few runs of xtu and prime my temps dropped another degree or two after a few hours.

I just ran a 5 minute xtu CPU test while writing this and max temp was 76* auto fans with turbo off and new undervolts. Lesser undervolts that is, because I've still experienced crashing even after hours of presumed stability. Currently near the end of a 5 minute ram xtu test as I write and peak temp is 74*. Consistent because my ram tests peak has always been a couple degrees lower than CPU.

I highly recommend gelid extreme or shin etsu if you can find it. I forgot the beat shin etsu model number but it was good stuff too. But gelid is much easier to find. I strongly making sure that your heatsink is making good contact on both dies. One pea sized glob on GPU and about 2-2.5 sized pea glob on CPU die since it's bigger and very important. A little too much is much better than not enough. Way too much can be bad, and potentially catastrophic. After you add your paste make sure to center heatsink with screw holes and then press firmly on heatsink with even pressure approximately where your globs were. I didn't follow numbers putting screw back in. What I did was two screws diagonal on each side. I only tightened it enough to be tight. Then add rest of screws. I then 'dogged' rest of screws. 'dogged' tighten the **** out of it, but these are tiny delicate screws. Use cautions dogging your screws. You definitely don't want to strip screw head or standoff female receptacle or whatever you call it on motherboards. And just be patien while doing this. No need to rush. And gelid is like dealing with spiderweb silk. Be cautious when you pull away from adding paste. Either twist the strand or use a qtip to cut it.

And as a personal note, I know ic diamond is good but it's expensive and more importantly it's abrasive. Not really a good paste to use if you're still figuring out your own little method or routine or you're a noob to doing this. I'd feel very comfortable using ic diamond now but still taking it off for a new repaste can slowly degrade your die and heatsink with scratches or pits. Worst case scenarios are noobs not k owkmg what they're doing losing the etching of their dies because they've not taken it off correctly and/or used too much pressure. Hope that helps. I'm on my lunesta, sleep aid pill, and sometimes I get flappy too much, especially writing posts like this. Hope I didn't offend your experience level. And if I haven't hope the info helps your temps a lot.

 

According to XTU it is 2133 I'll screen shot tomorrow as well as continue with heat updates.

 

I found Gelid Extreme to be better than ICD, and doesn't stain/scratch like ICD. Been repasting for years with ICD until tried Gelid and won't go back now.

@superkyle1721 Also thanks about 2133. Tempted to order some hyperx 2133.

 

Just making sure it is known, you would need 204-pin laptop DDR3L So-dimms. I am not aware of a hyper-x version in the laptop arena at 2133 from Kingston (but I could be mistaken). Here is a link to your two primary options:
http://www.newegg.com/Product/Produ...07609 600006178 600213067 600523465 600373148
Hope that helps!

 

Amazon has it but expensive.

Now thinking Ballistix instead.

I have prime.

 

Did you try newegg? Newegg is usually best place to buy ram. If you're patient too, put it in your carr or wish list and they'll send a coupon eventually.

 

ICD doesn't scratch dies if you take it off correctly.
Remove heatsink
Apply cleaning solution to CPU and Heatsink (such as Arcticlean or alcohol etc)
Wait about 3-5 minutes
Wipe off and have best time of life with the easiest-to-remove thermal paste I've ever seen.

 

@pukemon thanks for all the info. I did know that IC thermal can cause scratches etc. I have both the artic thermal paste remover and surface purifier. You are right in saying I am a noob to pasting laptops but I have built several desktop computers and have reposted several more. I have never bother with repasting my laptops since they generally did not run this hot. With that said It seemed as though the pasting of the laptop is straightforward. One thing I did do was maintain pressure on the GPU/CPU to ensure that air bubbles did not form. That was my main concern. I'm currently running some test and I will compile them and post them here once finished. I will then let you guys decide what you think about the pasting and if I should start over or not based on temps under different conditions.

 

@Zymphad here is the ram I am using. Installed perfectly first time and I have always had good luck with Gskill.

G.SKILL Ripjaws Series 16GB (2 x 8G) 204-Pin DDR3 SO-DIMM DDR3 2133 (PC3 17000) Laptop Memory Model F3-2133C11D-16GRSL http://www.newegg.com/Product/Product.aspx?Item=20-231-760

 

Also I'm curious what you guys undervolts are and which settings you used. I was able to change the dynamic voltage to -115mV with everything else the same and it would finish the benchmark but later crash. I reduced it to -100mV and it would run for a few hours and eventually crash. I think I'm settled around -80mV for stability purpose but I'm not sure if playing around with the other voltages can help stability. I've done some reading on it but for whatever reason undervolt using XTU is harder to find than other programs.

 

opened all of the compartments to find the loose screw and did not see one.

I did run Belarc and got the key so did licencingDiag.exe

It is a CORE license type, that means you can use the generic license key to install from 8.1 media.

 

I just bought 4x4Gb of the same type to allow both memory channels to run ganged to optimize bandwidth. http://www.newegg.com/Product/Product.aspx?Item=N82E16820231752

 

Yeah I did the same except went all out with 4x8GB over kill for what I need it for but so is this entire laptop

 

The undervolting isn't as important as the voltage. Each of three variants have different stock voltages and different stock offsets per core. If you want to compare, in xtu by the graph, click on the wrench and turn on the voltage monitor. That will tell you what your chip can really do and can be compared to desktop websites talking about Overclocking!

 

Thanks a lot for this info. Can you please explain a little more? Just by changing the dynamic voltage offset I was able to reduce temps by nearly 10 degrees. I'm almost finished compiling data so my results will make more sense soon but it seems as though by just changing that it made huge improvements. If changing that is not very important in reducing temps then what voltage do I adjust to do this properly? If you are using a 4790K do you mind taking a screen shot of all your settings so I can take a look? I think that will help me better understand what you are saying.

 

Same here with Gelid Extreme, found it to be the best traditional paste solution for an at home setting.

My process for mating a heat sink to processor is a bit extreme, but it is how I was trained.

Note: this MAY void the heat sink warranty but does not void the CPU.

If I perceive ANY roughness on the heat sink when I drag my fingernail across the heat sink, I polish the surface with jeweler's rogue.

NOTE: Practice on an old heat sink to learn the technique before trying it on anything. This takes me about 2 hours now to get a good polish.

After cleaning and inspecting with a 10x loupe, I "polish" Gelid into the heat sink pores. I then back drag a small film over the heat sink.

Next I clean the CPU as much as I can, gently "polish" Gelid into the CPU surface, you are not trying to remove material, just get the heat sink compound as deep into the metal as possible.

Then mate the two together to the specified torque.

Technically you want the thermal compound layer to be as thin as possible, the ideal is to just fill in the pores on each surface achieve a consistent mechanical connection across the contact area. this can get you an extra 2 or 3c if you are really good at it on top of the change in paste.

 

The dynamic voltage offset says how much less voltage you are using. The core voltage tells you the exact voltage being used. By doing a dynamic offset instead of setting the core voltage to a specific amount allows for extra voltage here or there when requested. A static voltage on the core means that it is set at that permanently, no more or less. If it needs more to run under static Overclocking, you crash your system, set voltage higher, try again (Overclocking of the 90s and early 2k). Here, dynamic offset allows more wiggle room and stable overclock with less voltage instead of setting it to a voltage necessary under all conditions. Even with this, knowing what voltage is going to the core and other components is important as a comparable value to understand your system's needs and how voltage effects heat in your system. Look up Overclocking the 4790k in desktops. They give you a core voltage amount set, not the offset. The offset varies by cpu as the voltage does, but voltage is a hard measurement, offset is not.

So undervolt like you're doing with dynamic offset, but post hard voltage as well for a true cpu efficiency comparison!

 

CPU core voltage peaks at about 1.15V but hovers between that and 1.105V I'm assuming this is the value that's best used in comparisons?

 

Now just add the multiplier settings related to that voltage and it tells me voltage at a specified ghz. I started by first finding the lowest stable under volt on cache voltage. Then the lowest stable undervolt on core voltage. There are guides out there that explain desktop Overclocking. Refer to them on how to set each voltage. This will get the best results. I will have up links and a manual soon, but you can Google quicker. I do a 20mv jump until crash, then back it off by 5mv. Others go by 10mv jumps. Xtu stable is a good start, then wprime, then prime 95 large set, memory, then small if you want to stress out for video encoding. Burnin test is decent too.

 

Ok here is a list of some of the things I have compiled. Let me know what you guys think