Jump to content

Recommended Posts

unreal25: Completely agree with the shooting of the designer that decided we have to pop the touch panel for access. I just broke the extreme right top clip of it by using what i thought was a safer method, popping it out from the hinge cover portion.

Huuy: I'm also experiencing the same. Ive checked the power status whenever the lights change and it does not show that the AC is plugged in even though it is. Although this usually happens when I'm running something intense. About the scratch, leave it alone. Get a replacement if you can. I tried to polish a scratch on the black border on my M15x using meguairs plastic polish and it only made it worse. If it is on the black part try this... take the tip of your index finger, rub it on your forehead and then on the scratch. Tada! Magic! :)

Ive completed the repaste on the CPU and GPU. New 0.5mm thermal pads on the DDR5 chips as well and now temps of the GPU dont go over 80C on Furmark.

Ive used Shin Etsu X23D on the CPU and the GPU. Problem is there is a 4 degree difference on the cores at load and idle on the CPU with a max of 81C. Should I be worried about the difference? Any way to reduce the difference in temps between the cores?

Core 2 is the highest even before when I used AS5 but there was only a max of 2C difference.

NB. I'm not using any retention mods.

Edited by Nospheratu
added info about no retention mods
Link to comment
Share on other sites

  • Founder

4C degree on the CPU cores can happen depending on which benchmark you use. If it's Prime95, you will see it occur. I wouldn't worry about it too much because sometimes it's out of your control due to how flush the heatsink is. Welcome to T|I nospheratu.

Link to comment
Share on other sites

Thanks for the welcome Brian :)

Awesome site you have here and cant wait for the R2/R3/M18x review! Your avatar is off the hook man, is that part of a video/screensaver or did you make it yourself?

Ive been using wprime 1.55 with 8 threads for max temp testing. I was thinking the same thing about the heatsink. I screwed in each screw completely by the numbered markings and not a little at a time following the numbers. Not sure if that makes a difference. Ill give it a try tonight without removing the heatsink and report back.

Link to comment
Share on other sites

  • Founder

You have to fine tune the pressure each screws applies to the heatsink and the CPU so you can make sure that both surfaces are as flat as possible to each other, (for me I had to use only 2 screws, the other two are just there doing nothing, using 3 ir more screws with my heatsink changes the geometry to the worse).

Contrary to Brian, I fine tune the screw pressure by leaving my CPU to idle, the reason I do it is cause the temps don't fluctuate that much so I have better feedback when I do a change, with load on the CPU the multipliers drop or distribute the workload to the rest of the cores in order to keep the temps down so it is harder for me to adjust the screws this way.

  • Thumbs Up 2
Link to comment
Share on other sites

I wonder if anyone else is having the same problem with the lights changing. I'm getting some stuttering in some games, wondering if that's the problem or my crappy SSD (waiting for dell to send the screws for the other HDD).

Link to comment
Share on other sites

  • Founder
Thanks for the welcome Brian :)

Awesome site you have here and cant wait for the R2/R3/M18x review! Your avatar is off the hook man, is that part of a video/screensaver or did you make it yourself?

Ive been using wprime 1.55 with 8 threads for max temp testing. I was thinking the same thing about the heatsink. I screwed in each screw completely by the numbered markings and not a little at a time following the numbers. Not sure if that makes a difference. Ill give it a try tonight without removing the heatsink and report back.

There is the CPU retention mod that you can also employ, have you considered that? Its useful for those with extreme processors. The avatar is from a screensaver I have, I used a screen capture program to convert it to a video file and then an animated gif :)

  • Thumbs Up 1
Link to comment
Share on other sites

StamatisX: Thanks for the wicked tip! Managed to reduce my max temps by 2 degrees and reduce the difference in temp between the cores to a max a 2C. Core 2 is still misbehaving at idle though. At load it seems to even out.

Huuy: I experienced micro stutters in Portal 2/3DMark11 when I had HWinfo32 running in the background. Check if you have anything running in the background that could be affecting the game. It might be something else but its worth a try.

Brian: I've considered the mod but I think I'll leave that for another day when I'm bored or run out of things to fiddle with :D

Link to comment
Share on other sites

You have to fine tune the pressure each screws applies to the heatsink and the CPU so you can make sure that both surfaces are as flat as possible to each other, (for me I had to use only 2 screws, the other two are just there doing nothing, using 3 ir more screws with my heatsink changes the geometry to the worse).

Contrary to Brian, I fine tune the screw pressure by leaving my CPU to idle, the reason I do it is cause the temps don't fluctuate that much so I have better feedback when I do a change, with load on the CPU the multipliers drop or distribute the workload to the rest of the cores in order to keep the temps down so it is harder for me to adjust the screws this way.

You know I can't agree more... although I didn't repaste I tried just adjusting the screws on the R2 and with Prime95 going and I noticed the same thing... that you couldn't tell a difference between the temps and what adjustments you made....

Huuy: you can also disable individual sensors in HWinfo32, if your not using all of the sensors... every one uses a little resources so its up to you what best suits your needs. HWinfo32 took some cpu time when all sensors were on and fan control was on... I turned the fan respin value as long as it could go without letting the fan die down... this saved resources too... hope that helps.

Edited by mw86
Link to comment
Share on other sites

  • 2 weeks later...
  • 5 months later...

Ok, I've got an M17xR2 inbound and am pretty excited to join the club!

Just wondering what your guys thoughts are on IC Diamod thermal paste. I was looking at grabbing this one but I thought I'd check and ask whether the one in the OP or another type was a better option. Thanks in advance.

Link to comment
Share on other sites

Just wondering what your guys thoughts are on IC Diamod thermal paste.

There has been quite some discussion about ICD here on this forum, the paste performs really well, it's one of the best TIMs on the market.

You can find a review here: Innovation Cooling – Diamond 24 Review | Tech|Inferno

Also check out these two threads:http://forum.techinferno.com/general-notebook-discussions/815-innovation-cooling-ic-diamond-giveaway-survey.html and http://forum.techinferno.com/general-notebook-discussions/913-ic-diamond-24-survey-results.html

I can only think of one slight disadvantage ICD has... if you wipe off the paste from the die in case of a repaste it might leave some scratches on it... very tiny scratches, but still visible. That's due to the diamond particles in the paste. But when you clean the die carefully you shouldn't have issues. And it doesn't really matter anyway.

Edited by svl7
Link to comment
Share on other sites

There has been quite some discussion about ICD here on this forum, the paste performs really well, it's one of the best TIMs on the market.

You can find a review here: Innovation Cooling – Diamond 24 Review | Tech|Inferno

Also check out these two threads:http://forum.techinferno.com/general-notebook-discussions/815-innovation-cooling-ic-diamond-giveaway-survey.html and http://forum.techinferno.com/general-notebook-discussions/913-ic-diamond-24-survey-results.html

I can only think of one slight disadvantage ICD has... if you wipe off the paste from the die in case of a repaste it might leave some scratches on it... very tiny scratches, but still visible. That's due to the diamond particles in the paste. But when you clean the die carefully you shouldn't have issues. And it doesn't really matter anyway.

Thanks for the reply svl7, I'll definitely grab some. I'd rather risk a few little scratches on the die for those sort of temps.

  • Thumbs Up 1
Link to comment
Share on other sites

I'd recommend getting Arctic MX4 instead of the IC Diamond - it's MUCH easier to apply and wipe off, and it's just as effective. ICD7 works perfectly for video memory chips, though - I recommend applying it instead of those Fujipoly pads. I bought the 1mm version and it's too hard - it's not easily squished into place like the usual gel pads and your GPU may be left without pressure.

  • Thumbs Up 1
Link to comment
Share on other sites

True, MX-4 is much easier to apply if you haven't repasted often, but I definitely had lower temps with ICD, just a couple of degrees, but still clearly noticeable.

An advantage of ICD is that it is more viscous, this makes it harder to apply but the paste will stay at it's place and won't get sucked out as easily as MX-4 when the chip / heatsink is heating up and then cooling down again. Disadvantage is that it is much more expensive.

  • Thumbs Up 1
Link to comment
Share on other sites

Thanks guys, I ended up going with MX-4. The price to get IC Diamond shipped to Australia was ridiculously expensive compared to MX-4 and it would probably be better to use as it is my first time repasting. Just wondering, would I be able to substitute MX-4 for the thermal pads? If so, how much should I be putting on?

My M17x should be arriving next week (so excited!) and I'll be repasting on the weekend after exams are over. Thanks again for the help. :)

Link to comment
Share on other sites

  • 5 months later...

Awesome, I just got some Arctic Silver 5 (not many options to choose from where I live) and will be performing this mod today or tomorrow... I have one issue, I'm using HWInfo 3.95-1600 (latest) and it will not show the sensors and temps of my GPU #1, it just shows GPU #0, Is there a setting I need to change?

Added after 5 minutes:

Please delete my previous Post, just as I clicked "Post" the info appeared on the sensor window. I'm attaching it. I have one question: Why are the clocks for GPU and Memory so different between GPU 0 and GPU 1?


post-280-14494993083483_thumb.jpg Edited by draiken
Link to comment
Share on other sites

  • Founder
Awesome, I just got some Arctic Silver 5 (not many options to choose from where I live) and will be performing this mod today or tomorrow... I have one issue, I'm using HWInfo 3.95-1600 (latest) and it will not show the sensors and temps of my GPU #1, it just shows GPU #0, Is there a setting I need to change?

Added after 5 minutes:

Please delete my previous Post, just as I clicked "Post" the info appeared on the sensor window. I'm attaching it. I have one question: Why are the clocks for GPU and Memory so different between GPU 0 and GPU 1?


[ATTACH=CONFIG]4180[/ATTACH]

it's weird indeed, iirc this shouldn't be happening.

Are you running the latest version of HWiNFO (although i doubt the problem is there it's better to always run the latest) ? Try running furmark for a couple of minutes and see how the clocks are then ?

Link to comment
Share on other sites

Actually the same also happens to me. The secondary GPU runs a little hotter. The difference isn't that big, on the full load I think GPU0 is ~72 C, GPU1 around 75-77C. This was every since I repasted, but I actually haven't checked before.

In your case however, it seems GPU1 was running at higher clocks for whatever reason. In my case they both ran at about 100% utilization with 700/1000MHz stock clocks.

Link to comment
Share on other sites

Not sure if this will help, but sometimes one card on my M18X seems to get stuck at low clocks while in game, firing up trixx and adding an OC (Even if it's like 5mhz) fixes it.

Link to comment
Share on other sites

Not sure if this will help, but sometimes one card on my M18X seems to get stuck at low clocks while in game, firing up trixx and adding an OC (Even if it's like 5mhz) fixes it.
- Man, I'm not familiar with OC tools, can you provide a link for trixx?
Actually the same also happens to me. The secondary GPU runs a little hotter. The difference isn't that big, on the full load I think GPU0 is ~72 C, GPU1 around 75-77C. This was every since I repasted, but I actually haven't checked before.

In your case however, it seems GPU1 was running at higher clocks for whatever reason. In my case they both ran at about 100% utilization with 700/1000MHz stock clocks.

- unreal, I was reffering to the GPU clocks, not the temps, my cards allways run different by a few degrees... even after re-pasting.
it's weird indeed, iirc this shouldn't be happening.

Are you running the latest version of HWiNFO (although i doubt the problem is there it's better to always run the latest) ? Try running furmark for a couple of minutes and see how the clocks are then ?

- Yes double checked also tried the 32-bit version and it gave me the same results. Also if I do not open GPU-Z before launching HWInfo the monitors on the second GPU do not show... and GPU-Z states stock clocks for both my cards 700Mhz for GPU and 1000Mhz for Memory... strange indeed, could this be driver related? I'm running AMDs 12.4
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use. We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.