Jump to content

[BENCHMARKS] The OFFICIAL Thread

Brian

By Brian
in Clevo

 Share

Recommended Posts

  • Replies 847
  • Created
  • Last Reply

Top Posters In This Topic

Popular Days

Top Posters In This Topic

Popular Days

Popular Posts

Vitaliy Jungle
Vitaliy Jungle

Thank you guys for inviting me to join your community! I am glad to be part of it.

Khenglish
Khenglish

So I name my card 980(m?). I got 1.2V to run. Here's 1466/6960 3dm11:   http://www.3dmark.com/3dm11/11427346   And firestrike same clocks:   http://www.3dmark.com/3dm/134

coolane
coolane

My P870DMG cooling mod: Added: Two short heat pipes on master; Two long heat pipes on slave; Two short heat pipes on CPU; Bunch of small heat sinks on top;  Glued with ar

Posted Images

Guest

Guest

Posted
Posted

Here's something for the Haswell owner's to aim for...

New XTU World Record for 4930MX - According to HWiNFO64, this benchmark drew 159W for CPU Package Power.

ekq2a1Ph.jpg

CgUOVoPh.jpg

I thought testing IC Diamond again, just for statistical "good measure" would be fun... wrong! I'm going to have to put the Liquid Ultra back before I do any more extreme overclocking. Nothing else is capable of handling the 4930MX blast furnace's asinine thermals well enough to do this kind of number chasing. That's not a slam against IC Diamond (good stuff), but it is a major body-slam against Haswell.

Link to comment
Share on other sites
Khenglish T|I Addict

Khenglish

Posted
Posted

[MENTION=119]Mr. Fox[/MENTION] have you tried liquid pro? It's more conductive than liquid ultra and doesn't have the chunks in it like ultra, letting you spread it thinner. I find I can get slightly better temps with it, but it is harder to get it to wet properly. I find it is best to apply a thin layer to both the heatsink and the CPU.

Link to comment
Share on other sites
Khenglish T|I Addict

Khenglish

Posted
Posted

Cotton buds are no good since they leave behind lint. The liquid ultra paintbrush also isn't very good. What I do is wrap kitchen plastic wrap around my finger and spread it with my finger just barely touching with no pressure.

Link to comment
Share on other sites
Guest

Guest

Posted
Posted
@Mr. Fox have you tried liquid pro? It's more conductive than liquid ultra and doesn't have the chunks in it like ultra, letting you spread it thinner. I find I can get slightly better temps with it, but it is harder to get it to wet properly. I find it is best to apply a thin layer to both the heatsink and the CPU.

I have not tried it yet. I apply Liquid Ultra to both surfaces as well. Applying to one surface is not effective. I haven't encounter the chunks you mention. The Liquid Ultra I have used paints on nice and smooth with the included brush. I have heard the Liquid Pro is difficult to remove if repasting is necessary. Have you found that to be the case? (Liquid Ultra is a breeze to clean up.)

I have used it for 5 months on the Alienware 18/4930MX with superior results and no evidence of degradation in cooling ability. That's amazing considering the machine runs a Haswell blast furnace CPU at a minimum of 4.3GHz with C-States disabled at all times. I am having the same experience with Liquid Ultra with the M18xR2/3920XM running at the same clock speeds for daily use.

I took apart the Alienware 18 on Sunday to inspect the condition of the TIM. Here is a series of photos showing my application of Liquid Ultra. The last 7 images are what it looked like on Sunday. This stuff is beyond amazing.

Link to comment
Share on other sites
Guest

Guest

Posted
Posted

Probably because it is applied very thin... essentially "painted on," so by having both surfaces wet with the liquid metal paste it ensures the bond is complete. I applied to the CPU die only on my first application and the temps were not effectively controlled. I used very little material because I did not want any of excess, so it probably was not enough paste to work well. Applying a very thin coating to both surfaces produces amazing results. By painting a thin layer on each surface it also minimizes the risk of getting the liquid metal on something you don't want it to get on. (Being it is electrically conductive warrants appropriate caution.) You might get equally good results apply a heavier coat to one surface, but I think the risk of having excess paste squish out and cause problems could be greater. The key to getting good results and not making a mess is to use only enough to get the job done effectively, so if both surfaces are wet when they come together, with no excess, that has worked really well for me. I used this application application method for CPU and GPU on the Alienware 18, M18xR2 and an XPS desktop CPU, all with excellent results.

Something I noticed when I disassembled the M18xR2 and the Alienware 18 for inspection is that the paste that is exposed to air hardens and forms a seal around the perimeter of the die, and the paste that is not exposed to air remains wet. The same is true for normal thermal pastes to some extent, but I think this is one reason the Liquid Ultra durability is so great. On both machines, my "inspection" was for the sake of satisfying my curiosity about what I might find. After 5+ months of heavy overclocking neither machine has shown indications of the Liquid Ultra breaking down and losing its cooling properties. (Photos from both inspections are in the album above.) In a way I wish I would have left them both alone to see how long the Liquid Ultra would last before it needs to be reapplied, but the mere fact that both have gone 5+ months without deterioration is still an amazing feat in itself. With other thermal pastes the most use I could get with extreme overclocking was about 60 days (using IC Diamond)... most thermal pastes I have tried to use (TX-2, AS5, MX-4, NT-H1) are effective for no more that 20 to 30 days of extreme overclocking in my experience.

I still find IC Diamond is excellent for use on my video cards... very effective and durable. But, for a heavily overclocked CPU the liquid metal paste has no match for effectiveness.

Link to comment
Share on other sites
Ethrem T|I Fan

Ethrem

Posted
Posted

You have to sand Liquid Pro off of the heatsink, it doesn't come off easily. It's much more permanent than liquid ultra. I have been investigating both of them and liquid pro sounds like a huge hassle.

Sent from my HTC One_M8 using Tapatalk

Link to comment
Share on other sites
Guest

Guest

Posted
Posted
I always have to sand ultra off my heatsink, but it always wipes cleanly off the die. I have not tried to remove pro yet.
I would be interested to know about the experience removing the Pro. From what I have heard, it's more difficult to clean up. Less than half of what I read turns out to be true, especially with respect to thermal paste experiences. I would trust what you have to say about rather than believe most of what I would read about it.

From what I can tell from my own experience with Ultra only the paste on the heat sink that is exposed to air doesn't come off cleanly. The area that mates with the die and the die remain liquid and are easily removed. I like using the scuff pad on the heat sink as it makes a very nice contact surface, so I would probably use it anyway.

Link to comment
Share on other sites
typarker001 Settling In

typarker001

Posted
Posted

hey im having trouble running 3dmark tests. All tests crash with an error "hardware device removed" and driver crashes aswell. My system seems to be unstable at higher temps. I've flashed the unlocked bios for my M18xR1 and the modded Vbios for an 880M. I have made no changes to any settings.However i know nothing about overclocking. Any help would be appreciated. Sorry wrong thread

Link to comment
Share on other sites
Guest

Guest

Posted
Posted
hey im having trouble running 3dmark tests. All tests crash with an error "hardware device removed" and driver crashes aswell. My system seems to be unstable at higher temps. I've flashed the unlocked bios for my M18xR1 and the modded Vbios for an 880M. I have made no changes to any settings.However i know nothing about overclocking. Any help would be appreciated. Sorry wrong thread

More often than not that crash is caused by a couple of possibilities...

Most frequently something is stealing focus from the benchmark as it is launching or switching between tests, or the voltage is too low. Often the voltage being just a shade too low for the overclock will cause an Windows OS lock-up, but it can cause this issue as well.

By "stealing focus" I mean something is launching in the background, including even a system tray balloon notification, which is interrupting the benchmark. You will need to figure out what is disrupting the benchmark. If you are using RTSS with OSD and the settings selected in RTSS are not compatible it can cause this as well. Running RTSS with OSD on DX8 and DX9 (3DMark05, 3DMark06) will cause the benchmarks to crash in the same way.

Link to comment
Share on other sites
  • 2 weeks later...

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.

×
 Share
Go to topic listing


Copyright (c) 2023 Tech|Inferno Media. All rights reserved.

×
×
  • 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.

 Accept Cookies  Reject Cookies