The Notebook Review forums were hosted by TechTarget, who shut down them down on January 31, 2022. This static read-only archive was pulled by NBR forum users between January 20 and January 31, 2022, in an effort to make sure that the valuable technical information that had been posted on the forums is preserved. For current discussions, many NBR forum users moved over to NotebookTalk.net after the shutdown.
Problems? See this thread at archive.org.

    best thermal paste for cpu/gpu

    Discussion in 'Hardware Components and Aftermarket Upgrades' started by hnijhar, Jul 6, 2012.

  1. hnijhar

    hnijhar Notebook Guru

    Reputations:
    0
    Messages:
    55
    Likes Received:
    0
    Trophy Points:
    15
    im looking to change my Asus G73JH-B1 thermal paste due to temps.
    what is the BEST thermal paste avaialable for cpu and gpu.
    thanks.
    i7 740qm idles at low 50's, gaming it goes to low 70's, and at 100 perfect cpu usage, it goes to high 80's-low 90's.
    5870M- idles at 63 degress, and while gaming goes to low 80's.

    Are these temps bad?
     
  2. Qing Dao

    Qing Dao Notebook Deity

    Reputations:
    1,600
    Messages:
    1,771
    Likes Received:
    304
    Trophy Points:
    101
    If the computer is not throttling the speed of the CPU or GPU due to high temperatures, then no, the temperatures are not bad.
     
  3. hnijhar

    hnijhar Notebook Guru

    Reputations:
    0
    Messages:
    55
    Likes Received:
    0
    Trophy Points:
    15
    whats throttling, and i just want to be on the safe side!
     
  4. sgogeta4

    sgogeta4 Notebook Nobel Laureate

    Reputations:
    2,389
    Messages:
    10,552
    Likes Received:
    7
    Trophy Points:
    456
    Your CPU temps seem a tad bit high but overall they're ok. Throttling means that when you CPU reaches a certain temperature, it limits the CPU speed. You can easily see this when you run a stress test and you see the max clock speed being less than what should be the max clock speed.

    As per your question, there is no single best paste, other wise no one would ask the question... there are several pastes that perform within a few degrees of each other. Check out a few reviews that have many different pastes, and look for the ones that are consistently in the top 5 at the store of your choice since you might not be able to easily get some of them. I haven't checked names in a while but I recall Arctic Cooling MX, Tuniq TX, Shin-Etsu X23, IC Diamond 7, as all being pretty good.

    Don't get hung up on which one is number 1 and number 2 since there are many variables that can make 1 better than 2 in situation X but 2 better than 1 in situation Y. The most important factor is not the paste itself but the technique that you use to apply it. You can have the best paste in the world but if you do a crap job applying it, it will perform poorly. Once you decide on a paste, youtube methods of applying. I typically do a small ball or line depending on the shape of the die (CPU/GPU).
     
  5. hnijhar

    hnijhar Notebook Guru

    Reputations:
    0
    Messages:
    55
    Likes Received:
    0
    Trophy Points:
    15
    kk. yeah thanks for the reply, i blew air in the exhaust fans and ill record temps again.
     
  6. hnijhar

    hnijhar Notebook Guru

    Reputations:
    0
    Messages:
    55
    Likes Received:
    0
    Trophy Points:
    15
    my new temps after blowing out the exhaust fans are: 50 degrees on idle, 67-69 while gaming for the cpu.
    the gpu: while gaming is 77 usually, and while idle its 63.
    im just gonna take it to a shop. is $37.50 a good price for someone to open my laptop and blow everything and reapply thermal grase to bote cpu/gpu.?
     
  7. __-_-_-__

    __-_-_-__ God

    Reputations:
    337
    Messages:
    1,864
    Likes Received:
    10
    Trophy Points:
    56
    thermal pastes are not a miracle. I really don't known why people think otherwise.
    if you buy the best TIM in the world you'll probably not even get less then 5ºC temp drop and that is if you are very lucky.

    anyway, the best is indigo extreme. I don't recomend it.
     
  8. Qing Dao

    Qing Dao Notebook Deity

    Reputations:
    1,600
    Messages:
    1,771
    Likes Received:
    304
    Trophy Points:
    101
    No, for a few of reasons.

    1. When they say, "reapply thermal grease," do they mean just white silicone grease? That is probably what is there right now.

    2. You can do it yourself for the cost of the thermal paste.

    3. If the CPU or GPU are not throttling and your system is not shutting off due to heat issues, you aren't going to see a single difference.
     
  9. maverick1989

    maverick1989 Notebook Deity

    Reputations:
    332
    Messages:
    1,562
    Likes Received:
    22
    Trophy Points:
    56
    Qing Dao is right. You should first check to see if your CPU is throttling or not.
     
  10. __-_-_-__

    __-_-_-__ God

    Reputations:
    337
    Messages:
    1,864
    Likes Received:
    10
    Trophy Points:
    56
    $37.50 for 5minutes of work, removing some screws and applying a paste?
    I can't understand why anyone would pay that much.
     
  11. Thaenatos

    Thaenatos Zero Cool

    Reputations:
    1,581
    Messages:
    5,346
    Likes Received:
    126
    Trophy Points:
    231
    Maybe because they dont have enough know how or confidence to do it themselves. Either way its not a terrible price if they are a qualified professional using good paste. Granted I too think its crazy to pay that for a re pasting, but thats because I have always done that myself. But if youre lacking the knowledge or confidence I dont think its a terrible rip off if they use quality paste.
     
  12. hnijhar

    hnijhar Notebook Guru

    Reputations:
    0
    Messages:
    55
    Likes Received:
    0
    Trophy Points:
    15
    and blowing out all the dust?
     
  13. TANWare

    TANWare Just This Side of Senile, I think. Super Moderator

    Reputations:
    2,548
    Messages:
    9,585
    Likes Received:
    4,997
    Trophy Points:
    431
    If you are removing the HS it is best to clean the system of all dust etc but also give the HS a bath. Oils and other debrise build up over time. you need to get these off as welll. Otherwise efficiency will be down and it will collect dust just that much quicker.
     
  14. Crimsoned

    Crimsoned Notebook Deity

    Reputations:
    268
    Messages:
    1,396
    Likes Received:
    3
    Trophy Points:
    56
    5 minute job? That's a joke right? Even with easy access heatsink laptops, any tech worth his salt would run tests on thermal before and afterwards.

    We charge approximately $85 for a thermal application. It obviously goes down if the laptop has easy access CPU heatsink (lowest is $65 for thermal job only, if bundled with other services it can go lower than $65).

    This includes:
    -Testing of thermals prior to thermal job if possible (temps 90c+ are considered critical, tests are skipped and thermal job is done asap).
    -Disassembly, reassembly.
    -Cleaning of components if needed
    -AS5 for the most part, or Shin Etsu 7783D. It's the two thermals that have withstood the test of times.
    -Thermal testing, including: vents/CPU/RAM/HDD using thermal diodes and using software like Hwmonitor on top of that. Idle/load.
    - It will also include reapplications if thermals aren't bettered by at least 5c.

    Typical results:
    -10c or better drop in temperature compared to: stock TIM application, and most competitor custom jobs.

    I personally have done approximately 90 or so reapplications for testing on a single system. Approximately 200+ applications on different laptops JUST for testing.
    I've blown quite a bit of money on TIM (heck just a few days ago, I blew $20 in Shin Etsu repeating some application tests because there were some aspects of shin etsu application that were bothering me namely in laptops where PSI is much lower. I only test on the thermal greases that I use, so it can get costly but there's no alternative if you want accurate results.

    You're paying for experience, and performance with us. The time we've spent testing and the knowledge we have (don't mean to brag) have yet to show up on our competitors' results This means we've yet to come across an optimal thermal job from manufacturer nor custom nor serviced desktop/laptops.
     
  15. Qing Dao

    Qing Dao Notebook Deity

    Reputations:
    1,600
    Messages:
    1,771
    Likes Received:
    304
    Trophy Points:
    101
    There is a sucker born every minute. You can talk about all the vent readings and repastings you want, but that is just an excuse to milk so much money out of poor saps who don't know any better. You are just trying to increase the perceived value of your services even though the actual value to the customer is no greater than just cleaning out the dust and properly applying a good TIM. At best, a 5 minute job. At worst, a bit over an hour.

    That doesn't even sound like it is within the realm of possibility.
     
  16. Crimsoned

    Crimsoned Notebook Deity

    Reputations:
    268
    Messages:
    1,396
    Likes Received:
    3
    Trophy Points:
    56
    I had written a long paragraph. But realized it was futile.
    You post is insulting, and quite frankly ignorant. 5 minute job? You're gambling with a customers PC by not allowing proper time for testing pre/after thermal jobs. I won't even get to it, it's obvious you probably haven't done this very much if at all.

    If you want a $15 quick job, go somewhere else we don't do half-hearted jobs, and hope for the best. We don't do temporary fixes.
    Jobs like that tend to get ugly with customers whom out of desperation for a cheap repair, think you'll give them the world while they pay you less than minimum wage. We'd rather charge the time it takes to test the system properly, do the thermal job properly and have them leave with a trouble free experience that won't come and bite them in the...

    Real life example of charging little and offering a repair to a very serious problem (not related to overheating): Warning, it's kind of long.

    We had a customer come in requesting their laptop be fixed. The laptop lcd was completely snapped off the assembly. We asked 3 times, what had happened. Each time the customer replied: Nothing, it just broke from use. First off hinges don't break off in that way from use on the HP DV6000 she brought in. Sure they will loosen from the base at times. Sure maybe the brass bolsters on the back LCD cover will break off. But the hinges won't snap off like twigs and the plastics wouldn't be broken off the way they were.
    He requests the laptop be fixed, he says he doesn't want to pay more than $25, and needs it that very day. We explain for $25 there's no way we can replace the hinges/plastics, etc. My father explains the damage on the notebook is not matching regular use and it looks like someone had either thrown the laptop, stepped on it, or dropped it. The customer says no, that it happened from regular use...
    The customer says he really needs the laptop for his wife because she needed it for school urgently.
    My father decides to do the man a favor (normally we wouldn't take this type of work in), and tells him he'll weld the hinges for now while her wife uses it for school. However to come back asap to get the hinges fixed, the $25 he pays now would count towards the future repair. The customer agrees.
    Since a diagnostics was refused, and the timeframe was short. My father welded the hinges best he could.
    Customer was contacted upon completion and picked up the system satisfied and very happy with the result (tbh even I was impressed it open/closed properly and looked pretty decent compared to before).
    Customer's wife calls 2 hours later claiming the notebook doesn't work. My father explains there was pretty serious damage on the notebook, and a diagnostics was offered at the time of drop off.
    The wife starts getting loud saying that we need to repair the laptop so that it will power on (not what we were paid to do). We tell her our job had no impact with the hinge repair. However if she wanted to she could bring the laptop and we could do a full diagnostics. She starts saying she can't and that we already know whats wrong. We're tech people. :rolleyes: Apparently by touching a computer, we should know immediately what is wrong with it.
    We offer a free diagnostics, she refuses and keeps blaming us telling us to replace her laptop right away. We keep explaining several times our job=did not require us to disassemble the laptop, period.

    She agrees to bring the laptop... Except she shows up with her, her husband, and 2 gangsters (one tattoed with gang stuff, and the other looking like he's ready to go fight in a club).
    My father maintained his cool and tried to explain what the job we did was, and that we needed to do a diagnostics on the laptop. She keeps demanding the laptop be repaired or for it to be replaced.
    She begins to curse, and one of the guys nearly put his hands on my father.

    I immediately called the cops as my father asked the two other people not involved to wait outside. I informed him several times the police were on their way.
    Police came in, asked what the problem was. Took the customer, his wife, and 2 rather aggressive people outside and asked the 2 gangsters to please stay out of it or they would be arrested. They informed the customer there was nothing they could do and advised her that if we really had wronged her they could file a lawsuit.

    The police once hearing our story told us not to worry and asked if we had any video surveillance (we didnt, but the building we were in certainly did, including audio).
    They took us to court. The judge after seeing the laptop, some hinges we brought in for example (pulled from the different but same model laptop) asked how the laptop came to be damage.
    The customer said her laptop had frozen. So she took out the charger, and tried to take out the battery. After a while she managed to get it out. However when she was trying to put it back, it would not go in.
    She demonstrated.
    So she placed the laptop upside down with the LCD open to form a tent. Then proceeded to hammer the battery with her hand.
    The judge soon, the ruling was in our favor.
    1. Her husband did not the test system before drop off nor during the pick up. This was alone reason enough for her to loose her case.
    2. She (the judge) knows little about computers but the customer's actions were clearly abuse to the laptop. Diagnostics had been offered to her at no charge, and she refused. She mentioned this was something she wouldn't have done had she been the one dealing with the customer.
    3. Did not appreciate her talking when she wasn't asked to, and that just because she was needy (she was working in McDonalds, and going to college)it did not give her the right to sue someone else and expect payment just because. (customer complained many times during the case that she was poor, etc)
     
  17. mambastik

    mambastik Notebook Enthusiast

    Reputations:
    41
    Messages:
    44
    Likes Received:
    0
    Trophy Points:
    15
    I guess it's all a matter of perspective. Of course talking about temp readings and repasting will sound easy; we're on a notebook - hardware subforum.

    "There is a sucker born every minute." You can probably read the same thing when you visit motorsport forums. How many of you would tinker around with your own car vs. taking it to a mechanic? It's just relative.
     
  18. __-_-_-__

    __-_-_-__ God

    Reputations:
    337
    Messages:
    1,864
    Likes Received:
    10
    Trophy Points:
    56
    sorry but the laws of physics say it isn't possible.

    the only reason why that should happen was if the first paste was very very crappy AND it was very terribly applied. and even that won't give that much differences.

    complete ripoff. it should be illegal. only extremely stupid people would pay that much. 15$ is tottally aceptable. I known because it was exactly the price I charged. and because it was an OK price and because we always did a good job we got hundreds of more clients.
    instead of having 1 client for $85 I had 50 for $15 each.
    It would take an average of 15minutes to change the paste and clean the cooler. sometimes more sometimes less depending on the model. it's $60 per hour plus premium paste cost that in large quantities it was like $1.50 per application.
    in a week we had an average of 70 clients for repaste. that's $945 for 17h30min of work.
     
  19. hnijhar

    hnijhar Notebook Guru

    Reputations:
    0
    Messages:
    55
    Likes Received:
    0
    Trophy Points:
    15

    thanks man, you actually left a proper reply, the other people think that there the worlds smartest people. lol. and i know it doesnt take 5 minutes, 5 minutes maybe for a proffesional but not a newcomer, what if something breaks? rather get it done by a proffesional, thanks for the reply.
     
  20. hnijhar

    hnijhar Notebook Guru

    Reputations:
    0
    Messages:
    55
    Likes Received:
    0
    Trophy Points:
    15
    my local tech said he uses 'stars professional thermal paste' ive never heard of it, is it good? i ordered mysled some ic diamond 7.
     
  21. Mobius 1

    Mobius 1 Notebook Nobel Laureate

    Reputations:
    3,447
    Messages:
    9,069
    Likes Received:
    6,376
    Trophy Points:
    681
    I would go against that thermal paste, because I've never seen the reviews online.

    Stick with your IC Diamond 7 ;)
     
  22. Crimsoned

    Crimsoned Notebook Deity

    Reputations:
    268
    Messages:
    1,396
    Likes Received:
    3
    Trophy Points:
    56
    Sorry you feel that way.
    And I call BS on your story, and your work. It's obvious why you closed or stopped doing those repairs. It wasn't worth while, and your job is just like fast food restaurants. Do it quick, if there's any problems customer can come back.

    If it really took you 15 minutes, you more than likely didn't test the system (quite irresponsible). Meaning your application of TIM=goes untested. You won't know if you hit an air pocket or not. What if there was a crack in the heatsink? Etc. (irresponsible)
    15 minutes sounds like you were a unsuccessful at home tech trying to scrap some lunch money, to be honest. Probably nabbed some memory/hard drives didn't you? ;)

    So far our clients have been happy with our work. Particularly in the assembly/disassembly aspects, as well as turn around times.
    Most came from at home techs, probably like yourself, whom charge very little. But end up: breaking parts, not double checking work, taking days or weeks because you have a full time job elsewhere, and of course stealing.
     
  23. TANWare

    TANWare Just This Side of Senile, I think. Super Moderator

    Reputations:
    2,548
    Messages:
    9,585
    Likes Received:
    4,997
    Trophy Points:
    431
    60-85$ on a system is reasonable for a pro job. Being retentive, as I am, It takes me an hour just to do my daughters wich is a single simple heat pipe. When I do it I pull the HS,clelan the system where I can access to be as if it were new. Then I do the same for the HS. Making sure all those fins are 100% cleaned off of oils etc takes time.

    Then of course before assembly a full visual inspection for cracks, holes oxides or other wear/damage. A dry fit to be sure nothing is warped and then pasting and being sure of proper preasure etc is better than a 5 minute task in itself.

    Now I'll agree for a production quality installation it can be done fast. When a pro is doing a custom high performance repaste it ends up costing a bit more. Is it over kill, maybe but that is part of the free market. We all get to pay for what we want!
     
  24. Crimsoned

    Crimsoned Notebook Deity

    Reputations:
    268
    Messages:
    1,396
    Likes Received:
    3
    Trophy Points:
    56
    @Tanware

    Thanks for posting a proper response. With proper methodology.
    Yeah cleaning fins is a pesky job but it has to be done.

    I really doubt Qing or -__- have done very stringent testing.
     
  25. Quanger

    Quanger Notebook Evangelist

    Reputations:
    42
    Messages:
    406
    Likes Received:
    2
    Trophy Points:
    31
    I've used both ic7 and as5. I really don't notice much of a difference. I would repaste AS5 yearly and am told IC7 last longer so we'll see.
     
  26. __-_-_-__

    __-_-_-__ God

    Reputations:
    337
    Messages:
    1,864
    Likes Received:
    10
    Trophy Points:
    56
    it doesn't take hours to test a system. to load a temp monitoring program and do a burn test is just 2 clicks in a bat file and pen drive exe. then the notebook does the job on it's own and give us the results. no need to keep staring at the monitor for hours doing nothing waiting for the test to end.
    all notebooks were tested and results were presented to the client. always.
    your accusations are at least stupid. and trying to promote your business on an internet forum is a sign of failure.
    sure I don't work there anymore. I decided to change business for something more flexible. I became a real estate investor. I sold 49% of the shares to others and now I just collect the profits. The company turned into a franchise and it oppened several other stores. clients are more then happy. actually other stores just stopped offering those services because they couldn't compete with such quality and good price. now they just sell hardware nothing more.
    claiming -10ºC is a lie and everyone that knowns just a bit about pc's known you are full of bs.
     
  27. __-_-_-__

    __-_-_-__ God

    Reputations:
    337
    Messages:
    1,864
    Likes Received:
    10
    Trophy Points:
    56
    stick with something that performs better and is cheaper then ICD7.
    never heard of that "start" thermal paste but any paste won't give more then 5ºC compared to premium ones.
     
  28. RainMan_

    RainMan_ Notebook Evangelist

    Reputations:
    180
    Messages:
    396
    Likes Received:
    11
    Trophy Points:
    31
  29. Crimsoned

    Crimsoned Notebook Deity

    Reputations:
    268
    Messages:
    1,396
    Likes Received:
    3
    Trophy Points:
    56
    About 2-3 hours~ for testing. 1 hour pre-testing.
    We don't stick to software monitoring software only due to inaccuracies. Thats why we load temp diodes on components this is used to cross reference as well as give us an idea of thermal dissipation on the notebook. We're a bit more stringent then you, obviously.

    Trying to promote my business? No hardly so. Just providing a real world example of pricing as well as real world practices and methodology for my business.
    I've not shared, nor will I share my business address, or phone number, or any of the sort in threads, nor private messages (despite being asked in another forum, OCN. I refused to give information).
    Please feel free to look at my profile/threads/posts. If you want you can have access to my private message. Not once have I promoted or obtained business from NBR, it's not what I am here for.

    If you say so. Personally I am calling BS on your work.

    I am glad you are successful in this economy. I am very happy to hear that. Hope you remain profitable and happy for many years to come.

    10c degree changes from stock TIM applications (From manufacturers) or bad applications (pretty much all local tech shops around me) are a very real reality for our thermal jobs. Does it happen every single time, absolutely not. But most of the times, we're looking at a 10c degree change.

    Would you mind posting a picture of a TIM application you've done?

    Here's part of the test I did regardling the Shin Etsu test that raised some concerns with me.
    These are 3 of 9 slides. 6 others were done soon afterwards.
    The three, are:
    B: Bead in middle
    L: Line in middle
    X: X lines
    [​IMG]

    We can see the bead method allowed for more even spread. And due to the method, takes less time to apply as well as allows for better control of the amount of time.
    The line method unfortunately ends up in an uneven oval shape spread, and is a tad by more difficult to control. But possible, just takes a little more time.
    X method is the more difficult. And can take a long time to apply the proper amount of time. Razors are necessary here. The TIM is spread unevenly most of the time unfortunately.
    Pressure applied: approximately 15 pounds, then 20 pounds for 15 seconds.
    The slides were heated then allowed to cool for a little before the pictures were taken to produce a rather rough representation of a thermal cycle.

    I apologize for the quality of the picture. It's poor.
    I would have used my photography halogen lamps but those things get hot and I don't feel like spending a few hundred on some budget CFL or LED photograph lights.
    For now I switched to some daylight 100w equi. CFL lights. It gives a much better result.
     
  30. hnijhar

    hnijhar Notebook Guru

    Reputations:
    0
    Messages:
    55
    Likes Received:
    0
    Trophy Points:
    15
    whats the best way to apply ic diamond? clean off the old paste and put a 5mm bead of paste in the center, and set the heatsink on and screw it down? any special instructions? and if i put excess, doesnt it not do anything because its not conductive? so excess is ok, but to less it bad. and how big is a 5mm bead?
     
  31. ellalan

    ellalan Notebook Deity

    Reputations:
    336
    Messages:
    1,262
    Likes Received:
    82
    Trophy Points:
    66
    Any thermal paste would do fine but make sure it's a brand name like AS5 or ICD 7, I would prefer not to pay $85 for repasting because it's more than what my CPU cost and prefer to do it for free by myself.
     
  32. misterhobbs

    misterhobbs Notebook Evangelist

    Reputations:
    715
    Messages:
    591
    Likes Received:
    9
    Trophy Points:
    31
    I recently ran an experiment comparing a few different thermal compounds and there really wasn't that much of a difference when applied correctly. They were all also relatively to apply.