PDA

View Full Version : Should I RMA G751JY help!



Gurendaiza
04-27-2015, 10:11 AM
Hello , ROG fans and owners,

I just bought my new G751JY yesterday and i have been using it for the last 6 hours . I'v used many ROG laptops before but i noticed the g751jy suffers a problem with high temps and fan not going above 3k RPM even on high load (where the fan actually needs to cool the internals)

i'v stress tested the cpu and gpu the same time using Prime95 and FurMark as someone suggested on this forum to get the temps and fan speed at full load.

i will be posting pictures of both idle and full load taken while the stress test running .

I'am reaching a max of 93c per core on full load which is above 92c and its funny that the fan is not pushing harder to go over 3k to cool off the cpu!

please help me out in making a decision whether to RMA this laptop or just stick to what i have , btw iam using 980m and bios updated to 205.

i really like the rog g751jy so far iam just worried that this will cause issues in the future. I fried rog laptop gpus before playing ffxiv and this is acutally the main reason i'am worried.

idle
48656
max load
48657

regards

QC_AGENT
04-27-2015, 01:19 PM
If the temp is your unique concern and everything else is fine (the screen, the sound, webcam, usb ports) I would rater send it to an authorised tech shop for a warranty thermal repaste. If there is none in your location, then, go RMA.

Personnally, my WebCam was faulty and I chosen a warranty repair. Even though I bought it from an Online store (NewEgg), a local dealer was kind enough to take care of it since he was an authorised seller.

The risk of sending it RMA is to get something worst in exchange.

Gurendaiza
04-27-2015, 01:50 PM
well thanks for your reply, i have a 7 days return thingy so I can still do that but first wanna make sure if this is normal or not.
i mean is this what everyone complaing about...

QC_AGENT
04-27-2015, 02:29 PM
I believe yours to be hot based on what I read before. Howerver, I did not even tested mine yet, since it feels cool to me. I'll install CPUID and run it tonight and let you know the results. Since they had to change the Motherboard to fix my webcam issue, I expect the Thermal paste to have been done properly by the tech who worked the issue. Lets see what happens.

Gurendaiza
04-27-2015, 02:31 PM
thanks man for helping out.. i really like the laptop but I don't want faulty one.

TeMagic
04-27-2015, 02:50 PM
Send it in for repair is my recommendation. It probably is a loose heatsink and/or thermal paste problem. Not a biggie but your temperatures are a bit high. Should be a small thing for the reseller to correct without messing anything else up in the process.

Clintlgm
04-27-2015, 03:03 PM
Your within the return period I would just return it and get another one or go to another retailer for another one. You never know what is going to happen at the repair center. You'll have a good chance of getting a better one with a replacement from the seller.

Gurendaiza
04-27-2015, 04:16 PM
thanks all for your time and effort, the reason iam asking is because iv sewn alot of complaints on the forum regarding this issue but no one mentioned RMAing or replacing cause of the chances of getting a nastier one I will see what others have to say about it or i might just refund it to get a desktop and save at least 600$

QC_AGENT
04-27-2015, 05:50 PM
Its not that bad. Keep in mind that only people having issue are complaining. The satisfied custommers do not show up here. Most Laptop are fine

Gurendaiza
04-27-2015, 06:51 PM
i just played csgo which is considered light gaming ,

cpu max = 86c
gpu max = 69c

i called them up , i will probably get an exchange tomorrow or a new desktop ,, but mostly a new rog!

QC_AGENT
04-28-2015, 12:49 AM
Ok, just tested with COD Advance Warfare for 30 min and I got 77c for CPU and 65c for GPU.

48676

Yours is definitely hot. For your info, my motherboard has been changed (therefore repasted) when I sent it for a webcam issue.

hmscott
04-28-2015, 02:04 AM
Hello , ROG fans and owners,
I just bought my new G751JY yesterday and i have been using it for the last 6 hours . I'v used many ROG laptops before but i noticed the g751jy suffers a problem with high temps and fan not going above 3k RPM even on high load (where the fan actually needs to cool the internals)
i'v stress tested the cpu and gpu the same time using Prime95 and FurMark as someone suggested on this forum to get the temps and fan speed at full load.
i will be posting pictures of both idle and full load taken while the stress test running .
I'am reaching a max of 93c per core on full load which is above 92c and its funny that the fan is not pushing harder to go over 3k to cool off the cpu!
please help me out in making a decision whether to RMA this laptop or just stick to what i have , btw iam using 980m and bios updated to 205.
i really like the rog g751jy so far iam just worried that this will cause issues in the future. I fried rog laptop gpus before playing ffxiv and this is acutally the main reason i'am worried.
regards

Gurendaiza, those temps are exactly as expected for a G751JY running Prime95 and Furmark at the same time :)

There is no need, or improvement to be gotten from re-pasting, returning, or RMA'ing - those temps are just right.

You are running the CPU/GPU 100% simultaneously and that is why you are pressing up against the thermal throttling limit - the CPU/GPU share heatpipes.

When you are gaming you won't hit anywhere near that, so not to worry.

If you are going to run rendering or CUDA compute batch jobs, and you hit some Thermal Throttling, you can detune the Core 3/4 of the CPU a couple of multiplier notches to stop it under 4 Core load. But, your temps aren't bad enough to worry.

You could run with hwinfo64 sensors only mode, and look at the Thermal Throttling lines to see if any of the No's become yes'es.

You have a keeper, have fun! :)

hmscott
04-28-2015, 02:07 AM
Ok, just tested with COD Advance Warfare for 30 min and I got 77c for CPU and 65c for GPU.

Yours is definitely hot. For your info, my motherboard has been changed (therefore repasted) when I sent it for a webcam issue.

QC_AGENT, your test gaming session isn't heavy duty enough to load up the CPU/GPU to see highest load/temps like Gurendaiza.

Try the test that Gurendaiza did, run both Prime95 (small fft's) and Furmark 1080p 15 minute Burn-in test - you only need to run for 5 minutes, and then you will see similar temperatures to Gurendaiza's results.

arrtaas
04-28-2015, 05:05 AM
I don't see why would you be running furmark and prime 95 at once, furmark gives you like 1 fps per second in this case obviously gpu is not used 100%. Run them seperately. I ran Prime 95 small fft for 15 minutes and max temp was 74C (room temperature 20.5C, notebook cooler as well)

Gurendaiza
04-28-2015, 05:38 AM
thanks again for sharing.. although my question is with this kind of temperatures now what would happen when dust accumulate.. plus csgo temperatures are really high? someone called cs go here light gaming lol.

Gurendaiza
04-28-2015, 05:40 AM
I don't see why would you be running furmark and prime 95 at once, furmark gives you like 1 fps per second in this case obviously gpu is not used 100%. Run them seperately. I ran Prime 95 small fft for 15 minutes and max temp was 74C (room temperature 20.5C, notebook cooler as well)


you need to run both to see how heat from gpu and the cpu correspond to each other.

Gurendaiza
04-28-2015, 06:01 AM
Gurendaiza, those temps are exactly as expected for a G751JY running Prime95 and Furmark at the same time :)

There is no need, or improvement to be gotten from re-pasting, returning, or RMA'ing - those temps are just right.

You are running the CPU/GPU 100% simultaneously and that is why you are pressing up against the thermal throttling limit - the CPU/GPU share heatpipes.

When you are gaming you won't hit anywhere near that, so not to worry.

If you are going to run rendering or CUDA compute batch jobs, and you hit some Thermal Throttling, you can detune the Core 3/4 of the CPU a couple of multiplier notches to stop it under 4 Core load. But, your temps aren't bad enough to worry.

You could run with hwinfo64 sensors only mode, and look at the Thermal Throttling lines to see if any of the No's become yes'es.

You have a keeper, have fun! :)

hello bro , i tried what you suggested and got to yes(s) lol Core # 0 and 2 Thermal Throttling >_< ...

hmscott
04-28-2015, 08:21 AM
hello bro , i tried what you suggested and got to yes(s) lol Core # 0 and 2 Thermal Throttling >_< ...

Gurendaiza, ok, so that isn't unexpected either. Many do see instances of Thermal Throttling under that load. Again, that is a really unusual load.

If you turn on hwinfo64 logging, and export to a file, you need to convert the ;'s to ,'s and load it into Excel to see *how often* it hits Thermal Throttling. Most important when monitoring real gaming or work you do.

Most times the Max Temp and Thermal Throttling only show up as an occasional peak with lower temps in between.

Only if your work benchmarks in the same thermal range, and you are Thermal Throttling doing gaming or real work would I recommend reducing the Core 3/4 multiplier to stop the Thermal Throttling.

Using Intel Extreme Tuning Utility you can raise the Multiplier for Core 1/2 to the maximum, and reduce Core 3/4 from max to 30x/30x to start, then re-run the tests / monitoring and see if that was enough / too much reduction, and increase/decrease again until you get the right multiplier dialed in for your work.

For most people, you can raise the multiplier for Cores 1/2/3/4 and Cache to maximum all the time, because the Prime95 (small fft's) and Furmark simultaneous is a load 99% of people won't see in real use.

But, it's good to know the limits of your hardware :)

I would recommend loading up your favorite heavy duty game and running for a while with the hwinfo64 logging on and see what temps you get under normal gaming conditions.

Please let us know what you find :)

hmscott
04-28-2015, 08:34 AM
thanks again for sharing.. although my question is with this kind of temperatures now what would happen when dust accumulate.. plus csgo temperatures are really high? someone called cs go here light gaming lol.

Gurendaiza, CS:GO can really kick up the CPU utilization, but it doesn't load up the GPU, so the cooling system should shove the heat out both the CPU and GPU cooling system better than an independent cooler design - that's where the G751 shared heatpipe cooling wins :)

Limiting FPS on CS:GO can be problematic, so a trick to use is to reduce the number of Cores CS:GO is using. Check out this method to use Affinity to stop using Hyperthreaded cores:

G750JM high temperature on CS:GO
https://www.youtube.com/watch?v=PVl8Eupbr_E&feature=player_detailpage#t=51

As it turns out, most games are happy enough with 4 cores, without Hyper-threaded cores or more than 4 real cores, and get the same FPS under all but less than 2 core CPU's

In other words, it might not be a bad idea to try running all games with only 4 Cores, and disabling the 4 Hyper-threaded cores for all gaming...

CPU Cores for Gaming: How many do you need? - Q1 2015 Update
https://www.youtube.com/watch?v=PVl8Eupbr_E

While exploring XTU, give undervolting a shot too, you can reduce temps by reducing the power used by the CPU.

Here is a sample setting for -45mV and maximum multipliers for cores/cache, what I am using with a 4700HQ:

48699

Gurendaiza
04-28-2015, 08:43 AM
Gurendaiza, ok, so that isn't unexpected either. Many do see instances of Thermal Throttling under that load. Again, that is a really unusual load.

If you turn on hwinfo64 logging, and export to a file, you need to convert the ;'s to ,'s and load it into Excel to see *how often* it hits Thermal Throttling. Most important when monitoring real gaming or work you do.

Most times the Max Temp and Thermal Throttling only show up as an occasional peak with lower temps in between.

Only if your work benchmarks in the same thermal range, and you are Thermal Throttling doing gaming or real work would I recommend reducing the Core 3/4 multiplier to stop the Thermal Throttling.

Using Intel Extreme Tuning Utility you can raise the Multiplier for Core 1/2 to the maximum, and reduce Core 3/4 from max to 30x/30x to start, then re-run the tests / monitoring and see if that was enough / too much reduction, and increase/decrease again until you get the right multiplier dialed in for your work.

For most people, you can raise the multiplier for Cores 1/2/3/4 and Cache to maximum all the time, because the Prime95 (small fft's) and Furmark simultaneous is a load 99% of people won't see in real use.

But, it's good to know the limits of your hardware :)

I would recommend loading up your favorite heavy duty game and running for a while with the hwinfo64 logging on and see what temps you get under normal gaming conditions.

Please let us know what you find :)

hello again bro, i was returning the pc and your reply is the only thing that changed my mind inwill just use the laptop and see what happens.. tho undervolting or underclocking won't effect performance?

hmscott
04-28-2015, 08:53 AM
hello again bro, i was returning the pc and your reply is the only thing that changed my mind inwill just use the laptop and see what happens.. tho undervolting or underclocking won't effect performance?

Gurendaiza, the undervolting doesn't affect the performance - make it too low and the CPU will BSOD Windows, so find that point and back off a bit and try again. Once you find the undervolt that works at heavy CPU load, let the laptop run idle for about 30 minutes or so, if it doesn't BSOD you have found the sweet spot.

My first G750JW was stable at -125mV !! No such luck for me since, though others have been lucky, so it's worth a try. My JX did -75mV stable, and my JH only did -20mV out of the box, but after 17 months it's stable at -45mV.

The underclocking of Cores 3/4 will reduce performance only for 3 or 4 simultaneous threads, most of the time your computer will be hitting 1/2 Cores so raising their Multiplier gives you better performance most of the time, and only slightly reduced performance of only Cores 3/4 occasionally - you won't notice :)

But, monitor your normal usage like you did for the Prime95 (small fft's) + Furmark 1080p 15 minute burn-in, and you will find you won't hit Thermal Throttling or high temps - benchmarks are only to find the limits of the hardware and don't represent typical every day use.

Good call on keeping the laptop, time for some gaming :)

Gurendaiza
04-28-2015, 09:13 AM
Gurendaiza, the undervolting doesn't affect the performance - make it too low and the CPU will BSOD Windows, so find that point and back off a bit and try again. Once you find the undervolt that works at heavy CPU load, let the laptop run idle for about 30 minutes or so, if it doesn't BSOD you have found the sweet spot.

My first G750JW was stable at -125mV !! No such luck for me since, though others have been lucky, so it's worth a try. My JX did -75mV stable, and my JH only did -20mV out of the box, but after 17 months it's stable at -45mV.

The underclocking of Cores 3/4 will reduce performance only for 3 or 4 simultaneous threads, most of the time your computer will be hitting 1/2 Cores so raising their Multiplier gives you better performance most of the time, and only slightly reduced performance of only Cores 3/4 occasionally - you won't notice :)

But, monitor your normal usage like you did for the Prime95 (small fft's) + Furmark 1080p 15 minute burn-in, and you will find you won't hit Thermal Throttling or high temps - benchmarks are only to find the limits of the hardware and don't represent typical every day use.

Good call on keeping the laptop, time for some gaming :)

thanks again i need to find the sweet spot now lol, will google to find guides and will report back with results . thanks again bro. btw i used rog swift monitor for 144hz while playing i hope this is not effecting my temps , and lid is always closed never used the laptop keyboard from day one lol.

hmscott
04-28-2015, 09:24 AM
thanks again i need to find the sweet spot now lol, will google to find guides and will report back with results . thanks again bro. btw i used rog swift monitor for 144hz while playing i hope this is not effecting my temps , and lid is always closed never used the laptop keyboard from day one lol.

Gurendaiza, I don't use the keyboard either, a better wired or wireless external to keep the laptop clean and preserve the keyboard for resale :)

You might want to monitor temps with the lid closed and open, I know the G750 keyboard/top is sealed so there isn't really any point in leaving the lid open, but I have heard the G751 is open all over, so leaving the lid open might help with cooling.

But, your temps look the same as others posting test results - and I don't think any I read were with the lid closed.

It's always worth actually testing, if you don't know, then you just don't know - and a guess is as good as a miss :)

Have fun :)

Gurendaiza
04-28-2015, 09:44 AM
Gurendaiza, I don't use the keyboard either, a better wired or wireless external to keep the laptop clean and preserve the keyboard for resale :)

You might want to monitor temps with the lid closed and open, I know the G750 keyboard/top is sealed so there isn't really any point in leaving the lid open, but I have heard the G751 is open all over, so leaving the lid open might help with cooling.

But, your temps look the same as others posting test results - and I don't think any I read were with the lid closed.

It's always worth actually testing, if you don't know, then you just don't know - and a guess is as good as a miss :)

Have fun :)

iam using your method now to undervolt i went -20mv now test all seems fine i got max of 73c per using xtu stress test i will try to push more and report back with screen shots.

Gurendaiza
04-28-2015, 11:25 AM
ok tried 100mv and 90 but csgo crashed pc restarted . so went to try out 40 and did prime and funmark run and max is now up to 85C for cpu and 70 for gpu , i can see improvements

arrtaas
04-28-2015, 08:34 PM
ok tried 100mv and 90 but csgo crashed pc restarted . so went to try out 40 and did prime and funmark run and max is now up to 85C for cpu and 70 for gpu , i can see improvements

Try -80mv, that is what I use. Very good temp improvements. I also use 36x 35x 34x 34x stable.

Gurendaiza
04-28-2015, 08:38 PM
36x 35x 34x 34x is underclocking or over I really dont want to loss performance.. and - 80mv on both?

arrtaas
04-28-2015, 08:48 PM
Please check this thread of mine out http://rog.asus.com/forum/showthread.php?60244-G751JY-Fan-Settins .

I also had heat issues. I believe the stock thermal paste ASUS uses is garbage or this models need thicker type of thermal compound due to weird gaps (my guess only).

If you could do the same testing I did, maybe we could compare results and see how they compare.

36 35 34 34 is overclocking. Better performance for sure.

hmscott
04-28-2015, 08:52 PM
36x 35x 34x 34x is underclocking or over I really dont want to loss performance.. and - 80mv on both?

Gurendaiza, yes, reduce both CPU/Cache voltage the same amount for now. And, make sure you are selecting negative offsets, it's easy to slide the wrong way into positive offsets that increase voltage - since we can't actually OC - change the clock or increase multipliers beyond the locked range - there is no need for an increase in voltage.

Different models of CPU can have different maximum multipliers - for my 4700HQ the maximums are 36x,35x,34x,34x Cores and 36x Cache, yours might be different - slide the slider up all the way and find out :)

You will find the tipping point of the voltage offset will be about +5mV, so I find the sweet spot by starting high, like -125mv, crash, halving it to -65mV/-75mV, and keep doing that until it is stable, then increase by -10mV at a time until BSOD and then drop back by -5mV.

It doesn't take long to get it right. And, even if the delta in temperature isn't huge, it's all additive, so every little bit contributes to the whole goal of temperature reduction.

arrtaas
04-28-2015, 09:10 PM
What do you guys mean by saying "both" ? I only offset Dynamic CPU Voltage.

hmscott
04-28-2015, 09:47 PM
Please check this thread of mine out http://rog.asus.com/forum/showthread.php?60244-G751JY-Fan-Settins .
I also had heat issues. I believe the stock thermal paste ASUS uses is garbage or this models need thicker type of thermal compound due to weird gaps (my guess only).
If you could do the same testing I did, maybe we could compare results and see how they compare.
36 35 34 34 is overclocking. Better performance for sure.

arrtaas, for most people it is too big an ask to get people to redo their paste, and most first time laptop re-pasters will make enough mistakes feeling their way through it the first time, that it will take a few passes redoing the paste to get it to work long term, so I strongly recommend to not do it for most people.

If their temps are bad enough that they are Thermal Throttling even on normal low stress application, then the problem is bad enough for an RMA to Asus to get them to fix the problem.

If they are finding this out right when they get the laptop, they can take it back to the seller and get another one.

Noone should have to go through the hassle of doing their own re-pasting on a new laptop. You pay good $ for a functioning laptop, and $ for support to fix any problems, so let support do the re-pasting.

There are many times that re-pasting doesn't help enough, and there is something wrong with the hardware - needs replacement - and an RMA is the only thing that can fix it - so now they have to explain to Asus that they took it apart and re-pasted but it didn't help, and can Asus please now fix it for them...

Asus is usually pretty cool about helping out in such cases, but you might get a tech on a bad day that could blame you for messing it up and charge to fix it...

Here are 2 typical threads on the process 2 different people went through, resulting in many re-pasting's before they were happy with the results. These are people adept at working with PC's that eventually got it right, but needed to persist for a long time, many hours, before getting there.

Repasted G750JX, shocking result.
https://rog.asus.com/forum/showthread.php?52198-Repasted-G750JX-shocking-result

Asus g750js thermal paste replacement
https://rog.asus.com/forum/showthread.php?58562-Asus-g750js-thermal-paste-replacement

What I am saying is you need to be prepared for the contingencies before starting, and be willing to follow through over a period of time, longer than you might think.

I am thinking of trying Gelid GC Extreme, Coollaboratory Liquid Ultra, or IC Diamond 24k next time :)

The Best Thermal Paste For CPU Cooling | 2015
http://www.coolpc.pw/?p=1

Not sure what pads/shims I will use - important to fill the gaps with something that doesn't let the air in and dry out / separate the paste from the surfaces.

Have fun :)

hmscott
04-28-2015, 09:56 PM
What do you guys mean by saying "both" ? I only offset Dynamic CPU Voltage.

arrtaas, the other one besides "Dynamic CPU Voltage Offset" is the "Processor Cache Voltage Offset", on the Controls page bottom right, or on the Cache tab, here are all the adjustables for the 4700HQ - the newer CPU's also have additional adjustments.

Changed settings are in Yellow
48744

arrtaas
04-28-2015, 10:18 PM
hmscott,

I completely agree on everything you said about changing the thermal paste by yourself, definitely not an easy job on this laptop and definitely not recommended for people who haven't done it before.

What i mean't i wanted him to use same software to test what he currently has and compare temps, without any hardware changes. However, RMA might take forever, and not necessarily be approved. I personally choose to work on my tech myself (sometimes i think it is a bad habit LOL).

The links you gave are definitely another example of how thermal paste can actually change the temps, as it did to my laptop, which was unexpected, I thought 2-3C change will be great, but got to 10-12.

I will change the cache voltage and see how it goes. Thanks again.

Also IC DIAMOND 24 is definitely worth a try. Kind of pricey, i had to heat it up couple times to make it somewhat useable, since it is really really thick.

arrtaas
04-28-2015, 10:54 PM
Ok so I tried to change the Cache voltage offset and it didn't make any difference in CPU temps but gave a slightly lower performance. So this is what I have running stable (basically just the clock undervolt):

48745

hmscott
04-28-2015, 11:04 PM
Ok so I tried to change the Cache voltage offset and it didn't make any difference in CPU temps but gave a slightly lower performance. So this is what I have running stable (basically just the clock undervolt):

arrtaas, the performance decrease was coincidental, dropping the voltage won't change performance - it runs at the same multiplier the same way either way.

There are variations in test results from run to run due to other processes starting/stopping/running at different times coincidentally when your benchmark/test runs.

There are lots of other things running on Windows at the same time, so some runs might show better results than others run at other times.

That CPU has a lot more settings to change, many for voltage - lots of opportunity for additional tweaking for reduced voltage / heat generation - wish I could help, but I can't test from here :)

Read the help text in XTU for those new settings that can be changed on your CPU, and play around with settings - be sure and do a lot of saved profiles - with description names - so you can keep track of what you are doing - maybe even take lab notes to preserve/track the results.

Then place come back and let us know what the optimized settings are for your CPU :)

arrtaas
04-29-2015, 12:08 AM
I agree that the performance difference was coincidental, however there was no temperature difference at all as well. Do you experience temp difference when undervolting cache?

hmscott
04-29-2015, 12:51 AM
I agree that the performance difference was coincidental, however there was no temperature difference at all as well. Do you experience temp difference when undervolting cache?

arrtaas, I always adjust them together, so I can't say which gives the most/any change. Doing both does make a difference. Reducing voltage reduces power used / heat dissipation , so measurable or not, it is an change in the right direction :)

Gurendaiza
04-29-2015, 05:48 AM
hello all again ,

i didnt get a chance to change the multipliers and voltage , i will do so and report back!

just to make sure , going -80 on both and 36 35 34 34 ..

see you soon !

Gurendaiza
04-29-2015, 10:36 AM
i tried -80 just now and btw stock is on 36 35 34 34 here is a pic

48752

arrtaas
04-29-2015, 10:39 PM
How did your temps changed? Changing cache didn't do anything for me, so you don't have to go that low there, but of course if it runs stable, why not then..

hmscott
04-30-2015, 01:10 AM
How did your temps changed? Changing cache didn't do anything for me, so you don't have to go that low there, but of course if it runs stable, why not then..

arrtaas, tracking voltages so they are the same isn't about reducing heat - if the negative CPU offset did the heat reduction but a negative Cache offset didn't make a temperature difference I would still implement it to match the offsets.

Historically it has been a good idea to match internal voltage offsets. Without a specific mention that it's ok to disconnect the tracking of the offsets, or that it is required, I would match them.

There isn't a specific mention of keeping "Dynamic CPU Voltage Offset" and "Processor Cache Voltage Offset" the same, but in the past it is best to keep offsets in the CPU matching - within a range at least. 80mV to me would be outside that range.

There may be a technical document from Intel that specifies the relationships internally with voltages, but I can't find it in a few minutes of searching.

I will continue to match the voltage offset's, if you don't then you don't :)

Gurendaiza
04-30-2015, 05:36 AM
i'v tried what hmscott suggested , on -80mV and while playing CSGO i notice 5-6c drop .. its not a huge different but believe me its why better than 86c ! another thing i noticed , is when temps are low fans still dont try to push harder which is really weird, i know its an issue that asus needs to deal with but i think people been waiting for more than half a year now. i dont want to use NFC cause i really dont wanna change the stock profile not to void the warranty . btw now one answered me on 36x35x34x34x part.

and thanks again for all your support specially hmscott

hmscott
04-30-2015, 06:02 AM
i tried -80 just now and btw stock is on 36 35 34 34 here is a pic

48752

Gurendaiza, your CPU can run faster than mine, so you don't need to limit your settings to the same as mine.

I see there is space to increase the sliders for all 4 cores and the Cache to higher than the 36x,35x,34x,34x 36x, if I had that option, I would use it.

Go ahead and slide them all up and let us know what the max settings are, very nice :)

Gurendaiza
04-30-2015, 06:33 AM
Gurendaiza, your CPU can run faster than mine, so you don't need to limit your settings to the same as mine.

I see there is space to increase the sliders for all 4 cores and the Cache to higher than the 36x,35x,34x,34x 36x, if I had that option, I would use it.

Go ahead and slide them all up and let us know what the max settings are, very nice :)

but increasing wont effect the temps right ? should i slide them all up and apply ?

kevinjb101
04-30-2015, 06:48 AM
Most of the G751JY users suffer from this very conservative fan including me. I used the software called Notebook Fan control. And there is a config for G751JY. Now my fan at full load can run at max 4200rpm at certain temps. Mostly 3500rpm during gaming. Try it. It dropped the max temps to about >10c

Gurendaiza
04-30-2015, 07:07 AM
Most of the G751JY users suffer from this very conservative fan including me. I used the software called Notebook Fan control. And there is a config for G751JY. Now my fan at full load can run at max 4200rpm at certain temps. Mostly 3500rpm during gaming. Try it. It dropped the max temps to about >10c

i mentioned before the reasons i don't wanna use NFC (notebook fan control), I don't want to miss with fan profiles made by asus in case they decided to update their bios and change fan speeds (adapting the temps) .

and running fans on full load most the time will wear them off as far as i know.

i wonder why asus dont spare some time to look into the issue , its easier fixed now than having thousands of people RMA or fixing , it would actually save them money lol.

Gurendaiza
04-30-2015, 09:21 AM
yet another crash on csgo on -80 lol what should i do next!

hmscott
04-30-2015, 10:42 AM
yet another crash on csgo on -80 lol what should i do next!

Gurendaiza, back off 5 or 10, to -75mV or -70mV, and run test for stable operation again.

Unlikely you would guess correctly the first couple of times.

Imagine me trying to get -125mV on my G750JH, and how long it took testing to realize that -20mV was the best I could do :)

Patience.

And, don't do your important stuff on the laptop while at a test setting... make a Profile in XTU for MaxDefaultmV - up all the multipliers to Max and set the offset's to 0 - and set that setting while doing the important stuff - then back to the current offset test Profile when you are ready to run test settings again.

And, if you are tired of testing for a while, run the MaxDefaultmV Profile while gaming without worry of crashing for a while.

Have fun :)

hmscott
04-30-2015, 10:48 AM
i mentioned before the reasons i don't wanna use NFC (notebook fan control), I don't want to miss with fan profiles made by asus in case they decided to update their bios and change fan speeds (adapting the temps) .

and running fans on full load most the time will wear them off as far as i know.

i wonder why asus dont spare some time to look into the issue , its easier fixed now than having thousands of people RMA or fixing , it would actually save them money lol.

Gurendaiza, fan speed / heat - this isn't an actual problem. The fan profiles work as Asus intended and designed.

Asus likes to ship cool quiet running laptops - to attain both goals Asus keeps the fans low until the last moment, and then ramps them up to keep the temps under throttling point - but Asus isn't out to make a chilly willy cool laptop - that would be too loud.

The laptop will run cool enough to not damage components. Apple runs their little nuclear piles called Macbooks near 100c most of the time under normal use - actual use, not poser time - much hotter than even Asus runs them.

Don't worry about the temps / curves, enjoy the benefits of how hot Asus / Intel know stuff can run, and not fail. :)

Gurendaiza
04-30-2015, 05:12 PM
Gurendaiza, fan speed / heat - this isn't an actual problem. The fan profiles work as Asus intended and designed.

Asus likes to ship cool quiet running laptops - to attain both goals Asus keeps the fans low until the last moment, and then ramps them up to keep the temps under throttling point - but Asus isn't out to make a chilly willy cool laptop - that would be too loud.

The laptop will run cool enough to not damage components. Apple runs their little nuclear piles called Macbooks near 100c most of the time under normal use - actual use, not poser time - much hotter than even Asus runs them.

Don't worry about the temps / curves, enjoy the benefits of how hot Asus / Intel know stuff can run, and not fail. :)

iam away from home atm but i will try what you suggested all multipliers up and test mV... so you are saying dont use nfc lol.. lets see what happens and i will report back as usual thanks again for sharing your time and efforts..

yea and another funny thing csgo didn't crash at - 80mV at the first day but it did on the second.. so its really hard to guess at you mentioned lol