Jump to content

Alienware M15 R2 - i7-9750H skyrockets to 100°C

HI everyone,

 

I am very very new to the overclocking and undervolting world. So I come here with a lot of questions, hoping that you could help me understand what I'm doing a little bit better.

Please keep in mind that I do not really know what should be considered as normal and what is alarming. I'm just guessing here.

 

So my first question is:

  • 1- What should I read/watch/listen to to learn about Hardware? is their any good online books, online courses, tutorials?

 

I have a brand new Alienware M15 (full specs are listed below), and it's running very HOT. It used to BSOD a lot after a extended period of gaming. But after restoring Windows to factory settings I have solved the problem.

I suspected that overheating was causing it so I installed XTU to track the temperature.

My CPU runs at 49-55°C when completely idle. It is at 55-65°C while surfing on Mozilla and spikes instantly to 85 - 95 even 100°C for the slightest operation. I feel that this temp is very unstable. While gaming it skyrockets to 100°C and does not go down the 95°C mark.

 

image.thumb.png.bc345f7d864f54ef026fa8de4ea84340.png

CPU Temp and thermal throttling while browsing the web (lowest 55°)

So here are my questions at this point:

  • 2 - Is the graph supposed to look like that? Is it normal for a CPU to lose or gain 10+°C every few seconds?
  • 3 - what is considered a good temp at Idle/common use/Heavy use? for a standard gaming laptop?
  • 4 - Same question for a Alienware M15?
  • 5 - Should I be alarmed?
  • 6 - is it normal to have thermal throttling? and power limit throttling? or should I aim to get rid of it?

 

I thought of undervolting my CPU as it seems to decrease the temp and the throttle. but something weird happened here.

When I undervolt on XTU, nothing shows up on HW Monitor.

1533309639_figure2.thumb.PNG.7be5f7d0a9c98195eb71d019e8e0f7bf.PNG

XTU and HW Minitor showing different Voltage offset

xtu.thumb.PNG.86fde4650fff8a9f07f7c27cd7761d91.PNG

List of Intel processors compatible with XTU

 

 

I have so many other questions (on the Alienware Command Center, power limit throttling, settings to lower CPU temp, the indicators I should follow etc...) but I have already been very long here!

 

 

 

CPU: i7-9750H - GPU: RTX 2070 max-Q - RAM: 16 GB - Bios: Alienware 1.3.1 - OS: MS Windows 10 Family

Link to comment
Share on other sites

Link to post
Share on other sites

7 minutes ago, Stalink said:

HI everyone,

 

I am very very new to the overclocking and undervolting world. So I come here with a lot of questions, hoping that you could help me understand what I'm doing a little bit better.

Please keep in mind that I do not really know what should be considered as normal and what is alarming. I'm just guessing here.

 

So my first question is:

  • 1- What should I read/watch/listen to to learn about Hardware? is their any good online books, online courses, tutorials?

 

I have a brand new Alienware M15 (full specs are listed below), and it's running very HOT. It used to BSOD a lot after a extended period of gaming. But after restoring Windows to factory settings I have solved the problem.

I suspected that overheating was causing it so I installed XTU to track the temperature.

My CPU runs at 49-55°C when completely idle. It is at 55-65°C while surfing on Mozilla and spikes instantly to 85 - 95 even 100°C for the slightest operation. I feel that this temp is very unstable. While gaming it skyrockets to 100°C and does not go down the 95°C mark.

 

image.thumb.png.bc345f7d864f54ef026fa8de4ea84340.png

CPU Temp and thermal throttling while browsing the web (lowest 55°)

So here are my questions at this point:

  • 2 - Is the graph supposed to look like that? Is it normal for a CPU to lose or gain 10+°C every few seconds?
  • 3 - what is considered a good temp at Idle/common use/Heavy use? for a standard gaming laptop?
  • 4 - Same question for a Alienware M15?
  • 5 - Should I be alarmed?
  • 6 - is it normal to have thermal throttling? and power limit throttling? or should I aim to get rid of it?

 

I thought of undervolting my CPU as it seems to decrease the temp and the throttle. but something weird happened here.

When I undervolt on XTU, nothing shows up on HW Monitor.

1533309639_figure2.thumb.PNG.7be5f7d0a9c98195eb71d019e8e0f7bf.PNG

XTU and HW Minitor showing different Voltage offset

xtu.thumb.PNG.86fde4650fff8a9f07f7c27cd7761d91.PNG

List of Intel processors compatible with XTU

 

 

I have so many other questions (on the Alienware Command Center, power limit throttling, settings to lower CPU temp, the indicators I should follow etc...) but I have already been very long here!

 

 

 

CPU: i7-9750H - GPU: RTX 2070 max-Q - RAM: 16 GB - Bios: Alienware 1.3.1 - OS: MS Windows 10 Family

I'd download intel XTU and tweak stuff in thee. CPUs do hit 100 in laptops it's not uncommon and it will ramp back down to protect itself. What is the rated base clock of your laptop and what is the clock when it's "throttling"?

Dirty Windows Peasants :P ?

Link to comment
Share on other sites

Link to post
Share on other sites

Those voltages don't mean much, you have to go for Vcore...

The temperatures are high but not impossible for a laptop

Main PC [The Rig of Theseus]:

CPU: i5-8600K @ 5.0 GHz | GPU: GTX 1660 | RAM: 16 GB DDR4 3000 MHz | Case: Lian Li PC-O11 Dynamic | PSU: Corsair RM 650i | SSD: Corsair MP510 480 GB |  HDD: 2x 6 TB WD Red| Motherboard: Gigabyte Z390 Aorus Pro | OS: Windows 11 Pro for Workstations

 

Secondary PC [Why did I bother]:

CPU: AMD Athlon 3000G | GPU: Vega 3 iGPU | RAM: 8 GB DDR4 3000 MHz | Case: Corsair 88R | PSU: Corsair VS 650 | SSD: WD Green M.2 SATA 120 GB | Motherboard: MSI A320M-A PRO MAX | OS: Windows 11 Pro for Workstations

 

Server [Solution in search of a problem]:

Model: HP DL360e Gen8 | CPU: 1x Xeon E5-2430L v1 | RAM: 12 GB DDR3 1066 MHz | SSD: Kingston A400 120 GB | OS: VMware ESXi 7

 

Server 2 electric boogaloo [A waste of electricity]:

Model: intel NUC NUC5CPYH | CPU: Celeron N3050 | RAM: 2GB DDR3L 1600 MHz | SSD: Kingston UV400 120 GB | OS: Debian Bullseye

 

Laptop:

Model: ThinkBook 14 Gen 2 AMD | CPU: Ryzen 7 4700U | RAM: 16 GB DDR4 3200 MHz | OS: Windows 11 Pro

 

Photography:

 

Cameras:

Full Frame digital: Sony α7

APS-C digital: Sony α100

Medium Format Film: Kodak Junior SIX-20

35mm Film:

 

Lenses:

Sony SAL-1870 18-70mm ƒ/3.5-5.6 

Sony SAL-75300 75-300mm ƒ/4.5-5.6

Meike MK-50mm ƒ/1.7

 

PSA: No, I didn't waste all that money on computers, (except the main one) my server cost $40, the intel NUC was my old PC (although then it had 8GB of ram, I gave the bigger stick of ram to a person who really needed it), my laptop is used and the second PC is really cheap.

I like tinkering with computers and have a personal hatred towards phones and everything they represent (I daily drive an iPhone 7, or a 6, depends on which one works that day)

Link to comment
Share on other sites

Link to post
Share on other sites

Ok great thank you.
So I should not be worried to have so much up and downs in the CPU temp.

while gaming (SCII and Civilization 6 mainly) it stays at 95°-100°C though. I'll send a print screen as soon as I can.

 

I'll try Vcore tomorrow morning and see how it goes.

Link to comment
Share on other sites

Link to post
Share on other sites

Here are the infos I got from the Intel website for my CPU:

 

From what I understand, this is the main indicator of the CPU performance. My goal should to have this value match my utilization. So the lowest possible when idle or on MS Word, and be consistently at 4.5 GHz while gaming. Is that right?

So I need to find the sweat spot where the CPU undervolt does not impact my frequency. Am I making sense?

 

Regarding the Vcore. How can I get the value, and what exactly I am looking for? Nothing called Vcore or Core Voltage appears in my HW monitor (See screenshot below). is the VID #0 the correct value here? I want to make sure that what I do on XTU has a real impact.

 

image.png.eed2c14d6cb51965e358c903f6917cc8.png

Just to make sure:

  • on XTU, the value I need to watch is the Max Core Frequency, right?
  • What is the Processor Cache Frequency? And should I monitor this value too?
  •  the Vcore is referred as Core Voltage on XTU
  • I should see a drop of 100mV on my Vcore if I go for a -100 mV Core Voltage Offset. Right?

 

image.png.3a77e7b723d3ff7835b1fe64320f01ca.png

Link to comment
Share on other sites

Link to post
Share on other sites

-AW laptops have a bad reputation of uneven heatsink surface and bad thermal paste

-undervolt BOTH core and cache, if you're not able to undervolt it might be due to Plundervolt (Google it)

 

1. There are many reputable websites and YT channels (I couldn't list down all for obvious reasons). For desktop PC related content, some of my favorite YT channels are HardwareUnboxed / TechSpot, GamersNexus, Jayz2cents etc. For laptop reviews, Jarrod's Tech, Bobs of all Trades, Ownordisown etc

2. Need to run a constant heavy load to really see what's happening

3./4. I would say 30-50C depending on ambient temp

5. The thermals you provided are not great but could be normal in AW laptops (since they suck)

6. Thermal and power throttling happen in every laptop depending on load, however by choosing a laptop model with better cooling you could avoid thermal throttling more due to more thermal headroom

7. HWinfo64 would always be my choice (though software sensors aren't accurate either)

8. Yes

Desktop specs:

Spoiler

AMD Ryzen 5 5600 Thermalright Peerless Assassin 120 SE ARGB Gigabyte B550M DS3H mATX

Asrock Challenger Pro OC Radeon RX 6700 XT Corsair Vengeance LPX 16GB (8Gx2) 3600MHz CL18 Kingston NV2 1TB PCIe 4.0 NVMe SSD

Montech Century 850W Gold Tecware Nexus Air (Black) ATX Mid Tower

Laptop: Lenovo Ideapad 5 Pro 16ACH6

Phone: Xiaomi Redmi Note 10 Pro 8+128

Link to comment
Share on other sites

Link to post
Share on other sites

Thank you!

As recommended I downloaded  HWinfo. It works very well.

But it does not show any changes on the voltage (screenshot below).

Is it possible that XTU is actually doing nothing? should I try Throttlestop instead.

From what I read, Plundervolt does not apply to AW M15 and i7-9750h, so this should not be the problem.

 

 

image.thumb.png.5716a0cdd179474dc83246c876dc372f.png

Link to comment
Share on other sites

Link to post
Share on other sites

I've been playing Starcraft II yesterday. My CPU jumps directly to 100°C and stays up there. and it goes immediately to thermal throttling.

My frequency is at constant 3.9 GHz; which is good I guess. And frankly the game still runs smoothly.

Every 30 min I undervolted my CPU to compare performances. I started at 0mv, then -100mV, -150mV, -200mV, -250mV through XTU. The performance is strictly the same every time. So I strongly suspect that XTU is not actually doing anything. I will try ThrottleStop today and see how it goes.

 

731801775_-200mv-30minSCII.PNG.8565ad5703e67db0934cb470a295b538.PNG

30min of Starcraft II

 

I've also stress tested my CPU with different voltages ( 0mv, then -100mV, -150mV, -200mV, -250mV). Same thing here: he performance is strictly the same every time.

Temps are at 90°C average peaking at 100°C. It thermo throttles immediately and then power limit throttles...

 

58135924_0mV-30minStresstest.PNG.6632c815c101d1363dc9146161e1bd6f.PNG

CPU 30min Stress test

 

 

  • All of this should not be considered normal right?
  • What exactly is Power Limit Throttling? and should I get rid of it?

 

Link to comment
Share on other sites

Link to post
Share on other sites

2 minutes ago, Stalink said:
  • All of this should not be considered normal right?
  • What exactly is Power Limit Throttling? and should I get rid of it?

-if you want further testing, PM me and I'll guide you

-CPU have power limits, in short if PL1 (long term boost) is set to 45W, it will not draw over 45W, however many Intel CPUs aren't able to achieve the highest turbo speed with the power budget, under multi core load (require more than that in order to run at max all core turbo, example: i7-9750H max all core is 4GHz)

Desktop specs:

Spoiler

AMD Ryzen 5 5600 Thermalright Peerless Assassin 120 SE ARGB Gigabyte B550M DS3H mATX

Asrock Challenger Pro OC Radeon RX 6700 XT Corsair Vengeance LPX 16GB (8Gx2) 3600MHz CL18 Kingston NV2 1TB PCIe 4.0 NVMe SSD

Montech Century 850W Gold Tecware Nexus Air (Black) ATX Mid Tower

Laptop: Lenovo Ideapad 5 Pro 16ACH6

Phone: Xiaomi Redmi Note 10 Pro 8+128

Link to comment
Share on other sites

Link to post
Share on other sites

  • 4 weeks later...

As a user of m15 R2 for 4months i ll share my experience.

- CPU will reach 100C while gaming or under heavy load.

-CPU temperature can increase-decrease very quick (within a second it can jump from 60C to 90C)

-Undervolting CPU will not give you much performance (other laptops have better success with this but not this model). I tried up to -0.145V and the difference is not noticeable.

-If you are playing demanding games you don't have much of options here. No matter what you do you will always get 100C on CPU because of Turbo boost. The only way to stop cpu hitting 100C is to change the thermal mode to "quiet" or "balanced".  This is a well written review for this model explaining all those profiles https://tech360.tv/reviews/alienware-m15-r2-review/ . 

- I contacted DELL support previous week and it seems if the laptop is under warrant you can ask for "repair" for free. Today a guy from dell came and changed the whole heatsink of the laptop and applied new paste (i bought my own paste since they will apply some cheap dell paste if you don't have your own). The difference was close to zero.. I still get 100C on CPU.

-My suggestion is to buy a cooling pad for laptop if you wanna play brand new game titles. If you are just browsing and for multimedia always select the "Quiet" thermal this is the only option to prevent 100C tempts. "Quiet" locks the cpu to 45W and after few min the cpu frequency will drop to ~2.6-2.8Ghz and the temperature drops to 75-80C.

-A trick i discovered on the web, since very often turbo boost hits even while idle, you can go to power options in windows and on your current plan reduce the processor power management  min/max to 99 (instead of 100). This will guarantee that the turbo boost will not trigger. 

Link to comment
Share on other sites

Link to post
Share on other sites

  • 3 months later...
On 5/8/2020 at 12:38 PM, chrzang said:

As a user of m15 R2 for 4months i ll share my experience.

- CPU will reach 100C while gaming or under heavy load.

-CPU temperature can increase-decrease very quick (within a second it can jump from 60C to 90C)

-Undervolting CPU will not give you much performance (other laptops have better success with this but not this model). I tried up to -0.145V and the difference is not noticeable.

-If you are playing demanding games you don't have much of options here. No matter what you do you will always get 100C on CPU because of Turbo boost. The only way to stop cpu hitting 100C is to change the thermal mode to "quiet" or "balanced".  This is a well written review for this model explaining all those profiles https://tech360.tv/reviews/alienware-m15-r2-review/ . 

- I contacted DELL support previous week and it seems if the laptop is under warrant you can ask for "repair" for free. Today a guy from dell came and changed the whole heatsink of the laptop and applied new paste (i bought my own paste since they will apply some cheap dell paste if you don't have your own). The difference was close to zero.. I still get 100C on CPU.

-My suggestion is to buy a cooling pad for laptop if you wanna play brand new game titles. If you are just browsing and for multimedia always select the "Quiet" thermal this is the only option to prevent 100C tempts. "Quiet" locks the cpu to 45W and after few min the cpu frequency will drop to ~2.6-2.8Ghz and the temperature drops to 75-80C.

-A trick i discovered on the web, since very often turbo boost hits even while idle, you can go to power options in windows and on your current plan reduce the processor power management  min/max to 99 (instead of 100). This will guarantee that the turbo boost will not trigger. 

This comment is extremely helpful. It just freaked me out this morning when I noticed that my CPU temp "dance" between 50 and 80 within a second and I was wondering if something wrong with the thermal system. It seems that the constant change is normal. And I just surprised that the fan mode will trigger various cpu level..... I guess the fan mode supposed to be just a fan mode. Can't understand why AW can't let me limit the CPU while choosing a fully load fan mode. 

Link to comment
Share on other sites

Link to post
Share on other sites

Heeey! i'm going to share my experience, i bought an AW m17 r2 about a month and a half ago, when iddle i was getting beetween 50 - 70 degress (this is when is conected, when is on battery i got 30-40 degres, 50 sometimes), i was very worry about that becouse i work a lot doing adobe stuff and i used the laptop almost all the day, so this what i do:

1. I installed cpu trhottlestop and i undervolt my cpu to -125 mv ( i know that i could go farther than this becouse i undervolt it to - 150mv and there was no blue screen problem but in my day to day i only do 125mv) i got a few degrees less on my cpu and this prevent the pikes on the turbo boost

2. i also set up the Turbo power limit in my cpu to 75 and 75 

2. i create a new profile in my alienware comand center and i used in iddle , this profile help me a lot becouse temperature are stable in 50-55 degress on iddle 1057632059_Anotacin2020-08-26111509.png.ce69590071e26047c1f4d6ebc40e24a1.png

4. In trottlestop i created a perfile with the turbo boost limited to 3.6 ghz, i only use the 4 ghz profile when i'm rendering an image, so while playing on 3.6 ghz and the fans at the fastest speed i get 80 - 85 degrees stable, y always play with the gpu overclock function to the max and gpu its not the problem becouse i usually got 60-70 degress at max

5. i recently realised that i was getting better thermals with the turbo boost off ( in trhottlestop i turned off the turbo boost and played with the fans at max) i was getting like 40 - 50 degrees max in the cpu, in quiet profile i was getting 60-65 degrees in the cpu ( and that doesn't give me bad fps) 

 

SO in sumary : undervolt -125 mv for iddle , undervolt -150 mv for rendering process with 4.5 ghz (90-95 degrees), set the max frecuency to 3.6ghz or 3.3ghz or 3ghz and Turbo boost power limit to 75 75,  it depends on what you are more confortable with , for better thermals play with turbo off becouse many games demands more gpu than cpu, create an iddle profile in AW comand center like mine, and thats all , this are captures of games with my ghz configurations ( they are in ultra and in 1080 with the vsync on and, dota 2 and assasins creed unity ) (Funfact: playing fall guys with vsync of i get 40 degrees on cpu with turbo off and ultra settings with 60-70 fps, if i turn vsync on, i got 144 fps and 60 degress i dont know why )

 

 

 

Captura_de_pantalla_(2).png

Captura_de_pantalla_(3).png

Link to comment
Share on other sites

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×