*OFFICIAL* Alienware 17 R5 Owner's Lounge

when having alienware r4 or r5 focus on one point that this laptop can not run as desktop replacement
so you will not be bothered by the endless problem of overheating until you can not feel the other features on your laptop
because from the factory it has been like that since alienware 17 r4 in launch. and until now no one can fix the problem heatsink it completely
and the company also does not care
you will run out of energy and time to solve this overheat problem. endless
Maybe.I think to solve this problem
the easiest way is to limit max cpu on power management. 90% or 85%
Lower fps but runs smoothly

if the component keeps at 90+ I think it will make the component's lifespan not last long
Sent from my CPH1723 using Tapatalk

 

Thanks Mate that fixed it. Playing bf1 with 4.4ghz allcore now. No 800mhz dip anymore

 

yup thought so
I was underwhelmed with my 1080p IPS screen coming from an old 1080p IPS 2012 XPS. Putting them side to side the alienware seems - blurry!

 

Temps during bf1 are high. Upper 8Xs to low 90s... but stable 4.4ghz. No drops. Definitly fixable with LM. GPU runs with 1950mhz @69 degree.

With 4.3ghz mid 80s degree and with 4ghz high 70s to low 80s

 

I called dell support today (posting in here for my 15r4 as that thread seems dry ) regarding my i9 / 1070.
Theyve said the same thing. Unless you have a sustained / average of 2 cores over 95 without overclock on - they wont repaste or send a tech out.
My range was 80's and lower 90's. Peak of 1 core at 96.
Going to run more benchmarks tonight.

 

No thanks brother ! glad I it helped

 

Is the PCH one chip or multiple? It is getting hot, outside of the CPU also getting hot. It appears its heat is creaping upwards over a long time and then the laptop simply stops (crashes, some diagnostic comes on from the BIOS I think, etc). I can understand that I may have a faulty installed heatsink allowing the CPU to overheat very quickly, but with the CPU throttled to x35, x36 why is the PCH's temps rising? Does it use a separate heatsink?

 

Unfortunately, they designed it such that the PCH is simply open to the air and does not even have any heatsink or thermal solution.
You can see a lot of people who have done the PCH mod (including me) to either add a 4-6mm tall (15 or 17) heatsink or graphene tape (what I did) from the PCH to the heatsink.

Sent from my SM-G950W using Tapatalk

 

Is there a way to get more than 110wPL ? Cinebench 4.4ghz maxes at 105w but 4.5ghz is hitting 110w and reduce clock to ca4ghz

 

By upping the voltage and frequency you create heat that a sensor registers. If 4.4 @Current voltage dies not produce more heat, you've passed the limit @4.4 Ghz of creating more heat with the given voltage so you have to up it again (by going to 4.5) which again will create more heat.

Try OCCT and lock your cpu @4.5 and see what it does.

 

Temps are ok. I definitly hit the 110w powerlimit @4.5ghz

Btw atm i run 4.3ghz@-175mv. Dont know if this is good or average

 

Someone on Reddit claimed to hava made a @4.3 with a -200Mv, so if that's true I think you're in the right ballpark. It is for sure a pretty hefty undervolt compared to my -75 on my 7820HK.

 

Thats why i was wondering... my 6820hk need +20mv @4.2ghz

 

Different chips, different voltages. Not all chips are created equal.

 

Sure. Was just wondering that the sixcore has so much uv Room left.

 

Just to update everyone Dell have proposed to send me a replacement laptop because their tech team want to "study" mine at their workshop! Today I received a new purchase invoice from Dell for the new laptop. Of course I know now that it will come from China and will take 3 weeks to arrive. Will it be better than the current one? The tech support people seemed to think that there is something seriously wrong with mine, so I hope so. If the new laptop also overheats then I will return it for a refund having given up on Dell.

 

IA AC DC Loadline VID boosting of about 140mv.
Same thing as the GT75 Titan (and previous models).
This can be disabled by unlocking bios menus on the MSI. Not so easy on the alienware.because of that Bios Guard thing. You can unlock menus and flash the APTIO capsule but it will just result in a brick and needing a bios recovery method or SPI programmer.

 

-182mv, 108w max, 4.5ghz, no throttling, stock paste, lifted Back

Update: Firestrike 19456P @4,4ghz CPU @1950mhz GTX1080

https://www.3dmark.com/3dm/26468292?

 

Hi guys! I received and installed the 32GB 2666MHZ memory (Corsair CMSX32GX4M2A2666C18 Vengeance 32 GB (2 x 16 GB) DDR4 2666 MHz) today. It is really running at 2666mhz! What a relief. I also installed a PM961 SSD. I just run a dell extensive stress test and here are my highest temps (only took screenshot of the red temps!). OK the PM961 needs a heatsink (highly expected) (I wonder why it has two temps reporting on HWinfo?) but what worries me is the CPU temp (CORE 2) which has reached 95 celcius and it actually throttled as Hwinfo says. Highest frequency reached just 4.392GHZ during the test (well I have overclock disabled). If CPU reaches this temp with just an extended Dell stress test, what is going to happen with heavy games and even more with overclock enabled? I do not want to change thermal paste on CPU or touch anything I should not....

https://imgur.com/a/j3MB0Xk

 

Drive Temp 1 is the memory chip
Drive Temp 2 is for the memory controller (it goes hotter under load)

There should be a thermal plate covering the ssd if you bought it with your Alienware (as per image). But it won’t make much difference when writing large files the drive will throttle after approx 15 seconds and the speed will plummet.

Re the cpu as others have said bad design, it’s really poor it’s like this new out of the box. Contact Alienware, and ask for a new heatzink or return. Your correct when you play a game the heat will rocket in all cpus due to additional gpu heat. What temps do you get on your gpu when playing a game say after 30 mins?

Do you know who the cpu/gpu heatzink is made by Sunan or CCI?

 

No matter how hot the CPU gets (and why) it throttles itself in multiple ways: frequency (multiplier), power, current and of course on temperature. So no matter which of the parameters get exceeded, your laptop will continue to work, and in many cases you would not even know it.

Try to monitor the PCH temperature. I have noticed that while playing a game (World of Warcraft) it creeps up, and keeps going up slowly (over say 15-20 minutes), and when it reaches 95C+ then the laptop reboots. I have seen no "throttles" on the PCH, in other words there is nothing to lower the temps. I think the GPU's lines pass through the PCH so as you can see on the right hand side graphs, there is a direct relationship between GPU usage (power and temperature) and PCH temperature. In those graphs when you see the GPU/PCH going down is because I have alt-tabbed to the desktop (so the game goes into the background). Also, note that there is hardly any CPU usage, this is all down to the GPU/PCH. While the GPU runs cool at 69C / 170W, the PCH temps keep increasing, until, a hard reboot!

 

Something is very wrong with how your Cpu perfom. All too low 3DM FS physics scores. And too low Cinebench scores as well vs. clocks.

I can see nothing has changed from Dell. Boiling Cpu as well boiling ssd’s. Probably the same for PCH chips also. And this is from the smaller brother AW15R4.

 

You should buy AW r5 for your kids and use it at stock to keep everyone warm.

 

@Papusan How high should it be? Be aware: i have only 2400mhz RAM. Got same score @4.4ghz yesterday. Maybe it was cache clock throttling @4.5...
Saw a similar 3DM Firestrike score in the results list from a msi book

 

+-1400cb. +-18500 in FS physics. Both stock clocks (4.3GHz all cores).

 

Its very strange behvaiour, since the PCH location is the same as the previous gen and in the previous gen it didnt get hot at all.

 

Oh'no. Same for PCH chips in previous gen as well. Here is a few... http://forum.notebookreview.com/threads/new-vbios-and-pch-temps-through-the-roof.802383/

http://forum.notebookreview.com/thr...4-owners-lounge.797457/page-476#post-10502817

http://forum.notebookreview.com/thr...4-owners-lounge.797457/page-527#post-10519884

And there have been created how to do guides to lower PCH temp as well, due the temp mess.

 

only the top link shows high pch temps after a custom gpu bios?

In general i never saw the pch creep over 80c in previous gen or the one before that. But a too high pch temp can be a cause for the cpu to drop to 800mhz.

 

Read in the posts... You will find many more with a search (And all model sizes is affected). But most part of the images in the forum is now broken and only show IMG. Have seen many above 80C. For me is temp like this too high although you don't always run into throttling.
There is several PCH mods out there.

 

My observations are

1) the CPU overheats on its own, within fractions of a second, every time it tries to raise the multiplier, and in my case this may well be a case of a badly mounted heatsink. I have posted images with very little or no CPU % yet the temperatures oscillate between 40C-90C (and then temp throttle) every second or so.

2) the CPU overheats (over 100C) when the GTX-1080 is busy - it seems the interconnected heatpipes somehow tip the i9-8950 over the edge.

3) the PCH overheats to over 93C when the GTX-1080 is busy - it may be because of G-Sync the PCH is used more? No idea why. When the PCH reaches 95C or more the machine reboots. I have found no way to slow down the PCH other than not using the GTX-1080.

4) the keyboard reaches 52C on two areas, around the left shift and right shift keys.

5) the SSD reaches 90C as well...

6) the CPU fan (on the left hand side) expels air from two large vents, one on the side and one at the back. While the air coming out the rear left vent is very hot, it is of mild strength. In contrast there is a much stronger flow of air coming out the side vent, ****still serviced by the same fan****, and this is much cooler, almost cold. Why are we wasting cooling energy by expelling cool air from the laptop? It should all be diverted to the rear vent where it is obviously needed!

 

i've been testing this all day and it seems very crazy but it seems to work (for me) if you have time could you also verify this.

i've been thinking for a while now to tape off the left side of the vents to force the air to get out from the back (passing over the heatsink fins) rather from the left side and backside.

i've observed a 3 to 5c temperature drop due taping the air off. i think because it forces the air to exit from the back forcing the air to pass over the heatsink (and heatsink fins).

 

I did the same using the palm of my hand trying to stop the air coming out and hoping it would then go over the back. I did not observe any differences but did not think to tape it! Will tape it now and will let you know. Stay tuned...

 

.

 

I did tape over the vent with electrician's tape for a few minutes and did not notice any appreciable change in the amount of hot air coming out the back or to the temps. Not sure how it works inside since I am not going to open it to find out!

 

The GTX 1080 runs at an almost constant 180W / 70C. The i9 runs at 20W-40W / 100C / throttled. How can they remove 180W of heat so perfectly well to keep the GTX 1080 at around 70C, when at the same time they cannot remove a puny 20-40W of heat from the i9? Same laptop, same enclosure, same fans, same designers... Plus, the i9 instantly spikes to over 100C with no CPU%, which means bad heatsink or NO paste? They may have forgotten to add paste! I cannot explain it otherwise.

 

The gpu fan is bigger, so can push more air over the heatsink. The heatsink has 1 pipe more then then cpu (cpu has 2 copper pipes, gtx 1080 3)
And tripod heatsink, while the gtx 1080 is fastened with 4 screws. So it has a more even pressure than the cpu. And they use a shared heatsink. So the heat from the GPU can be transferred to the CPU making the cpu overheat/throttle. Those are fairly logical problems that contribute to the CPU spiking so much.

 

my PCH/GPU temps did not even go red during Dell extended stress test. Only CPU (one of the cores). Regarding the SSD, I bought it to try the speeds but did not install a heatsink yet. I ordered one to try. Hopefully (according to some people), it is going to sort things out a bit. Again I have no overclocking selected.

Well, I guess if HWINFO says that there was a throttle because of the CPU, then I guess it did throttle because of the CPU. I wont touch the CPU thermal pad. If it causes issues (eg. throttling or even worse) during gaming, then I will contact Dell.

 

Well, you can open it and check. Maybe the thermal paste is just not good! Replacing thermal paste with high performance ones used to improve things since M17X R2 times (my previous Alienware). I cannot warranty it will make any difference on R5 and i9.

 

@Papusan loaded deafults in bios now. Activated CPU performance mode (unlocks 110w PL, Multis @ 43,44,45,47,48,48, so stock), undervolt -130mv

Cinebench 1376 with 4,3ghz. Ok now?

2 Pages ago you said ca.1400p cinebench and 18500 3dmark physx score. So my stock Performance is in that range now.

I know you dont like bga Hardware. Bit its still fact that the combination of i98950HK, and GTX 1080 is a powerfull configuration, even if its bga. So i would be happy if you could help to get the best possible performance

https://www.3dmark.com/3dm/26497966?
this is stock

 

Is there a way to get the 8750 with a 1080?

If so, doors it come with the 240w psu or the 330w psu?

 

If the GPU is running at a lower temp, it will not transfer heat to a higher temp object. Heat transfer works the other way. If anything, heat will be transferred from the cpu to the gpu side.

 

Hey, I just order the new alienware 17 r5. I was wondering if anyone has repasted their alienware 17 r5 yet, and what sort of results they saw? Also what paste you used.

 

It is not available via Dell currently. I do not think they offered that configuration on the 17R4 either, so it may likely never be available. You would still want the 330W brick even with the 8750H. MSI uses it with the 8850H + GTX 1080 configuration of their GT75VR.

I don't think anyone here has tried repasting yet, but it will probably give the same results as it did the last time.

 

Yes but in this case does the gpu side has more copper than the cpu side. Making the copper from the right side transfer more easily to the CPU. That's what you see in action when you use your cpu and you at the same time.

If you can display a graph, you'll see how hitter your GPU becomes, how hotter the CPU becomes.

A cpu only has 2, 4, 8 and so on. A gpu has a few 2560 cores working underneath and texture units and so on. That's also true if you go look at the application of GPU's in cryptio mining, AI and so on. It has more cores to divide and asynchronous compute solutions, then any cpu solution. And the more calculations can be done, the hotter it becomes.

 

@iunlock does every gtx1080 have the unlocked vbios now? Iplayed GTA5 1h now and max was 189w. Average about 165w. Shoudnt it be maxing at 200w? and i think the average should be higher?

 

Every gtx 1080 sold from February 2017 has the oc bios from factory. It says 180 watts, but I've pulled 225 without problems. Maxing out at 68C

 

mhm... my average is at 165w... how can i check if the vbios is the correct one? how have yoou pulled 225w? maybe we can do the same test and compare

 

Don't worry you have the oc bios, your max was 189, while the old vbios was limited at 150watt.

butt if you really want to check:
http://www.dell.com/support/home/nl...nware-17-laptop&languagecode=en&categoryid=vi

here is the oc vbios. if you last bios numbers/letters are currently higher then 86.04.5B.00.8F you're on the oc bios. note that there are multiple OC bios revisions. i for instance have the 86.04.5B.00.92 from factory. while yours can be 86.04.5B.01.xx or 86.04.5B.xx.xx

 

ok. but if the temp is ok the card should consume those 200w...shouldnt it? why is it only unsing 165 and not 200 capped?

 

no a cpu, gpu, ram, gsync/freesync display uses what it can use. for example there are scenes where it only needs 165 watts to render it at 120 fps (or the max of the engine) while some scenes are more graphical intense, that they require more power a.k.a more watts to render to scene. making it spike to the 180+ region and then go back down to it's avarage power usage till it needs the full power again.