*OFFICIAL* Alienware m15 Owner's Lounge

No offense taken at all, and yes I’m a noob! Oops yes I meant -.158. I found that CPU was stable at -150.4 and cache about -.125.
I used Intel XTU and stress test at first and switched to Throttlestop because of better reviews. I reversed all the XTU settings then shut down. Then installed Throttlestop and kind of went from there.
Where I get tripped up is Turbo Boost power limits. I went from 55long power/90short power to 35/50. I think I got turned around after that by limiting the cores at 35.


Sent from my iPhone using Tapatalk

 

I remember now, I used Aida64 extreme. I am going to try and run Prime, after I have set back everything to -.125 for gpu and cache. Set the Iccmax for CPU as recommended to 65, 140,40,&30amps. Set the speed shifts as reccomended. Set the Intel GPU back to where it was.
So from here should I only work with the CPU undervolt first, to find it’s sweet spot, then try and lower the cache the same way? Of course test after each time I change it.


Sent from my iPhone using Tapatalk

 

Yes, do one setting at the time, if something goes bad, you will know what it has be, and you can bring that setting up a notch to get stable. By any means, do not use the profile you are testing as your start up profile. For that, use one that is tested for stability, or you may end up with start up instabilities that will be a pain to address

Do you use some laptop cooler, or have you rised the back of your laptop a little bit (1" will suffice)? it will help a lot with the airflow

 

The laptop cooler did a good job reducing temps 10-12 degrees. It’s a 30$ Targus. Originally I was using wood blocks to lift it. I was playing Halo Wars 2. I will try again. Maybe me messing with wattages and not having the Speed shift numbers right and other things didn’t help.
For sure on the instability. For now I suspended it in task startup. Also I will set the right profile for startup.


Sent from my iPhone using Tapatalk

 

Hi everyone I had a question. Is the cpu fan fixed yet with the latest bios? The issue where the fan would ramp up to 50% any time the cpu temperature went over 40°C and ACC profiles couldn't change it.

 

Hey, just upgraded from 2.1.1 to 2.3.0. The fans are fixed in the sense that they are completely silent on idle with CPU temp around 50C. The downside is that Cinebench 20 score went down 100 points for me (in Performance fan mode). I'm kind of OK with that, given that I don't run CB-20 level CPU loads and the upgrade includes some security fixes.

 

Can anyone please share the dumped by gpu-z vbios of RTX 2080 max-q? Thank You!

 

Here you got it. What are you going to do with it?

 

May be have you 90.04.52.00.31 version?

 

Can't really tell exactly what did it, but I solved my M15 waking up every 3 hours even with the lid closed and on battery. Finally can sleep overnight and wake the machine up in the AM with only a minor 3-5% loss in battery. I deleted some wake timers, but confused as to why any wake timer would do anything as the energy plan said to ignore wake timers on battery, and all the wake timers I saw did not have the option checked to run while on battery. Anyways - annoying and confusing.

What I can't solve is how to get battery life on this machine. I'm basically at 2.5 hours. This morning it ran for 27 minutes and burned through 20% battery with screen at 50% brightness, all AlienFX off, doing light browsing in Edge, and in POWER saver mode! I realized that Steam was running in the background, but still, should be getting a bit more than this.

I'm considering switching to the AERO 15 w/OLED - anyone compare between the two here?

UPDATE: This may all be my fault. Even though I turned on Battery Saver, in AWCC, I was set to "high performance" mode instead of balanced. I guess I'm a dunce. What's annoying is that I will often lose the performance slider native in windows when you click on the battery so I can't see this at a glance unless I open the AWCC app.

 

That's good to hear. May I ask which cpu/gpu you have?

 

No i don't, that's the dump of my gpu, from gpu-z

 

It's in my sig: 8750h,2070mq.

 

Would love to hear from those of you who have repasted.

The actual process of repasting looks straight-forward, but I am looking to purchase better screwdrivers as the five small M2.0 screws will not work with my M2.0 screwdrivers and I don't want to risk stripping them.

Replacing the wireless card was super easy as the screw was larger, but these smaller screws are a real pain.

I asked Dell what screwdrivers they recommend and they replied that a standard screwdriver would work; which simply isn't true as my screwdriver works with other M2.0 screws no problem (I may call to see if a technician can help me better)

What screwdriver set is everyone else using for the tiny screws in their Alienware M15?

 

Not sure if this was reported a while back, but I finally solved another major issue with my machine. It was waking up with the lid-closed at a very regular interval that was awfully suspicious. I thought I had solved it earlier by deleting some wake timers, but that wasn’t the case. Two nights back I happened upon the machine sitting on my desk, lid closed, but noticed the little Alienware head was glowing on the top. In some ways, I felt like I just stumbled on one of my kids doing something wrong!

I opened it and ran powercfg /lastwake and it reported “ACPI Wake Alarm” was the culprit.

I had to dig around the internet for the specific instance path reported, and it turns out many other folks were having this problem. The conclusion is that this may be a widespread Windows 1903/Laptop bug (how well known is unsubstantiated) - but basically, the machine is waking briefly to enter hibernation, but fails to do so. The default hibernation time is 180minutes after asleep - so thus my “every 3 hours” battery drain going on. Since it doesn’t hibernate, it remains awake until your default time limit is reached for going back to sleep.

Here is one such thread:

https://www.reddit.com/r/windows/comments/burul7/desktop_keeps_waking_itself_from_sleep_acpi_wake/

The solution? Disable hibernation (set to never in the power plan). I then subsequently changed it so that, on battery, when I press the physical power button, it will hibernate. The only issue with this is that if you are in a situation where the unit goes to sleep at 5% battery left before you hit hibernate, you may lose data if you can’t get to an outlet in time.

SO this is not ideal, but it’s an acceptable work-around for now.

I’m surprised I haven’t see more complaints about this, but I take it the majority of gaming laptop owners actually leave their device plugged in most of the time?

 

This is a widespread issue. Read the last 5 pages of tongfang qc7 on this forum. Intel and microsoft are aware of this issue.

This issue is going on for months apparently.

 

Good to know, I didn’t find that user in a search of this thread, but I’ll take your word for it. What comes around, will come up again...lol. Because of this bug, I complained to order support and convinced them to give me another 120 bucks off my already steeply discounted M15. I think that’s a legit complaint - no way this kind of major bug should exist out of the box.

 

Hello all,

i updatet the bios from 1.4.1 to 2.3.0. I installed every version between, because the support told me, I should do this. The fan works and with no load they are not running. The fan profiles are working too.
Maybe someone needs this information!

The cinbench R15 results are:
1210 [bios: 1.4.1]
1179 [bios: 2.3.0]

both results with undervolt 8750H [-0,120 V]

 

Is anybody using this https://www.amazon.com/gp/product/B07THHFD7F ? Any cheap alternative for 240W with signal (so it will charge)? Size doesn't matter as the point is to have the original as a portable AC adapter and this one as the desk-replacement adapter.

 

I've got the original 180 watt adapter.

Have people got better performance by using the 240 watt version?

Sent from my ELE-L09 using Tapatalk

 

Mine came with the 240W version, I think that because of the OLED screen

 

Answering to myself FYI.

After ordering and receiving the one I linked, I found that even if it is detected as a 240W at the BIOS, it seems that the internal chip is not compatible enough and the CPU will get throttled to 800Mhz under the BD PROCHOT condition. I can use Throttlestop to disable it but booting is extremely slow at 800Mhz. So I returned it and tried craigslist. I found a pair of used AC adaptors that I could test before buying. They were as new and original Dell parts (or at least I didn't get any throttling), and at $30 each, I think it was a good deal. So my advice, don't trust any seller, check yourself if you can. Also, probably it will be easier to get cheap authentic adapters in the used market.

 

Just an update on my previous posts.

I had problems getting the heatsink off my Alienware M15, however the iFixit Manta driver kit did the job. Once I got the heatsink off, repasting the CPU and GPU was an easy task.

It managed to improve my 3D Mark and PC Marks by a few hundred, but the biggest improvement is noise. It runs a lot cooler than before and the fans are not kicking in as early as they used to.

Would highly recommend others do it too as it's a cheap upgrade and will help you get more from your laptop

 

So the speakers on my M15 have stopped working. It seems to be a driver issue related to the Realtek Audio Drivers. I remove the drivers and restart my laptop and audio works fine as long as my laptop is not connected to the Internet but as soon I connect to the internet Windows automatically reinstalls the drivers and the audio stops working again. Anyone know of a solution? I've disabled automatic driver updates too but it still somehow keeps updating the drivers and breaking my audio

 

Have you tried using the drivers from the Dell page when you put in your Laptop tag and it brings up the drivers for your specific model? If that doesn't work, you can use something like DDU to disable automatic driver updates. Still seems like a annoying workaround instead of a fix

 

Is 1.4.1 supposed to be the best BIOS for the m15? I just got mine in October and I doubt its on the latest BIOS but I also doubt its on the 'best' since I think many people recommend a old 1.4.1

 

guys I am planning to upgrade my RAM to 32GB anything i should look out for and do you have any recommended brands/models.. Thanks

 

I believe these are highly recommended due to HyperX being a amazing brand and it having very good timings (better than most, if not all? sodimm available right now) with the max speed this CPU supports.
https://www.hyperxgaming.com/us/memory/impact-ddr4?Speed=2666MHz&Module Capacity=8GB&Kit=Kit of 2

2666Mhz is recommended. It has slower timings than 2400Mhz but it performs better

 

thanks will check this out

 

To anyone that cares, I updated to the latest 2.3.0 and it works way better than the 2.0.X version it was on. Never tried 1.4.1, but I am satisfied with how much more quiet/silent it is during light use.

 

.

..

.

.

.

.

.

Thanks bro~ M6600 is so awkward, it only support optimus, but only support newer AMD cards lol~

Hey guys, does M6800 mb fit in M6600 chasis? Is it possible to build a fake M6800 with M6600?

 

Only 17C difference between coldest and hottest core

 

Yeah that is annoying because those 2 cores out of 6 are making it throttle 7C sooner. Whereas if it was aligned properly, it would have more than 7C worth of temp headroom

 

Maybe post this link to Dell support next time you talk with them. From Intels's enginers perspective... At home, the lab engineers consider a load temperature above 80C to be a red alert, meaning that's the no-fly zone, but temps that remain steady in the mid-70’s are considered safe.

Inside Intel's Secret Overclocking Lab: The Tools and Team Pushing CPUs to New Limits. Intel's engineers explain oc'ing, voltage etc..

PAGE 7: 'Safe' Overclocking Voltages and Techniques

 

.

 

That is such trash of them to say that. You literally bought a 9750H and you do not have 9750H performance you paid for. Since they are lying to you and have screwed you of money by selling you a underperforming laptop, I would say use Throttlestop to overvolt it and increase the temps and use some program to hide that it is running. A Dell tech isn't gonna know how to do anything non-hardware related for the most part. But if they are acting like your already terrible temps are fine, make them worse until they agree they are terrible lol

 

Why should they fix it? Dell's engineers are amazing... They designed the system to withstand 100C I wonder why they didn't do this before T JUNCTION temp have been 100C now the last years after Ivy.

https://www.dell.com/community/Alienware/Coffee-Lake-temperature-concerns/m-p/7251354

 

They should just blame Intel for it. I mean, they really don't do much for their processors except refresh them and increase clocks and voltages, so of course the 9750H vs 8750H is gonna be even hotter, they use the same HSF but with a different CPU. Not that the m15 has an amazing HSF, but it works well for being a thin laptop.

 

.

 

Well I don't doubt the using old parts to 'fix' laptops, but I had screwed up my heatsink trying to get off melted thermal pad plastic off of it and the replacement the tech put on worked just as good or maybe slightly better than when I got the m15. But I have the extended warranty so they came to me to fix it. But still, if I was sold a laptop with such problem I would give them the choice of, "come fix this/replace this" or "I want a refund". Nowadays all laptops are about the same, though Alienware are the only ones with AGA support. So unless you need that, you could go with a Gigabyte Aero, MSI, ASUS and still have the same hardware + performance, of course they are like +-1-3% in benchmarks

 

.

 

Honestly the problem is simple, they use dumb mounting for their HSF. Many people have proven that their CPU + GPU HSF mounts are trash and it causes uneven pressure of the HSF which can result in huge temperature differences like you are experiencing. It's not a terrible design, though some people have bad luck and they receive very bad heatsinks.

If they would simply use 4 screws for GPU and 4 for CPU, it would work way better for them. But they always use 3 and 4 (7) and dumb little mounts.

Often times using a copper shim helps a lot with evening the core Temps and getting a nice HSF mounting pressure

 

.

 

Why did you use 2133Mhz instead of the default 2666Mhz memory?

 

I'm getting the same differences with dynamic voltage no matter what non-LM paste, but static volategr makes this irrelevant because temp never goes above maybe 95C in CB20. Something's b0rked with this dynamic voltage implementation.

 

Find @kahuna0k and @CptXabaras posts earlier in the thread and change voltage to static in TS as per instructions, it's a world of difference (but a bit less stable in my experience).

 

.

 

@fermeskop, why 4 posts for one answer? Use the Edit-function.

 

,

 

CL19 2666 (the RAM provided by Dell lol) is indeed slower.

Should have been more specific I guess, I was referring to HyperX Impact 2666MHz (as per my sig), which is CL 15 and about 8% faster than 2133@13. OK, so you just reused some old slightly slower memory, fair enough.