*OFFICIAL* Alienware m15 Owner's Lounge

Too late for me

Sent from my Moto Z (2) using Tapatalk

 

@etern4l

Haven't payed to much attention to it. during the last CB run i've used the performance profile on AWCC and with XTU i've set 75W on both short and long Turboboost. Got a flat line on the trend with CPU boosting constantly at 3.9Ghz on 6 cores without throttling. I'll check out onto it with more attention and let you guys know.

 

Yeah, me too .

I tried disabling automatic driver installs in Windows update.... and it's still doing it. First issue I've really had with Windows, but it's not exactly their fault the BIOS is trash.
Not sure if I should call back in to support to complain about it again, but that doesn't seem to have gotten me anywhere before. Can't really take my mobile PC with me anywhere if it's going to be whirring away all the time, would be kind of awkward in meetings.

I can't imagine this helps battery life either: 2:45 estimate with the 90whr doesn't seem great. I was getting that with the 60whr that was in my first M15.

 

I don't think that's all the fans, check to make sure Optimus is working correctly and that the dgpu isn't always on.

Sent from my Moto Z (2) using Tapatalk

 

Ah you're right, another bug for 1903, everything is launching on the dGPU.

 

It's not present on mine, I assume you just have to open control panel and redo the launch settings for your apps through Nvidia control panel.

Sent from my Moto Z (2) using Tapatalk

 

Yeah.. I would have my laptop sent back to HIDevolution sooner or later.. once I get things sorted out with the local Bureau of Customs in my country.. These guys just keep giving me the run-around and just an overall HARD time trying to get Import Tax Exemption... >.> The most corrupt government institution in my nation living up to their reputation... >.>

 

You can try disabling "BIOS updates through UEFI capsule update packages". I had no idea this was a "feature" now, it seemed to finally stop Windows from auto installing the newer BIOSes. Fans curves are still wonky so I'll try BIOS recovery tonight.

 

I was able to downgrade bios with no issues and now my fans idle. I asked Frank on Twitter about this and he prompted me to let Umar know so I think they are working on a fix for a future bios soon

Sent from my Moto Z (2) using Tapatalk

 

Has someone tried to seal the gaps between the fans and the fins? Now that the R2 manual has been released, i see that those gaps disappeared.

Sent from my SM-G975F using Tapatalk

 

same for me... is it possible to install the 1.4.1 Bios with the provided .exe file form DELL?

 

Could you clarify what gaps you are referring to exactly (maybe draw on a picture or something) ? Why would they need to be sealed?

 

I have and it didn't really improve temps by anything, maybe 1 C less.

Sent from my Moto Z (2) using Tapatalk

Yes, it works

Sent from my Moto Z (2) using Tapatalk

 

Hello, i tried it and it works, thank you!

I got a message that it is not recommended to switch to an older bios, but I ignored it and it works

 

Hey anyone here with the 500nits 100% adobe RGB 4k display option? Can you check what the model number of your panel is on hwinfo, I believe Dell sent me an m15 with the older 100% sRGB 400nits display, just wanted to confirm my suspicions

 

8950HK in an AW 17 + LM makes sense. Too bad Frank Razer killed that and AW15 off At least there is the A51m in the 17" department, for those who don't need much battery life, but now with m15 R2 they no longer have a viable portable 15 inch model.

Yep, def getting some mileage off your LM!

 

Only crappy screens available for order on R2.

 

How fitting!

 

in the R4 it was needed already, and the 17 inch chassis is quite bigger than the 15 inch. the 15r3 throttled way sooner than the 17r4, same as the 15r4 against the 17 r5. Sadly AW went Off track and went full retard oh i mean Apple on a "gaming" laptop. they won't get my precious $$$.

 

so, we now have confirmation from Dell that the UK will not be getting the M15 R1 with i7 9570h but will go straight to R2...

 

There is very little practical difference between 8750 and 9750 anyway.

 

Agreed. On paper it looks like it's a decent spec bump. But in practice it doesn't amount to much. With so many of these laptops power limit and thermal throttling it's a much better buy to grab a 8750h laptop at a nkce discount.

 

In all honesty guys, I think this product refresh is in wake of the new Intel 10th gen that will be 10 nm. It will produce less heat and have more power per Watt. I'd say buy an 8th gen m15 with numpad OR wait until 10th gen with new legend design Alienware. I don't think 9th gen will be worth

Sent from my Moto Z (2) using Tapatalk

 

Yeah, pretty satisfied. Did a run, increased long TB to 80W and SB at 90W with XTU, no throttle and temps are still good. I consider this 8750H hot temper tamed (by the way, i keep it at 45W for gaming, don't feel like more of that is needed) Still, i get this kind of core differential only in CB20. On games, the core differential is within 2 - 4 degrees °C

 

Still hotter than the one Ultra Male got before he returned it http://forum.notebookreview.com/thr...iew-by-ultra-male.828258/page-2#post-10892840

 

Cool! so i have space for further improvements

 

Yeah, I hope so Because Ultra male wasn't happy with the thermals. One of the main reasons he returned it back

 

well, since is not overheating nor throttling, it runs every game i throw at it at max settings (the very reason why i purchased it)...even if i leave it as it is now, without trying to improve it further, i don't plan to return it or to sell it. Will see what will be available in 2 years from now, and then will see.

 

You mixed it up bro, he had 2 m15s. The thermals above are from the 8750 one with LM (still a bit unreal TBH) . He returned the second 8950HK unit which he tested stock so sure enough he wasn't happy with the thermals in CB20, the LM benchmark

Of course, your temps are fantastic.

Same here, I only experience significant core diff in CB20. I can live with that. One thing I can't see on your screenshot is the CPU Package power. With around -160mV UV I wouldn't expect to see more than 65W so pushing TDP above that has no effect, which is fine - as you said, running the CPU at much higher than standard TDP might not be ideal despite low temps.

 

there is one big difference which I care about, and that is the HW protection against most of Spectre and Meltdown variants

 

Not something all the recent microcode patches have already taken care of, e.g. https://support.microsoft.com/en-us/help/4465065/kb4465065-intel-microcode-updates ?

 

Yes 8750H with discount maybe worth to consider. But, personally if I’m on laptop market now i won’t get last gen cpu if new one is already out. It was like skylake 6700Hq’s to kabylake 7700HQ, better performance and more power efficient, even on some gaming may be not noticeable, but cpu hunger titles like SoTR you’ll enjoy the extra boost (maybe also extra heat lol ), but hey this’s forum for repasting and tweaking masters right. Power limit throttling would be “cracked” by users here, like @hackness on this MSI forum, @David Kirchik on this Razer forum etc.
This maybe not fair comparison due different brand but we can see some advantages having new cpu.

 

Yes, you should have the same level of protection, just performance will probably be not the same and I'd rather have as much implemented in HW as possible and leave microcode fixes to future issues

 

Microcode is hardware my friend.
https://en.wikipedia.org/wiki/Microcode

Some reading material regarding the 9750H's improved performance:
https://www.notebookcheck.net/9750H-del-vs-8750H-vs-9750H_11343_9576_11352.247596.0.html
https://www.notebookcheck.net/That-...tperform-a-Core-i7-9750H-laptop.421361.0.html

 

You won’t get a proper fix till Intel makes a hardware change/architectural change to how SMT works. Unlike AMD’s SMT implementation where the cached data for parallel threads are isolated from each other I think the issue is that with Intel’s Hyper Threading the parallel thread caches aren’t isolated from each other which is what’s making the situation worse in terms getting a proper fix.

 

Microcode update is a software patch to modify hardware-level instructions, not a hardware fix.

For more info on the protections implemented in the last revisions, see
https://www.intel.com/content/www/u...ngineering-new-protections-into-hardware.html

I'd choose a HW fix *even if* that would mean losing 5% performance (because the CPU itself is less efficient for other reasons, or it throttles more, or whatever).

Furthermore, hardware solutions are generally more thoroughly tested than any solution that "we can fix later, if the fix turns out to be suboptimal"...because of obvious reasons. You can usually still also fix a hw mistake with more microcode updates but...you know what I mean

 

Wow that’s 23% performance boost on Razer Blade cb r15 lol.

From that notebookcheck result, if doing math from this man LM and bios tweak translate 9750H can get 1484cb

Lol

 

Minor semantics, personal biases and preferences notwithstanding, the main takeaways from the Intel doc are clear:

Q1. Are there any differences in the level of protection provided by software mitigated and hardware mitigated versions of these SKUs?

A: No. We expect that the level of protection equivalent whether you have microcode update (MCU) based or hardware-based mitigations in place. The hardware-based mitigations are part of our ongoing commitment to advance security at the silicon level.

Q2. Are there any differences in performance between software mitigated and hardware mitigated versions of these SKUs?

A: For application based workloads, representative of typical usage, such as SYSmark* 2014 SE, PCMark10, WebXPRT 2015, and 3DMark Skydiver Physics the data confirms that the performance between steppings is the same within the normal run to run variation. For some synthetic I/O workloads, we have observed a performance difference between steppings. These synthetic I/O workloads are not representative of mainstream usage.

 

To be honest, I don't understand your tone.
I am aware of the performance differences and how the fixes are implemented, and I formed my opinion after researching the subject and also based on Intel's docs.

The protection level is the same, the performance differences are *usually* minor, unless you do a lot of I/O.
If you do a lot of I/O, then I bet you will see the impact, and that is confirmed by Intel.
That impact probably applies to the HW fixed silicon too, I would not be surprised. However, I would expect the perf decrease to be smaller.
I agree with @Aivxtla that a HW redesign is needed, and that will come in the future, hopefully
A good read: https://www.anandtech.com/show/1365...ith-spectre-and-meltdown-hardware-mitigations

I still believe a HW implementation of the fix is preferable even if for any unforeseen reason it were to lead to a lower perf and I can't see any argument pro 8th gen (beyond the price), but I respect everyone's opinion.

You don't think the HW fix is worth the higher price? That's totally fine, I respect your opinion.
My opinion is that it does (if you're buying a new laptop) and I provided a few arguments pro that.
Does it justify an upgrade to 9th gen, if you already have 8th gen hw? Probably not.
Different people have different needs.

Let's agree to disagree, it's totally fine

 

Well sure. Plenty of people want the new shiny. Myself included. But sometimes you can see a few hundred dollars difference in price from one cpu generation to the next. So it's just smart to bring up that if you want a good deal the last gen cpu is very similar.

I personally take the new one myself but, I hate be out of date.

 

I simply reject smoke and mirrors. There is no concrete argument in terms of strength of protection against Spectre and Meltdown, nor any benchmark results confirming a significant performance advantages of the 9750H. If someone offered me an exchange for the 9750H free of charge I would pass because it's unknown whether it will undervolt as well as 8750H does.

 

That's totally fine, you'd rather have a higher undervolt over a hardware fix.
It's completely understandable and I might agree, depending on how lucky your unit is

In general, I'd choose the hw fix, there is no faking the fact that the protection is implemented in hardware. Unless they are just lying, of course

 

If you look at the table of various exploits, most fixes are still software + MCU even on the latest processors, so, as @Aivxtla said, there really doesn't seem to be much to fuss over in terms of "hardware fixes". It's still the same architecture.

Based on other user's reports here (small same size obviously), I'm just slightly above average lucky at - 160mV.

 

Indeed the HW protections do not cover all variants yet, as initially said, and more vulnerabilities will be discovered in the future, I'm sure. One just gets the best he can at that point in time.

 

Not sure what you mean with I mixed it up... The results if from one of his m15 with the weaker i7-8750H who performed better than the $600 Mobile i9 Scam from Intel. LM or not doesn't really matter. All I said it run colder. And for the records... He returned all of them None of them had good enough thermals! Here is the results from the more expencive and so called faster i9 mobile.

Of course the unlocked mobile won't perform much better in the AW m15 chassis. Dell use same heatsink for both the locked down i7-8750H and the unlocked i9-8950Hk Dell advertice with 5.0GHz overclock profile. Amazing engineering.

 

I meant the truth is somewhere in between, Intel’s mitigation’s aren’t a true fix due to the architecture which does result in negligible differences for the average consumer post “fix” but under certain io/virtualization enterprise conditions it can result in a decent loss in performance. At the same time not sure if the current soft fixes themselves have holes and have truly addressed the issue completely rather than “mitigate” to an extent.

I’d say as an average consumer myself it shouldn’t have too much of an effect in daily use. If you want full hard fixes you might not see them till gen coming after Sunny Cove.

 

Ah, so max 75C in CB20 wasn't good enough? LM doesn't matter at all (I thought you guys just argued LM is awesome)? OK, I doubt that was his thinking but no more questions lol

I was just making the rational assumption that he returned the spectacularly (for the chassis size) cool 8750H because he was hoping to get more performance out of the 8950HK, and to be fair he likely would have using LM.

 

I meant Liquid metal doesn’t matter in this sentence (why he returned the laptop). Nothing beats Liquid metal if the cooling capacity has headroom to remove more heat (The heatsink grills still not fully saturated with heat - See... can remove more with better thermal paste).

You can’t remove more heat if the cooling capacity is at the border and can’t manage the tasks. Even heatsink soldered directly on die with perfect fits wont help. Nothing is stronger than the weakest link.

If Dell’s engineers have done its job properly they had designed better cooling for the unlocked chips.

 

I think it's funny how every post I've seen from this user is always negative about dell. It's not that he is wrong, he is right but it's funny cause Everytime my phone notifies me theres a new post from this user I know someone tried to be positive about this laptop XD

Sent from my Moto Z (2) using Tapatalk

 

Poopoosan is an X Alienware fanboy who has been betrayed by Mr. Azor's decision to go the BGA route and offer crippled hardware (compared to the past beastly Alienwares), thus, you will NEVER see his username in a Dell/Alienware thread with anything positive to say. Just take a look at this lovely avatar .

This is the ugly truth my friend.