*OFFICIAL* Alienware Area-51M R1 Owner's Lounge

If I want to use ThrottleStop to undervolt, do I have to do anything with Alienware Command Center other than turn off the OC profile in there?

 

Nah. Same PSU they've been using for many years now.

 

Disabling OC in AWCC is all you need for being able to use ThrottleStop or XTU

 

Thank you for sharing all this information. That info makes me wonder if Dell engineers have been able to figure out what's the root cause of the burn issues and how to solve them, but I guess they still have no idea... Also you mentioned Dell had an action plan, could you share that information with us? Thanks

 

Theoretically, yes... I think so. But, they would still have to enter into some type of licensing agreement with NVIDIA like the desktop AIB vendors do. I am pretty sure they could not just arbitrarily decide to start making laptop GPUs and do it without jumping through some hoops first. And, in the case of Dell's proprietary form factor, which is likely patented, they would need to do something with Dell to produce aftermarket parts for the Alienware Area 51m. I'd be shocked if Dell would ever agree to that. But, for MXM that might not be quite as much rigmarole.

That being said, since Clevo, MSI and ASUS are all marching to the beats of different drummers, that could be an issue. None of them are adhering to any kind of standard now that the Green Goblin is AWOL with respect to MXM standards. A venture of this sort probably would not be very profitable because there would need to be too many variations to accommodate the plethora of stupid MXM form factors we see today.

Laptops have always been kind of a pain in the butt, but never so much as they are today. The good old days when you could easily buy a Clevo MXM card to upgrade an Alienware, or vice versa, have unfortunately come to a sad and bitter end. This is one of the reasons I abandoned the idea of owning high performance laptops almost 2 years ago. It's just not worth the hassle any more.

 

https://twitter.com/Nomadsan/status/1131339162974605312?s=19

 

I removed the AW OC controls app and I lost access to the fan settings in AWCC, that's weird. I had to install it back just to be able to regain control over the fans from AWCC app. That sucks as I don't want the OC controls to suck resources from my system, I use TS for undervolting and OC and the AW OC app clearly relies on XTU as it installs its framework, services and there is a process called xtu running and sucking resources. Is there any way to get rid of this app without loosing control over the fans from AWCC?
By the way, I'm on latest version of AWCC.

Thanks

 

None at the moment

 

Everybody wants to be a benchmark champ. For me, that is absolutely and utterly pointless and borderline ridiculous. The sky won't fall if your machine performs 5% slower or 5% faster than someone else's. Use the darn thing to do what you bought it for i.e., play games! And while doing so, it's one's own duty to look at thermals and make sure the thermal pads are making good contacts. I have noticed that there are minor differences between heat sinks where some can take 1mm thermal pads perfectly while others need a bit thicker, say 1.5 or 2mm thermal pads. 1 size fits all is a little tricky in this case and needs extra vigilance on behalf of the machine owner.

I am talking about the nvidia chips, not the DGFF boards. Please re-read my comment.

I feel that nVidia's GPU chips are behaving out of specs and there is little to nothing that Dell could do other than speculate and install beefier components just to avoid disaster, but there is a limit to even that considering the fact that not all components have the same dye size and package dimensions. They cannot re-design the entire DGFF at this point in time. May be they will have to do it in the next iteration. In any case, re-testing someone else's chips to ensure good QC is in itself a big pain in the rear end, especially when doing at such a large scale, and Dell will have to rely on nVidia's given spec sheet to design the board based on it. That's unfortunate but there is no other realistic way of doing things. nVidia has been caught delivering chips in the past that were eating up a LOT of power. I once had a 1080Ti that alone ate 375 watts on the wall (without ANY OC) which was insane! And that card didn't live long. The person who bought that from me told me a couple of months afterwards that he had to RMA the card.

 

Hi SK. Currently, I'm still worried about my own area-51m gaming gpu temps at stock( without undervolt).
With balance or performance fan profile, I got about 86 ℃ throttled in graphic intensive games with 1800 mhz 98%u sage
With max performance, I got 83-84 C
Ambient 20 c
Could you do me a favor to test both conditions with your machine at stock? A curve plot would be much appreciated.

 

I never run my machine without undervolt because there is no point in giving more than needed voltage to the components. When I do, my GPU temps peak out at 84-85 degrees Celsius with the gpu hovering between 1980-1995 MHz (fans at full speed). I will highly recommend that you spend a little time with MSI Afterburner and create a nice set of profiles that are stable and prevent your components from baking up unnecessarily. With my normal gaming profile (937mV at 1950 MHz), I am crushing everything with max 76-77 degrees GPU temps with full speed fans. When I switch to balanced, it touches 81-82 mark and with performance fan profile the temps are around 79-80. I also run an undervolt on the CPU (-170mV at stock clocks, sometimes OC with -95mV) and it helps keep the heat profile down. This is important because there is a shared heat pipe involved here.

 

I always had temps under 80 with full load, and mine just passed away lol. The sad part is they wouldn’t let me do a full refund from HID.

 

When I say "temps", I talk about temperatures of ALL those components that DO NOT have temperature sensors monitoring them. That "GPU temperature" is just the temperature of the nVidia chip. Memory chips, mosfets and VRMs also need proper cooling and you can only ensure that with an accurate thermal pad job. If you haven't opened it up once, you are absolutely unsure wth actually went wrong. GPU and CPU temps alone mean nothing.

 

i saw someone here got only 69c max doing firestrike without undervolt. I am afraid my gpu is als a dying one

 

No it sounds perfectly healthy to me. Just have to tame these things. And I do not buy that person's temperatures at all. 69C is simply not possible on an RTX 2080 with a good enough load. That was either a typo or a purposefully created wrong piece of information to boast about it.

 

alright, thank you for you info

 

When I load windows in safe mode and disable the 2080 Then reboot . The computer runs just fine.

 

You could have a very lousy thermal pad job inside and it could be possible that one (or more) thermal pads are just not in the right position or have slid aside a little bit from their desired place due to being undersized. I'd open it up and see if all the thermal pads on the DGFF board are making nice and tight contact with the heat sink. If not, that could very well be your problem. A dead card will not boot at all. It could be shutting itself down to prevent permanent damage to one of the components.

 

Well, the flaw in that rationale is not understanding that there are people that intentionally buy (or build) a system for the express purpose of overclocked benching, and gaming is just a nice fringe benefit. They are as passionate about it and derive as much pleasure from it as gamers do from gaming. And, if you cannot do some really insane overclocked benching with it, owning it is totally pointless for those folks. It's only ridiculous if you're a gamer that doesn't care about overclocked benching. Thankfully, there are Clevo monsterbooks with @Prema firmware available to meet that expectation. We were kind of hopeful there would be a stronger contender in the Area 51m, but it doesn't seem to have worked out that way. Although it is worth recognizing that this is the first notebook that has given Clevo a respectable challenge since around 2013, when the P570WM dethroned the mighty M18xR2. Between then and now, everything Alienware had on offer was rather laughable if you were an overclocking enthusiast. Hopefully, they'll have an Area 51m R2 that gets the job done without the inordinate gimping, clock-blocking or catching on fire they have been famous for.

 

Well in that case, one has to be somebody who understands what are the consequences of benchmarks if your machine is not ready for it physically. I wouldn't even think about benching my machine if I am not sure what thermal components it is running and how good is the installation. Benching on machines that come straight out of the factory is pushing your luck a little too hard because those machines are generally not carrying the best thermal pads and TIMs. I wish that changes going forward but for now that is not the case and extra vigilance is needed on behalf of the owner of the machine. I agree that some people live and breathe benchmarks (which I personally feel pointless lol, but that's my opinion and i'm fully entitled to having one), but they do their homework before giving their machines a beating. As a gamer, I bench my machine just once or twice to see if it is stable, after doing all the mods. But if all you do is benching to share your firestrike scores on twitter then you will run into problems, especially if you are doing so without any thermal mods. I am willing to put any money on that.

 

Yes, absolutely. Everyone is entitled to have an opinion and whatever that is they should be passionate about it if they want to be taken seriously.

Catching on fire probably doesn't have anything to do with overclocking or running with a higher than spec power limit. I suspect it is an engineering defect. They were catching on fire--albeit in fewer numbers--back in the M18xR1/R2/AW18 days as well, so that part hasn't changed that much other than frequency of occurrence seems to be far greater now. And, back in the day it wasn't due to overclocking when they did... just defects from shoddy engineering.

 

Hey @Mr. Fox , I've seen your videos about the Eurocom 700W PSU. I know that the peak to peak ripple voltage on that thing is a minimum of 200mV! How is that thing holding up and have you experienced any coil whines with prolonged usage of that PSU? I consider 200mV ripple to be a silent killer and don't like anything higher than 50mV on my desktops. 20mV or lower is ideal.

Could be nVidia chips running out of specs as I mentioned in my previous comments. I have seen multiple times when nVidia's graphics chips are drawing more power at stock speeds than they are supposed to do according to the datasheet. And if a manufacturer relies on their provided spec sheet to design their circuit for the chip, they could eventually run into this kind of problems because when you rely on specs provided by the manufacturer of the chip, you are essentially trusting the part that this chip will not exceed this given voltage threshold and max instantaneous spikes will not exceed a certain given limit. This is Dell's first go at designing their own proprietary GPU solution so they'll eventually iron these glitches out. But for the first try, I'd say they've done a pretty awesome job with it.

 

I never had any real issues with anything related to Clevo. Sure, a random dud GPU (which isn't unique to Clevo), but I don't own any high performance laptops today. I moved to desktops about 2 years ago because I did not like the direction the industry was headed. So, I cannot speak to the longevity of the 780W PSU. I haven't heard anything about people having issues with theirs. When I moved back to desktops, Dell/Alienware was no longer offering anything I could respect. Neither were MSI or ASUS. And, Clevo was pulling too many shenanigans with musical form factors. I decided enough was enough. I should have done that a couple of years sooner. Couldn't be happier to have kicked the notebook habit. My only laptop now is an extremely well built, but modest (borderline pathetic) performing HP ZBook 15.

Yes, it could be that. But, since we are speculating I would be more inclined to believe it is an engineering defect or a production QC issue since that would be more consistent with their history. We haven't seen this with NVIDIA chips on the whole. But, there will always be a few stray examples like what you experienced. Given the insane volume of GPUs (mobile and desktop) that are sporting NVIDIA processors, I'd say they have a truly superior track record and it would be too odd of a coincidence for all of their defective chips to end up in Alienware Area 51m notebooks.

I have modded my NVIDIA GPUs on purpose to force them to run out of spec and they haven't balked in the least. My 1080 Ti that I had before is still kicking and being pushed almost daily to more double its rated power limits, and both of the 2080 Ti cards have held up marvelously to handling almost triple their rated TDP without so much as a whimper. The Clevo 980M, 980 and 1080 notebook GPUs I used to bench were also pushed to around 300% of their rated TDP almost daily without ever missing a beat. The 780W AC adapter did a nice job of accommodating that with the Clevos.

 

The big problem with this... No other of the ODM's models run into smoke and fire. Put the blame where it should be... On Dell for this "smoking" problems!

 

Yes we are speculating But that's all we can do with so little information available at hand. I'd still give Dell props for pulling it off in the first try. So far they've been sticking to nVidia's provided design sheet for the GPU circuitry and all their engineers did was port those schematics into their motherboard design layout and call it a day. But now they have to do all of it from scratch and I'd say the failure rate is still not that high. Going forward, I want Dell to continue creating higher variants of DGFF boards. That will tell nVidia that you can't just go ahead and kill a standard just because you want to. Also a healthy competitive trend will begin regarding who makes the best modular GPUs for laptops which is what I'm interested in for the future.

 

Yup, I totally agree with you. Everyone word in the quote above is spot on.

Yes, this is why I am reluctant to say it is NVIDIA's fault. It may not be Dell's fault either. It may be the company manufacturing the GPUs for them. But, it's got their name on it and the situation isn't good for the Alienware brand. They need to put this in their rear view mirror, pronto.

 

Can someone please post a MSI AB curve shot for undervolting the 2080?
I just realized now that pulling the points down actually does the opposite of undervolting I still might be the first user on the forum to manage to burn his GPU by his own actions.

Is -125mV a reasonable offset for the GPU? I don't have a ton of time for benchmarks and stability checks right now so I'd like to start with the popular vote

Is there a way to offset everything without pulling each Freq/Voltage graph point manually?

Thanks

 

Bro, you keep defending Dell/AW like you've a stock invested there or something like that.

Look, this are the hard facts of this situation:

1. No one knows for sure why this thing is burning out, not even Dell/AW who are supposed to have designed the damned thing for God's sake. It's not nvidia fault here, other 2080s have had no issues with this, not even the ones with 200W rating. And it's not just the card burning, like Donald confirmed above, even power inlet is at fault here. It's just a downright rushed product and were the Guinea pig here.

2. Alienware themselves have designed this system to be overclockable and have included overclocking by default in the Bios and their proprietary software ACCC as well. I've seen settings in AWCC where you can set the 9900k at 5.5GHz by AW themselves! Why would the end user not overclock the system when it is advertised and marketed as such from the very outset and we've paid $4,000+ for it? Your argument makes no sense to me.

3. As you've read many posts in this forum, this system is crapping out "even at stock clocks" and while idling at desktop or watching a simple "game of thrones" video, do you think the end user is at fault here or may be this "beast" couldn't handle the intense fight scenes?

Don't get me wrong, my system is in transit as we speak and I chose to buy it in spite of Donald advising me not to and wait for it to be fixed before moving forward. I want to see first hand what the fuss is all about. I want to run my processor at 5GHz on all cores and my RTX 2080 at 2Ghz for every single title I play and do the same every single day. Why, you may ask, because I've bought a "overclocking desktop replacement with desktop parts" and it was because of this expectation I'm shelling out $4000 for it. I'll stay on my pursuit no matter how many times this sytem craps out and how many replacements they'll have to send me. I know it's only a matter of time before they find the culprit and iron it out. Am I expecting too much? No! Because this product is advertised and marketed as such and that's why we bought it in the first place.
You keep taking defensive stance for Dell in this forum and keep blaming the end user for "overclocking" and what not, deflecting the blame towards nvidia who had no part in this whatsoever and refuse to accept any wrongdoings on the chief culprit here, Dell/Alienware themselves.

I do not understand your logic. Not even one bit. Open your eyes mate and hopefully you'll see it.

Just my two cents.

Edit: typo

Sent from my SM-G965F using Tapatalk

 

This video should help you get that sorted. Rather than trying to copy someone else's voltage curve, this should help you learn to get it dialed in to find the ideal spot for your GPU undervolt. But, it probably won't fireproof your system.

 

Thanks a lot
I actually saw some other clips but this one is a lot clearer on the whole process.

 

I

If you read my posts, I am NOT defending dell like I own Dell. All I am saying is that this world is not black and white. Every company does good things and bad things. Ignoring what they are doing right and only bashing what they have done wrong will not do it well for us as consumers because we want modular notebooks. They are going to be perfectly happy to sell us a shell that is not upgrade-able and you have to buy new stuff when yours get outdated as a complete package. I would not bicker about a burned card because Dell doesn't tell me, hey, you have to buy one more. They are actively helping and replacing them for the customers who had a problem. I have a background of circuit design and I know, first hand, what a bad batch of components can do to your product line. My way of looking at things might be a little different from the majority out here who just buy and use their stuff without understanding what is actually going on and what could be the underlying possibilities of a hardware failure. Bottom line is this:

If this is a design failure, it should happen to each and every single laptop out there indiscriminately (which is clear at this point that it is not, especially after Dell upgraded power components). So, design being wrong part goes out the window right here. The only possibility that remains open then is a bad batch of components whose QC is supposed to be tested by the manufacturer. If I am designing a circuit, I will not QC test each and every chip that I buy form a renowned manufacturer. That will be utterly silly on my behalf and the cost of bringing that product to the market after that will increase 3-4 times the actual cost of components.

 

Awesome, glad it was helpful. At the time I made the video there were lots of people with severely overheating notebooks that needed to be undervolted and it seems to have been very helpful to a lot of people. It makes me very happy when I can do something to help people like that. It's one of the more popular videos in my YouTube channel.

 

U passed ur refund window or still within ?

QED!

But u gotta give it to “Drogon” , it can burn anything from anywhere , 51M can’t even survive that .... lol

 

Quick question. When checking GPU Z, it says my card is running at PCIe x8 3.0 when it's capable of running at PCIe x16 3.0.

Is this correct? When I hit render, it still stays at x8 3.0. Is it supposed to run at x16 or no?

 

x8 only, AGA holds the other x8

 

Ahh okay. Thought so, just wanted to be sure. Thanks.

 

Within but I would charged a restocking fee lol, on a device that doesn’t work.

 

Just make sure you guys get as much warranty as you can afford as it is a $4000 unit after all, keep a spare web-surfing device in case your system craps out and you feel the need to curse Azor and his bandwagon online and vent your frustration and then when you're all set; PUSH this laptop as much as you possibly can. I'm sure by the end of your warranty and after many revisions, you'll have a very capable system that will get upgraded by Dell themselves along the way.

And then when you finally get a fully capable machine which you know is a gem and keeper, extend your warranty again and use it for the next 10 years! Follow this Mantra and you can thank me later.



Sent from my SM-G965F using Tapatalk

 

how much is the fee ?

 

I think it was 5-15%

 

Hey guys... so the IT dept at work decided that they would reset the BIOS and then lock it with a password, pffft.

I can't even boot to my SSD because I can't touch the boot options. I've asked them to provide me with admin access since I'm a power user and not some chump... but if they decide not to... is there a way to reset the BIOS without a password?

 

10 years? Wow that's a long time! I'd be replacing this within the next 3 years.

P.S., I am expecting a response to this post: http://forum.notebookreview.com/thr...-owners-lounge.826831/page-1154#post-10914178

Take the battery out and unplug it. Underneath that you will see a tiny 2 pin connector that connects the battery cell to the motherboard. Unplug that for 10-15 minutes and then plug it back in. The bios should be cleared (I'm not sure if they store passwords on a non-volatile flash though so will have to try it).

 

Not all of us can afford to replace the system every three years brother, that's the sad truth!

May be that's why you're so chilled out with the whole debacle?

I've used my m15x and M17xR3 for 8 years! So believe me when I tell you this, I will use this laptop for 10 years and buy just the warranty. Nothing else!

Sent from my SM-G965F using Tapatalk

I don't think a BIOS reset will clear the password with the CMOS trick S.K. mentioned. Worth a try but I think you'll need a password for sure.

Sent from my SM-G965F using Tapatalk

 

Sure, but I wouldn't go that far ahead to assume that somebody is having a stake in a company just because he doesn't bash them like everyone else is doing here. That'd be completely out of line.

 

That was just a joke so I hope you'll take it like one.

But there is nothing you say here that will justify Dell's QC and lack of comment on this whole saga. I'm not bashing Dell at all, in fact, I just bought the damned thing even after all the issues it has. You know why? Their warranty outside of US is just unbeatable. I know I can rest assured knowing whatever happens with my machine, I'll expect a part replacement withing 48 hours of creating a service ticket as long as I'm covered under warranty. That's why I bought it! Also, I've been waiting for a BGA-free desktop replacement for many years now. I had been bothering Frank in his Twitter DM for 3-4 years now for such specs. You can check my post history. I had created a thread sharing our conversation and none of us believed it at the time that AW would do such a thing because their focus was only on "thin and light". There are mode AW fans here then you probably realise but most of them have got fed up and moved on to something else. I an one of the last remaining AW supporter here.

However, if their warranty was like any other manufacturer, they can take this "fastest laptop in the world" non-sense and shove it down their own throats because at this point of time regarding this system, we are all sitting on a $4000 ticking timebomb and that's a fact!

Sent from my SM-G965F using Tapatalk

 

Well I'll agree to that when my system blows which I know it won't. But as you said, we agree to disagree. And I agree with the warranty part completely. Nobody beats Dell in warranty services, especially none of the "gaming" laptop brands out there.

Could you guys do a quick test where you measure how well different heat sinks make contact on both the CPU and GPU components keeping the thermal pad thickness same across all the test machines? I have observed that my old heat sink took 1mm thermal pads on DGFF memory chips but the new one takes 1.5mm and if I stuff in 2mm, it could even take those since the gap is larger. Since you have a larger set of laptops at your disposal at HIDevolution, you could actually see if we have varying clearances between different heat sinks and this is a heat sink QC issue that is leading to these failures? I'll be keenly waiting for your response.

 

I tried as follows with the revised heat sink and gpu i believe.
Clean all the thermal paste, put a piece of paper inside the gap between heat sink and the die, screw down. The paper cannot be pulled out in that condition, so I believe the gap is narrow enough for good thermal paste

 

I referred to thermal pads, not thermal paste. Not talking about contact to nvidia chip or CPU here.

 

Bro, I think no need to argue against each other both of you
From my point of view, you stand here more like a true notebook enthusiast or pro caring more about issues and solutions.
Vodoochild just as a normal customer paying more attention to the value, pros and cons of a certain product .

Haven't measured that yet.

 

100% right.

Yes, you guys bought version 1.0, or as it turns out, version 0.9 alpha.

Now hold on, this is how much I paid for the AW17R5, so the A51m is not that far off. I am also have same/similar expectations as you, but at least my AW17R5s do not catch on fire while watching Netflix

Dell ought to have done a mass recall, especially with the fire issue. It could be construed as criminal negligence if your house and family burn down to a crisp because of this and Dell knew all along.