*OFFICIAL* Alienware X Series Owners Lounge and Discussion

If Alienware pulls the same move on this more as the Area 51 m1 and 2 aftermarket ram may not work well.

 

I have an m17 r4 and its fans are not heard in games and tests. you just had to choose a quiet profile for fans in AWCC
look, in office mode and when browsing the Internet, the fans do not spin at all

 

For those that know - I briefly had a Razer Blade 15 Advanced 2021 edition earlier this spring. One thing I found distasteful was that Advanced Optimus forced the display run at 60hz when on battery with no choice for HFR. I am savvy enough that I tried to see if there was any setting to drive HFR via the iGPU but it did not change anything. I'm wondering if that was an early BIOS or driver issue, or if indeed the cost of advanced Optimus is no HFR while on battery and using the iGPU (of course you can always manually disable and use the dGPU, but then battery life blows). This may be a deal breaker for me on the X15 as I wanted the 1440P display. Anyone know? Advanced Optimus is still in infancy.

 

Isn’t that true for most motherboard vendors? I’ve always liked Asus and Asrock motherboards but they don’t always work well with certain brands of memory. Memory is usually far more finicky then any other component so I go by the manufacturers recommendations when choosing. But so far the only SDIMM I’ve found online for DDR4-3466 has been Dell.

https://www.dell.com/en-us/work/sho...sodimm-3466mhz-superspeed/apd/ab640684/memory

 

TPM Angst. So not an issue with the X series, but I don’t like it when these giant software companies force more control measures into their code. Waiting for the girl with the white rabbit tattoo to show up on my doorstep.

https://www.tomshardware.com/news/windows-11-leaves-computers-out

 

The gpu temps are pretty toasty ngl

 

HAH! The hottest I’ve ever been was near Ballad Iraq in August 2005. 59C in the shade. 70C is hot but these silicone chips should be able to push 80+ as long as it’s not for extended periods. Just don’t keep it on your lap.

 

I already mentioned and showed toasty temps in May. And the FPS is capped at 60 in the youtube video.

http://forum.notebookreview.com/threads/official-alienware-x-series-owners-lounge-and-
discussion.835993/page-11#post-11098157

LOOL

 

Indeed, there are some old online references to G.Skill 3466 SODIMMs

https://www.guru3d.com/news-story/g...4x16gb)-so-dimm-kit-at-ddr4-3466mhz-cl17.html

however, if you look at G.Skill website, the only SODIMMs faster than 3200 are 3800 and 4000 (and those are EOL).

https://www.gskill.com/products/7/2/197/Ripjaws-DDR4-SO-DIMM

Either Dell pulled a 180 degree turn to become the industry leader in high speed laptop memory or it's another marketing ploy. In theory you could get 3800 SODIMMs (if you can find them) and they should downclock to Dell's 3466, however, you know where my money is on that actually working

 

These might work? 2x16 and 2x32 F4-3200C18D-32GRS / F4-3200C22D-64GRS

 

Slim to no chance sounds about right. Laptops have their use but unless your buying one that uses desktop components, then overclocking becomes very limited. I recently read a review for one like that on tomshardware. It ranged in the $3000- $5000 range depending on options, but the storefront is in Europe, and you kinda have to ask yourself why you’d buy something like that when you could buy an similarly configured Alienware PC for much less. Your paying top dollar just to have desktop components stuffed in a laptop frame. If you want portability buy a laptop, and when your at home use a desktop. I prefer bigger screens anyways.

 

That's just 3200 CL22, rather than 3467, right?

https://www.gskill.com/product/2/197/1540866339/F4-3200C22D-64GRS-Overview

 

It barely took about 10 seconds to get to 80C on the GPU even with Vsync turned on. This thing is gonna thermal throttle hard out of the box, on both the CPU and the GPU. Now what will Dell write on their support page of X17? “As long as the GPU is running above the base clock, it is fine. It doesn’t matter if TGP drops below 150W” I can pretty much see the trend, given how crazy people are for this X17, they’ll buy this real quick and show off their X17 on reddit on day 1. Few days later they’ll report 87C with GPU dropping to 120ish W from 150W. Anyone interested in the X17, mark my words. You have been warned. It is inevitable, don’t waste your hard earned money. Dell is sure to come with interesting ways to cover this up, and they’ll try every cheap move in their arsenal to deny any warranty claims arising out of this thermal design.

Honestly though, Dell hasn’t learned anything from the burnt Area 51Ms. Conclusions can therefore, easily be drawn on what to expect from AW. And what was it btw? Element 31, Quad Fans?

 

Nothing stopping you from using large monitors with a laptop. Such a setup could displace a desktop, for all but most performance-intensive tasks, but not if there are too many performance compromises.

 

2x32GB is 3200 CL22
2x16GB is 3200 CL18.
Not sure what Dell's 1x16GB 3467 is listed as but the user would need to buy two of them to get 2x16GB.

https://www.dell.com/en-us/work/sho...sodimm-3466mhz-superspeed/apd/ab640684/memory
These are 16GB sticks ( unknown CL value )

 

Only one way to find out about the Dell RAM....

The 2x16GB 3200 CL18 G.Skill would almost certainly be an improvement over Dell-supplied memory if it worked...

 

I agree, and most of the time when I use my latitude it’s connected to my 64” 4k Samsung TV that I sit 15 feet away from with a wireless keyboard and mouse. Also, that is another reason why I’m not overly concerned with heat.
If anyone hasn’t looked notice the max temp on the 3rd page. Unless your living in a hot climate with no AC I wouldn't be to concerned about hitting 100C. The thermal protection is pretty good on these CPU's.

 

It would be pathetic if production units can’t keep the GPU under 75C at 165W. The Legion 7 does it easily in a 15 inch chassis with no LM and “just 2 fans”. Lol

 

Direct impact of heat on the user is not the only concern, although if you are going to avoid it by not keeping the laptop on your laps, that begs the question why you care about thinness so much. There are other serious issues with the CPU running at 100C though:

* The most obvious one - the CPU is throttling!
* Since these thin laptops have a unified heatsink, the GPU tempersture is hjgher, potentially also hitting its much lower throttle point
* Temperature inside the chassis raises affecting the lifetime/reliability of internal components
* Fan noise, and the sound of small fans tens to be unpleasant

I know it's a cliche, but heat is the number one enemy of electronics (after water I guess lol). What's worse is that if temps are poor out of the box, there probably won't be much one can do to fix it, apart from standard undervolting and perhaps trying to remove e31 and replace it with a different LM (whereas the previous thicker AW laptops typically respond very well to repasting).

 

I've never seen any substantive studies about these assertions...I've only seen anectdotal stories in subreddit and forums from people who claim their systems caught fire or something.

1) Industry specifications for 3 year life of CPU/GPU is assessed...I'm not sure what duty cycle but it is assessed at Tj-Max (not the clothing store lol)
2) Yes, heat caused failures occur (and in many cases it may be because a part slipped through QM that couldn't handle it's own spec'd temp and power limits) - I have no doubt, but at what true percentage of total gaming laptops sold experience a thermally caused failure prematurely?
3) Yes, heat, over time may eventually lead to an EM failure in the chip - but again, this is generally calculated on an industry standard over time, the question is life expectancy of a product at specific duty cycles (which I have never seen what intel or AMD expects as far as duty cycle but I haven't really extensively searched for this - I may be working at Intel next year so I'll pass along what I learn at that time lol).
4) If you want to get 5 years out of a gaming laptop, then yeah, you may want to lower the temp to improve reliability over that span, and buy the extended warranty
4) How many failures were caused by exacerbated circumstances? Like gaming on a blanket or something? Direct sunlight? Incredibly dusty fans? People lie all the time - not saying that thermal failures don't occur, but again, see point 2

My problem is this - people panic over temps that are within industry spec - it's what we spend all our time over and I've never seen anything but anectdotal proof that running close to Tj-Max is going to cause premature failure of computer components (other than what would be within industry expectations for such issues - probably sub 10%?). Perhaps if you buy a thin and light and plan a 100% duty cycle (as measure by time at max frequency) then you probably have the wrong product for your specific need.

 

My thing is they sell these 4k laptops with the intention of these machines having failures after a year. Thats whats wrong with these companies like dell. They dont care for the customer only for the money. Look at Asus they may have horrible customer service but they make great products and they listen about what ppl want in their machines unlike deal. You would think dell would do this but they dont......a new laptop and they take away more and ask for more. Seriously 420 bucks for 2TB drive??? thats the price of a 4TB

 

@Papusan said it best awhile ago. . .Dell wants to be apple. And they’ll do what they have too in order to get there.

They want to sell us expensive disposable systems that cost an arm and a leg. Cost a liver to upgrade. And you can’t service yourself.

 

I am not sure who you think is panicking. Certainly not those of us who are planning on giving these "limitlessly" (another obvious marketing trick - the limits are painfully obvious) thin devices a miss.

Not clear what your point 1. is about? By reference to my first point, was that an attempt to argue with the fact that extreme temps entail throttling?

You do understand that lower temps are strictly more desirable? You can try to rationalise this fact away by saying that a bit of throttling is fine, scorching chassis heat is OK if the device is used on a desk with an external keyboard, or that reliability doesn't matter because there is warranty or you are only planning to use the device for a short period of time. The straightforward fact still remains: lower temps are preferable, and the difference is particularly important at the higher end of the spectrum mainly because of throttling.

 

The RTX 3060 mobile is the exception, it has more cuda cores than the desktop part. The rest are still nerfed compared to their desktop namesakes, particularly the RTX 3080

I don't care how thin they make a laptop if it can cool 300w of heat, it doesn't necessarily have to be a DTR to perform great. The Legion 7 2021 has the most capable cooling for its size this generation, it is proof that thin and lights can have good cooling performance as well. And we are far from the limits of cooling systems.

 

My point is that chipmakers assess a specific duty cycle for assessing warranties and expected life of a chip (to calculate MTTF specs).

While temperature DOES indeed play into the equation for assessing MTTF - it does not mean, in the least, that if you put a set of machines running at different temps, consistently for years that the hot one is guaranteed to die prematurely or earlier than the others.

Some discussion in this thread I found interesting - i couldn't get to the TI paper cited in this thread, but interesting debate there on conclusions.

https://www.reddit.com/r/hardware/comments/5l3ufj/is_there_any_actual_data_supporting_the_idea_that/

"The current generation of TI industrial grade embedded processor products is designed to support a useful lifetime of 10 year operating at 105°C junction temperature TJ. The 10 year lifetime assumes a worst case situation of 100% powered on and run at a constant 105°C TJ temperature."

I've been on the internet long enough to know the dialogue and the general panic that occurs in most places because of temps in the 90s.

Will a PC running under 80C while on load really last longer than one closer to Tj-max? My assertion is NO - absolutely not and we should stop worrying about temps.

Things that may be more important are thermal dissipation as a function of time - and oscillations in temp - if you use a gaming laptop with a 10-20% duty cycle in a given 24 hour period, and your system goes through big thermal swings (regardless of the max temp it reaches) that may be more the cause of failure than a 10-20C different in max temps (a 50C swing in 10 minutes verse a 65C swing really matter all that much between two machines?).

Also - if AMD/Intel, and other Semiconductor manufacturers decided to just leave performance on the table by artificially limiting Tj max to some other number - well they just won't do that. You don't think Dell and others are sending product back to Intel when it fails prior to MTTF expectations?

My main argument is that no, "lower temperatures" don't largely matter - it's far more complicated than we make it, but for the average user, they can completely ignore temps. If temps and thermal throttle in result in bad performance, that's a different matter (framerate spikes, latency spikes, etc) - that's where I've seen much better chipset level PL and TL controls over the last few years.

 

Higher temps don't guarantee a failure, they increase the risk. It's not just about the main GPU/CPU chips, but - and perhaps primarily - about all the other electronics in the laptop: VRMs, VRAM, caps, resistors, battery, PCBs etc.

https://www.semanticscholar.org/pap...arak/624204572a8b7ee7a1f3a69e5c5e300b44227778

Ultimately, this plays into the presumed agenda: to produce disposable notebooks in need of frequent replacement.

BTW you mentioned you enjoy keeping a laptop on your laps. It better not be slim and toasty since there is a chance this could be relevant:

https://www.webmd.com/infertility-a...08/laptop-computers-may-affect-male-fertility

 

Oh I couldn’t agree more about heat being the number one enemy for electronics. Although I’d like to add dust and sand as a close second. Speaking from experience after the 39 total months I fought in Iraq during my 3 combat deployments. But during my 8 year break in Service I worked as an engineer for AT&T Wireless. That was from the mid to late 90’s and for a single cell site outage back then every minute offline cost the company $70,000 in lost revenue. I can only imagine how much more it would be today. Take into account an outage could occur for a multitude of different reasons from the central office up to the cell site. A single T1 card in the 90’s cost $25,000. A T1 card US standard can process 24 individual channels and with CDMA each channel could process 7 individual calls. A cell site might have enough radios which process calls from a cell phone to handle 2 or 3 T3 lines. A T3 on a GSM switch could process 28 T1’s. GSM originated in Europe until 1996 when it was first introduced into the US by Sprint Spectrum in Washington DC. I was one of the 30 or so techs involved with that project, and it was the first digital switch in the US. Before that everything was analog and things like caller ID or voicemail for cell phones didn’t exist. I was hired a year before that network came online for the public and about 6 months after coming online I was poached by AT&T Wireless in Sacramento California to be a part of their transition with the same Ericsson switches. GSM technology also introduced the smart chip into cell phones and a few years later the banking industry adopted the tech into credit and debit cards. My older brother has his master’s in IT from University of Maryland and my BA is from Columbia. He might have more formal education but i have little doubt my telephony experience trumps his technical expertise. Telecom switches are far more sophisticated then their PC cousins.

 

Today my fully equipped x17 was shipped. Maybe I can give some insights next week

 

When I was first learning how to make my own power cables for PC’s I learned the hard way that using insufficient wire gauge for the current was a rookie mistake. I had a couple instances of smoke before I corrected that mistake but never a full fledged fire. To their credit this happened twice with an EVGA T2 1600 watt power supply and although I fried a couple cables I made it did not damage the power supply or the component it was connected to. In this case an Aquaero 6XT for controlling water cooling peripherals. The Aquaero allows several different types of components which some draw power through the Aquaero. I underestimated the power draw. The Aquaero is fed power from an old school 4 pin peripheral connector. And because of the distance between the Aquaero ond the power supply in my Caselabs case I had no choice other then buy or make a custom cable. Any gauge less then 16 like 18 was not enough. I had 37 smart fans attached to 4 rads drawing power through mine. At first everything was fine, but when I sped them up the current spike scorched the cables. Lesson learned and the only cables I make now that use smaller gauge wire are for things like USB or SATA drives. Here’s some of my work.

 

Looks like production is speeding up. Delivery times for the X17 are listed today for July 20th. The order I placed last week says the 23rd.

 

Yes, my point is that temp is part of the equation used to calculate MTTF - but a difference between a machine running at sub 80C on CPU/GPU is not going to make an effective difference in MTTF for the average user verse one running closer to Tj-Max. They are designed to run at or near max for a long time.

Also - I've 4 kids deep in 9 years of marriage lol - I'd be fine with an extra dose of infertility at this point.

 

So does anyone know if the advanced optimus panel on the X15 will be locked to 60hz when using the iGPU? I think this will be a deal breaker for me if so - based on my experience with adv optimus on the razer blade advanced, and also reading a review on the MSI GS66 with the same 240hz QHD panel. Yes, I know you can "disable" optimus and drive the panel to max HFR on battery if you use the dGPU, but this is not acceptable for battery life.

Right now I have a Zephyrus G15 (also had the G14) and it's a standard optimus panel. I'll take the 10-20% framerate hit to enjoy 165hz in iGPU while on battery and still getting 8+ hours of battery life. I can always bypass optimus when hooked up to my UWQHD.

 

Just ordered a couple sticks of 3466 for when mine shows up. I'll benchmark and show results in a few weeks. I also ordered a pair of 1tb Samsung 980 pro's. Yes I'm aware that the 2nd slot won't go that fast but I'm curious about RAID vs non RAID configs and want to see for myself the difference. I have been using RAID for years with HDD's, SSD's, and both AHCI and NVMe M.2 SSD's for both primary and storage. I'd love to buy a few 8 TB SSD's like the 870 QVO for a RAID 10 setup but I'm not ready to spend $700 a drive just yet. Later though.

 

Are you sure MTTF is calculated as an average of all parts within a system, or on a component by component basis?

I'm currently employed with the railroad and we've had a number of vendors try to sell us field equipment with MTTF calculations based on individual components in certain environments, not as an installed unit. Using MTTF calculations, we've had these same vendors mathematically demonstrate that we shouldn't see component downtime in 99.98% of service scenarios, for a ludicrous 50,000 years. In reality, this doesn't happen. Crap breaks all the time, and did for this vendor too.

My point is that MTTF for an Intel CPU running at Tj MAX may not be inclusive of MB traces, chokes, MOSFETS, and VRMs that are embedded nearby. In fact, if you look at the AW failures on Reddit and here, rarely do we see CPU/GPU component failures; we see MB failures instead. Granted, we can't see the exact failure modes encountered but I'd hazard to guess that Intel's MTTF figures are intact and it's downstream components that die due to having lower heat thresholds or insufficient contact with the cooling solution.

 

Great point, of course it's not.
A lBGA aptop is basically a serial system. There are hundreds of components and there is no redundancy, the effective MTTF for the entire system will be much lower than that of individual components.

http://answers.google.com/answers/threadview?id=390140

 

Yes, reliability calculations are often based on the weakest link - I've been through plenty of lifecyle reliability evaluations for complex weapons systems where we try to hunt down on a per-spec basis the weakest component in a system, especially when as stated, building in redundancy.

Everyone here is missing my entire point of this argument. While heat is part of the equation, two identical machines, one with a CPU running at 95C and the other at 85C - will likely see no difference in MTTF along a 3 year life expectancy, nor any meaningful increase in probability of failure.

Now everyone is backpedaling to just say, CPU/GPU doesn't matter - it's something else in the system. My argument would be that people who have experienced failures, would have experienced those failure regardless of CPU temp being 10-20C less. Failures happen...the internet can aggregate the 5-10% of failures into one discussion board or virtual "area". So what we see in any forum is likely skewed from what the real failure rate would be, or to draw any meaningful conclusions from posted failures.

To me, what matters more than a chip running close to spec Tj-max - is fan noise, and consistent performance...and yes, I would love to reduce skin temps too, but not for reliability, just because it makes a portable device more comfy to use. But seeing how even the 7-15W M1 chips in my macbook air run to 100C - not sure how we will be avoiding this until quantum computing comes along with photon powered transistors.

 

Again: throttling, reduced comfort, reduced reliability (plus unpleasant smell of overheating electronics as a bonus). You locked onto the least significant part of the moderately important side of the argument. A 15W CPU running at 100C in a Macbook Air... brilliant. That, as ever, must have been a serious inspiration for AW here!

Anyway, the first units are shipping - time to get popcorn ready

 

Their a a few advantages to living in San Antonio. friendly people, Cheap gas & energy, low cost of living, nice whether, booming economy with more jobs then people to feel positions, and an Amazon distribution warehouse. Some things take a few days but every day more items are added to free same day delivery for prime members. And the Samsung 980 pro’s I ordered this morning fall into that category. Now I get to stair at them for a few weeks until Dell delivers. I also want to point out that although Dell has the option to choose 64gb of DDR4-3466 they do not sell anything higher then 16gb sticks. That said I doubt the vast majority of people would ever run any software that would bottleneck at 32gb. By all means please correct me if I’m wrong. I’ve bottlenecked at 4gb and even 8 gb ram but never with at least 16gb installed. And that stands for even slower CPU’s.

 

Hey could you let me know whats the part number for the RGB Trackpad & The board that its connected to under the battery i believe (if you dont mind doing me the favor) on your machine im need to order one but dont want the hassle from dell reps

 

No problem but it’s gonna be a few weeks. I was looking at my packing slip to see if they were listed but they aren’t I think the RGB trackpad comes standard as long as you order the RTX 3080. I didn’t specify it when I ordered mine. Are you going to get a 3070 or 3060 GPU and add it yourself?

 

?

 

You need to stay on point mate. This is a forum about the Alienware X Series of Laptops. I don't know about anyone else here but I am not interested in reading pages of miscellaneous information about your work history, location, military experience, mobile devices purchased, software purchases, used laptops, or divorce proceedings. Not trying to offend you in any way, just trying to keep the forum about the topic at hand.

 

im doing a mod project on a M17 R4. I have a 3080 in mine but im sure the 3080 is not required for the pad just that dell markets it that way to make more money

 

Lmaoo the dude cant help himself. I started laughing when I got the part about the custody of his son. Like what???

 

Hopefully you can find the part. I suspect Dell is having the same problem as consumers with parts shortages. The reason I think that is because last month they took away the OLED option for the R4's and didn't add it to the R5's or the X series. The new XPS 15's and 13's offer it but I think they shifted what they had on hand in that direction. I suspect the reason is because most gamers don't like OLED and staring at it for several hours could cause headaches. Personally their are only a few action games I ever really liked. Battlefield and Fallout. But I've wasted hours on Civilization since version 1. I prefer strategy games so if it was offered I'd have chosen OLED. I also take issue with people who think 4k on small screens is pointless because Apple and Samsung are always second to none. My 12.4" ipad pro just look beautiful at 2048 x 2732 res. The refresh rate sucks though and I'm always having to refresh Safari's screen when I go back to a previous page. But with a dedicated GPU like a 3070 on up it shouldn't be an issue. Who knows though. Maybe they'll return the option later on. When I bought the ex's XPS in 2009 Dell had just refreshed the model and turn around time was like 2-3 months because even then they had shortages. All the outsourcing that these venders do just screwed everyone. Samsung is building a new PC factory in Texas so hopefully that lowers cost for memory and storage. This morning when I bought the 980 pro's I did consider the Western Digital Blacks, but stuck with Samsung because I know they've always worked in every Dell I've ever owned. And I didn't want a $400 gamble with the Western digitals. And another thing. Before I sharpshoot Dell about dropping the 2230 port in the X17 I wanna open it up and see for myself. Then I'll shapshoot them after I'm better informed.

 

Have you ever seen the big Lebowski. I'm from Texas so please don't refer to me as dude. Secondly stop the **** talking because y'all are doing the exact same thing you got on me about. Stay on point. WTF ever. online tuff guys

 

Maybe don't leave so much ammo lying around ?

 

I personally love OLED for gaming. For me the colours and black levels are unbeatable. I have used this on an AW 13r3 which has a 1440p OLED display and also a 4k OLED TV. If I had a complaint it would be that neither of those had high refresh rates. For me OLED + 1440p + high refresh + Gsync should be the gold standard for laptop displays. I was hoping the X series would be packing that but will probably have to wait for the r2 version.

Never experienced a headache from OLED that I can recall and wasn't aware that was a thing.

 

Welp I'm gonna mod my m17r4 with the backlit mouse pad and I'm determined to find a 1440p screen with at least 144hz that fits

 

When Dell took the OLED option before I ordered am M15 R4 I went searching to see if anyone had them thinking I'd just replace it later. Come to find out that had I found an OLED screen it would not have just been a simple screen replacement. The reason being that the OLED screen has a slight difference in dimensions on the lid. Also, giving credibility to my theory of OLED shortages they're were none to be found. And ultimately it didn't matter because I canceled the order. If I may offer a suggestion. Before you spend money for a new mousepad you might wanna make sure that the sizes match up. Including the controller board you mentioned underneath the battery. Taking into account the complete reconfiguration under the hood between the two models. Just a thought though. I ordered a 2mm thick copper heatsink for my new drives and haven't a clue if they will work because the service manual doesn't specify. If it works great but if not it was only $20 and I can always use them for something else.