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

Does anyone here have an intermittent issue where their USBs are interrupted for a second or two and it causes the entire computer to freeze while they reconnect, or is this just an issue with my usb hub?

It's really annoying because it causes Discord to lose audio source and I have to restart the program for it to return to normal. Happens often enough to be a hinderance.

 

So in ECO mode it's basically an 8C/16T CPU in MT benchmarks? That's pretty unfortunate, and somewhat defeats the purpose. They should set the PPT to 95W or whatever the heatsink is spec'd for instead of just setting it to ECO.

(or use PBO as mentioned above)

 

Lies again with the SD card slot!!!!!!! Hahahaha

 

I have never even tried to run my 3950x in ECO mode. I did not purchase it for anything like that. I'm not even curious. But I wouldn't be surprised if ECO mode just disables one of the CCD's.

Too bad that's unfortunate for you.

So much negativity in this forum. I think I've learned everything I need to know about my 51M here and will just monitor this thread from now on.

Wish list? None really. This laptop checks all of the boxes I purchased it for. Anything above that is just gravy.

Papasaun, Mr. Fox, I'm not sure what your roles are here either, but I suspect your motives. And I also think your posts just breed more negativity.

And yes, the 51M should have it's own area for folks that don't own one, but love to discourage anyone else that owns, or is considering owning one..

 

Probably best to wait for real laptop results. The benchmark results I posted earlier show + 50% in GeekBench vs 9900KS in ECO mode. No CB20 unfortunately. Probably best to wait for some laptop results, but the potential is obvious.

What's a CCD in a CPU? Various articles on the Web suggest ECO mode TDP results in just lower clock frequencies as you would expect.

https://www.tomshardware.com/uk/reviews/amd-ryzen-9-3900-review-eco-mode

Of course, people would be hard pressed to use this in a proper desktop, it's designed for small form factor devices and exploits the fact that increasing CPU power offers diminishing returns.

 

I have
it also. Drives me nuts

 

That's why I posted about the new XMG lol. Because 3950x makes the once impressive 9900K/S not that impressive anymore, certainly for people like me who wants a lot of cores.

Though I do question why they use 15" so hard pass for me.

All I want to see now that the bar is higher, is 3950x, 17" 4K mini led screen, 128GB RAM, 2070 and up, vapor chamber and nice chassis like 51m, or msi, or some Asus or even gigabyte, anything but the 80s bland chassis designs.

10th gen intel looks silly to me now.

 

15 inch laptops are better on the go - and arguably many laptop owners require this. Which brings me to battery life, an unknown here. Can 3950X operate reasonably at a few W? How efficient is the graphics switching, if it's even enabled?

OTOH Intel introduced Optimus 2 which supports gsync, so hopefully this feature will make it to the 51M.

 

That was a engineering sample. It was on the infamous version with the extra heat pipe that fell decided against after showing the notebook.

 

What is the keyboard shortcut for the FN + F1 do? I see it looks like 3 double sided plates. I tried pressing it but nothing happened.

Also, is there no keyboard shortcut to sleep?

 

See Keyboard shortcuts in setup and specs https://topics-cdn.dell.com/pdf/alienware-17-area51m-laptop_setup-guide_en-us.pdf

 

Enables/Disables the Graphics amplifier if I'm not mistaken

 

Thanks for your help!
I did all exactly, but when I "Reconnect AC adapter and power system on" laptop didn't start beeping. Power button just lighted green. And after 30 seconds started blinking 2 red - 1 blue.
Any ideas?

 

Do you work for Dell?

@Mr. Fox

You're right man, Unfortunately, there is no dedicated Sleep button. Why leave the F6 and F7 empty?

 

The standard power button behavior is to sleep. If you changed it, you can just reapply it in the power options menu.

 

Is there anyway I can downgrade my bios to 1.5 for the higher GPU temp throttles?
Throttling at 76c is horrible. I am currently running the latest stock bios.

I tried using the method in the first post where you power off hold CTRL+ESC etc but I just get the error message below:
ROM Image is not Loaded
ROM Image update denied

Any help is deeply appreciated.

 

Totally forgot about that. Thanks for reminding me bro.

Send me your TeamViewer ID/Pass so I can help you create the USB properly

 

I might still have my EFI folder somewhere but don't think that has anything useful.

Beautiful list. I see lots of common laptops between us

I'm gonna be hella mad if R2 has SD slot. I literally use that slot on my other laptops almost every day.

 

Downgraded @oni222 to BIOS 1.5.0 successfully

 

@Spartan is making sense of everyone NOT on 1.5.0...

A true warrior

 

Thanks again for all your help!

 

Not trolling or disagreeing, but unless you are doing hardcore rendering and multiple apps, why on earth would you ever want 128 GB RAM? I mean unless I have like 6 VMs' up on my desktop I never get close to it, much less utilize 16 threads... I mean power is _nice_.... but ... what would you do to hit 128 GB ? Are you doing a lot of graphic work, folding or something?

 

I am a senior fx td (vfx artist) who worked on Hollywood movies like Aquaman, X-Men: Dark Phoenix, Star Wars Rise of Skywalker, etc, so I know the hardware requirements to achieve this kind of vfx work. I am not using my laptop to do these of course, but the studio workstation and the farm, but it's still beneficial for me to be able to work on similar stuff for personal projects.

A simple case of a water sim, say a river simulation you can do easily sim 30-40M particles using 12 GB roughly, but once you wanna go 1 level up in resolution (1 level down in voxel resolution or particle separation), you are looking at 96 GB so more likely you would have to have 128 GB to continue. (coz everything is cubic 2^).

So while having a lot of cores speeds things up a lot, not having enough RAM is what limits a lot of things. Also the app I use, can solve things on the GPU but often times GPU memory is severe bottleneck so have to set OCL to use CPU via intel drivers to bypass this. That's why I was excited about RTX 6000 on a laptop.

That's why I would love to have these on a laptop as I move often times between contracts.

 

No problem bro !

 

Nice. You must have thought about dropping an RTX Titan into the AGA?

 

I guess each person has is usage scenario.

For example, on my previous Area-51m, I configured it with 128GB of RAM. You know, just to feel that I'm top of the food chain! Then I checked my RAM usage, I barely use 6GB during my normal activities which is browsing the web with a few tabs, watching YouTube, whilst having SVP in the system tray, Antivirus (VIPRE), Internet download Manager, skype, OneDrive.

So even if I game or launch Photoshop or Premiere, I won't even come close to using even 30% of that. That's why this time when I bought the Area-51m. I left it alone with the stock config and I'm still using the same amount of RAM so I won't need more (personally)

 

As an owner of Area 51M, you can sleep well knowing you can expand your memory up to 128GB, which is the way it should be. It also increases resale value (since the buyer might actually need more RAM), which you might appreciate as someone who changes laptops every quarter at the latest

 

@Papusan @Mr. Fox @jc_denton

What can I smoke to be able to come up with such BS?

 

I'm all for maxing out for the hell of it, I was just curious what would utilize that much memory... 16 GB was just uncomfortably low, 32 GB would have been fine... but 64!

 

Yes definitely though don't know too much about these. Do they support any gpu or are they by wattage since each gpu had different power req?

Also not sure how big and heavy these things are. Rtx Titan or rtx 8000 would be great coz of memory.

But I can't drop 5k on a gpu it's too much. Gpu rendering would be really good using redshift but major studios don't use it but cpu based renderers still.

But I am mainly interested in sim capabilities of these hence I need more memory on these also. The app itself can't use multi gpu for sims only one.

 

They will support any common GPU by Nvidia and AMD up to 460 Watt, esp. if its a reference spec. I have used one with a GTX Titan, 1080 and 1080 Ti. I have heard of people changing them to have better fans and PSU IIRC. That being said....

https://www.dell.com/support/articl...mplifier-supported-graphics-card-list?lang=en

EDIT: I have to run mine "hotrod" style - the top is angled open due the height of the card : /

 

Can I use non ref card? Like the ones with big fans 3x etc? Also this is what AGA allows in 51m? I mean all high-end laptops should allow external GPU no? I will do more research.

 

Depends on the size of the card. But you can use a saw or Dremel tools if you need more space and mod it as this one... Not pretty but it works

http://forum.notebookreview.com/threads/alienware-m15-r3-wish-list.832156/page-5#post-10998603

 

I see but AGA seems Alienware only so I will pass that one hehe.

But I saw on Jarrod tech channel he said if you use laptop display with egpu there is a performance cost? Not sure why that is?

I assume related to Optimus and such?

 

If your system supports G-stink you won't be able to use the notebook display with the GA. The signals go throghout the iGPU and to the notebook display. This is the bottleneck. For me is Optimus in any form or shape a disgusting tech. I avoided it with 3D panel on my old AW and iGPU power rails isn't connected at all on my Clevo.

 

You can Google graphics cards officially supported by AGA. It includes older Titans but not Titan RTX. It could be due to the default PSU (450W AFAIR), which is very easily upgradeable, or it could be due to card dimensions.

We know that AGA doesn't disable the internal dGPU, but is it possible to connect the AGA to 51M, yet still have the display driven by the internal dGPU? For many applications that would be enough.

Optimus is awesome, because it greatly extends battery life, which you know - many laptop users care about.

Gsync sounds compelling until people realise the experience is similar to Vsync. VSync introduces very noticeable input lag (it's completely useless in FPS games), even at 60fps. If the reason Dell didn't include Optimus in 51M was to allow Gsync, that's a very poor tradeoff IMHO.

 

LOL, well at least don't inhale the blue magic smoke coming from hardware They might be able to limit FPS with the driver, that could in theory limit power/heat. But mostly it reminds me of the "download more ram" adds.

 

The AGA also unofficially works with many other cards. I've used a Titan RTX, Titan V and even a Tesla V100 in mine without any significant problems.

Conflicting drivers are usually the most frustrating thing to fight with. I've found this to especially be the case with Quadros. It's much easier to add an external Quadro card to a laptop that already has an internal Quadro like the precision series, which means you're stuck with TB3 enclosures. This isn't necessarily a bad thing--if your work load requires the large VRAM of something like a RTX 6000, I'm guessing a lot of the work you are doing is done exclusively on the card with minimal communication with the CPU. TB3 is plenty sufficient in this case. The TB3 bottleneck is worst when a lot of info is crossing the PCIe lanes. This is why external GPUs perform worse with high refresh rates and lower resolution (1080p + 144hz) compared to lower refresh rates and high resolution (4k60). In my case, using a V100 (32GB VRAM) over TB3 performs almost the same as when in my desktop for performing deep learning tasks. Fill the VRAM with work to do and let it go.

Yes, this is how I prefer to use my AGA with the 51m and why I prefer it as my mobile work device. If you run monitors off both the dGPU and the AGA, you will get high GPU utilization on the desktop window manager task and some odd stuttering but it works.

I prefer to let the internal dGPU run my multiple external displays and I can use the GPU in the AGA to perform my work tasks. I've commented multiple times to my colleagues and our Dell rep that I think this configuration is a huge opportunity for the precision series. I would love to have a less flashy Precision with ECC memory support and an AGA port for connecting a GPU but I totally acknowledge this is a very niche market. Until then, I'll keep using the 51m.

 

The new command center somehow improves the graphical performance, still capped out at 78C any custom bios to increase thermal limitations to 85 or 83?, also found this on reddit somehow someone managed to push area51m to 12000 score! (graphical)

https://external-preview.redd.it/9h...bp&s=c61927fecc91e6f7f7c39a4a8abb8a82f9609af8

 

Fantastic info, many thanks!
For completeness sake - any particular TB3 eGPU enclosure you would recommend?

 

Hi there,

I want to repaste the laptop (i9 9900K and RTX2080). I have repasted laptops with laptops CPU even with Liquid Metal. What do you think is best for this one? Thermal Grizzly Conductonaut Liquid Metal, Phobya Nanogrease Extreme or Thermal Grizzly Kryonaut?

I have right here the 3 pastes/LM so I can use one of the three.

Thanks for the advices!

 

Don't use Kryonaut in a laptop.

 

this will damage copper heat sinks, you guys never get it.....

 

Then, should I use Phobya? What do you suggest?

 

Haven't hear any bad opinions about Phobya NGE. Used it myself recently after Kryonaut pumped out - 6 months later all good. Worth trying before plunging into LM.

 

Gelid extreme and phobia are really good, Liquid metal should be used under the IHS and on bare intel die only, if it contacts with copper, it will react with it and eats it away with time, copper nickle is resistant to liquid metal...

its not about risk reward or perforamnce anymore, this is surely going to create a gap between the IHS & GPU die with the heatsink unless heat sink screwes are adjustable.

Area51m doesnt have a vapor chamber, it cant kick heat fast enough from the heatpipes to the fins to outside,

 

It will react with copper and create an alloy.

 

I've exclusively used the Razer Core series and love them. My first gen core X worked great and I've now upgraded to a Chroma model that has been completely trouble free. I admittedly don't have experience with other TB3 eGPU enclosures.

While I'm thinking about it, another overlooked benefit of the Area 51m and a few other Alienware laptops is the dedicated CPU PCIe 4x lanes to not just the AGA but also the thunderbolt controller. Instead of hanging thunderbolt off the PCH like most other manufactures (and thus requiring it to share DMI bandwidth with all other PCH devices like USB, NVMe, etc), it gets a dedicated 4x CPU connection. It isn't quite as fast as the AGA port because of the Titan ridge controller overhead, but it is faster than most other laptops. This allows for the possibility of two powerful eGPUs.

 

and the new alloy is it thermally conductive?

 

The best combo is Thermal Grizzly Conductonaut on the CPU + PHOBYA NanoGrease on the GPU.

Here is mine, Liquid Ultra on the CPU and PHOBYA NanoGrease on the GPU, only because I ran out of Conductonaut and I had a new tube of Liquid Ultra so I used it:

PHOBYA NanoGrease is the safest non-metal thermal compound and the best out of them so it's safe to use for both the CPU and GPU even if you were to move the laptop a lot.

Yeah we never get it, that's why we've been using Liquid Metal for more than 7 years and enjoying the lowest temps possible.

LOL this guys has figured it all out eh? Thank you master for the invaluable info!

Liquid Metal is ONLY bad for heatsinks that don't fit properly, the only bad fit heatsinks I experienced in over 30 laptops that I've owned is from Clevo, MSI and Alienware always have perfect heatsinks.