*OFFICIAL* Alienware X Series Owners Lounge and Discussion

I am starting to have doubts, should I stick with the 4K 120hz or go down to a 1080 300hz / 165hz
I will probably play games in 1080p anyways but I would like the 4K for videos and photo editing.

Does anyone have a 4K monitor pls chime in.

 

The 4k model has adaptive sync, just an fyi.

 

I have the 4k model.

The 4k model has adaptive sync, so screen tearing is not an issue. You can enable VESA adaptive sync in the Intel Graphics Command Center or under Windows Graphics settings under Variable Refresh Rate. I have it enabled under both, but that's probably unnecessary.


The comments below pertain to playing games at 1080p.

Also the penalty on Optimus will mostly be less than 10% with high graphical presets, if not negligible in most games. At low graphical presets you will most likely be maxing out the refresh rate on the display so any excess frames you would get without Optimus can be effectively ignored. For competitive FPS games you'll easily max out the refresh rate of the display when playing at 1080p, so it won't even matter in that case.

The comments below pertain to playing games at 4k

At 4k the effects of Optimus on the FPS can be largely ignored. The bottleneck will the the discrete GPU as opposed to the iGPU.



For a fairly reliable set of benchmarks concerning Optimus please see Jarrod'sTech videos.

(2) External Monitor Boosts Laptop Gaming Performance!? - YouTube

For more comparison at higher resolutions(here are some at qhd, I can't find benchmarks at 4k).

(2) Zephyrus G15 RTX 3070 NVIDIA Optimus - Time Spy & Games at 1440p - YouTube

 

Nice. My 8th gen laptop is clocking 64ns... Area 51M does better still with Impact RAM.

 

The BIOS hack for connecting the discrete graphics card directly to 4K120Hz internal screen works! You just need to adjust two values and save it. Afterwards your internal screen is directly connected to RTX GPU. Optimus will be bypassed completly. Unfortunately you will loose adaptive sync feature. Performance boost is great!

 

Gear 2 allows the OEM have higher RAM speed and bandwidth, but worse latency. Maybe the owners of the X-series should ask Dell support for an unlocked BIOS so they are able to change it to Gear 1? But will probably say nope due the MB quality. The lack of choices in bios stop the fun.

The high or better say the slightly lower latency showed from @JinTexas last screenshoot... High latency ain't the only problem. Look at the write and Copy speed who is also awful. Not faster than my cheapo 2133 OEM ram clocked at 2800MHz with lower timings and much lower latency.

 

The Corsair ram kit CMSX32GX4M2A3000C16 does not look bad. 3000mhz CL16. I wonder if they perform better than the crappy high latency dell packs with x17.

 

Do you know any results from other 11th gen intel mobile platform vendors, like GE76 or Legion i7? Are they significantly better in mem speed than X17?

 

That's great to hear. Would you mind pointing me what two values need to be changed. I think I missed out on the discussion talking about the bios hack and how to do it.

 

Not many enjoy benching nowadays. This one is already posted on the forum.


Here is the last one from @JinTexas with the faster 3466MHz sticks running on the new bios which offer lower latency.

 

Will do this at the beginning of next week with screenshots. Unfortunately I have no time this weekend for that.

 

Thanks. I am also guessing you did something like they did here.

Forum - [REQUEST] Checking for MUXed/MUXless or Fixed/Dynamic VGA option for BIOS (win-raid.com)

 

I got a response from a Precision 7760/7560 owner: 96ns latency with stock 3200 ram.... Dell.

 

Check voltage. HyperX/Fury Impact is 1.2V, same as Dell stock RAM.

 

Yeah, voltage is 1.2V.
Also G.Skill haves a 2x8kit @3200mhz CL16 --> F4-3200C16D-16GRS and 2x16 kit @3200Mhz --> Cl18 F4-3200C18D-32GRS.

Both 1.2V too.

 

Double check return policy and let's go I would just get Impact and be done with it, but if you have the time these are interesting experiments.

 

Yeah, as long as I can buy it from Amazon, no worries. That's why I risked with Ballistix, nothing to lose because they have a good RMA policy.

But I think there is a big shortage, at least in Spanish market, because most top tier ram are out of stock.

If I get one of these kits, I'll keep you guys updated, if it works, performance, etc.

 

Stock L7i ram. Ok, but not great. We've seen ~5% CPU bump after adding impact CL20 ram (L7i doesn't support XMP)- I'm waiting on mine to get here.

 

Thank you very much, I knew if someone can find values to compare then you, +2 rep. So it´s pretty clear, it´s again a Dell thing. While MSI runs the mem at CR1 and Tear 1 the X17 runs at CR2 and Tear 2. So whatever kit you put in the X17 it´has no chance ahainst MSI. But there are even worse products, take ASUS with it´s soldered x16 mem. The X17 has a better cooling for the CPU compared to the GE76 (which has on the other side a much better cooling for the GPU). It´s somehow sad that AW slowes down the mem and not giving one the option to change the values in the bios because it can deliver higher TDP and better cooling to the CPU than the GE76. At least my Kingston 3466 kit showes a litlle better values than JinTexas unit but still the same high latency under W11.

 

O.k., so the X17 seems to be right in the middle.

 

Here my most recent run with bios 1.2.1. Not good... L3 Cache seems slower under the new bios. Not the first time AW bioses get slower with each revision...

 

Another kind Precision user posted his result with the 3466 memory gimmick: 106ns

 

Img doesn't work.

 

Basically same exept for worse L3 cache results. I wonder if there is a difference between the Kingston and ADATA 3466 kit Dell is selling. I think JinTexas has the ADATA kit while I have the Kingston.

My L3 write and copy values are way too low. Seems to be a W11 problem. It´s a BETA, I know...

 

Best AIDA score so far for for Kingston DDR4 3466

 

Commie A hole thinks he's Joseph Stalin. Who does he think he is in the headline "Way more laptop then you need". Why do the people running the tech industry have such a disconnect from most people who use the hardware?

https://www.pcgamer.com/this-fully-...op-than-you-need-but-its-also-dollar1100-off/

 

Does anyone know why there is a large discrepency in max brightness between Pop!_OS 20.04 and Windows 10?

Pop!_OS peak brightness: 530 nits

Windows 10 peak brightness: 433 nits

Measured using an X-Rite Display Pro Plus

I've already unchecked every single power saving option in Windows but I'm getting much lower peak brightness.

 

For those who may be interested. Here are measurements of the Alienware X17 4K display's color coverage using an X-Rite Display Pro Plus.

 

Sorry to mention the obvious, but L3 Cache has nothing to do with RAM modules. Most likely an issue with how AIDA detects cache sizes, or perhaps a problem with background activity.

 

So advanced advanced optimus works on the 4k X17 when you use an external monitor. I'm using an HDMI to displayport connector cable. Colors look rich and vibrant. Much better pic pic on my 4k TV then when using HDMI.

 

colors on my external screen look amazing, Much better then my latitude which also had USB C monitor support. The response is amazing and my mouse flies around the 64" screen with no issues, The only downside is that it only supports 60hz which for me isn't an issue because I prefer strategy games live Civ which don't require anything more.

 

That would be:
1. A first
2. A bit pointless since when the laptop is using an external monitor, it tends to be plugged in, making Optimus redundant

In your case though, you can see that System Information clearly says "Advanced Optimus No"...

 

It's not 12000 but this score is from the external monitor at 4k and I forgot to turn off adaptive sync. The GPU test is faster then an older test I ran on a single Titan Xp 4 years ago.

 

Fair enough. I shoulda looked closer. Regarding external screens I almost always have mu laptops hooked up that way.

 

Something is very wrong. Look at the CPU graph and score. A train wreck. I wonder if the ram contributes to this.

 

Installed a secondary SSD on my x17. It's a cheap 1tb nvme v3 Kingston KC2500, but more than enough for my games and VMs.



The screw was tighten as heck.


----

By the way, I noticed the original SSD that came with my x17 makes no proper contact with the heatsink because of how thin is the thermal pad. I mean, this SSD only reaches like 70ºC during stress test but what's the point of having a heatsink if its not properly used. So I'll order thermal pads and see if I can improve thermals.

 

There are 3070 laptops that can almost beat this score that cost less than half the price. Really, really dismal especially on an external screen. 15% under performing from where it should be- a 3080 and 11980HK reliably do 12.5-13k from MSI, Legion, Tongfang, Clevo with max 13.5k etc. Tongfang have primarily adaptive sync too and don't take such a dump in performance (Eluktronics Prometheus).

 

There is something wrong with his score:

 

There seems to be an ongoing bug with 3dmark if you set your desktop resolution to 4k. @pitha1337 highlighted this in an earlier post of his. Also a stock 3080 running at 165W should be benching ~12600 in Time Spy graphics with the Tigerlake processors. The 13k you are referencing at the top of your range is because they're running a mild overclock(some turbo/performance settings on those laptops apply an overclock or the poster is not being forthcoming about what settings they have applied).

*OFFICIAL* Alienware X Series Owners Lounge and Discussion | Page 112 | NotebookReview

Here are what actual stock scores look like on an internal and external monitor. Note that I have the i7-11800H and an RTX 3080.

INTERNAL MONITOR







EXTERNAL MONITOR

 

Just by replacing the thermal pad of the ssd headsink cover my main drive went from 74ºC to 62ºC during heavy loads (at idle 35ºC-40ºC now, before 50ºC).



I had lying around 1.5mm thermal pads from EKWB (the ones they sell for the graphics cards) and seems to work just fine. So I sugges you all to check if your ssd heatsink are doing proper contact with your SSDs.

 

Hey guys!

I know that for this generation trfc can affect ryzen cpu performance greatly, but what about intel cpus? I ask because I’m getting a 7i soon, and since it doesnt support xmp, this ram is the best i can get (cl 18, 20, 20 40) : https://www.amazon.com/OLOy-2x16GB-260-Pin-Laptop-MD4S1632180BZ0DH/dp/B09338C27Y?th=1

Here are the zen timings:
https://i.ibb.co/VQbtr1r/244-D7-B51-FF34-42-EF-A7-DD-688-AFD30825-A.jpg

One thing I noticed between this ram and the next best alternative is the trfc which is a bit high (stock ram should come with trfc 550). Thoughts on this?

 

The OLOy ram has great specs on paper, the slow trfc as you noted, and seems to fail to perform better than stock ram in some of the benches around here and reddit. One head to head in reddit the OLOy ram was 3-7% slower across the board in a variety of benchmarks than impact ram. Stick to kingston impact ram if you want the small performance bump. May have more options if you get XMP support like the X17 for ram, but compatible voltages will still be a big challenge- L7i has no XMP unless someone figures out how to get into the advanced bios.

 

Are you referring to this thread here?: https://www.reddit.com/r/GamingLapt...urce=share&utm_medium=ios_app&utm_name=iossmf

While I agree with the testing methodology, its important to note that he tried the 16gb variant (2x8) which is single rank, and compared it to the 32gb (2x16) hyperx dual rank variant, which I suspect is the cause for the discrepancy between the two. The oloy has much lower latency @ 11% less than the hyperx, and I wonder what kind of scores they would have gotten with the 32gb dual rank variant of the oloy ram.

 

Someone was asking for Tomb Raider benchmarks so here they are. These were done on an external monitor.

Benchmark 1:
i7-11800H: Stock setting except a 90 mV undervolt on the Core and Cache.
RTX 3080: Stock settings



Benchmark 2:
i7-11800H: +400 Mhz turbo, 90 mV undervolt on the Core and Cache
RTX 3080: +120 Mhz Core, +1200 Mhz Memory

 

Which Kingston 3200 CL 20 mem do I have to buy? There are single rank and dual rank kits (2x16). Does single vs dual rank even matter on Intel platform (on Ryzen it does)?

 

HyperX/Fury Impact - I doubt there is more than one variant per capacity, but if there are dual-rank 2x are preferred. Every time I looked in the past, the 2x16GB kits were dual-rank, but do double check on kingston website.

 

The original Kingston 3466 modules are single rank according to CPUID. Dell cheaped out again... Maybe this is an explanation for the poor performance/latency. Ordered Kingston Fury dual ranks 3200 CL20 (careful, also single ranks are sold). Should be here in 4-5 days. will report how it goes.

 

Good luck. FYI @alaskajoel has these modules in his Precision, and reported approx. 90ns latency too. Looks like there is a bottleneck due to HW limitations or BIOS settings (the 2T command rate perhaps) which is responsible for these poor memory latency results in latest Dell/AW laptops

 

A bit dissapointing that the new flagship X17 with 3080 and CPU + GPU OC (bench 2) can only match A51mR2 stock performance

stock


OC

 

Yeah, unfortunately dynamic boost 2.0 is holding the laptop back. The 210W combined total power budget between the CPU+GPU causes strange power throttling behavior when the laptop exceeds those thresholds.