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

I believe this is the challenge for notebooks

https://edgeup.asus.com/2017/kaby-lake-overclocking-guide/3/

DDR4 frequency range Required CPU VCCIO Voltage range Required CPU System Agent Voltage range
DDR4-2133 ~ DDR4-2800 1.05V ~ 1.15V 1.05V ~ 1.15V
DDR4-2800 ~ DDR4-3600 1.10V~1.25V 1.10V~1.30V
DDR4-3600 ~ DDR4-4266 1.15V~1.30V 1.20V~1.35V

Throttlestop can control system agent voltage, but we don't have VCCIO controls

Due to this our theoretical worst case maximum is around 2933-3000Mhz for the average CPU, with some lucky people being able to do 3200+.

 

Latest update: I have a replacement motherboard with me from Dell and I am cross comparing it to my current motherboard. So far I have managed to accumulate some VERY interesting findings!!! I am amazed at the differences to be honest! Stay tuned, I will post a detailed comparison of power components on both later today.

 

Ram speed had a huge impact on fallout 4.

 

You THE man!

Sent from my SM-G965F using Tapatalk

 

Voltage doesn't show up in Afterburner, is this normal even after unlocking voltage control in settings and reinstalling ?

 

I guess there will be a revision number stamped on the DGFF and on the motherboard so you could tell ?

 

And that is the only sensible approach and it is fortunate that Dell give full refunds.

 

And you can wait for higher resolution panels to come out at the same time.

 

It has been mentioned multiple times that they have fitted incorrect/insufficient VRMs and those exact VRMs tend to blow up. I presume they burn because they work outside their SOA by very bad motherboard design.

 

What is VRM and what is SOA?

 

My 9900k turning at 100 ° C in intensive, and am beginner in settings. can you please tell me what exactly are the settings to modify and with which software to lower the temperature of the processor?

Speaking French, and not understanding well tutorial videos in English on YouTube, if you can be as clear as possible so that I do not make mistakes, it would be very nice of you.

 

What are your temperatures when idle?

Did you repaste your CPU and GPU?

What are your fan settings?

What are your power settings?

For software undervolting, you have a choice of Intel XTU or Throttlestop. Undervolting is a process of its own and someone else can explain it better than me.


Sent from my iPhone using Tapatalk

 

Hey guys, anyone tried removing the i9 sticker on the palm rest and if so, how did you do it?

Sent from my SM-G965F using Tapatalk

 

You can apply a little bit of heat with a hair dryer to aid in removal. If you don't care about tearing it you can just remove any residue with a good orange cleaner and then clean that up with some rubbing alcohol.

 

Got it. I will use the hair dryer method to remove it and use alcohol wipes to clean it up. Thank you for your suggestion. I personally don't like stickers on a laptop.

Sent from my SM-G965F using Tapatalk

 

CPU: 48-50°C idle and 100°C intensive.
with all factory default settings, and CPU/GPU factory Paste.

My configuration is based on i9 9900K and RTX 2080 (Latest Bios 1.5.0 and VBIOS 200W).

I have 32GB of MICRON DDR4 2666 RAM (16ATF2G64HZ-2G6E1) according to the CPU-Z software.

The RTX 2080 is based on GDDR6 chips (Samsung) apparently according to the GPU-Z software.

What to do exactly please?

 

So here we go guys. Today I got a call from Dell and instead of just sending me an audio daughter board, they sent over a replacement motherboard as well just in case, which is the one that I retained in order to compare with mine. It turns out that mine that came from HIDevolution was the weaker one, so I have upgraded the motherboard on my machine before writing this. The major difference is the power mosfets close to the power connectors on the back of the motherboard (those 2 notorious tiny chips beside the CPU fan that usually burn out on these motherboards). The first batch lemons have this part:

Lemon power inlet mosfets:


The upgraded motherboard that came from Dell had the following power inlet mosfet:


Difference between both these chips is HUGE! First, take a look at the lemon datasheet snapshot. Pay close attention to highlighted parts:

Lemon part AON-6380:


Now look at the upgraded part:

NIKOS-PK616BA:



Comparing Lemon vs Upgraded parts, we have 24A vs 50A continuous drain current (ignore the Ta ratings), 88A vs 120A Pulsed Drain Current, 60A vs 24A Avalanche current (lower is better) and 0.001mH vs 0.1mH Avalanche energy (higher is better). (Full datasheet links: AON-6380-Full-Datasheet vs NIKOS-PK616BA-Full-Datasheet). This is quite significant!

There were some other differences as well including a different Bios flash chip but those were not as major as this one. Also, I noticed that the upgraded board was actually hand repaired and I could see the hand soldering clearly, although it doesn't matter as long as it works. E.g., this bios chip on the upgraded motherboard is clearly hand soldered.




While I was at it, I noticed that the replacement heat sink that the engineer had brought, had an additional thermal pad for 2 more power components on the DGFF card that were hidden underneath the beauty cover. So I cut a portion of the beauty cover on mine and placed a thermal pad there!!!



This module Z5038QI5 is basically a 50A continuous DrMOS Power Module that has 2 asymmetrical mosfets and an integrated power driver and can be seen on both DGFF as well as on either sides of the CPU socket. Nothing changed here though.

 

Good find.

Sent from my SM-G965F using Tapatalk

 

@S.K post of the year !!! Cannot rep you yet.

(pure avalanche)

 

Seems my laptop was made and shipped from Dell on April 8th. HID sent me the unit by the 25th. Guess I will have to open my system up after work to see which MB I got.


Sent from my iPhone using Tapatalk

 

want me to call out Azor again ?

 

I think we should give Dell some time before we go on and call them out. The best way to let them know that they need to improve their ways is to call their support (in case you have a low spec part) and demand replacement. Eventually they'll know that people can see what they are doing, even if they don't say it out loud.

 

Alright cool bro. I'm more than happy to call him out, as it will show how he tries to turn everything around. Having seen the spec sheet, gives me the idea that they did not know what they're doing when building DGFF. Whenever you're ready, I'm ready !

 

you should pls go and re educate DELL

 

I'd certainly like to re-educate them about how to give a nice ultrasonic clean to a motherboard after replacing parts on it so that it doesn't look that messy as that bios chip looks, although with the naked eye it looks really nice and clean. It's just with a 5X zoomed in camera that you see the real work.

 

If I owned this expensive laptop, I would likely want my motherboard/GPU exchanged, especially when/after Dell has enough revised spare ones, which should be around now LOL

It is maybe possible that the problematic parts can be re-soldered by Dell quite easy and Dell would not have to worry about avalanches waiting to happen.

The MB and GPU could be exchanged on-site before it burns, then being re-soldered and ready to return to producition. I mean ... 3-4000 USD vs. 100 USD paid to repair guy VS. Alienware reputation.

Also, there are many people waiting for official report from DELL that all is fixed to buy the laptop without fearing that it breaks.

 

The Eye is a Cheat...

The amazing thing is, some with lemon parts may never have issue for years, if the yield on theirs is in the 100% perfect category

 

One more update: I noticed a significant temperature drop with the new motherboard in my DGFF temps, probably because of the reduced overall signal noise with the new, more rugged power components on the motherboard power intake. My 2080 is maxing out at 71 degrees now with 1950 MHz core, 7000 MHz Memory and voltage set to 937mV while running Heaven benchmark at extreme settings. Before the motherboard swap, the same afterburner profile was giving me around 77 degrees. This is significant and I am happy that Dell sent over the new motherboard!

 

I am glad ur happy after a long battle..... well deserved

 

Dell support has been REALLY good so far with me. I called them multiple times for different components and service was arranged the very next day or max 2 days after I called. One possible reason could be that I live in Singapore and Dell has a regional office in Malaysia where they manufacture some of their parts. But overall, I am really pleased with their support. Got to give credit where it's due.

 

Ugh. I got 10-11 more hours to go until end of my work day. I then have to go pick up the recently released G502 Lightspeed (right after I wished for wireless equivalent right?) at a local Best Buy before I even get home to open my 51m while also fine tuning my Matebook X Pro for professional use.


Sent from my iPhone using Tapatalk

 

That wait can kill. lol

 

your real lucky, am in another planet Nigeria.... What is Dell ?? The Area better behave until another 51M Monster comes to take its place.....

 

Umm, pretty sure I’ve checked those two mosfets before their death, clearly remembering the model NIKOS on them, so your explaination might be incorrect..

 

On the other hand, is the avalanche cure lower better? I can’t agree with that when it comes to the mosfet death itself..

 

Nikos makes plenty of them. The specs matter and they're tied to the part number, not the manufacturer name.

You need to study how mosfets work. Lower avalanche current means mosfet doesn't go bananas all of a sudden when it's max specs are exceeded, pushing it into the avalanche state. And more important than that is the avalanche energy. Higher energy means the mosfet is more rugged and can take a lot more abuse before it breaks down.

 

Any idea about this? @iunlock @Rocktaze @Ultra Male
Thank you very much in advance

 

But mosfet burning is not always caused by avalanche.In 3C use, heat problem is often the cause.

 

BTW. I got my motherboard swapped too two days ago, directly from China Kunshan where I live nearby. On it still is the 6380 module mos.
I’m really frustrated about this since dell is using two different kinds of parts with different specs at the same time.

 

L=0.1 mH is its test condition under the 0.1 mH inductance, you can not just compare it or the Eas. Eas power comparison should be taken under the same condition

 

The iGPU is only used though the thunderbolt port. External monitors should be plugged into the HDMI or DP on the RTX GPU in the back.

Undervolting the iGPU is not necessary for better temps

 

So HID got back to me about swapping over my dell replacement back under the HID warranty I had. They said they would do it but won’t cover anything that dell gave me as free upgrades. Pretty crappy on their part being that HID told me to go through dell for faster service. I don’t really see the point of having HID support anymore.


Sent from my iPhone using Tapatalk

 

But you live in the US, right? They are covering your PC with warranty, correct (minus the Dell parts)?

 

OK, thank you very much.
And regarding the other question about ThrottleStop, do you know what's the purpose of the "overclock" checkbox under FIVR/Turbo Ratio Limits? I have seen some videos where some people check it and others where they leave it unchecked. I have tried checking it and apparently there is no difference.

 

Does nothing. My 8850 is partially unlocked and can do 4.5ghz on 5 cores and 4.4ghz on 6 cores. It requires xtu to do that as throttlestop only goes to 4.3.

Also I notice everyone doing an offset undervolt. I get much better results doing static voltage. As in a 5-7 watt difference.

For example I do 4ghz at .9756v 6 cores. 9700k and 9900k should be even better.

 

Yes but it’s still annoying that HID won’t cover anything to do with my CPU, RAM, or storage devices since those were upgraded by dell


Sent from my iPhone using Tapatalk

 

Thank you.
I'm using also static voltages for full performance profile, but prefer adaptative for low load profiles as I get better temps in low frequencies with very low voltages. The main disadvantage I see with adaptative is that it can cause crashes when idle for being too low sometimes and requires much more time for tuning than the static profile where you only have to check that it does not crash under load. With adaptative you have to ensure that it does not crash with high load and also on idle.
I wish there was a tool for tuning the whole freqs curve, that would let you go to the limit under extreme load and also get very good temps with low load or idle.

 

Steve one of the King’s of overclocking

 

There is no denying the fact that the Nikos mosfet used in my motherboard is *more* rugged and *less* prone to burning issues than the AON one. Overall specs decide which part is plain better and not individual characteristics that do not even mean anything when considered alone. apart from the Avalanche energy in this case. What causes burn out of these mosfets is how much current flows through them during avalanche mode, which results in burning out of the mosfet. I'd seriously recommend that you deeply study the construction of mosfets and how they are driven under certain conditions. Unless that, you might find it VERY hard to reach a definitive conclusion when comparing multiple of these devices.