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

That's more the 5m less then I paid AFTER credits and GC. Feel great about that purchase

You will find zero heat issue with the 2070.

I would be interested in your temps though

 

I've done more tests:
Each test was run for 1 hour in Metro Exodus (on Ultra RTX, 1920x1080p no DLSS, Extreme Graphics preset in the same main menu - 1st menu with the train in the depot).
All the tests were run using the Balanced Fan Profile.

From what I saw in the metrics there is no difference between 1.5.0 and 1.7.3 / 1.8.1 Bios (except the forced throttle @ 79C - which can be bypassed with HwInfo64 in 1.7.3/1.8.1 bios, or it can be avoided using an OC profile in Alienware Command Centre)

My results are as follows:
VBIOS 180W -> 69-70 FPS - max temp 78-79 Raw OC of 130Mhz core and 500Mhz VRAM. Actual results: 1950Mhz Core, 7500Mhz Memory -> Power Draw 178W stable. (1 hour run)
VBIOS 200W -> 69-72 FPS - max temp 84-85 Raw OC of 130Mhz core and 500Mhz VRAM. Actual results: 1980Mhz Core, 7500Mhz Memory -> Power Draw 197W stable. (1 hour run)

VBIOS 180W -> 68-70 FPS with Temp Limit set via Alienware command centre to 76 Celsius and in thermal throttle -> 1935-1950Mhz Core, 7500Mhz Memory -> Power Draw 166W stable. (in around 10 minutes we get to 75C).
VBIOS 200W -> 70-71 FPS with Temp Limit set via Alienware command centre to 76 Celsius and in thermal throttle -> 1950-1975Mhz Core, 7500Mhz Memory -> Power Draw 176W stable. (in around 10 minutes we get to 75C).

As a result, I can say there is no actual benefit in running the GPU at higher temperatures, with more power draw as the results in game are the same. (1-2FPS at 70FPS avg. can be ignored). I did not measure the abs. low FPS and the abs. max FPS. There, I believe we would see better differentiation. However, in a real case scenario, it's irrelevant.

Both VBIOSes were tested on BIOS 1.5.0 and 1.8.1.

Yes, people are right!
You can run with a max temp of 80-ish Celsius on the GPU based on the VBIOS. However, in Bios 1.5.0 we lose G-SYNC support on newer monitors and I really don't see any benefit to it, except more power usage and higher temps for the ~ same RL performance.

I am really curios to understand what benefit would be to run the GPU at temps above 75 Celsius, as I am really struggling to see

Cheers,
Helifax

PS: for a moment there I thought there was a hardware difference between the earlier a51m models and the current one, but based on the tests, I don't believe there is anything "gimped".

 

Nice test you made there. I've never even bothered with the new bios because really it is not needed for me because i keep g-sync and don't win anything over a new bios. I just remember upgrading the bios and instantly getting a useless laptop. Do keep in mind that these results maybe perfect for your gpu but other gpu's can be totally different in terms of thermal headroom/performance.

 

@Prema I have Thaiphoon Burner, a So DIMM adapter and a PC to flash the dimms. Could you please give us advice, what to change, to get an 2933 CL18 profile on those dimms? What Info can i provide to ,,help,, you helping us? ^^

 

Great post and test results. My only small concern is running Far Cry 5 /Division 2 on the highest settings, I observe constant thermal throttling from about 5-10 mins into the game. Using Riva I can see the GPU clock doesn't drop below 1700mhz so it seems fine. But does running it constantly whilst thermal throttling not cause any issues to the GPU?

I'm on BIOS 1.8.1 and vBIOS 180w, I use GPU Tweak to set the thermal limit at 78c as AWCC messes with my Throttlestop settings for the CPU which are set at 5ghz all cores and -100mV. GPU is completely standard as I have no idea what I am doing. I attempted setting a curve but the voltage still spiked way above what I had set it to so just went back to standard.

 

I hate situations like this... should I go with the model I ordered, or wait and see?

:-(

 

do you need the performance NOW, then buy NOW.
do you need it LATER then buy LATER

There are not even trustful rumors about an R2 for now.

 

No, it's exactly the opposite. There are multiple throttles/limits in place for the GPU to act as safety guards for the GPU, VRMs, etc:

- Power Limit - When the maximum set TDP is reached (in this case 180W or 200W based on the bios), the GPU will stop increasing the boost frequency in order to stop drawing extra power. It will try to maintain the current frequency with the maximum allowed power draw.

- Current Limit - When the GPU tries to increase boost frequency but it will require more current to achieve that frequency. If the current required is outside of the specification, it will limit the current to the max allowed value and the boost clock for that frequency. (If you open the frequency/voltage curve in MSI afterburner you can see what frequency corresponds for what voltage).

- Temperature Limit - The GPU will follow the frequency/current curve, if power & current & temperature allows it. If the maximum temperature limit is reached, the GPU will stop: drawing extra power, increase the current and increase the boost clock. It will stay at the currently achieved frequency/voltage.
If the temperature still goes up, the GPU will scale down the frequency/current to a previous value in the effort to cool itself, even by 1C. Once it is "cooled" it will ramp up the frequency/current again. In practice, once the thermal limit is hit, it will hover around a few close values for frequency/current.
A simplified example: no power/current limit, core boost: 2050Mhz. Temps go up. It reaches temp limit of 75C. It down clocks to 1975Mhz. Temps still go up. It down clocks to 1950Mhz. Temps go down. It increases clock back to 1975Mhz. Temps still go up. It down clocks to 1950Mhz and so on and so forth.

It is actually safer to run the GPU with the temperature limit than without as the temperature and frequency/voltage are optimised for best results.
The only downside is you can't achieve maximum clocks (in my example from 2050Mhz the frequency dropped to 1950/1975Mhz. In practice this translates to 1-2 FPS loss). Ideally, you want to watercool the GPU and always run at maximum clocks if you want to push exotic resolutions like 4K, 32:9, 21:9 etc where every FPS matters.

Hope this helps!

Cheers,
Helifax

 

Oh don't get me wrong, I feel the same. OTOH, if there is and effective solution, at least that helps a good bit to take the sting out, perfect or not. My point is just its unlikely that a "fix" will ever come forth; if they didn't solve this frankly by Summer, its never happening in my book. OTOH, a solution as described is not a bad place to be.

 

I would think that if a 1440p panel existed they would have by now started offering it after the new year or in time for Christmas sales, Dell has offered new panels before during a revision I'm pretty sure.

 

I really appreciate your taking the time to explain in detail how this works,

And I HAVE seen it appear to work that way, however too many people ( Bob of all trades for one in his smack down video on YouTube)

And many of us in this forum see the other extreme all too often

The GPU thermal throttles down to 300-500 MHz and stays there long after the GPU and CPU are both cool, your game turns into a stuttering slideshow.

It would be really nice if all our machines worked as well as yours appears to work.




Sent from my iPhone using Tapatalk

 

All machines work like this if they are setup.
Currently there is a bug in the Alienware command center OC profiles. If you have no overclock profile, there is no "soft" thermal throttle setup. As a result the GPU will keep on warming up until it reaches the hard temp. throttle limit of 79C. At this point it will revert all the clocks to 500Mhz until it cools.
Every GPU has this. Normally is set to 87C (or what is the max one according to specs.) This is done to prevent your GPU from burning (and killing other components in your PC as well). This safety mechanism is there in case the cooling solution fails.
Alienware decided to move it from 87C to 79C without implementing a default soft lock throttle at 75C (which is enabled when an OC profile is set). This is a BUG that Alienware SHOULD HOT PATCH ASAP!!!

Like I said before, alternatively you can make a dummy OC profile (without any overclocking) and just set the GPU max temp to 76/77Celsius. Then everything will work perfectly fine as described above.

Hope this sheds some light and avoid any confusion.

Cheers,
Helifax

 

With this in hand I might one day when the prices fall upgrade from the 2070 to the 2080 then... but I'm holding out some hope that a 3000-series might appear for the R1... but I bet it doesn't.

 

Ok thank you so this explains why I was getting confused by different results when trying out different configurations and going back and forth between using AWCC or Afterburner to control the GPU.

I knew something weird was going on but I didn’t understand that AWCC was imposing the 75c throttle, I beloved it was the VBIOS or the BIOS.

So just to make sure I understand correctly, setting a dummy profile with thermal at say 77c with 0 Core and 0 Memory AVOIDS the hard thermal throttle at 75c ?

And best to just use this dummy profile and control GPU with Afterburner for reasonable overclock.

Which you find eliminates the hard throttle issue and you never run into the 300-500mhz throttle down issue ?





Sent from my iPhone using Tapatalk

 

I see a lot of talk on here about the 2080 throttling to 500mhz with newer bios...does the 2070 share this problem?

 

no it's safe

 

Yes, that is the idea! Just making a dummy profile with a target at 75/76 Celsius will avoid the problem.

And now it get's interesting... I am on bios 1.7.3 and I checked for updates today using the Alienware Update program. It says I am up-to-date. Which is strange as I know bios 1.8.1 is around.
So I went and both Googled and logged into my Dell account and used my service tag to search for updates. There is no bios 1.8.1 anymore?! I wonder why it was removed...

I made a video showing the GPU exactly before the nasty 79C hits with the huge thermal throttle. I also show, how you can make the profile in Alienware Control Center to avoid the hard thermal throttle, as it is clearly seen in MSI Afterburner OSD.

I am running the latest Bios (now) 1.7.3 and VBios 1.0.0.5 (180W) - with GSYNC. You can see from the video that putting the temperature limit to 76C will actually make the GPU use higher clocks while maintaining a lower temperature (75C) and power draw.



Hope this helps and please let me know if you have questions!
Cheers,
Helifax

PS: I apologize for the lack of mic sound. I didn't realised it was muted until after
PPS: If it not obvious, any time you change a profile in Alienware Command Centre it will reset the values set by XTU or MSI Afterburner. Thus, you will need to re-apply them.

 

Thank you for the video! And your experiments I see you have an excellent taste in games

I am not sure when the “ throttle “ was introduced was it with 1.8.1 ?

Maybe there is benefit to running 1.7.3 ?

I definitely think you have a golden GPU there you’re lucky to be able to push the memory above 300 I start getting flashes and crashes above that.

Everything I read says that being able to push the memory higher gets you more gains then pushing Core.

As far as Dell update goes lol I think we have all experience with that software not finding anything / most of the time.

And Dell drivers, software, BIOS all tend to go missing all the time on Dell site, personally I think it’s just another sign of the chaos going on there, not just a Dell problem but everywhere now these days sadly, shoddy no attention to detail left hand does not know what the right hand is doing etc.


I actually don’t have any of that Dell stuff installed, nothing gets installed by anyone including Microsoft unless I go out and grab it myself, all those bloatware background processors are removed from my system.






Sent from my iPhone using Tapatalk

 

I noticed 1.8.1 was removed from Dell's site yesterday too! Very strange!!!

 

Sorry for not waiting enough in the video to show the horrible throttle that appears at 79C. I will try to make a new video (while keeping it to a minimum, as it takes some time to get it to happen).

I just Googled "Area51m Bios 1.8.1" and google finds a link to the dell website. However when I follow the link I get this:
https://www.dell.com/support/home/uk/en/ukdhs1/drivers/driversdetails?driverid=t8gpn

" Get the latest driver
The driver you are looking for is no longer available. To view the latest driver for your device, please enter your product details."

I find it very strange... I've never seen this before (I have 3 Alienware laptops and I've been using them since 2011 - The Nebula Red is still my favourite colour ^_^). I always kept the dell bloatware around (but disabled all the auto-starts and periodic checks as they are annoying and resource consuming).

The hard thermal throttle apparently was introduced with bios 1.5.4 (personally, I haven't tested it). My unit came with 1.7.3 and later it forced 1.8.1 on me (with all the Intel Security fixes - which made my scores in benchmark lower quite a bit, so I reverted back to 1.7.3 and ignored the "Urgent Update" on the Bios). I did test 1.5.0 which does have the hard thermal throttle set to 88 C (unlike the 79C in the newer Bioses) but I also lose G-Sync with that bios, as my unit is from January 2020.

I do wonder if we are going to see a newer bios in the future and more importantly what will it contain... (Hopefully a fix that doesn't allow the GPU to go directly to hard throttle mode, completely bypassing the software thermal limit - which currently happens if no profile is selected).

Cheers!
Helifax

 

Whoo maybe that means a new BIOS is on the way?

We will have to wait and see, why is it always a mystery lol everything they do no one at Dell knows about it even after it happens oh. Well wait and see.


Sent from my iPhone using Tapatalk

 

Is lunar white best avoided if you plan on using the internal keyboard a lot? Or is it ok as long as you take care keeping it clean?

 

Its not a mystery. Azor did it to screw with 51M user's.

Come on.

You guys were thinking it...

 

Well I just now tried to reproduce your results
I’m back on VBIOS 180w and BIOS 1.8.1 and I setup AWCC 0 Core 0 Memory / launched afterburner and applied a moderate OC 80 80 / hard throttle to 300 after 5 mins on the Metro Exodus loading at 1920x1080.

So I went back and reset to 0 and 0 in afterburner tried it again same hard throttle 300 after 5 mins and it never recovers the GPU immediately drops to 53c but MHz never goes back up.

Something different between our systems I’m going to try it on 1.7.3




Sent from my iPhone using Tapatalk

 

Put the limit to 76C ( On 77/78C sometimes the GPU will spike to 79C and will trigger the hard throttle.)
Make sure you test and save the profile and that is enabled (you can see from the main menu of the Alienware command center).
I see no reason as this should not work (unless the OC profiles are not applied by the software?! which could be another bug in the AWCC )

Cheers!
Helifax

 

I tried similar I set to 80 thermal in AWCC and now on 1.7.3 and 180 w VBIOS same issue even on performance fans hard throttle to 300’s around the 5 min mark.


Yes I always check that the profile is selected on the main screen as well I have seen that behavior all the time. AWCC is wonky.

I’m glad it works for you, but some of us have this issue out here and sadly it’s not simply a matter of copying your settings to make it go away.






Sent from my iPhone using Tapatalk

 

They even stopped update this Download site... Product support for Alienware Area-51m

Always save files even if you don't use them right away!

 

That is the problem. You are setting the soft thermal limit to 80C which is right on edge (or above the 79/80C hard thermal limit). Set it to 76C and you will see you don't get the problem (as I've shown in the video).
There is no real benefit between running the GPU at 75C or 80C. You will get the same FPS in both cases (with lower temps and lower power draw/usage).

There is no reason for this not to work.

Cheers,
Helifax

 

I made a video which can enable 2933 on area 51m

 

Oh I did I tried 76c as well , it hard throttle at 3:32 mins lol

I just rolled back to 1.5.0 and have been running Metro for over 12 min no throttle.

It could all be down to how wonky AWCC is even between uses on each machine I really think you can’t expect it to do the same thing twice in a row.





Sent from my iPhone using Tapatalk

 

What you need: 3000/3200 or higher sodimm memory sodimm to dimm adapter thaiphoone burner spd file.

How to: to download this spd file https://drive.google.com/file/d/1OjmU... open thaiphoone burner, flash xmp and change Manufacturer to A-DATA Technology, Part Number to AO1P32MCST2-BW4S, Serial Number to B3710100h, DRAM Die / Stepping to B.

Enable 2933Mhz: put a flashed memory to slot 1 boot up and enter in bios then ctrl+alt+del to reboot, and you will find the pc stucked on bootup press power button 5 seconds to shut it down then boot it again and enter in bios. Done!

 

Thanks, and someone find a way to passthrough 75℃ gaming and 78℃ furmark without throttle.

 

I doubt you'll find a 100% viable fix. It's about the choice between Intel's security patches and for some G-stink vs. the old firmware from spring last year(Pest vs cholera - some need the Intel security patches due workplace requirements). Dell should fix up in own mess before they release the new refreshed models. + the new fixed firmware should have support for latest 9th gen processors who is intended for the Z-390 chipset in this upgradeable notebook. Everything else means you never got what they promised with the beginning of last year. If you accept nothing, then expect nothing at next crossroad as well.

 

to find a way to avoid throttle under bios 1.8.1

 

I know The older haven't the problem. Neither have the rest of Alienware models. The thin one (m15/m17).

 

high frequency memory need 1.8.1, make your choice, haha

 

Good guy Apple in 2020, the company that prioritized thin design over actual function and purposely makes it more and more difficult to the point of lawsuits for 3rd parties to fix their machines.





Apple is more nefarious than the majority.

 

Anyone got bios 1.8.1 saved somewhere ?

If we only could find a throttle fix for the 1.8.1 that would be the holy grail for everyone owning the 51M -

 

Ive posted a quick guide here, you can follow it.

http://forum.notebookreview.com/thr...bration-overclocking-tweaking-presets.831874/

 

https://1drv.ms/u/s!AslCY6JaX-EljbkY3MMmEgEug2p2SA?e=hzJZbe
1.8.1 BIOS there, let me know if it doesn't work.

If editing the SPD data is really all it takes to get the RAM working, then hopefully they push a fix for this. Someone should be losing their job over this entire thing. Hopefully Umar pulls through on this.

 

Is bios 1.7.3 current?

 

I might be wrong but I think it can throttle, but the GPU runs so much cooler and with the relative beefy cooling on the A51M, you rarely see it.

1.8.1 is current.1

 

Was current - until they pulled it.
1.7.3 is now the "latest"

 

Oh they pulled it - any word why ?

 

I am sorry to hear that. Out of curiosity did you try to see if you have the updated AWCC and OC controls installed? I saw there are quite a few updates on the Dell site. My machine came installed with the latest versions.

Could this be the reason it doesn't work for you, but it works on my machine? It sounds as the OC profiles don't do anything in your case. Did you try to overclock the CPU/GPU just to see that what you specify in the profiles actually applies?

Cheers,
Helifax

 

Interesting that they pulled 1.8.1, maybe there's a bug in there after all.

Or *tinfoil hat*

Maybe they're mad that people are so close to getting their ram to run over 2400mhz

 

I am very curious as I am on it currently as are probably quite a few folks...

 

Not here for the re hash of the old tired PC vs APPLE wars friend.


Sent from my iPhone using Tapatalk

 

I appreciate the follow up thank you,

Yes I’ve been diligent in updates at Dell/ Alienware as well as here.

I actually began reading this forum since shortly after I ordered my first Area51m ( 9700 - 2070 which I returned when I realized I would forever wonder about 9900 - 2080 lol )

I actually have read EVERY post in this thread took me weeks, and I think I have tried every trick that had been thought of by all of us.

Sometimes things get a little better for a while and I think OK that’s that it works ok now, but then I’ll get the itch to try and wring a little more performance out of it , or try to figure out and overcome the throttle issue,

Usually I end up spending a day or two wasted on reboots and benchmarks / downgrade and upgrade BIOS /

I’m sure everyone here is familiar with the drill


The thing that really gets in my way is the inconsistency of it all, the way Dell has this laptop locked down makes it really hard to troubleshoot ( at least for the way my mind works

I keep getting slightly different results, I’m thinking that to really do this right it’s probably best to pull the CMOS battery out each time you want to make a major change, it’s looking to me that Dell has some sort of unusual deep integration between their custom BIOS and AWCC and it messes up when you try to change it.

I also think the way they designed this they actually expect that everyone will use AWCC and create a profile for each game, and the default profile is Performance fans, I expect they consider performance fans to be required for games.




Sent from my iPhone using Tapatalk

 

Haha, I can relate...

Thanks for all the info (it was very interesting) but I prefer not relying on an AWCC bug to make my rig work correctly. I'm currently back to 1.5.0 (thanks to NBR) and with thottlestop to OC / UV, everything is running smoothly without those terrible fps drops when playing with 1.7.3.
CPU is @ 5GHz, -100mV and GPU is +150 memory and core and I never get over 80°C for the GPU, so I really don't see the issue here.