*OFFICIAL* Alienware 17 R5 Owner's Lounge

I feel bad for all you guys having troubles, this is what happens when AW uses a flawed design to cool significantly more powerful components. They literally used the same design (with a few minor tweaks), unbelievable. Lol when people said they'd change the cooling for the R5…

 

In my case it is weird - GPU-Z and HWinfo both show VREL ALL THE TIME when the GPU is doing something productive, like playing a game... I think the VREL they have got it upside down, it should be ON unless something is bad and then it is OFF. I have not proven this yet, and anyway maybe a little bit academic as my GPU works well (touches wood).

 

Yes, I think I can. My laptop was working great, until I decided to go to NVidia and download the latest drivers. After installation all my games would stutter horribly, WoW was freezing for seconds at a time, Far Cry 5 was unplayable at all. I thought the usual "thermal" issues and whatnot, in the end I went to Devices, uninstalled the GTX 1080, rebooted maybe, I cannot recall because I was not keeping notes at the time, and then back to Devices again and installed the drivers that Windows chose and the driver version was a whole bunch of releases back, not the latest. Been working fine since. I do get the occasional stutter, but it is rare, like once in 30 minutes maybe, in WoW, and I think it is when it goes to disk to load something new, but it could also be something else, like thermals, as I have not proved it yet.

 

I actually still have the stock drivers. I may just return if I can't figure this out. I still can't believe this thing is hitting 64db when gaming and still stutters. If anyone else has any tips I'd be much obliged

 

What version do you have?

The latest from NVidia is 398.36.

Mine is 389.12 or 23.21.13.8912 if you prefer. Check yours.

 

I appreciate that. Lol just got out of bed and turned it back on. I’m on the same driver version as you

 

When the machine boots, it displays "Alienware" press F2 (quick) to get into the BIOS, check the fan settings. Then, if you have installed Dell Command Centre it has its own "profiles" which may be turning the fans on.

OK then I can only start guessing.

In order of ease I would do the following:

1) I would install HWinfo to monitor the CPU and GPU temps. If the GPU is very hot for example, it may be throttling.

2) Then I would go to Microsoft and download "Create Windows 10 installation media" https://www.microsoft.com/en-gb/software-download/windows10
which burns a USB stick with Windows 10. I would do that, go to setup, zap the hard drives, delete all partitions, re-create the partition and install a fresh Windows 10, not update, not repair, install fresh. Windows 10 would not know about the NVidia and about a dozen other bits on the laptop (like Tobii) but we want a clean install at this stage. You will need to have the Killer Wifi drivers on the stick as well because Windows 10 cannot use the wifi, this you download from Dell name is "Killer-Wireless-AC-1550-WiFi-Driver_YF1T4_WIN_1.6.1858_A01.EXE" or similar. You also download the Nvidia drivers from Nvidia but an old version like 389.12, on the USB stick too.
With such a "lean" Windows install you can then try your games and see how it goes.

3) You can call Dell Premium support and see what they say. It is their job actually and they may fix it for you. However, if they advise to send the techie out to replace your heatsink or motherboard then I would just return it and at the same time order another one to see if it works better. That is what I did and the second one is so much better.

 

I will definitely check out the fan option for future reference. I wish it was a heat problem. I should’ve let you know I posted the temps on the previous page. If it was the heat at least then I’d know. I think I’ll just return it at this point. I don’t have the time or energy to go through having support come fix the issue or dealing with this on the first day after spending $2300. Maybe I’ll get another one if they do a major update in a few years. I appreciate all the help thanks for detailing out a response

 

Try nvidia drivers 391.35, all drivers after this version on my r4 cause stuttering in far cry 5 ac origins. I belive it may be a gsync issue. And at the moment I can’t update windows to 1803 to see if that resolves it unless I force the update which I won’t be doing.

 

I re-ordered my system with some upgraded components eg. 4K screen, killer 1550, 16GB ram as default (previously default was 8GB) but to be upgraded with my purchased ram modules and backpack. It was same price as previously (in fact a few less) using discount code and dell offers. It should be received within two weeks. All I now is that this time I am going to try CPU undervolt and repaste with MX-4 (yes MX-4 and not thermal grizzly) before trying anything else (LM and thermal pads). And this time, I wont accept any heatsink replacement or anything. Test the system extensively within the two weeks period and return again if needed (in that case it will be the end of AW for me).

 

I had a similar issue on my pc and I found that the resolution scale was set to 1.5 in the advanced settings under video settings in the game. turned that down to 1.0 and it fixed it.

 

You're all going to have issues and it's all due to unbalanced heatsink and pad placement. If you don't do anything about this yourself, you'll still have a problem

Sent from my SM-G960F using Tapatalk

 

@ everyone who's having stuttering

What Pete said above me, and also have GPU-Z open when you play a game that stutters. I spent literal hours upon hours looking for a reason my 7820HK/1080 system did this, and it was the VRMs overheating. If vREL is displayed when you mouse over the "PerfCap reason" bar in GPU-Z and your GPU is also 0% in the usage bar at the same time it says vREL, you have an issue with the power delivery components.

Now, I didn't have the time or patience to unscrew my laptop for the 50th damn time so I just contacted Alienware support who gave me an old vBIOS that didn't OC the GPU. Kaby Lake machines and all R5s ship with components factory OCed, yet they all use the same horrible cooling design, paste, and pads.

Downgrading pretty much solved the issue totally, at the expense of giving less power to the card; I'll happily make that trade-off. Since there is no version of the R5 vBIOS that isn't over clocked, you can try manually reducing voltage to the GPU and see if that helps.

“ Those who cannot remember the past are condemned to repeat it.”
http://forum.notebookreview.com/thr...gtx1080-in-game-random-stutter-freeze.804499/

 

My card is permanently on vREL anytime it's doing anything involved, from 30% upwards. It does not stutter, and it consumes up to and over 180W happily.

I am not convinced that vREL is to be taken at face value. Maybe vREL with a concurrent drop in utilisation % - then maybe yes. I quote my post from the R4 forum:

https://hardforum.com/threads/gpuz-perfcap-reason-vrel.1935504/

it says:

"VRel just means the core is unable to step up to the next voltage bin on the Boost curve because the core will become unstable, thus limiting maximum Boost clock. Generally it's nothing to be worried about, as all cards will get this perfcap when running games because the Boost curve is defined beyond the maximum voltage that is reasonable. For example: The Boost curve on my Titan X doesn't stop until around 1.4V, a voltage which would probably kill the chip in very quick order."

 

If GPU usage drops to absolute 0, that means voltage or lack thereof is causing the card to completely stop outputting frames at that point in time, which is 99% of the time a power delivery issue especially with laptops. But, of course, if it's not affecting anything and it says vREL, it's fine. But it wasn't for me.

 

There are various reasons why a game might stutter:-
Windows update / software
Nvidia drivers -
Actual game
Gsync issues

The issue with the vrams getting hot on the r4 1st revision was heat, however dell corrected that with the 2nd revision motherboard on the r4.

Also the symptom of the vram heating was the game froze for 2-4 seconds and then carried on.

Vrel being displayed is not a big issue at all in gpuz, you shouldn't be worried about Vrel and VOp. They are both quite common.

 

Don't think so, mine was manufactured 2018 and it still stuttered/froze. I have seen R5s with i9/1080 doing the same in person, and sure enough it was vREL. Like I said, it's only significant if accompanied by the marked freezing you mentioned, and a GPU utilization of 0 at those points.

 

yes but do not forget that different manufacturers, use different vbioses. on top of that is that Dell is a OEM with it's own ways (like locking the memory for overclocking in the bios, while other manufacturs have the option to do it in windows or bios). did you forget that the debacle begin in the first place after uploading the OC vbios. the R5 is no different from the R4 except some minor changes (like an extra two cores) coffee-lake is built on the same 14 nm skylake and kaby-lake where build, and is effectively called 14nm++ . my chip is also under a constant VREL @75C when benchmarking/gaming but it does not stutter. VREL is just a flag for a bunch of different problems, the real reason can only be known if someone can extract to VBIOS, stuttering can also be from the PCH, but when it's your PCH you cpu is also boiling under the keyboard before the GPU does. so if your PCH does not go above C81, and your cpu does not thermal/power throttle, it has to be the VRM's

 

Not strictly true as far as I'm aware (happy to be proven wrong) but from my experience it's only when you experience VRel and GPU load falls to 0% at the same time, that you have an issue.


Sent from my SM-G960F using Tapatalk

 

I meant 0, not 100

 

Unfortunately it was already on 1.0

I’m at the end of the rope here. The temps are fine and I can’t figure it out. Here’s another screenshot of gpuz while playing. If I can’t solve this in the next few hours it’s a definite return for me

 

Undervolt the gpu and see if it helps, if it does its the gpu vrm. If you have throttlestop disable bd prochot and see if the stuttering goes away. If it does, then it could be a few things but we'll see.

 

I tried it without using throttlestop altogether and unfortunately still had the issue. I appreciate the gpu advice. I actually just got the return shipping label. I'd rather not deal with this problem. I can still undervolt the gpu if you want. It doesn't seem like that is the issue though

 

Little confused here. What did you try if you didn't disable bd prochot in throttlestop or undervolt in msi afterburner? If you try these it will take you 5 minutes for each task.

 

Bought one of these to test. I can literally watch the upper layer of the keycaps move up and down when pressing keys, especially in the ten-key area and the tab/tilde/escape area. Just like the R4 I had and returned. I dunno how they can advertise a keyboard with a steel backplate that does this and feels like the Clevo keyboard. Ah well, back to the store it goes.

 

Sorry I misinterpreted that. I tried the bd prochot disable in throttlestop and that actually ended up making the entire computer a little slower. Super weird. Tried undervolting with afterburner and still getting the stutter. Any other idea what it could be?

 

That's odd that it would make things slower. It shouldn't at all. When it was disabled did you try gaming?

 

I did try gaming as well. The stutter stayed the same. I meant outside gaming it made things slower

 

Known issues latest nvidia driver:

• Using power monitoring in GPU monitor tools causes micro stutter. [2110289/2049879]

So if you have stutter try without osd/ monitoring. I had this stutter when using Afterburner and importing GPU Power via hwinfo.
Maybe an minimized hwinfo can cause it too. So be sure its closed before you try

 

Thanks for letting me know! That's very interesting. Unfortunately it was happening before I got any monitoring tools, which is why I got them :/

 

Oh ok thanks. Strange. Did you ever post a hwinfo64 full screenshot?

 

I didn't, but I can fire it up and post one if you'd like

 

Huh. I must confess when I always went to check temps and vREL presence I never considered that the tools I was looking at could themselves be the cause of it!

 

yeah it was just an idea from me that hwinfo cause my stutters and tried without... and I didn't saw any stutter since I don't use hwinfo to monitor ...today I saw the known issues point and now I know why I hadn't any stutter and was happy that my thoughts didnt play a joke on me

 

I have the same around the ESC key corner, and nowhere else. I think it is not bolted in correctly. I tried to take a video of the problem and show to tech support but the movement is so slight does not really show.

I decided to keep it. If/when I open it I might have a look.

 

Can it be the actual game causing it? For example in WoW when the game tries to load anything from disk then it stutters / pauses, maybe 1/2 a second. So I try to load the map, or look into my inventory, and there is a small stutter. I can more or less do it at will. This is the game's fault, considering that these disk loads should happen seamlessly in the background. When I connect the Bluetooth earphones the game freezes for over a second, maybe two seconds.

Unfortunately we do not have a software method to measure the VRM's temps...

 

Has anyone purchased recently the Alienware 17 r5 Core i9, GTX 1080, with 1440p 120Hz screen?
I want to buy one, but need to be sure if thermal issue has been fixed.

 

The thermal issues are a lottery. Mine bought 3 weeks ago had thermal issues and I returned it. I didn’t really intend keeping it tbh, I just wanted to see for myself the differences between the r5 and the r4. The thermal issues made the decision easy.

But really if you can keep your monies and wait for the refresh - now isnt the time to buy a serious gaming machine with the new gpu architecture - try to wait it out til September ish .... or buy the discounted r4 if you can’t wait ... I did a mini review of the r5 in an earlier post in this thread if it helps

 

Most annoying bug ever. Now I know why I never encountered it before until someone mentioned it then I set my TDP back and sure enough....yuck. I"m using an unlimited TDP 1070 so it never hits the power limit.
Not using monitoring tools greatly reduces it but still doesn't eliminate it fully. The best way to to reduce it to the far minimum is to either:

1) set Nvidia CP to prefer max performance.
2) Lock a voltage point on the MSI Afterburner graph with "L". This will also prevent the card from downclocking. You can try combining this with using monitoring tools to see if that helps.

 

VREL means nothing for us in the end. You can ignore it. However there have been cases of desktop cards not triggering VREL when they are supposed to.
It simply means that the card's boost speed is being limited by the maximum voltage being sent to the GPU. Instead of power limit preventing the boost clocks from going higher, the current voltage (reliability voltage) prevents the boost clocks from going higher.

The difference between VREL and PWR is that PWR flag (Power limit) causes the card to jump to a lower voltage / clock point on the graph, which reduces clocks and GPU voltage until it can boost up again. Note that it's possible to trigger power limit without a decrease in clocks (this may be a vbios bug) if the TDP gets very close to the maximum TDP without actually exceeding it.

One thing I found out about VREL is, and this may be yet another bug:
If you "flatten the curve" or lock a high voltage point on a certain driver, you may not have VREL flagged. But then if you switch drivers and use the exact same configuration but a *NEW* Profile with the exact same curve or locked point, you DO have VREL flagged. This seems to be a complete bug, as switching from the "old" saved profile that was created with the old driver, but using the NEW driver, and the new profile created with the new driver causes VREL to either vanish or reappear (e.g. old profile vs new profile with identical configuration), although absolutely zero performance differences or changes. The curves were 100% identical, exact same configuration. Go figure.

In other words, don't worry about it.

 

Few questions.

Do you guys think the R5 is worth getting? I understand some have thermal issues and some don't. Also what hardware comes with the R5? I was a previous owner of an R3 which had disappointing stock SSD drives. Does the R5 come with a samsung ssd? And in your opinion, would the UHD(IPS) be better over the QHD(TN)? I do game but not fps gaming. I mostly play single player games and take part in graphic/video editing. If I do play multiplayer games it's at most league of legends. This laptop will be used for a ton of editing(photo, video, graphical) along side single player games and watching movies and other projects.

 

Yes. It is nice hardware. But I wouldn't pay for the i9 version personally (few extra MHz ain't worth that money for me), particularly the work you will almost certainly need to do to stop it going full thermal-meltdown Chernobyl China Syndrome burning a hole to the Earths's core DeathStar mode on you.

Depends on your config. AW don't use Sammy SSDs. Get a stock HDD specification and buy your own Sammy to clone. Far cheaper than paying the AW tax.

The UHD is worth getting even if you DO play FPS. For your use cases, it would be a winner I think.

Best of luck.

 

Does anyone know which driver/application is responsible for Fn+F11 (switching the trackpad on/off) ? I have installed a very lean system and the Fn keys work except the trackpad and would like to be able to switch it off if possible?

 

The touchpad and pointing device drivers maybe???
https://www.dell.com/support/home/uk/en/ukdhs1/product-support/product/alienware-17-r5/drivers

Failing that, why not just disco the touchpad in Device Manager?

 

A few reviews... http://forum.notebookreview.com/search/20392249/?q=Review&t=post&o=relevance&c[user][0]=612304&c[thread]=815492

And nope. You have to ask what ssd’s will be added.

 

My previous AW17 R5 was reaching 52C on the keyboard, right shift key, this one, the worst I have seen it is 45.5C, and it's 27C right now in this room.

My AW17 R5 with i98950HK and GTX 1080 is working well not getting too hot.

The UHD panel is a must have IMO, it is an AUO B173ZAN01.0

The SSD nvme is SK Hynix 512GB

 

Depends on the ,,Love,, you want to invest... do you want to use it out of the box? Then no.
Do you want to undervolt cpu and gpu and/or repaste then yes, its worth a try

 

Hey. Sorry for the late reply. In all the excitement, I didn't manage to get a picture of the new heatsink D:

I will try to get a picture as soon as possible. But yes, it seems like Dell has a new Heatsink Manufacturer ROBIN, I didn't see ROBIN heatsinks for the R4.

Though to be clear, my R5 is much more stable out of the box than my R4.

So far, didn't run into any issues. Got a Cinebench score of 1236cb (x39 on all cores, -0.100mV). Throwing this Alien in an air con room made it run as cool as a desktop.

 

IMO, if you're just a casual gamer, go for the i7-8750H and 1070 model, though, you will need to do a repaste. Don't get the i9 and 1080 since the stability of the units isn't worth the extra money trade off, especially you will still need extra money for LM repastes, repadding and etc, which doesn't fully guarantee stability, yikes.

I am using the QHD TN 120Hz panel. I was pretty skeptical about TN, but Alienware chose one of the best TN panels. It almost looked like an IPS. Plus, if you're new to 120Hz, trust me, u won't go back to 60Hz. As for the 4K, I guess it depends on you, but I personally don't recommend it since 4K isn't really optimal to date, not to mention, a 1080 may easily struggle in the future. But if you're gonna do productivity (Photoshop, etc) then yea, 4K. The 1440p is more than enough (don't see any SDE and the jaggyness is better than 1080p)

As for the SSD, I don't think it will be Samsung. So far, I have gotten SK Hynix SSDs. Not the best, but they are far from worst. However, despite Dell claiming these are "Class 50 NVMe SSDs", they don't perform as well as maybe a 960 EVO. Still, its better than a system with just a spinning HDD.

I am planning to get a 970 EVO in the future. The SK Hynix will be converted into a storage SSD.

 

Sure....buddy....justify by any means necessary to feel good about it.


Enviado de meu Pixel 2 usando o Tapatalk