XPS 15 9570 Owners Thread

I know my info is a bit fragmented, I work in a busy call center and do this stuff in my spare time between calls to keep busy. I am having to pull info from several sources spread across several days and bring it all together on short notice to give you all the most complete picture of what has been tried and confirmed thus far.


MY RESPONSES TO YOU:

1.) yes, this I know and concur is standard behavior

2.) I am doing much of this fully remote and relying on the patience of the person on the other end trying various test cases. For the most part his machine doesn't make it longer than 30 minutes most times if it is going to freeze, so we have been calling 1-3 hours of up time a relative success. However there was at least one test case we let it go overnight and it froze after several hours of up time. If I had physical possession of the machine I would do things my own way but I don't.

3.) I will work on pulling one from his machine.

4.) Setting NVMe Adaptive Link to "Lowest" has yielded the best power savings and C states in my experience.

5.) Not sure what you mean by "disabling C states is not an option"... there is a setting in BIOS to disable them at the BIOS level.

6.) One might think, however his old SSD exhibited the same behavior, so failed/failing SSD is highly unlikely imo.

7.) I don't think it is any of his PSUs for the fact that it happens on all 3: 1 - his Genuine Dell 130W barrel jack charger, 2 - the 240W external PSU that comes with the TB16 dock, 3- his 60W aftermarket (TB3 iirc) charger

8.) Failing / intermittent VRMs / Mosfets / motherboard have not been fully ruled out yet, but I would suspect them to consistently fail under heavy loads, which has not been the case... it tends to fail at or near idle... literally doing nothing just idling, or even something small like just streaming a YT video or doing light office type work

9.) We tried with Microsoft NVMe driver first and then with the Samsung NVMe driver... both have resulted in crashes.

 

Why did you change the NVMe link state, youre not allowed to do that with modern standby laptops. I asked a few questions and you didnt answer any of them. Who changed the link state and why and how, you cant do that normally via the energy profile, it is hidden normally with modern standby. Disabling the c states mostly could also be a workaround of it not triggering lower link states and somehow disconnecting the SSD. Tried to remove the SSD and check if all contacts are fine and insert it again, you said the issue happened with another SSD too?

What was the state of the laptop after a clean AHCI installation? Why did you bug around with it already after a clean installation like changing link state ect?

So the laptop also froze after clean installation and NOTHING changed NOTHING installed, no undervolt, no drivers ect?

 

I followed this posts:
https://www.reddit.com/r/Dell/comments/asdr64/dell_xps_9570_new_realtek_high_definition_audio/
https://www.reddit.com/r/Dell/comments/bbpuwb/9570_no_audio_output_is_installed/
https://www.realtek.com/en/componen...-codecs-high-definition-audio-codecs-software
And now audio works without any problem. Thank you very much for suggestions!

I have also found that generally usb hubs with embedded Ethernet port causes some computers to freeze. I guess I have to find any other workaround. But anyway, no need to update bios beyond 1.2.2

 

Anybody getting battery drain on their Dell XPS 9570 when the machine is completely shutdown? (I'm 100% sure it's not in hibernate or standby mode)

 

Hi everyone,

last week I had an accident with my xps 15. Liquid was spilled all over the laptop when it was running. When the liquid was spilled the screen went black. I immediately shut down the xps and let it dry for 48 h. Unfortunately this did not happen at home, so I did not have the tools with me to open up the xps right away. Two days later I opened up the laptop and checked for remaining liquid. Removed the fans, heat sink and mainboard. I did not find any liquid. Only two places looked like something crystallized there. This spots I cleaned with rubbing alcohol. After I put everything back together I dared to try to start up the xps again. It booted up normally an I thought I got lucky. Did some testing and everything seemed ok. Until I tried to start a game. There I realized that the game did not run as usual. After some additional checks I found out that the Nvidia 1050 is not recognized by the system (does not show up in the device manager). A Diagnosis in the boot menue did not reveal any problems. So I guess the GTX 1050 probably died when the screen went black after the liquid was spilled.
So now I am thinking about my options. Is there anything I could try to revive the 1050?

I am also wondering about a replacement mainboard. I have the i5 mainboard. Is it possible to also use the i7 mainboard as a replacement part?
Or maybe even the 7590 i7? Is it possible to get replacement mainboards directly from dell?
Does it make sense at all to try to to get a replacement mainboard or are they so expensive that a new 7590 makes more sense than trying to get a replacement?

Hopefully someone has some good ideas.

 

Well, you can watch videos from Louis Rossmann for DIY general laptop water damage repair tips. You really need to go over the board with a good loop or microscope else it is easy to miss damage. Some of those soldier joints might need touch up. Some of the components might be damaged. And it is difficult to assess any downstream damage without schematics. I suppose you could test some of the components in the damaged area on the board (and some might need to be removed) but then you have the challenge of determining and sourcing replacement parts. And getting SMD tools. If the GPU contacts were damaged, I don't see how that could be reballed DIY. There is a possibility the GPU is fried also.

There is a repair shop in Germany referenced in these forums a few months ago to consider if you are in Europe. This guy in California seems to repair several XPS models


Motherboards pop up on eBay but the quality will be "haphazard". Below is a US shop with a deep Dell parts catalogue. Either way, at these prices you might consider a new laptop.
https://www.parts-people.com/index.php?action=category&id=143&subid=1201&refine=motherboard

 

Thanks pressing for the reply.
Do you happen to have the link to that german repair shop?

 

No but if you search the XPS forums here there were posts in the last ~6 months with the details.

 

Here it is:

https://www.electronicloop.de/Startseite-1

 

Does the latest BIOS still stop undervolting?

1.8.1

 

Why would it stop?

 

How do you even get that new BIOS? It's not available on Dell's website...
I've turned off automatic BIOS update option (I forgot the exact name of the option) in BIOS, so I don't get it in Windows update...

 

Hey guys, I have a question - already some time ago I noticed my charging speed was limited to 40 watts. If the battery charge was below 35% it would charge faster, up to 80 watts, but when once it reaches 35% the charging goes down to 40 watts, even if the Express charge option is on in the Dell Power app.

I just reinstalled Windows, thinking something was wrong with my Windows, but again the same thing.
Do you have the same charging limit of 40 watts?
Maybe it was introduced in some recent BIOS (I'm on 1.16), or there's something wrong with my laptop/battery?

 

This is the italian Dell website.
I think he would refer to 1.18.1, not 1.8.1

 

Slow charging generally extends battery life (due to lower temperatures) so is a good thing for all your devices IMHO.

 

Yeah, I know that, but sometimes I need the fast charging. I just know the laptop used to charge much faster, and would slow down when the battery charge reached around 75%.

Do you still have your 9570? How fast does it charge when on standard charge?
I just need to know if my laptop/battery is faulty or this charging behaviour has been introduced for ALL in some prior BIOS.

I just tried to turn on the express charge option when the battery is below 35% and the charging speed even goes down all the way to 35W! That's slower than my phone charging. The standard charge option charges at 75W when the battery is below 35% and slows down to 40W after that.

 

A lot of nice reading Maybe enjoy it in coming weekend.
BLACK LIST. Adjustable voltage control/turbo ratio limits are locked out

 

I got if the Aus Dell site https://www.dell.com/support/home/e...t&oscode=wt64a&productcode=xps-15-9570-laptop

 

Ha thank you. actually I have read that some time ago and I'm familiar with Pkundervolt. Just made a mistake here and thought he is asking about resuming of undervolting in last bios. My bad, sorry.

 

Someone was able to install latest ThunderBolt firmware update for 9570?

https://www.dell.com/support/home/d...n&oscode=wt64a&productcode=xps-15-9570-laptop

It says for me cant install because I have to install latest Thunderbolt drivers first... which are installed. Or incompetent didnt update the drivers for the 9570 on their site.

 

I use the Dell update app and everything has always worked fine with that.

 

Anyone else please?

Latest TB drivers are installed obviously:
https://www.dell.com/support/home/d...9&oscode=wt64a&productcode=xps-15-9570-laptop

Version: 17.4.79.510, A12



Firmware update doesnt work:



Over the amazing Dell Updater it also doesnt work it says afterwards "failed:



And then it doesnt find the update anymore afterwards So thats another nice bug in the crappy Dell updater, that it just flags the update as done even it failed and then wont show it anymore.

Update: was able to fix it by removing the driver and re-install it again, even the same version was installed. Then it worked.

 

When it updated for me using the app it also updated the Intel Management Engine Components Installer.

 

New BIOS released.

 

I noticed mostly all latest Dell bios updates for all models had this change:

- Updated the video controller name in the BIOS to match with the video controller name in the operating system

what does that even mean? Anyone have an idea?

 

Can you still undervolt?

Anyway, the real reason behind this update (and all the other dell firmware updates right now) is to propagate a patch to the firmware installer which contains critical vulnerabilities.

https://www.dell.com/support/kbdoc/...ver-insufficient-access-control-vulnerability

 

I am still on 1.16.2 and wont update anymore. risk is too high, undervolt is forever blocked with a new bios.

If anyone has updated to latest 1.19.0 and can check if you can still enable undervolt by loading factory settings in bios, I would appreciate it.

And nope, those bios updates have nothing to do with DSA-2021-088, which is all about a driver, dbutil_2_3.sys, which I dont have on any of my Dell machines, because I dont install crappy Dell software on my machines. The link you posted says that explicit and how to remove the driver if you have it on your laptop.

All I want to know is what "Updated the video controller name in the BIOS to match with the video controller name in the operating system" means.

 

I've updated the BIOS of my Precision 5530, the business sibling of the XPS 9570, this week and undervolting via XTU ist still working like before (undervolting must be applied after every system start). So I'm pretty sure the same should also apply for the XPS but no 100% guarantee from my side.

 

It definitely doesn't sound like anything important... I can't see in what scenario this would be important or affect performance. But I don't know...

 

@Rippchen How did you obtain undervolt on those laptops, via the "load factory settings" trick? Or did you patch the EFI bios itself via another way?

I had access to a 9560, and the "load factory" trick wasnt working anymore after the last bios update. So I am not so sure to update my 9570.

 

Lol aren't you the one that argued with me here? Seems like you should be able to figure it out.

 

You have some serious problems.

 

This time I had to do nothing after BIOS update from version 1.18.1 to 1.19.0 to retrive undervolting capabilty. Everything was working fine after automatic update via Dell Command Update application and as to be expected without resetting the bios to factory settings. I own that Precision 5530 now a little bit more than a year and so I went over a number of BIOS updates. For this device I'm note really sure if I had to reset the bios even once going through different BIOS versions. I had to do so with some of my previous machines (Precsion 7510, 7710, 7520, 7730) but I can't really remeber, because finally I was always able to retrive undervolting functionality.

 

I did the update to 1.19.0 and can confirm, you keep undervolt, if you come from 1.16.0 with working undervolt enabled and update to 1.19.0.

 

Man, wasn't that a risky thing to do?? If it didn't work, you wouldn't be able to go back?
I'm also on BIOS 1.16.

 

Yeah it was, though I asked a few people on reddit too over the past days, who told me they did the update and kept undervolt. There is still the possibility to revert back bios cvars manually via efi boot console, though this could change in the future too of course.

 

@maffle What is the current situation with the S3 sleep? Can it be enabled on the latest Windows, and would it work at all?
I switched my laptop setup to using the laptop with closed lid and an external monitor. I haven't been using modern standby because the laptop doesn't sleep AT ALL when plugged in (especially with a thunderbolt USB+Lan hub). Fans are running all the time, wifi is on, emails are coming through with the sound notifications, etc. But the Windows battery report says the laptop is in standby, lol! So in practice, Windows standby = screen off. No difference.

Although, when the laptop is on battery it really goes to a proper disconnected standby and fans go off. But it still uses 6-10% of battery per hour, and it hibernates automatically after 2-3 hours, so it's useful only for a short standby...

So, the only way around it for me was to use hibernate instead of sleep, and that's what I've been using for quite some time...
But, with the new closed lid setup, in order to bring the laptop from hibernation I have to pull out the laptop, open the lid and press the power button, and put it back. Which is now a hassle...
That's why I'm thinking of trying to switch to the old S3 sleep from the modern standby, so I can wake the laptop with a keyboard or a mouse (Logitech USB receiver, bluetooth wouldn't wake it up).
If I remember from the last year, the main issue with switching to S3 was that you had to turn the bluetooth off and on every time you restart the laptop? Is this still the only drawback?

 

OK, I found a few articles and a reddit post by @maffle (thanks!) that explains how to activate the S3 sleep.
It was quite difficult/impossible after Windows 2004 because Microsoft completely removed the S3 sleep registry entry, but they enabled it again in Windows 20H2.

So now the S3 sleep can be enabled just by entering this line in the Command Prompt (as admin):
reg add HKLM\System\CurrentControlSet\Control\Power /v PlatformAoAcOverride /t REG_DWORD /d 0

If you want to go back to modern standby just delete the PlatformAoAcOverride registry entry.

This is documented by Microsoft here.

 

So I tried the S3 sleep and it works wonderfully. A proper sleep, like it should be. No complaints there.

However..... For the love of god, I can't manage to wake the laptop from S3 sleep with a keyboard or a mouse (Logitech USB wireless receiver, which is basically like a USB keyboard/mouse).
I tried many different things in Windows power settings and Device manager, but nothing helped. The only way to wake the laptop from S3 sleep was to open the lid and press the power button.
Btw, the laptop is docked, with a closed lid. Waking it from the modern standby with a keyboard/mouse works fine.

Anyone knows what the issue might be with waking a laptop from S3 sleep with an external keyboard/mouse??
I think that used to work before, or at least for sure on my older laptops...

 

@splus Sorry for writing back a bit late I was kinda busy. I dont recommend S3 on the 9570, the issue I reported years ago is still present and even worse with latest bios, yes, I will talk about this in a bit. I noticed a weird behavior after updating to bios 1.19.0 from 1.16.0, making things even worse if you use S3.

So yes, technically you still can activate S3 on the 9570, with the above registry entry. It will work "fine" entering S3 and also waking up from it. Yet there is this glitch after waking up from S3, which causes a "runaway" task on one of the cores of about 9%, not showing up in task manager, so it has to run internally on the EC or bios itself. I always reported it on this post over the years but also here:

https://www.reddit.com/r/Dell/comments/91313h/xps_15_9570_c_state_bug_after_s3_sleep_and_modern/ (scroll down to the mid part)

So if you use S3, you always had this 1.5W drain afterwards not going away anymore, until you rebooted Windows. Entering S3 and waking up, and you will have the problem again.

And now comes the super amazing thing. I tested S3 again after updating to 1.19.0 and... the drain is now DOUBLED. So you will have around 3W drain, that is a lot, after waking up from S3, and I also noticed a runaway task now on TWO cores. You can see it on here:



Before I updated to 1.19.0 I regularily tested S3 once in a while and it was before at around the usual 1.5W. Normal idle drain is around 0.4-0.6W.

That is on idle btw, nothing else is running. The two abnormal activies you see are caused by the S3 sleep after waking up. 99% idle shown in task manager. What is going on inside you cant see with any tool I am aware of. It must be some stuck behavior inside the CPU or bios itself.

That kinda totally ruins S3 for the 9570.

Also. That you cant wake up from S3 via usb devices is normal, and it was always like this with the 9570. For that to work, you need a bios, which supports "wake from USB" which the 9570 does not support. At least the setting is hidden in the Dell bios.

There is a "wake from usb" option in the bios, yet that just works for the Dell Dock you can put on the TB port. Those Dell Docks have a button, which you can press to wake up a closed 9570. I am not sure if that works though for S3, but maybe just for modern standby. I dont have a dock so I couldnt ever tested that.

Modern standby is as garbage as it always was, still randomly breaking and going nuts, with latest Windows 10 21h1.

The fan problems are still present, always triggering when the dGPU wakes up. Which btw always happens if you boot the laptop, put it to sleep, and wake from sleep. So the fans start to spin every time you put it to sleep or wake it from sleep, for no reason.

This for sure was the last laptop I ever bought from Dell. It is just disgusting how ignorant and incompetent a company can be.

These issues would be super easy to fix, and would give thousands of customers a good experience and happiness, which would result in them totally would buy Dell again.

I cant understand how incompetent managers could be at Dell, not realizing how important these little things are, and how easy they would be to fix.

 

The XPS is close to excellent. . . but there's always something!

 

It was excellent until they stopped undervolting.

 

Hi, I've recently gotten into an issue with my XPS 15 9570. The dGPU doesn't show up in the device manager. I've tried to do a hard reset on Windows but I see no changes. Does the BIOS system information show dGPU for any of you guys? I want to know if this is somehow a hardware issue or if a recent windows update has screwed me over. Thanks!

 

Where is dGPU supposed to show? or are you saying under Display Adapters? This is what mines show:
Intel(R) UHD Graphics 630
NVIDIA GeForce GTX 1050 Ti with Max-Q Design

Also, BIOS v1.20.0 is out, Restore to Factory Reset Settings is still available. I am running Windows 11 Home 22000.120.

 

Yeah, my dGPU is not showing under device manager and I've tried pretty much everything to fix it but it doesn't allow me to update the drivers because it cannot detect my dGPU. I really doubt it is a hardware issue. This is so weird.

 

Does the dGPU show up in your BIOS? I can't remember but I think the dGPU can be enabled or disabled in the bios.

 

No, I don't see the dGPU in System Information tab in the BIOS, that's why I was wondering if someone could check if it shows up for them, or does it just show the integrated graphics. Also, I don't think there is an option to enable or disable dGPU in my XPS bios, because there is no option for that. I'd appreciate if you could check that for me, even better a solution!

 

We don't see many failed GPUs on the XPS models. So hopefully just a driver issue. Is your laptop under warranty?

I would try the following:

0. Make sure you have everything critical backed up.

1. Go to windows update and try to download any updates. RESTART the computer. See if that works.

2. Else, fully uninstall the current drivers using DisplayDriverUninstaller. That might work best in safe mode. Restart per DDU instructions.

3. Go to dell.com, type in your computer serial number and download the latest nvidia driver from Dell. Install and restart per instructions.

4. Else, get the latest W10 driver directly from nvidia. Note they changed the deployment method last year.

 

Yes, you are right that the BIOS only shows the Intel GPU but not the nVidia. What you could do is use one of the Linux on a USB Flash Drive distributions and then see if it shows the nVidia at all since if it doesn't, then you know it's another issue. You may also want to look at DriverStore Explorer at https://github.com/lostindark/DriverStoreExplorer and see if it shows anything with nVidia at all in it and then if it shows nVidia, delete those items and then reboot the system and see if it autodetect the nVidia stuff or not.