Docking XPS 9570 to a USB-C dock, drops max CPU to 800Mhz

How much power does your USB-C dock supply? The XPS 15 9570 expects 130W to perform properly. Anything less and it will throttle CPU and GPU performance and/or in extreme cases continue draining the battery in order to offset the power shortage. Since the USB Power Delivery spec’s official max is 100W, the only docks I know of that will work with the 9570 properly are the Dell WD15 with the larger 180W power supply option and the Dell TB16 with the larger 240W power supply option. Both of those can deliver up to 130W over USB-C to Dell systems, so Dell is doing something proprietary there.

So you can either use one of those docks or else keep the system’s AC adapter plugged in even while you have the USB-C dock also connected.

That does not make sense though. How can a connected charger be worse than being on the battery, even if the charger is of low wattage? Since it does not lock the CPU to 800Mhz when I am on battery, why should it lock to 800Mhz when I connect a non-Dell 60W charger?

BTW, most of the times it does connect the charger correctly and not lock the CPU at 800Mhz.

---

@devsk wrote:

That does not make sense though. How can a connected charger be worse than being on the battery, even if the charger is of low wattage? Since it does not lock the CPU to 800Mhz when I am on battery, why should it lock to 800Mhz when I connect a non-Dell 60W charger?

BTW, most of the times it does connect the charger correctly and not lock the CPU at 800Mhz.

---

Because when you're on a charger, the system gives priority to avoiding battery consumption (and ideally charging the battery) rather than maintaining max performance.  That means running the CPU at a lower frequency to keep the system's power consumption within what the charger can provide.  The system only continues to drain the battery while connected to a charger as a last resort under very heavy load and/or with a severely undersized charger -- and if you think about it, this prioritization strategy makes sense.  Suppose you waited until your battery was almost dead to connect a charger, and the only charger you had was undersized.  If the system continued draining your battery to prioritize max performance, then your battery would die even though you had a charger connected, so you would be unable to continue working.  By reducing its power consumption instead, it allows you to continue working while you have a charger attached, even if it means operating at a lower speed.  For most people, that would be a preferable outcome to being completely unable to work.

Here's another situation: Suppose you had an undersized charger and wanted to charge your battery while you had a wall outlet readily available so you could work on battery later in the day.  Once again, if the system optimized for performance, you wouldn't be able to charge the battery, and in fact you might continue draining it.  By having the system reduce its power consumption instead, you'll be able to charge your battery a bit while connected to the undersized charger so that you can run on the battery later, even though you'll have reduced performance while charging.  Again, for most people, that is a preferable outcome.

As for why it doesn't always happen, I don't know the details of the firmware well enough, but I do know that batteries charge faster when they're at certain charge levels, so it could be that your CPU locks to 800 MHz when the battery allows high-speed charging and is freed up when the battery would not be charged as quickly.  But either way, the undersized charger is the source of your problem, regardless of how often it happens or whether you think this design makes sense.  The bottom line is that the system is designed to operate with an AC adapter that provides a certain wattage level, and if you choose to run a setup that does not provide that, you should not expect your system to run optimally in all situations.

---

@devsk wrote:

I don't think its by design. I think its a firmware bug in Dell BIOS or some hardware issue with their USB-C soc. 60W charger is not an undersized charger by any measure. It provides enough juice when the firmware detects it correctly and uses it correctly.

60W is nearly half of the power delivered by the barrel charger. 800Mhz is the minimum frequency in the frequency table. Give me at least half the performance....:-)

---

You can believe whatever you want in terms of whether it's intended or appropriate, but if you do some research, you will consistently find reports of this behavior on the XPS 15 9550, 9560, and 9570 when it is running from a 60W charger.  A 60W charger is indeed undersized for this system's hardware.  As you say, it's not even half of the capacity delivered by the charger that the system was designed for and ships with, and somehow you don't think that qualifies as undersized?  The H Series CPUs that come with the 9570 are designed to draw up to 45W all by themselves, and that's before considering the power requirements of the display, the dGPU when active, various controller chips on the motherboard, the SSD, WiFi adapter, and of course any power you might want to use to actually charge the battery while you have that charger connected.

As I said earlier, I've explained to you what's responsible for this behavior and provided options for alleviating it.  So if you wish to actually solve your problem, you now know how to do so. If you instead wish to continue arguing that your setup shouldn't cause this behavior, then that is of course your prerogative, although it seems wholly unproductive to me, and I don't see any value in me continuing to discuss that aspect of this.  Best of luck.

Please stop answering if you don't know. I appreciate you taking the time out but please don't give a wrong answer.