#undervolting

♻️ Can GPUs Be Eco-Friendly? Yes—Here's How. ⚡

From power-hungry mining rigs to massive AI workloads, GPUs drive the modern digital world—but at what environmental cost?

In our latest blog, we dive deep into: 🔋 Tools to measure GPU power draw ⚙️ Undervolting & underclocking for efficient mining 🗑️ The fate of old GPUs and global e-waste problems 🌱 Full lifecycle analysis—from chip fabrication to recycling

💡 Whether you're a tech enthusiast, eco-conscious gamer, or data scientist looking to reduce your carbon footprint, this 4000+ word guide has actionable insights and tools for sustainable GPU usage.

I rarely do this - this, being posting about practical and personal things-, but I have received so much grief from my recent (poor, I will add, in retrospect) choice for a new Dell laptop, namely the *XPS 9570*, that I already identify and sympathise with imaginary others that might be experiencing the same grief as me and want to reach out. I also would not post something like this for the extra reason, that I am not an expert on any of the following topics. But the fact alone that I had to go through these steps, just to make use of my equipment/investment, denotes that these might no longer be considered expert material for the present day consumer by the manufacturers. . . (although, I do feel I have lots to post about manufacturer responsibility, reliability, quality control, product life expectancy, and general priorities in life.) Finally and hopefully, it might prove helpful to get a fast, high level overview of the process, so that you might research more into each topic afterwards and follow actual expert advice.

So, imaginary twin, I hope it might benefit you, if you stumble upon this:

0) The trigger: If you are feeling like your laptop can boil your tea and burn-scar your desk along with it: be alarmed. That ain’t normal. 1) Take the test: Install a software for temperature monitoring. Look online and you’ll find plenty free options. I went for ‘Core Temp’ because it was tiny, cute and customizable. 2) Confirm your fears: If indeed you see higher than normal temperatures on idle (depends on the build, but let’s call it anywhere between 50-80°C) or temperatures frequently close to 100°C when stressed, you need to take action. I saw spikes of 97 at start-up with only a few start-up services, averages at 85 and was throttling all the time while working. Action means: 3) Gettting a cooling pad with fans: Dell should hand these out at purchase time. My laptop has the vents on the bottom and a tiny distance from the desk it is placed thanks to low ridges on the chassis. I can’t see how this air venting system was ever supposed to work without a pad. Even one with no fans would help. I saved 3-4°C’s with this. 4) Undervolting: Before getting out the screwdrivers for the hard part, let’s change a few settings to see how the situation can be improved on the soft side. Undervolting means sending less voltage to the processor and graphics chipset and is possible only on models that allow for it by the manufacturer(yeah, you can be *that* unlucky). Again, there are free to use options and I chose the Intel eXtreme Tuning Utility. You can temper the values until you blue-screen. I was more reserved and stayed happy at -0.150V for the core unit and -0.70V for the graphics chip. I saw significant improvement and much less thermal throttling, but, to my mild disappointment, overall temps still stayed pinched. *Please keep in mind that every now and then, the settings get reset and need reapplying. I have observed that the reset is related to Dell system updates, but cannot confirm. Also, I do not know if this is the case only with the Xtune software, but would expect not. In any case, when they get reset, you ll soon feel the heat, so no chances it can slip your attention. 5) Repasting: removing and reapplying the thermal conductor paste applied on your processor and GPU unit to direct heat to the heatsink. Normally, this is something one needs to do anyway after a few years of use, since the thermal conductivity of the material decreases with age (gets clunkier too). But for new laptops, you might hope that it has been sloppily applied at the factory and that reapplying will improve things. Many useful guides out there that demonstrate the details of how this can be done, links below.   6) Adding thermal pads on the MOSFETs: It proves that processor/GPU are not the main heat sources in the XPS 9570! The volt regulator module that powers them causes huge overheats. Get some low conductivity thermal pads, stack em high until they touch the chassis (which is really hard to tell, I went with 3 layers of 1.5mm) and now you can hope to use the chassis as an heatsink (boosted with extra cooling from the fan-base).

Ta-tan! I have no further experiences to share. 

If you are like me, you bought this or a similar model being a long, hard Dell fan, and basing your decision on the quality reputation that Dell enjoy(ed?) and proved on all the previous models you have had, plus the excellent customer support service. I was immensely surprised to find out that:   - it is not my instance of the XPS 9570 that was overheating/faulty, but the whole class itself! :O   - it is not just the XPS 9570 that has a faulty design, but previous models too! :O  This is the third (I believe) addition to a series of overheating laptops that Dell has released trying to pack much computational power in a very thin container.

The fact that this issue has been known, not addressed, and on the contrary, repeated -> my faith is revoked and the hard lesson, that illustrates how outdated some of my mentalities are, is: CROSS-CHECK first.  

P.S. While you are at it, please enjoy this wonderful video that is -both visually and audibly- a celebration of my troubles: link P.S.2 From my research, I found these are the most explanatory articles about the whole process: -UltraBookReview  -NoteBookCheck

P.S.3  When I started my search, I found this video useful: $12 Hack To Boost Your Laptop Performance! 

P.S.4 Don't bother with Dell forums about this. Just saving your time.

It’s broken. Just writing this quick post so that someone out there might not waste an hour looking for info like I did.

Little bit of context: “p-state overclocking” is better than regular overclocking, because it doesn’t leave your clock speed and voltage completely fixed. So your processor will still ramp down when it’s (mostly) idle. On Ryzen, this can save you dozens of watts.

But here’s one thing: on the x399 Taichi motherboard, according to the information I’ve gleaned from various sources, it is partially broken, unless you have a 1950x.

How so? The Vcore will stay fixed, no matter what the current clock speed is. This means you’ll still be using ~1.35 volts at 2.2 GHz, which is extremely suboptimal and a tremendous waste.

Several solutions were pointed out, such as having to leave the pstate VID to be the exact same and modifying voltage using the general offset instead, etc. etc. but nothing works. I’ve reset my BIOS settings and then modified pstate0 by +25 Mhz, this being literally the only change done from the default settings.

It still made Vcore fixed at ~1.22 volts, and it wouldn’t budge.

From what I gather, this behaviour is broken since about 1.92 or 2.00, but pstate tweaking should still work properly if you have a 1950x...?

Nowadays, I use an Alienware 13R3 as my laptop. Before that, however, I was on the i7 model of the Surface Pro 4. It had a funny quirk, you see: it was powerful hardware... too powerful to be contained. The i7-6660U present in the device was rated for a TDP of 15 watts, but it turned out that merely using the CPU at a decently high load could blow past through that power allowance.

Thankfully, Intel processors come with two power limits. Beyond the TDP, you also have the “boosted” limit, which, in this case, was 25 watts. It still wasn’t enough to sustain both the CPU and GPU at full load at the same time, though. And consuming 25 watts generates a lot of heat, enough that the system will be like “hmm, things are toasting up in here, let’s throttle things down back to 15W.” While you could play games on the machine, that flaw made it annoying.

There were two things that you could do:

Pointing a small USB fan at the back of the device allowed it to significantly cool down (the insides and the chassis acted as heatsinks)

Using software to reduce the amount of power that goes through the chip; less power means less heat, which means less throttling, and so on.

A lot of people in the Surface subreddit back then recommended using something called Intel XTU (which stands for Extreme Tuning Utility), and while it’s probably a great tool for people who go overclocking and fine-tuning every low-level variable of their hardware, it’s very overkill if you just want to reduce the current that goes through your processor. (It also had a few very annoying quirks that needed extensive workarounds.)

We only need to reduce the current that goes through the chip: undervolting.

Radiator for Socket 478 on K7 Orion

A note on the installation at one time loved mnogomi (including Doors4Ever :) first boxed cooler from Pentium4 for s478 slot on the AMD Athlon processor 750 having the characteristics

The frequency of 750MHz, the bus from the Alpha EV6 100MHz (200MHz effective), constructive Slot A

512K Level 2 cache running at 2/5 core frequency (300 MHz in the case of this model)

Announced at the end of 1999.…