Seen a lot of people talking about thermal issues with it. I'Ve noticed that it can run quite warm under load, even with the fans running a full speed power limit, throttling and thermal throttling was happening so let's take a closer look and find out if we can improve things with some under vaulting. This laptop has Intel's new. Eighth generation coffee leak, i7, 8850, H CPU, which has six cores with 12 threads and supports overclocking by default, it's clocked at 2.6 gigahertz, but the single core turbo boost speed is 4.3 gigahertz, while the all core boost speed is foggy. Gertz out of the box. My unit was actually overclocked to 4.3 gigahertz on all six cores, so we're going to need good cooling to actually get good performance with this by under bolting. The CPU were essentially telling the CPU to use less power to perform the same tasks and less power results. In less Heat, it's, quite simple to do using intel's extreme tuning utility, also known as x2, you put simply you slowly decrease the core voltage level run a stress test to confirm stability. Then, once confirmed to be stable, you drop the voltage, some more and repeat until you've got a stable system. If you take it too far, you'll blue screen in Windows. If the CPU isn't getting enough power with my worse x5, I was able to under bolt the 88 58 by minus 0.15 volts, and this is what I used in.

My testing. Testing was done with an ambient room temperature of 18 degrees Celsius, it's approaching winter here. So expect warmer temperatures in a warmer environment at idle, both the CPU and GPU set below 40 degrees Celsius. Otherwise it was running fairly. Warm. The gaming result in green was done by playing pub G at high settings with the fans on default speeds. The yellow bar just above this was also from playing pub G, but with the CPU under vaulted and CPU overclock done all caused to 4.4 gigahertz and GPU overclocked by 120 megahertz under faulting. The CPU alone was enough to drop the temperatures a little, even while overclocked and we'll see how this affected clock speeds in the next graph. Full load was test with aida64 and the heaven benchmark running at the same time, and with the fans of both default speed or maxed out, the temperature was the same as the fans were very close to full speed already, but you'll hear this for yourself shortly. Finally, the purple bar is the same stress test, but with under vaulting and overclocking applied, and we can see that this gave us just a slightly better GPU temperature at 90 degrees. The CPU was thermal, throttling it didn't go above this and the clock speed suffer as a result. Here are the average clock speeds while testing, so you can see any throttling. While gaming, with pub G at stock settings no throttling was observed, we can see all CPU cores averaged around 4 gigahertz, which is the all core turbo speed of the ad 850 H, while overclocked all cores ran just under 4 point 3 gigahertz, despite having them set To 4.

4 gigahertz in this test – and this appeared to be due to power limit throttling – this was improved a bit with the under vaulter. In the stress tests, there was some thermal throttling taking place even with the fans on maximum. Despite this, under vaulting, the CPU did still improve things quite a bit as shown in red. The CPU cores were around 500 megahertz faster as a result of the CPU under volt, and only just barely throttling. If the GPU load was removed completely, the CPU temperature would drop quite a bit into the low 80s face of the clock speeds. I got while just running CPU only stress tests without the under fault. The CPU was power limit throttling, rather than thermal throttling thermal throttling was only happening with both a combined CPU and GPU workload under faulting. The CPU not only dropped the temperatures, as shown earlier, but also reduced the power limit throttling resulting in the higher clock speeds shown here so it's helping out in more than one way. I did attempt GPU under vaulting, but I wasn't able to see any noticeable difference. I tried, starting with small changes, then larger using MSI Afterburner, but it didn't seem to be helping. As for the external temperatures, where you'll actually be putting your hands at idle, the body of the laptop is sitting in the low 30s, so it's quite cool while gaming. This increases to about 44 degrees and is a little warm to the touch in the centre, but it's not too bad, and I saw a very similar result while running the stress tests.

However, the wrist rests and right. Side of the keyboard stayed fairly cool in comparison, and I didn't notice any differences with the undervolt overclocking applied. As for the fan noise produced by the laptop I'll, let you have a listen to some of these tests. Music at idle, the Fen was still audible, but it was fairly quiet while running the stress test and while gaming, the total system volumes don't change much and there were only just behind the full fan. Speed going to full fan. Speed, didn't really change the temperatures, as it was basically already close to maxing out anyway, and it was fairly loud great. So how does this performance boost actually translate to games and practical applications? I'Ve used Cinebench to just quickly show the differences in CPU performance again, as mentioned earlier, the undervolt removed power limit throttling the CPU wasn't. Actually thermal throttling, with the CPU. Only workload as under bolting makes the CPU use less power it removed to the power limit throttling that was taking place, thereby improving performance, i've just tested a small handful of games with and without the minus 0.15 volt cpu Wonderbolt applied with 4.3 gigahertz on all six Cores in rise of the Tomb Raider with the built in benchmark we're, seeing approximately a 4.5 improvement, while under vaulting compared against running a default, as expected, the lowest settings show a better result. Pub G shows a similar result, almost no difference at highest setting levels with a small boost at lower settings.

Dota 2 is behaving the same way. The results are almost the same here in orbit. The lowest setting levels, Ghost Recon with the built in benchmark is once again showing us the same results, except with better 1 lows, this time for a synthetic benchmark, I've used the heav'n benchmark from Unigine and again fairly similar results, as shown the improvement by under vaulting. The CPU does seem to translate to slightly better gaming performance. Nothing spectacular though it depends on the game. As expected, we tend to see a larger boost in performance at the lowest setting levels, which makes sense, as with lower graphical settings, you're more likely to be CPU limited, as this less work for the GPU to actually do. Where is at the highest setting levels, the GPU is able to stretch its legs, so the gap closes a bit more realistically with the high end hardware in the Auris, x5 you'll probably be gaming at higher setting levels, so even without under vaulting it's performing pretty well. But this may depend on the game, although we saw some clock speed differences under a full stress test. The fact is, most games don't fully max out both to see bu and GPU to cause this behavior. It does run loud and warm, but, as we've seen it still runs great in games. You could always put on some headphones if the fan noise is too much for you. These differences in performance shown aren't hard and fast rules.

Temperatures will vary between laptops depending on factors such as how well paste has been applied, and even the silicon lottery. This basically means each CPU or GPU, probably has different limits with regards to the amount of voltage required to run so just using my numbers with your own laptop, probably isn't, a good idea, the results will vary, but if you're experiencing thermal throttling that attempting under bolting Could give you a performance improvement as we've seen here? Replacing the thermal paste is another option to try and improve this issue, but unfortunately, this isn't something I can do with the review unit for a number of reasons. Generally, a company won't really want you to tamper with the laptop as it needs to work in the same way for the next person to receive it. If I go ahead and replace the stock thermal paste, I can't really put the old paste back, so the next reviewer would experience something different from what you'd actually see with the product and report incorrect results under bolting, on the other hand, isn't physically intrusive and as We'Ve seen it did improve the performance in this particular unit. Granted the temperatures didn't really change much. We were able to get some more performance out of it. Using a cooling pad could possibly help to but I'm hoping to investigate that more in a future. Video I'd be interested in hearing your results with under vaulting and how much of an improvement you've seen down in the comments and leave a like if you found the information useful thanks for watching and don't forget to subscribe for future tech.