I would suggest to take a look at the CPU/GPU thermal paste: I've checked mine and it was pretty old. Mine T14 has high temperatures now, but I'm sure that when the paste will be replaced with fresh one, then they will go down to 45/48°C @ max power.

Obviously do this if your notebook is not under warranty, or it will be voided.

Cheers

Quote from: Benjamin Herzig on July 15, 2020, 16:47:13
A word about the rating: The T490 got the exact same score. Since the T14 is extremely similar and identical in many ways, the rating is also very similar.

You can make an argument that we should punish the heat problem by taking away a few points from the overall score. However: This problem only happens during the stress-test, which is a load way beyond everyday use.

Would it be right to reduce the overall score because of something that many users will never experience? I am not so sure. What is fair I think is mentioning it in the verdict and not recommending the Intel version of this product.

Also: While I am German, Notebookcheck is an Austrian company. ;)

The T490 had lower temperature. Depending on which version much lower.

As for the argument of it being a stress test, I said this before and I'll say it again. For the gaming category, you have a Witcher test as a balance between idle and max. For the other categories, you have nothing to simulate heat during typical loads. How about doing that? Create a benchmark to use that would simulate typical load and use it as a balance between idle and max.

PS I understand the verdict aspect, but that score is still good enough to get it into the top 10 of the category... a laptop not recommended making it into the top 10 is odd.

A word about the rating: The T490 got the exact same score. Since the T14 is extremely similar and identical in many ways, the rating is also very similar.

You can make an argument that we should punish the heat problem by taking away a few points from the overall score. However: This problem only happens during the stress-test, which is a load way beyond everyday use.

Would it be right to reduce the overall score because of something that many users will never experience? I am not so sure. What is fair I think is mentioning it in the verdict and not recommending the Intel version of this product.

Also: While I am German, Notebookcheck is an Austrian company. ;)

Good job Lenovo, a Thinkpad runs much hotter than the Dell XPS 13 7390

Quote from: A on July 15, 2020, 08:31:29
Because notebookcheck has to consider other uses for the laptops, like a grandma baking cookies.

This is the flaw in notebookchecks rating system, the weights do not have tresholds. So while I love notebookchecks reviews I can't help but facepalm at the rating system.
...
They would stare at it and go... yeah, it catches on fire but our weight score says 96%. So to be reasonable we'll go with 95%.

This is typically German. For example, their car magazines can have very elaborate methodologies for measuring all sorts of things, trying very hard to quantify "goodness". And try balancing shoulder space with acceleration or quietness. Plenty of people then whine when their favourite car doesn't win on points against a car they don't like, or the gap isn't big enough. The thing people need to remember is that in these systems, points are awarded even for the trivial, basic things that we take for granted and wouldn't bother mentioning. And they work with fixed priorities which might differ from ours. The things that matter to us and are a big deal as they distinguish the product are small in the grand scheme of things. Just by doing the bare minimum, you can have a relatively high score. And you can see it in the distribution of scores. It's possible to deal with these problems. You could say, for example, that if a laptop wants more than 70 % total, it needs to perform at least this well in thermals. Even if it's strong everywhere else, you're capping it. This is what EuroNCAP does in crash testing. You're setting a minimal standard for the different levels so that nothing can be too bad. One weak point can then trash the score completely.

I'm pretty sure it would still get 96 %. And the verdict would say that both samples caught fire and they don't recommend it. Or they would refuse to give it a score based on not finishing. Unless their methodology included penalties, they wouldn't reduce the score just because they felt like it.

Personally, I don't get upset about scores because I don't care. I only register the totals because they're prominently positioned. It's always the text that matters to me and actual measurements. Not the points. For example, I always read the keyboard section. I never look at the score for input devices. What the heck does 91 % mean? I would have no clue whatsoever.

Quote from: RinzImpulse on July 15, 2020, 08:01:33
And yet, still get 88%. Still wondering how your test percentage come that high

Because notebookcheck has to consider other uses for the laptops, like a grandma baking cookies.

This is the flaw in notebookchecks rating system, the weights do not have tresholds. So while I love notebookchecks reviews I can't help but facepalm at the rating system.

I mean just you wait, one of these days they will get a laptop that gets a perfect score on all their tests but it gets so hot it catches on fire. And they would be like...hmm... rating system says 96%. Then one voice of reason would say, lets ask for another laptop to test it again, they do that and it also catches on fire.

They would stare at it and go... yeah, it catches on fire but our weight score says 96%. So to be reasonable we'll go with 95%.

When laptops get that hot, it destroys all the components, especially the battery. So the impact is more than simply based on it being hot to touch but the lifespan of your product that you paid for.

Makes no sense to buy anything intel. More expensive, more heat, and crap performance.

And yet, still get 88%. Still wondering how your test percentage come that high