By using this site, you agree to our Privacy Policy and our Terms of Use. Close
haxxiy said:
Chazore said:

But in this case, Linus boasted about "getting the facts straight and ppl acknowledging them", and yet GN pointed out how wrong he was with his own findings which turned out to be true, thus fact.

So really it's Linus bullshit and borked data vs GN's actual data and sources, along with a dev who got fucked over, because Linus was too lazy to shell out another 400.

Linus needs to just take the L and course correct before shit gets worse (believe me, it will with this kind of information, and a dev who's been fucked over, it will spread over time). 

Except for the cooler thing, the rest are non-points. Reviews of PC components are all over the place over the internet because measurement can be done in so many different ways with so many different conditions, some of which are beyond the reviewer's understanding or control.

Of course, if methods change, or something is measured again, they should be open about it, but I don't think GN understands enough of the null hypothesis to realize this proves nothing (well, it does, just not the damning case he thought he had).

The issue isn't whether or not Linus makes mistakes or their reviews are different than everyone else, the issue is LTTs unwillingness to correct those mistakes and learn from them but even when they correct them, it's either very late or they only correct some of them. In the 4090 video, there were several issues. First the 4090 being 3x faster than 3090 Ti in Cyberpunk and the second was a 3090 Ti being 60% faster than a 3090 in Shadow of Tomb Raider. Anyone with any technical knowledge about Ampere should have been able to catch that easily. People in the comment section pointed out both but not only did it take several days for LTT to correct the Cyberpunk issue, they never bothered correcting the Shadow of Tomb Raider issue. With Youtube video, time matters especially because the initial days gets the most amount of views. And there's plenty more examples in the video of such cases where it really should have been obvious that there was something wrong with the data.

The other issue is they keep giving bad advise regardless of the data or end up with the wrong conclusion based on bad data and they do not redo testing to correct it. For example the NH-D15. Linus contentiously says it can take on any CPU while his own data shows that a 13900K thermal throttles to 13700K speeds. Yet his conclusion doesn't change on repeated videos. If a casual user buys a NH-D15 and pairs it up with a 13900K thanks to his advise and sees that it thermal throttles, it's a bad experience. And the mouse review that was mentioned where they didn't take the tape off. The list goes on.

The key thing is that this isn't just one video, this happens continuously even to this date. Even the last video they posted yesterday had issues if you read the comment sections. Reviewers can make mistakes but they need to correct and learn from them. They shouldn't just do what LTT is doing and rush videos out with mistakes after mistakes with the unwillingness to correct them and learn from them.

JEMC said:
Chazore said:

What other complaints?, you mean the data that was off?.

Given that I haven't watched the video, I'm using Cyran and Yuri's comments as a basis:

As far as I know, LTT didn't insult neither GN nor HU. He merely said, during the 7600 or 4060Ti 16GB review (I don't know which one), that they had retested all the cards with the latest drivers, unlike the other two sites. That is factually true. Hardware Unboxed answered stating that they do a couple of retest and, if they don't see any changes, they reuse the old data. I don't know if GamersNexus said anything about that before this video, but they post the month and year of the data in their reviews, proving that they don't repeat all the tests every time.

And it's fine, you don't need to repeat all the tests all the time because, more often than not, the results will be the same, but sometimes they change, and you can check the re-review of the 7900XT that HU did a couple weeks ago. So, if some site has the resources and manpower to repeat the tests, then all the better for us.

Another one, using the success of LMG as an argument against it because it has over 100 employees and a CEO is ridiculous. So what, if someone does well and succeeds in life then he no longer can be trusted? What kind of BS is that?

The amount of videos that have to be done in a week surprised me. How can videos do they post on their other site, the plane one? Because they don't post that many videos in Youtube per week.

My last argument would be less of a nitpick and more of a warning. GamerNexus has zero evidence that Linus, LTT or the new CEO are being dishonest because of their past employers or partnerships. Expressing his "concerns" about it in a video where he criticises other LTT practices may give the wrong idea that this is a reality, not a simple possibility. It's a dangerous suggestion that not only damages LTT's credibility, but also puts in question GN's real intention behind this video.

GamersNexus isn't perfect, and they've also edited videos with some text overlay to fix mistakes, although not as many times as LTT (obviously given that they also make a lot less content), and they should be cautios when makes certain claims, or it could backfire and damage their own reputation.

I think you should really watch the video because like I said, the summary doesn't do it justice.

No one is really asking them to redo all their tests, the issue is that there really shouldn't be insane outliers like the ones they produce. In the 4090 video, there were several issues. First the 4090 being 3x faster than 3090 Ti in Cyberpunk and the second was a 3090 Ti being 60% faster than a 3090 in Shadow of Tomb Raider. Those are very obviously incorrect so why not take the time and rest just two GPUs in those two categories? They don't have to retest much, they only need to retest the 4090 in Cyberpunk and 3090 Ti in Shadow of Tomb Raider. That's it. Anyone with any technical knowledge should have seen those outliers and realized there was something 110% wrong. And the issue is that this continues to happen and their process of correcting, fixing and learning from their mistakes is clearly terrible. And that is because of the fact that a) Linus himself openly admitted to the fact that he doesn't want to retest things because it costs money and b) Their staff clearly doesn't have any time to do proper videos as admitted by their own staff. That is the type of attitude that really shouldn't be acceptable.

The issue isn't the fact that Linus has 100 employees and ceo and etc. The issue is that even though he has 100+ employees while many other review outs have less than 10, LTT continues to make bigger and more obvious mistakes than review outlets who have less than 10 employees. And most importantly, they do not learn from them and continues to make said mistakes.

Last edited by Jizz_Beard_thePirate - on 15 August 2023

                  

PC Specs: CPU: 7800X3D || GPU: Strix 4090 || RAM: 32GB DDR5 6000 || Main SSD: WD 2TB SN850