It's really worth the watch. The main thing everyone is upset at is LTTs treatment of a small two man startup Billet Labs that sent LTT their one of a kind prototype watercooler. It would cool a cpu and gpu at the same time. LTT used a 4090, which it was not designed for, and they shit all over it.

Billet Labs sent them a 3090 ti, build guide, the cooler, and talked to an employee there about design choices that was not communicated to Linus as Linus criticized a lot that should have been covered. If you watch the LTT review video of the cooler under the context of this being a two man group startup showing off their prototype, the video comes off as very flippant and unprofessional. It was a LTT shit show video where the entertainment came from the use of metaphorical duct tape and bailing wire to make something work with no regard to how their opinions could break or make Billet Labs.

On the WAN show Linus said he didn't want to spend the $500 bucks in man hours to do the tests correctly and the results, even if significantly good, wouldn't change his opinion that "no one should ever buy this."

Billet Labs needed the prototype and video card back and was told twice they'd get it back. LTT sold the prototype at an auction for charity.

Linus put out a response, after Steve's video, making it sound like they were already in talks to pay Billet Labs back for the cost of the prototype. In a follow-up video Steve reached out to Billet, Billet says that they had not heard anything else from LTT until Steve's video.

That's just the core issue that everyone is mad about, there is a TON of stuff covered in the video and its worth watching. It's like a respectful professional nerd "Content Cop" video with the amount of depth he goes into everything.

There are basically 3 major points:

-LMG is pushing videos out too fast. Not enough editorial review to catch glaring errors and can be misleading to consumers.

-LMG was lent a custom prototype of a GPU waterblock by Billet Labs for testing and review. LMG was told it was for a 3090 specifically but decided to test it on a 4090 instead. It ended up performing terribly and was reviewed as essentially a bad product that "no one" should buy. He doubled down on this even after people pointed out he used the wrong GPU. Then to add insult to injury, he auctioned off the prototype where Billet's competitors were in attendance (not sure if they got a hold of it, but the issue still remains) instead of returning it as promised. So now Billet is not only sitting on top of a very negative review, they are missing essentially their best prototype.

-Lastly, some ethics concerns around their review process such as offering favorable reviews to certain vendors like Framework (which Linus had invested in), Asus, and a few others. As well as how they treat their employees (grind)

It's actually entertaining, but as a (rather long) summary (mind you that I'm doing this from memory)

• Clip of an LTT employee taking a jab at GN and HUB while presenting their labs by saying something along the lines of "our reviews are more precise because we retest everything everytime", which is a lie.
• Lots of stupid and big mistakes that should have been noticed by anyone looking at the data, like the 4090 being 300% faster than the 3090Ti just in one game/test, when it should not.
• Many of those mistakes not being corrected at all, and if they are, they are done in the laziest of ways and as barebones as possible. For example in the previously mentioned 4090 review, they only "corrected" the aforementioned problem in a comment (which was not even pinned until the community complained), but ignored other problems that were present in the video.
• Many other vids are so lazily done that even if they know they have made a mistake, rather than reshooting it (the whole vid or just that fragment), they will just include a callout in the vid. Not a problem if this were in 1 or 2 videos from time to time, but as of late, it is a frequent problem.
• Conflict of interests/Biases. In a video GN put out as an example, LTT implied that all Asus products are good. A more egregious example is that they like Noctua products so much, that even when their own testing shows that the CPU is thermal throttling with Noctua products, they will still recommend it over anything else.
• Delaying a review of an AMD cpu because his results weren't similar to the rest of reviewers nor AMD's numbers. He didn't want to be "the black sheep".

And now the 2 worst things:

First, they tell you about a mouse LTT reviewed in ShortCircuit, which they essentially bashed because it had very poor gliding (funny because they then chose that mouse anyway for some kind of event)... just to discover that they hadn't taken the plastic sheets off the pads. LTT, of course, rather than amending the error, decided to blame the company for making them "hard to see" (like any other mouse company does, duh).

And the worst of all. They did a review on a prototype of a waterblock done by an indie company/startup called BilletLabs. It was their best prototype so far and the only unit they had, which was going to be used to give it to other reviewers and then continue working with it themselves (this is important). This waterblock was only tested to work with the 3090Ti, but LTT, lazy as they are, decided to do it on a 4090 anyway. Surprise surprise, it didn't fit well at all, so their video was, essentially, pure bashing of the product because, as it turns out, it is hard to fit a waterblock on a card it wasn't designed to work with. And of course, it had poor performance as a result anyway. After community backlash, however, Linus decided to double down, saying that the product was, essentially, garbage anyway (not with those words, but it's easy to read that when he says that he would not recommend it to ABSOLUTELY nobody, no matter how well it would perform), Luke seemed to be... unwell during Linus' rant, btw. He also said that he would not be wasting money on having someone shoot a proper review of a review HE HIMSELF BOTCHED. So basically, he bashes a small company, all while doing a shitty as review even my grandma would do better and then he is out of fucks to give. Not content with that, BilletLabs asked a couple of times for the waterblock, so they could, as said, give it to other reviewers and then keep on working with it. LTT agrees to give it back. What does LTT do, however? They fucking sold it at an auction during the LTX. THEY FUCKING SOLD IT WHEN THEY WERE SUPPOSED TO GIVE IT BACK. The best (and only) prototype of a startup made by 2 dudes and, not content with bashing them, they fucking sell it at an event to which competitors were in attendance. The fucking audacity.

Same. I can’t actually watch a full GN video. Though I definitely skim through their reviews to get the data I need because they are good at testing.