Hardware Unboxed Goes on Twitter Rant in Response to LTT Labs Potshots

[LAwLz]

20 minutes ago, mariushm said:

The power supply equipment is industry standard, chroma units etc...  if seasonic told them something, it's probably what current / power rating the modules should have, how many modules - basically programmable electronic loads sold as modular units -  ( how many modules they use to test 12v in their labs, how many modules are used for 5v or 3.3v etc) and maybe guides them in how to configure the loads.

But the testing method is sort of standardized, and there's the ATX specification easily accessible and that can be used to configure the testing equipment.

 

I think the Jonnyguru guy also gave them advice, and he works now at Corsair, and unless I'm mixing things horribly, Jonnyguru guy also helped Gamersnexus sort of like a consultant, so it's not a big deal.  

 

Personally I'm not worried about the electronic loads and all that, but I'd be more worried about properly using those super expensive oscilloscopes that often have fancy expensive probes and for which you really have to know your shit to get things right... there's lots of "gotchas" with scopes and high frequency testing.

That's all good and fine.

I just think the fact that they said Seasonic specced it for them, in addition to Seasonic being a sponsor of the video sets off several red flags to me when combined.

 

I might have missed it (didn't watch the video too carefully), but I couldn't find anything in the video about them also contacting others for advice or help. Maybe that's because Seasonic sponsored the video and demanded that they withhold some information. Maybe there is some other reason for not mentioning it (if they didn't mention it). But the fact that it's a possible scenario, that information is being withheld in order to appease sponsors, is the exact kind of bullshit I do not want to even be a possibility from a "lab" that people will look to for proper information.

 

 

It kind of seems to me like Linus was also surprised that they decided that Seasonic should be a sponsor on that video.

I mean, just listen to what Linus himself said and tell me this doesn't sound like a red flag, trying to be spun into something positive:

Quote

[Seasonic] Man, those guys never cease to amaze me. If they didn't have the confidence to know this thing is gonna make them look good rather than bad, why would they have gotten involved in helping us spec it? 

 

Now, imagine if the same thing was said about Intel sponsoring software to test CPUs:

Quote

Intel, man, those guys never cease to amaze me. If they didn't have the confidence to know this software is gonna make them look good rather than bad, why would they have gotten involved in helping us buy and download it?

 

 

Even if Intel, or Seasonic, just want the best and most fair testing methodology possible, it still looks incredibly bad to have them sponsor (I assume that means paying Linus money) and decide what to use for testing competitors' products. You'd have to be incredibly naive to hear that and think "yeah, this gives me confidence because clearly Seasonic/Intel just want a fair testing methodology to help consumers".

The whole "Seasonic thinks this will make their products look good" is in and of itself a massive red flag. You don't want companies to determine how things should be tested because it "will make their products look good".

 

Don't get me wrong, I think Seasonic PSUs are usually great, and I think their PSUs would look good in pretty much any test, but this is incredibly bad optics and just causes some unnecessary conflicts of interest that may or may not undermine the legitimacy of the tests. Trying to spin Seasonic designing a test to make their products look good into something positive is just Linus trying to save face. I suspect that he also thought it was a bad idea to have them sponsor that video. Switching the sponsor of that video with the sponsor of some other video would have been much better and wouldn't have needed this weird mental gymnastics that (shouldn't) fool anyone.

[ImorallySourcedElectrons]

1 hour ago, LAwLz said:

Back when I watched them, they constantly did just that.

Like in this video, which is like 90% incorrect information and shows a complete lack of understanding of the fundamentals.

That video is 90% cringe and feels like something from an early 2000s infomercial shown at some school event, no argument there, but please do tell how that is 90% incorrect? It's indeed a bit fudgy about how they're used at a hardware level and they mix up a couple of t erms, but that's a good enough explanation when aimed at a layman. They do mix up packets and ethernet frames, but that distinction is hazy at best because of how ill-devised the entire OSI networking model is in general. And they are in fact used at a hardware level for ethernet frames, it's been a couple of years since I've had to deal with those but it's literally something along the lines of: preamble (basically for clock recovery/PLL syncing), destination MAC address, source MAC address, etc.

 

1 hour ago, LAwLz said:

Maybe they have gotten better since then, but the amount of topics I see pop up here where someone quotes some really strange things from LTT videos indicates to me that they haven't.

Please do link, I'm bored out of my mind and stuck at home with a broken bone in my foot due to going full Linus on a wood beam.

[LAwLz]

1 minute ago, ImorallySourcedElectrons said:

That video is 90% cringe and feels like something from an early 2000s infomercial shown at some school event, no argument there, but please do tell how that is 90% incorrect? It's indeed a bit fudgy about how they're used at a hardware level and they mix up a couple of t erms, but that's a good enough explanation when aimed at a layman. They do mix up packets and ethernet frames, but that distinction is hazy at best because of how ill-devised the entire OSI networking model is in general. And they are in fact used at a hardware level for ethernet frames, it's been a couple of years since I've had to deal with those but it's literally something along the lines of: preamble (basically for clock recovery/PLL syncing), destination MAC address, source MAC address, etc.

Almost the entire thing is wrong.

 

A large portion of the video is decided to talk about NAT, which is completely unrelated to MAC addresses. NAT is not done using MAC addresses, it's done using port numbers.

They never explain what a MAC address is or what it's actually used for. Saying "you need both a MAC address and an IP address" is not an explanation, and that statement is not even correct.

They say that IPSs use your MAC address to do things like block you if you don't pay your bill, which might be true in some cases but isn't true in 99,9% of cases.

 

They don't even mention "switch" a single time in the video, even though that is the primary use for MAC addresses.

 

 

If you watch that video, you will have no idea what a MAC address really is. You will at best have an incorrect idea of what it is. The problem isn't just that the video is total bullshit, another problem is that someone who doesn't know any better will find that video, don't realize that it is awful, and then after watching that video they will think they know what a MAC address is, but in reality they don't.

The entire premise of the video is built upon incorrect assumptions. If you think it is mostly good then you don't know what a MAC address is. It is not good for a layman. It's bad for a layman because a layman won't even realize that they are being fed incorrect information.

 

Also, the distinction between a packet and a frame is not "hazy". It is very clear in the OSI model. If you think it's hazy then maybe you have watched too many LTT videos and as a result, have a bad understanding of the subject.

In layman's terms, an IP packet contains IP-related information. An Ethernet frame doesn't. An IP packet can contain a full Ethernet frame, but the reverse is not true.

 

Here is a proper video about what a MAC address is. It also goes over what an IP address is, why we have subnet masks, what a router is and what the difference between an IP address and a MAC address are and why we have both. It also does so in pretty much the same time as the Teckquickie video fails to explain any of those concepts.

[Rex Hite]

Maybe the drive for more in-depth product testing and buying expensive testing equipment is; flexing; generating content; and re-investment because there's too much money sloshing about looking for a home lest the tax-man claims it.

 

Hard to imagine choosing a case fan based on whether Linus' Lab has  better anechoic chamber than Gamer's Nexus. 

 

On the business and corporate side of purchasing, the vendor warrants their  claims of product durability and suitability, not some YouTubers with more money than sense. Imagine if a YouTuber had to warrant their own claims.

 

And in the case of enthusiast PC parts, it's little more than the asinine chasing the infantile for the sake of views. We are mostly talking about toys after all, and so a bit of controversy is really part of the appeal of the whole thing.

 

 

[ImorallySourcedElectrons]

1 minute ago, LAwLz said:

Almost the entire thing is wrong.

 

A large portion of the video is decided to talk about NAT, which is completely unrelated to MAC addresses. NAT is not done using MAC addresses, it's done using port numbers.

They never explain what a MAC address is or what it's actually used for. Saying "you need both a MAC address and an IP address" is not an explanation, and that statement is not even correct.

They say that IPSs use your MAC address to do things like block you if you don't pay your bill, which might be true in some cases but isn't true in 99,9% of cases.

 

They don't even mention "switch" a single time in the video, even though that is the primary use for MAC addresses.

 

 

If you watch that video, you will have no idea what a MAC address really is. You will at best have an incorrect idea of what it is. The problem isn't just that the video is total bullshit, another problem is that someone who doesn't know any better will find that video, don't realize that it is awful, and then after watching that video they will think they know what a MAC address is, but in reality they don't.

The entire premise of the video is built upon incorrect assumptions. If you think it is mostly good then you don't know what a MAC address is. It is not good for a layman. It's bad for a layman because a layman won't even realize that they are being fed incorrect information.

 

Also, the distinction between a packet and a frame is not "hazy". It is very clear in the OSI model. If you think it's hazy then maybe you have watched too many LTT videos and as a result, have a bad understanding of the subject.

In layman's terms, an IP packet contains IP-related information. An Ethernet frame doesn't. An IP packet can contain a full Ethernet frame, but the reverse is not true.

 

Here is a proper video about what a MAC address is. It also goes over what an IP address is, why we have subnet masks, what a router is and what the difference between an IP address and a MAC address are and why we have both. It also does so in pretty much the same time as the Teckquickie video fails to explain any of those concepts.

Before we continue this discussion, can we agree you need different levels of abstraction for different publics? You very clearly get a "this is an explanation aimed at your mom" vibe from most techquickie videos, and that's also very much the intended public from all the ones I've seen so far. For that sort of public, this is a perfectly ok solution, and talking about how things like how ethernet was originally developed and why MAC addresses were absolutely necessary back then would be quite counterproductive in most cases. This is enough to explain to a relative why that Wi-Fi router uses a MAC address and not an IP address. As to NAT, don't even try to explain that to a non-technical person, so skirting around those sort of issues and accidentally explaining something as NAT while it isn't, I don't see an issue with that. The only thing I would change is maybe clarifying that MAC addresses are very much a local thing, while IP addresses are more of a global address thing. If you were explaining this to an older person you could quite easily make the analogy with old phone area codes and having a local PBX.

 

And no, I disagree, OSI is ill-defined and it'd be awesome if we could go back in time and throw some of the terminology and layers out of there. Folks can (and have) just called it ethernet packets, and no one except the neckbeard in the room actually cares about that mistake. And that's without getting into the fact that the OSI model introduces a lot of layers that are quite artificial in nature, and that almost none of the hardware and software on the market actually respects the applicable OSI model when it comes to implementing network communication. It's an abstraction layer matryoshka doll that could afford to shed a few layers.

 

As to that CBT Nuggets video, if I linked that video to my mom, her eyes would cross and she'd just be staring at the screen blankly asking me what the hell that even was. If I link the Techquickie video to her she'd probably go "a MAC address is something different than an IP address, and each computer has its own address and it never changes". A big part of explaining technology is knowing who your public is, like I'm fairly certain I'd probably lose you if I were to get into the gritty details of why you had to space 10BASE5 coax taps from a transmission line theory point of view, and you'd most likely lose me quite quickly if you were to get into the gritty details of something like SSL.

[Electricity Taster]

22 hours ago, ImorallySourcedElectrons said:

I find it hilarious that you're speaking in favour of statistical validation of measurements, and then use veritasium as source, arguably one of the worst offenders on YouTube when it comes to grossly misrepresenting things in favour of making "wholesome" clickbaity content.

 

In any case, I don't think what you propose is realistic due to other reasons: there's no actual guarantee that the quantities you're attempting to measure follow a known distribution, unless if you go to very large sample sizes, this is due to the way we tend to bin dies during the manufacturing process. So you'd need to get quite a few devices per manufacturing lot to even stand a chance of getting a statistically meaningful measurement, otherwise the odds are that what you're actually measuring is the variability of your test method.

 

Yeah, Veritasium has made mistakes in the past, but that specific video is correct and I believe it is a very good visualization/explanation. 

As I mentioned before: Central Limit Theorem. The required sample size might "only" be 30 to achieve sufficient statistical power. Either way, my point is not that this is realistic (I actually said it wasn't), my point is that whether to re-run a test or not is like arguing if you should rinse dirty dishes with water once or twice. You need dish soap! In other words, re-running a test is still anecdotal evidence. It does not reach the bar of what can be considered a scientific test because you cannot attach a confidence interval over 90% to that result. 

I think LTT Labs are still going to be very important to improve the industry, but let's not pretend it is scientifically rigorous. And, potentially, if some butthurt manufacturer took LTT Labs to court, it might backfire.

[LAwLz]

6 hours ago, ImorallySourcedElectrons said:

Before we continue this discussion, can we agree you need different levels of abstraction for different publics? You very clearly get a "this is an explanation aimed at your mom" vibe from most techquickie videos, and that's also very much the intended public from all the ones I've seen so far.

Yes, but the problem with the TechQuickie video is that it is wrong. It's not just "explaining things in simple terms", it is clearly confused and talks about unrelated things.

It's like if a video was called "what is a CPU" and it started talking about "the GB of a CPU determines how much data you can store in the motherboard modem". MAC address has literally nothing to do with NAT, even though a large chunk of the video talks about NAT for some reason. 

 

This is not a case of "they are dumbing things down so that it is easier to understand". This is a case of "they had no idea what they were talking about so they started talking about something vaguely similar because the writer was confused". It's the blind leading the blind. It's an awful video because it does not even talk about the subject it thinks it talks about.

 

 

 

6 hours ago, ImorallySourcedElectrons said:

This is enough to explain to a relative why that Wi-Fi router uses a MAC address and not an IP address.

But the video does not actually explain why "a Wi-Fi router uses a MAC address", which is also a statement that is completely wrong.

 

 

6 hours ago, ImorallySourcedElectrons said:

As to NAT, don't even try to explain that to a non-technical person, so skirting around those sort of issues and accidentally explaining something as NAT while it isn't, I don't see an issue with that.

If they didn't want to explain NAT then they shouldn't have dedicated a large portion of the video to explaining it, and then also got it completely wrong (NAT has nothing to do with MAC addresses, it does not use MAC addresses at all).

 

 

6 hours ago, ImorallySourcedElectrons said:

And no, I disagree, OSI is ill-defined and it'd be awesome if we could go back in time and throw some of the terminology and layers out of there.

OSI is not ill-defined at all. If you think it is then you just don't understand it. Just because you don't understand something does not mean it's ill-defined. It is very rigorously defined. Maybe if you watched fewer LTT videos you wouldn't think it was ill-defined, because you wouldn't have been taught a bunch of misinformation.

 

6 hours ago, ImorallySourcedElectrons said:

Folks can (and have) just called it ethernet packets, and no one except the neckbeard in the room actually cares about that mistake.

Maybe a channel pretending to be "educational" should actually educate people on proper terminology.

Maybe you think this is confusing because you and the people you listen to use terms incorrectly.

 

6 hours ago, ImorallySourcedElectrons said:

As to that CBT Nuggets video, if I linked that video to my mom, her eyes would cross and she'd just be staring at the screen blankly asking me what the hell that even was. If I link the Techquickie video to her she'd probably go "a MAC address is something different than an IP address, and each computer has its own address and it never changes".

I think my mom's eyes would cross and she would stare blankly at the screen at the TQ video as well. Which would be good, because getting taught wrong information is better than not being taught at all.

 

 

I don't think I will be able to convince you that the TQ video is bad so I will give up on that. But I want you to know that the TQ video is pretty much wrong in everything it says. That's not just me being a pedantic neckbeard either. It fundamentally shows a lack of understanding for the subject it talks about, and gets confused with other somewhat-related subjects and starts talking about those instead. It is an embarrassingly bad video. It makes no sense and does not achieve its goal at all. Again, it's like if a video about "what is a CPU" ended up talking about how many GB you can store in a motherboard modem. It makes absolutely no sense. That video I linked really is on that level of "the person hosting this has no idea what they are talking about". But since it comes across like they do have a clue, people will eat it up, and some people like you will try and rationalize it through mental gymnastics.

[ImorallySourcedElectrons]

39 minutes ago, Electricity Taster said:

Yeah, Veritasium has made mistakes in the past, but that specific video is correct and I believe it is a very good visualization/explanation. 

As I mentioned before: Central Limit Theorem. The required sample size might "only" be 30 to achieve sufficient statistical power. Either way, my point is not that this is realistic (I actually said it wasn't), my point is that whether to re-run a test or not is like arguing if you should rinse dirty dishes with water once or twice. You need dish soap! In other words, re-running a test is still anecdotal evidence. It does not reach the bar of what can be considered a scientific test because you cannot attach a confidence interval over 90% to that result. 

I think LTT Labs are still going to be very important to improve the industry, but let's not pretend it is scientifically rigorous. And, potentially, if some butthurt manufacturer took LTT Labs to court, it might backfire.

I'd say quite recently as well, I'm not ready yet to forgive him for that awful speed of electricity video. I've seen similar argumentation used to explain why sound waves can travel at the speed of light, completely disregarding the fact that energy and information transfer is what we're actually interested in.

 

Let me rephrase my point: when you apply the central limit theorem you need to be careful what you're actually measuring. You also see many research papers where they make this mistake (can't think of another word for it, sorry), they fail to realise what they're actually measuring and contribute the measured effect to the measured quantity. The simplest example of this is when you're measuring a fundamental constant or material property, you might be tempted to think that the average measured quantity is your target value, but it can also be the average measurement error. So I'm always a bit hesitant to simply work with target confidence intervals, and instead I aim to understand what the actual statistical distribution should be of the quantity I'm measuring. This is especially an issue with output from manufacturing processes with in-line quality measurement and culling of faulty products. But I fear we're going to be diving into the statistics of manufacturing and process analysis if we continue on this topic, and that's not a good subject for a Sunday evening or a web forum.

 

And I don't disagree that it's not scientifically rigorous, but that also doesn't mean the data is worthless. I think the key thing for LTT would be to pay attention to sudden shifts or odd trends in measured data, steady progress in the same direction with multiple combinations of hardware would probably indicate a trend for a specific part. But then we get into the entire causation vs. correlation debate.  

 

14 minutes ago, LAwLz said:

Yes, but the problem with the TechQuickie video is that it is wrong. It's not just "explaining things in simple terms", it is clearly confused and talks about unrelated things.

It's like if a video was called "what is a CPU" and it started talking about "the GB of a CPU determines how much data you can store in the motherboard modem". MAC address has literally nothing to do with NAT, even though a large chunk of the video talks about NAT for some reason. 

 

This is not a case of "they are dumbing things down so that it is easier to understand". This is a case of "they had no idea what they were talking about so they started talking about something vaguely similar because the writer was confused". It's the blind leading the blind. It's an awful video because it does not even talk about the subject it thinks it talks about.

 

But the video does not actually explain why "a Wi-Fi router uses a MAC address", which is also a statement that is completely wrong.

 

If they didn't want to explain NAT then they shouldn't have dedicated a large portion of the video to explaining it, and then also got it completely wrong (NAT has nothing to do with MAC addresses, it does not use MAC addresses at all).

 

OSI is not ill-defined at all. If you think it is then you just don't understand it. Just because you don't understand something does not mean it's ill-defined. It is very rigorously defined. Maybe if you watched fewer LTT videos you wouldn't think it was ill-defined, because you wouldn't have been taught a bunch of misinformation.

 

Maybe a channel pretending to be "educational" should actually educate people on proper terminology.

Maybe you think this is confusing because you and the people you listen to use terms incorrectly.

 

I think my mom's eyes would cross and she would stare blankly at the screen at the TQ video as well. Which would be good, because getting taught wrong information is better than not being taught at all.

 

I don't think I will be able to convince you that the TQ video is bad so I will give up on that. But I want you to know that the TQ video is pretty much wrong in everything it says. That's not just me being a pedantic neckbeard either. It fundamentally shows a lack of understanding for the subject it talks about, and gets confused with other somewhat-related subjects and starts talking about those instead. It is an embarrassingly bad video. It makes no sense and does not achieve its goal at all. Again, it's like if a video about "what is a CPU" ended up talking about how many GB you can store in a motherboard modem. It makes absolutely no sense. That video I linked really is on that level of "the person hosting this has no idea what they are talking about". But since it comes across like they do have a clue, people will eat it up, and some people like you will try and rationalize it through mental gymnastics.

I think you see it far too negatively. it's not the most robust of explanations, but it does the job well enough. If this was a video aimed at network engineers, I'd be worried, or if it was someone overconfidently spouting nonsense to a camera (see Veritasium's electricity video), but this was kind of ok.

 

Quote

Maybe if you watched fewer LTT videos you wouldn't think it was ill-defined, because you wouldn't have been taught a bunch of misinformation.

Thank you for the somewhat insulting statement, but I finished my engineering degree quite a while ago. But no, I think it's an ill-defined system to define network architectures and have thought so ever since I was exposed to actual networking hardware at a low level. I mostly see it as an artificial attempt to abstract highly interwoven things into separate layers, it's a prime example of design by committee. It's a bit like a design pattern but for network engineers, it sounds like a good idea until anyone actually tries to rigidly apply everything it proposes in real life. It's funny what folks try to define as presentation layer, I have yet to see a session layer implementation that actually does its job like OSI envisioned it, the practical implementation of the logic link control sub-layer is quite frankly a non-standardized mess, and don't get me started on guaranteed delivery. It's also not uncommon for hardware to casually ignore the OSI model and take steps it's not supposed to take according to the layered model, but it does so in an attempt to improve the chances of successfully delivering messages, etc. So no, it's not misinformation, I'd prefer to catalogue it as experience-induced cynicism.

[starsmine]

8 hours ago, LAwLz said:

That is true. But in order to inspire me with such confidence they would have to have dropped the sponsorship (so that Seasonic isn't giving them money or other compensation), and preferably have talked to others as well. Not just, in their own words bought something "specced by Seasonic".

 

I am not entirely convinced that isn't exactly what they are doing, and I am worried that if that happens people will still believe their numbers are accurate because "they spent so much money and it is called 'the lab', so it must be true!". That's why threads like the one I posted earlier, where the results are very strange and we have no idea how or what they actually tested (because they don't/didn't disclose that).

Maybe that won't happen again, but they have a very big uphill battle to fight if they want me to take them seriously. The reason for that large hill is their countless fuckups in the past.

Seasonic isn't giving the labs ANY sponsorship in such a way.
LTT consulted with them to build the test equipment, not get money from them. Consulting the industry you are testing is not only normal, but NECESSARY.

No one is lowering any standards here.

The Electrical Engineers at Seasonic have the potential to lose licenses if they specifically were knowingly involving themselves in a conflict of interest. Professional EE's are BOUND by https://www.ieee.org/about/corporate/governance/p7-8.html 

[leadeater]

16 minutes ago, starsmine said:

Seasonic isnt giving the labs ANY sponsership.
LTT consulted with them to build the test equipment, not get money from them. Consulting the industry you are testing is not only normal, but NECESSARY.

No one is lowering any standards here.

The presented issue is that Seasonic sponsored the video, the counter point to that is Seasonic is sponsoring the video not Labs and the testing programme. LTT has a policy of trying to make sponsors of videos relevant to the video subject/topic i.e. it wouldn't make much sense to have Dbrand as the sponsor for that video.

 

Making a video about power supply testing has a risk to it, that being low viewership so those videos tend to have video sponsors to offset that risk. Those brands may also want to be associated with the video due to high audience crossover and engagement. If you are watching a video about power supply testing then you are probably interested in Seasonic products or the brand.

 

Also I'm fairly sure Labs is a subsidiary company like Floatplane is and finances are separated, they are actually trying to run things by proper business practices for a while now. For example these subsidiaries actually charge each other for work/time carried out. So while LTT is making a video about something Labs is doing LTT itself is interested in the video production and revenue and has nothing to do with the Labs. Think of it like LTT doing a tour of a manufacturing facility or company, sure it's a little incestuous being part of the same organisation group but that is the applied concept here.

 

Seasonic sponsored the video, not Labs.

[ImorallySourcedElectrons]

16 minutes ago, starsmine said:

Seasonic isnt giving the labs ANY sponsership.
LTT consulted with them to build the test equipment, not get money from them. Consulting the industry you are testing is not only normal, but NECESSARY.

No one is lowering any standards here.

And even if they did, it's pretty normal for industry to fund the independent testing labs in their field or for the standards they use. We most certainly do that at work, and we rarely if ever have direct contact with these labs except for the test reports.

 

Just now, leadeater said:

LTT has a policy of trying to make sponsored videos relevant to the video subject/topic i.e. it wouldn't make much sense to have Dbrand as the sponsor for that video.

Knowing Dbrand, they would most likely sponsor a video about testing their products, and then ask LTT to put all the competing tested products through a wood chipper as condition for them to get paid, or something ridiculous like that. It is Dbrand after all, they'd lean into possible corruption allegations, we're talking about the same company that has a CEO with a pyramid shaped PC after all.

[LAwLz]

53 minutes ago, ImorallySourcedElectrons said:

I think you see it far too negatively. it's not the most robust of explanations, but it does the job well enough. If this was a video aimed at network engineers, I'd be worried, or if it was someone overconfidently spouting nonsense to a camera (see Veritasium's electricity video), but this was kind of ok.

The problem isn't that the video doesn't have "the most robust of explanations". The problem is that half the video is about a different subject entirely that has nothing to do with what the video tries to explain. I don't think you understand how much wrong stuff is in that video. The person who wrote it fundamentally doesn't understand what a MAC address is or why it is needed. That's why they end up bringing up things like NAT, which DO NOT USE MAC ADDRESSES AT ALL.

 

NAT uses port numbers, not MAC addresses. The video from TQ says that the router uses MAC addresses to do NAT. It doesn't. It simply does not do that. That's not how it works. It's not even remotely correct. It brings in an entirely separate concept that has nothing to do with MAC addresses and then goes "this is what MAC addresses are used for". No, they aren't. Not even close. 

 

Again, this is like making a video about CPUs and then half of the video says "CPUs are used to determine how many movies you can save on your motherboard modem, also known as a hard SSD drive". It makes absolutely no sense whatsoever. It's just a bunch of random networking terms thrown together. The video is garbage, and so are plenty of their videos. It does more harm than good existing, because everyone who watches that video will either laugh at how piss poor it is or get a completely wrong impression of how MAC addresses work or what they are used for. Someone watching that will probably think it has something to do with NAT, which it doesn't.

Stop trying to pretend like the video explained MAC addresses in a more casual way because it didn't. It explained something factually wrong that wasn't even related to MAC addresses.

 

It really is "someone overconfidently spouting nonsense to the camera". That's all that it is. The problem is that people who don't know any better will look at it and don't realize what it actually is. 

 

 

 

53 minutes ago, ImorallySourcedElectrons said:

But no, I think it's an ill-defined system to define network architectures and have thought so ever since I was exposed to actual networking hardware at a low level.

Sure...

Ask anyone working with networking what the difference between layer 2 and layer 3 is and you will get crystal clear definitions that matches both the implementations in software/hardware as well as the specifications from IEEE. It really isn't confusing at all. 

I could draw you a diagram that illustrates the differences if you want. I'd just have to draw up the various headers and it would be incredibly easy to see that once the packet is formatted to an IP packet it contains things like the IP header, while the Ethernet frame does not.

 

 

34 minutes ago, leadeater said:

Seasonic sponsored the video, not Labs.

If LMG was serious about presenting itself as unbiased and reliable then I don't think such a distinction would need to have been made to begin with. It seems like a very fuzzy and arbitrary line to draw just because it makes the situation look more favorable to LMG. 

I think it was a very bad idea to let Seasonic sponsor the video about their testing equipment, especially since they themselves admit that Seasonic specced the gear they are going to use for their tests. I mean come the fuck on... You can't seriously say that the quote I posted earlier doesn't sound like trying to twist a potential conflict of interest into something positive.

It was a pathetic attempt that I think fell flat on its face. "wow, they must be confident to sponsor our testing equipment lol". If Gamers Nexus had said that about Intel deciding what software to test AMD CPUs using then people would have been up in arms. 

[Thaldor]

@ImorallySourcedElectrons

I put more off-topic stuff into the spoilers, just about the msitakes LTT make.

Spoiler

You want LTT video with frogs:

 

Thing that is huge frog is that they at no point even mention the most important thing about 18650 batteries. That 18650 is just size standard with usual nominal voltage of 3.6V and you should ALWAYS make sure to check the amperage of the 18650 battery to know what you're actually putting into something. It's not once or twice I have seen "well educated people" going and blowing the shit out of things by buying genuine 18650 batteries (as in, not Trustfire or similar Chinese explosives graded "fake" batteries which marketing speeches are full of lies) like Panasonic NCR18650B cells and putting them into things that can draw over 20A even 30A from a cell and the NCT18650B is only rated for 10A and has a great possibility to go nuclear. Not to mention that just any spotwelder in battery work is a no no.

 

I grade this as one of the LTTs most dangerous videos ever and mostly it comes from that the only person really underlining that you shouldn't do any of it, like do not even attempt to take a battery assembly apart, just NO, is the editor with the brief moment under the intro logo "Don't try this at home (really tho)". Probably would have been better if even Linus didn't try any of it because he is so far away from his waters that he is probably trying to row a boat in a desert.

 

More recent one with more usual LTT slip:

 

The circuit they're saying that is the PSU and which they change is just backlight power controller and the real PSU in this case is external and in the cable. People were screaming this at the comments enough that LTT actually made a comment about that and pinned it to the top, yay, which is good thing because now everyone coming over month later get the actual fix for the problem from the comments.

 

But everyone has mishaps from time to time. Personally my biggest problem with LTT is that they always seem to underline how they are this and that and if you follow them a bit more, no they are not that at all. The IMO best example is when the Facebook Quest became and it came out that the device is locked to FB account, in the first WAN show after that big talking point was how shitty that thing is and how that makes the whole Quest questionable product. Next week Facebook sponsored video about Facebook Quest and that account locking is a small little sidenote and it's the best VR device for it's price and you should buy it, Linus takes flag for it and the next WAN show is about Linus screaming how the sponsors don't affect the videos and his opinions and how he is unbuyable.

The next one is the "trust me, bro" warranty... One thing to take away from Linus as the owner of LMG is that he isn't your friend, he is doing business and just like any business, he also isn't your friend. Then they make expensive screwdriver and very expensive backbag and people start to question the warranty policies of LTTstore and Linus goes full "warranties doesn't mean anything because everyone breaks them but LTTstore is different and we don't need warranty because I will make sure we always make things right". No one asked for lifetime warranty but at least the minimum warranty policies as written so people know what to expect and at least that isn't Linus running away with the cash since he isn't our friend.

 

But generally on the topic.

I believe the whole testing everything every time again is just wasting money. If there isn't a clear performance tuning done for the tested GPU's in the newest drivers, nothing performance aimed in the Windows or game patches, it's just waste of time to test the GPUs and CPUs again and again just to be different. There can be huge changes but just testing one or two older parts to see are the results still in the quite large error margin is enough. If they want to go more deeper the first thing to do would be growing the sample size, one (at the worst manufacturer cherry picked) unit is far from enough to draw any true conclusions if we go to the lengths of caring about the Windows version and game patch version regardless is there actual performance upgrades done because the silicon lottery is a bigger variable than those especially if the parts are taken to the limits.

 

From entertainment viewpoint the always testing everything is probably better idea because there won't be the same games and stuff tested. But then again that is bad for anything more general data, like publishing huge GPU or CPU database based on their testing would be waste of time because putting products from two different test groups to be compared wouldn't be the best practice, especially after the lead tester statement that the tests are always run again because everything affects the results. GN has gone in depth about having a big database of different products and why you need to standardize the testing so every product goes through the same tests. There is things like better optimization for older parts with drivers but that goes down to the methodology where the sidenote "parts tested with the first supporting drivers unless there has been a problem" is made and the playing field artificially leveled. As in, even if the RTX 2080Ti would run 20% better with the more optimized drivers that are the first ones which support RTX 3080Ti, that isn't taken into account when testing because that isn't the "as launched" performance, except if there has been huge problems with the drivers than later testing with better drives can be used, but that isn't year or couple after.

 

Then there is the pure stupidity testing. Like if Labs is going to remake every test every time, it would be complete resource waste to every time test the plethora of CPU coolers with artificial load because there isn't anything else than ambient temperature that may change even in years. If you put Noctua NH-D15 today on top of 200-600W thermal load to test it's cooling capability, it will have the same cooling capability for the same load a decade later, testing it 5-10 times per year because comparing it to X, Y and Z coolers in different projects is just wasting resources.

[dogwitch]

1 hour ago, Thaldor said:

@ImorallySourcedElectrons

I put more off-topic stuff into the spoilers, just about the msitakes LTT make.

  Reveal hidden contents

You want LTT video with frogs:

 

Thing that is huge frog is that they at no point even mention the most important thing about 18650 batteries. That 18650 is just size standard with usual nominal voltage of 3.6V and you should ALWAYS make sure to check the amperage of the 18650 battery to know what you're actually putting into something. It's not once or twice I have seen "well educated people" going and blowing the shit out of things by buying genuine 18650 batteries (as in, not Trustfire or similar Chinese explosives graded "fake" batteries which marketing speeches are full of lies) like Panasonic NCR18650B cells and putting them into things that can draw over 20A even 30A from a cell and the NCT18650B is only rated for 10A and has a great possibility to go nuclear. Not to mention that just any spotwelder in battery work is a no no.

 

I grade this as one of the LTTs most dangerous videos ever and mostly it comes from that the only person really underlining that you shouldn't do any of it, like do not even attempt to take a battery assembly apart, just NO, is the editor with the brief moment under the intro logo "Don't try this at home (really tho)". Probably would have been better if even Linus didn't try any of it because he is so far away from his waters that he is probably trying to row a boat in a desert.

 

More recent one with more usual LTT slip:

 

The circuit they're saying that is the PSU and which they change is just backlight power controller and the real PSU in this case is external and in the cable. People were screaming this at the comments enough that LTT actually made a comment about that and pinned it to the top, yay, which is good thing because now everyone coming over month later get the actual fix for the problem from the comments.

 

But everyone has mishaps from time to time. Personally my biggest problem with LTT is that they always seem to underline how they are this and that and if you follow them a bit more, no they are not that at all. The IMO best example is when the Facebook Quest became and it came out that the device is locked to FB account, in the first WAN show after that big talking point was how shitty that thing is and how that makes the whole Quest questionable product. Next week Facebook sponsored video about Facebook Quest and that account locking is a small little sidenote and it's the best VR device for it's price and you should buy it, Linus takes flag for it and the next WAN show is about Linus screaming how the sponsors don't affect the videos and his opinions and how he is unbuyable.

The next one is the "trust me, bro" warranty... One thing to take away from Linus as the owner of LMG is that he isn't your friend, he is doing business and just like any business, he also isn't your friend. Then they make expensive screwdriver and very expensive backbag and people start to question the warranty policies of LTTstore and Linus goes full "warranties doesn't mean anything because everyone breaks them but LTTstore is different and we don't need warranty because I will make sure we always make things right". No one asked for lifetime warranty but at least the minimum warranty policies as written so people know what to expect and at least that isn't Linus running away with the cash since he isn't our friend.

 

But generally on the topic.

I believe the whole testing everything every time again is just wasting money. If there isn't a clear performance tuning done for the tested GPU's in the newest drivers, nothing performance aimed in the Windows or game patches, it's just waste of time to test the GPUs and CPUs again and again just to be different. There can be huge changes but just testing one or two older parts to see are the results still in the quite large error margin is enough. If they want to go more deeper the first thing to do would be growing the sample size, one (at the worst manufacturer cherry picked) unit is far from enough to draw any true conclusions if we go to the lengths of caring about the Windows version and game patch version regardless is there actual performance upgrades done because the silicon lottery is a bigger variable than those especially if the parts are taken to the limits.

 

From entertainment viewpoint the always testing everything is probably better idea because there won't be the same games and stuff tested. But then again that is bad for anything more general data, like publishing huge GPU or CPU database based on their testing would be waste of time because putting products from two different test groups to be compared wouldn't be the best practice, especially after the lead tester statement that the tests are always run again because everything affects the results. GN has gone in depth about having a big database of different products and why you need to standardize the testing so every product goes through the same tests. There is things like better optimization for older parts with drivers but that goes down to the methodology where the sidenote "parts tested with the first supporting drivers unless there has been a problem" is made and the playing field artificially leveled. As in, even if the RTX 2080Ti would run 20% better with the more optimized drivers that are the first ones which support RTX 3080Ti, that isn't taken into account when testing because that isn't the "as launched" performance, except if there has been huge problems with the drivers than later testing with better drives can be used, but that isn't year or couple after.

 

Then there is the pure stupidity testing. Like if Labs is going to remake every test every time, it would be complete resource waste to every time test the plethora of CPU coolers with artificial load because there isn't anything else than ambient temperature that may change even in years. If you put Noctua NH-D15 today on top of 200-600W thermal load to test it's cooling capability, it will have the same cooling capability for the same load a decade later, testing it 5-10 times per year because comparing it to X, Y and Z coolers in different projects is just wasting resources.

correction . pugent system. Standard the testing .

 

[leadeater]

3 hours ago, LAwLz said:

If LMG was serious about presenting itself as unbiased and reliable then I don't think such a distinction would need to have been made to begin with. It seems like a very fuzzy and arbitrary line to draw just because it makes the situation look more favorable to LMG. 

The distinction does not need to be made that's how it actually is. That's just not how you are or want to see it but none the less the fact remains Seasonic sponsored the video and that even happens to be what is stated in the video right at the start.

 

Also the most current video was not sponsored by them and covers talking to PSU companies not Seasonic.

 

 

The issue here is you are commenting on things you simply are not informed enough about and don't seem to care enough to be informed about. Your view point is twisted. Like I said already, you're jaded and finding anything to criticize and what you think is valid to you doesn't make it valid to others. While it may appear odd that a PSU company is sponsoring a video if you want to properly raise criticisms then put in the effort to know more about the situation  

 

3 hours ago, LAwLz said:

If Gamers Nexus had said that about Intel deciding what software to test AMD CPUs using then people would have been up in arms. 

You mean like Gamers Nexus did for their own PSU testing? Yet it's only a problem for LTT? 

[ImorallySourcedElectrons]

18 hours ago, starsmine said:

Seasonic isn't giving the labs ANY sponsorship in such a way.
LTT consulted with them to build the test equipment, not get money from them. Consulting the industry you are testing is not only normal, but NECESSARY.

No one is lowering any standards here.

The Electrical Engineers at Seasonic have the potential to lose licenses if they specifically were knowingly involving themselves in a conflict of interest. Professional EE's are BOUND by https://www.ieee.org/about/corporate/governance/p7-8.html 

IEEE is about as corrupt as it gets these days... They made themselves irrelevant by being pro-corporate douchebags in recent years. No one cares about their code of conduct because the board and other higher ups themselves don't respect it either, their publishing arm has switched entirely to pay to publish crap and they've mostly switched to questionable Chinese editors that themselves publish highly questionable papers, etc.

 

17 hours ago, LAwLz said:

The problem isn't that the video doesn't have "the most robust of explanations". The problem is that half the video is about a different subject entirely that has nothing to do with what the video tries to explain. I don't think you understand how much wrong stuff is in that video. The person who wrote it fundamentally doesn't understand what a MAC address is or why it is needed. That's why they end up bringing up things like NAT, which DO NOT USE MAC ADDRESSES AT ALL.

 

NAT uses port numbers, not MAC addresses. The video from TQ says that the router uses MAC addresses to do NAT. It doesn't. It simply does not do that. That's not how it works. It's not even remotely correct. It brings in an entirely separate concept that has nothing to do with MAC addresses and then goes "this is what MAC addresses are used for". No, they aren't. Not even close. 

 

Again, this is like making a video about CPUs and then half of the video says "CPUs are used to determine how many movies you can save on your motherboard modem, also known as a hard SSD drive". It makes absolutely no sense whatsoever. It's just a bunch of random networking terms thrown together. The video is garbage, and so are plenty of their videos. It does more harm than good existing, because everyone who watches that video will either laugh at how piss poor it is or get a completely wrong impression of how MAC addresses work or what they are used for. Someone watching that will probably think it has something to do with NAT, which it doesn't.

Stop trying to pretend like the video explained MAC addresses in a more casual way because it didn't. It explained something factually wrong that wasn't even related to MAC addresses.

 

It really is "someone overconfidently spouting nonsense to the camera". That's all that it is. The problem is that people who don't know any better will look at it and don't realize what it actually is. 

 

Yes, you keep repeating all of that, but I simply disagree with your assessment. Ideal world, sure, reality, no.

 

17 hours ago, LAwLz said:

Sure...

Ask anyone working with networking what the difference between layer 2 and layer 3 is and you will get crystal clear definitions that matches both the implementations in software/hardware as well as the specifications from IEEE. It really isn't confusing at all. 

I could draw you a diagram that illustrates the differences if you want. I'd just have to draw up the various headers and it would be incredibly easy to see that once the packet is formatted to an IP packet it contains things like the IP header, while the Ethernet frame does not.

This is actually getting grossly insulting. You keep repeating "it's clear" and completely ignore any counter argument, while I literally pointed out several known issues with the OSI model. And that's without even getting into interesting side issues like the fact that communication standards based on the OSI model literally led to stupid shit like websockets their horrible attempt at quality of service.

 

17 hours ago, Thaldor said:

@ImorallySourcedElectrons

I put more off-topic stuff into the spoilers, just about the msitakes LTT make.

  Hide contents

You want LTT video with frogs:

 

Thing that is huge frog is that they at no point even mention the most important thing about 18650 batteries. That 18650 is just size standard with usual nominal voltage of 3.6V and you should ALWAYS make sure to check the amperage of the 18650 battery to know what you're actually putting into something. It's not once or twice I have seen "well educated people" going and blowing the shit out of things by buying genuine 18650 batteries (as in, not Trustfire or similar Chinese explosives graded "fake" batteries which marketing speeches are full of lies) like Panasonic NCR18650B cells and putting them into things that can draw over 20A even 30A from a cell and the NCT18650B is only rated for 10A and has a great possibility to go nuclear. Not to mention that just any spotwelder in battery work is a no no.

 

I grade this as one of the LTTs most dangerous videos ever and mostly it comes from that the only person really underlining that you shouldn't do any of it, like do not even attempt to take a battery assembly apart, just NO, is the editor with the brief moment under the intro logo "Don't try this at home (really tho)". Probably would have been better if even Linus didn't try any of it because he is so far away from his waters that he is probably trying to row a boat in a desert.

 

More recent one with more usual LTT slip:

 

The circuit they're saying that is the PSU and which they change is just backlight power controller and the real PSU in this case is external and in the cable. People were screaming this at the comments enough that LTT actually made a comment about that and pinned it to the top, yay, which is good thing because now everyone coming over month later get the actual fix for the problem from the comments.

 

But everyone has mishaps from time to time. Personally my biggest problem with LTT is that they always seem to underline how they are this and that and if you follow them a bit more, no they are not that at all. The IMO best example is when the Facebook Quest became and it came out that the device is locked to FB account, in the first WAN show after that big talking point was how shitty that thing is and how that makes the whole Quest questionable product. Next week Facebook sponsored video about Facebook Quest and that account locking is a small little sidenote and it's the best VR device for it's price and you should buy it, Linus takes flag for it and the next WAN show is about Linus screaming how the sponsors don't affect the videos and his opinions and how he is unbuyable.

The next one is the "trust me, bro" warranty... One thing to take away from Linus as the owner of LMG is that he isn't your friend, he is doing business and just like any business, he also isn't your friend. Then they make expensive screwdriver and very expensive backbag and people start to question the warranty policies of LTTstore and Linus goes full "warranties doesn't mean anything because everyone breaks them but LTTstore is different and we don't need warranty because I will make sure we always make things right". No one asked for lifetime warranty but at least the minimum warranty policies as written so people know what to expect and at least that isn't Linus running away with the cash since he isn't our friend.

 

Regarding the battery repair video, this is literally the introduction:

Sorry, but if that's not clear enough, you probably shouldn't be allowed around a soldering iron in the first place.

 

On the monitor, I presume you're talking about the bit at 6:30? Calling it a power supply when you first see that PCB  isn't a particularly egregious error, that'd also be my first assessment if someone showed me this thing:

That sort of coil with what smells like a transistor and some filtering caps next to it is quite indicative of a switch mode converter, so power supply is entirely fair. And overall, the presented strategy of just getting a supposed known-good PCB from eBay isn't a bad strategy either. I'm not saying they're perfect, but a lot of what I'm seeing here is the sort of nitpicking that strongly discourages folks from showing the things they work on themselves. Some of you guys are even more vicious than last year engineering students.

 

17 hours ago, Thaldor said:

But generally on the topic.

I believe the whole testing everything every time again is just wasting money. If there isn't a clear performance tuning done for the tested GPU's in the newest drivers, nothing performance aimed in the Windows or game patches, it's just waste of time to test the GPUs and CPUs again and again just to be different. There can be huge changes but just testing one or two older parts to see are the results still in the quite large error margin is enough. If they want to go more deeper the first thing to do would be growing the sample size, one (at the worst manufacturer cherry picked) unit is far from enough to draw any true conclusions if we go to the lengths of caring about the Windows version and game patch version regardless is there actual performance upgrades done because the silicon lottery is a bigger variable than those especially if the parts are taken to the limits.

 

From entertainment viewpoint the always testing everything is probably better idea because there won't be the same games and stuff tested. But then again that is bad for anything more general data, like publishing huge GPU or CPU database based on their testing would be waste of time because putting products from two different test groups to be compared wouldn't be the best practice, especially after the lead tester statement that the tests are always run again because everything affects the results. GN has gone in depth about having a big database of different products and why you need to standardize the testing so every product goes through the same tests. There is things like better optimization for older parts with drivers but that goes down to the methodology where the sidenote "parts tested with the first supporting drivers unless there has been a problem" is made and the playing field artificially leveled. As in, even if the RTX 2080Ti would run 20% better with the more optimized drivers that are the first ones which support RTX 3080Ti, that isn't taken into account when testing because that isn't the "as launched" performance, except if there has been huge problems with the drivers than later testing with better drives can be used, but that isn't year or couple after.

 

Then there is the pure stupidity testing. Like if Labs is going to remake every test every time, it would be complete resource waste to every time test the plethora of CPU coolers with artificial load because there isn't anything else than ambient temperature that may change even in years. If you put Noctua NH-D15 today on top of 200-600W thermal load to test it's cooling capability, it will have the same cooling capability for the same load a decade later, testing it 5-10 times per year because comparing it to X, Y and Z coolers in different projects is just wasting resources.

GPU and CPU drivers do have a major impact on game performance and both AMD and NVidia frequently release drivers with updates to make specific games work better, even for some of the older games, so if they're planning to run automated tests with games, rerunning them is a pretty valid strategy. Just check the changelog for the latest NVidia driver: https://us.download.nvidia.com/Windows/531.68/531.68-win11-win10-release-notes.pdf

That first game is literally from 2020, and if you continue you even find a list of open issues that refer to both Halo Wars 2 and Watch Dogs 2:

(For reference: Halo Wars 2: 2017, Watch Dogs 2: 2016)

 

You're basically blaming them for trying to cover all their bases and calling it a waste of money, while significant performance improvements can be gained with some specific driver versions, and sometimes specific manufacturers do make changes which can impede particular models of a GPU from taking well to a particular update. It's been less common in recent years with NVidia taking more direct control of driver releases and hardware designs, but it still happens sometimes. So testing multiple GPUs is in fact the responsible thing to do.

[LAwLz]

I give up.

You can't wake someone who is pretending to be asleep. Any and all errors or issues with LMG can and will be brushed under the rug by mega-fans, or through Olympic-tier mental gymnastics be spun into something positive.

 

Getting ports mixed up with MAC addresses? Nahh, totally fine because... ISO layers are confusing, even though it is perfectly clear and well-defined what the difference between an Ethernet frame is and an IP packet is (hint, one has IP information in it, the other only has MAC address information in it) and because the IEEE is corrupt... somehow... and layer 7 something something...

 

Isn't it weird that even though the layers are apparently so "fuzzy", I will be able to exactly pinpoint which layers it has been processed through if shown a diagram of a data unit? I don't find it hard at all to differenate between an Ethernet frame and an IP packet. I find it kind of weird that we are even having this conversation since it was not one of the many issues I had with the video, yet somehow that is all that this conversation seems to boil down to now. Could it be because all the other criticisms of the video were valid so the conversation was forced into a red herring argument? Nahh, that can't be...

 

 

It's totally fine to have a video about MAC-addresses not actually talk about what MAC-addresses are used for, and instead talk about something that uses port numbers, because the person writing/hosting the video has no fucking clue about basic networking, yet is assigned the job of teaching others about it. Not an issue at all...

 

 

18 hours ago, leadeater said:

You mean like Gamers Nexus did for their own PSU testing? Yet it's only a problem for LTT? 

Stop straw-manning.

I never commented on GN's PSU testing and I don't watch their videos so I don't know what potential conflict of interests they have. My analogy was about AMD and Intel CPU testing.

[dizmo]

On 8/4/2023 at 11:29 PM, Vilacom said:

Found the not lawyer! Knew you’d show up somewhere lol, yeah they’re gonna throw the guy in jail, perp walk tomorrow. LOL holy shit thats amazing. 

Found the guy that doesn't fully grasp the English language!

I said can, not that it will.

[Vilacom]

1 hour ago, dizmo said:

Found the guy that doesn't fully grasp the English language!

I said can, not that it will.

Oh that’s my bad man!  So you are a lawyer, just not one anyone should ever hire.

Or actually, lets be honest, you’re not a lawyer at all, because no actual lawyer would ever even bring up the slightest chance of an actual criminal defamation charge because it would be similar to a meteorologist telling planes they shouldn’t fly because there is a chance they get hit by a meteor, which is still more likely than LMG getting charged with criminal defamation. 

Even bringing it up makes you sounds foolish to anyone who actually knows what they’re talking about, and anyone who believes you is gonna sound like an idiot when they say something similar around someone else from the first group.  



 

[dizmo]

4 hours ago, Vilacom said:

Oh that’s my bad man!  So you are a lawyer, just not one anyone should ever hire.

Or actually, lets be honest, you’re not a lawyer at all, because no actual lawyer would ever even bring up the slightest chance of an actual criminal defamation charge because it would be similar to a meteorologist telling planes they shouldn’t fly because there is a chance they get hit by a meteor, which is still more likely than LMG getting charged with criminal defamation. 

Even bringing it up makes you sounds foolish to anyone who actually knows what they’re talking about, and anyone who believes you is gonna sound like an idiot when they say something similar around someone else from the first group.  



 

Ha. Ha. Ha. You're so funny.

[AlwaysFSX]

I don't understand how so many people choose the wrong hill to defend vehemently. HUB is right about their data handling and testing methodology.

[IkeaGnome]

17 hours ago, ImorallySourcedElectrons said:

This entire thing should have been an email from someone at hardware unboxed to Tim saying "hey, I saw what you said there, I kind of disagree, this is why: ..."

Private response to a public comment? 

[05032-Mendicant-Bias]

I think this has been overblown.

Linus adressed the issue fairly. Tim from labs is not a PR guy and he could have been worded it better, the statement itself is needlessly comparative, but strictly speaking valid. Linus saying that competition is not only for manufacturer but also game reviewer is again valid, but I think it could have been avoided. I agree more with the statement "It's not reviewer against reviewer. It's reviewers making manufacturers accountable and protecting consumers."

 

I personally trust Harbor on Box and Gamer Nexus to get more accurate review. They just have more experience and better history in that regard. I especially love HU pricing trend videos, and price to performance charts. From GN I really enjoy the case and thermal reviews.

LTT I see it as less accurate and more entertaining, that just covers more stuff. Maybe one day LTT will get over that gap with sheer resources and labour, who knows. But it'll take a consistent record of that. Trust takes time to earn, and GN and HU to me have a proven history of being very accurate.

[leadeater]

On 8/8/2023 at 9:08 AM, LAwLz said:

Stop straw-manning.

I never commented on GN's PSU testing and I don't watch their videos so I don't know what potential conflict of interests they have. My analogy was about AMD and Intel CPU testing.

Absolutely everything I say has a real reason and point to be said.

 

So time and time again when we get here I have to teach you the same lesson, if it's said it ain't straw manning and you 100% absolutely were in involved in whatever I said was talking about just like now. Don't like being called out for your ridiculous comments, silly examples, and completely fair discussion reminder that you probably aren't being objective and unbiased then be more careful with what you say. Your comment here about AMD and Intel is one of those, it's so far removed from what is happening and what is being talked about I do not view it as being relevant nor valid thing to say. But hey just like you think your criticisms were valid you might think differently, you said it so you must have thought that.

 

"I don't watch Gamers Nexus videos so I don't know any potential conflict of interest"

"I don't watch LTT videos" *but I'll happily run my mouth and shoot from the hip with little to no information*

 

Seems like a fair assessment of the situation?

 

 

On 8/7/2023 at 11:53 AM, LAwLz said:

If Gamers Nexus had said that about Intel deciding what software to test AMD CPUs using then people would have been up in arms. 

 

 

Gamers Nexus literally consulted with Corsair and JonnyGuru which included testing equipment and testing methodology as well as posting videos about it but this is not a problem however it is when LTT does it? The one and only difference is one had to fly multiple hours and paid for it themselves to make the video yet still received direct industry advice (also outside of the video via email etc too) from a brand they will be testing and the other took a walk across the road to a different building and had the video itself sponsored by a brand that will be getting tested but no money went toward the testing/Labs aspect.

 

Both did and are taking advice from industry brands that are going to be tested and reviewed, only 1 has a conflict of interest? 

 

Whatever your opinions are be honest with yourself first then with everyone else, you are not bias free, you are not treating everything the same, that is impossible and unfair to expect from anyone but don't ignore that you are subject to that like everyone else.

 

We might finally start getting more data driven reviews of power supply testing with more detailed aspects and you want to actively discredit it before even seeing it. Sure LTT has lots of problems with data presentation and catching data errors/outliers but that doesn't mean that the actual work going on in the background is utterly flawed. Most of the common errors that make it through to videos isn't errors in data either it's errors in graphs and script and B roll etc, all very important things to the video and information trying to be presented but it doesn't undermine the actual data that was gathered during the testing. Personally I think a lot of the problems come from too many cooks in the kitchen, too many people involved with each video and not good enough oversight of the whole process from a single person intimate with every phase who is more likely to actually catch basic errors than someone with basically no involvement with the majority of the video subject and testing. This is supposed to be the "writer" but I'm personally not sure of the exact involvement all the way through.

 

Every video with an onscreen correction at the bottom left, or a substantive correction posted in the comments should be seen video production wise as a release failure and review conducted as to why and how it happened and aim for zero. This should also be addressed before trying to add in more detailed data in to videos.

[ImorallySourcedElectrons]

4 hours ago, IkeaGnome said:

Private response to a public comment? 

Which was never meant to become a public comment... And if they hadn't responded like this, odds are almost no one would have noticed.