WHY is Everyone Buying This Power Supply??

[LeapFrogMasterRace]

The hell are u asking me for

[FlyingPotato_is_taken]

@ImorallySourcedElectrons Do you use Rigol? They have downsides, especilly in the UX [and bugs], but they are dirt cheap for what they offer.

 

For tek the MSO5 is fine. What I don't get is the new MSO-2. Either they sell at extreme discounts or they are just not worth it at MSRP.

Keysight?  They are in a wired place for the general purpose scopes. There UI is fine but just fine. At the asking price you might as well just buy a scope with nicer UX. 

[raesorbleid]

Made an account specifically to suggest something that I don't think was addressed in the video, since Linus specifically asked for feedback from the community (3:35 in the video).

I think that, for the test results to be useful, they should not only fill in the holes in the 80 Plus certifications as mentioned in this Tech Quickie video, but make an effort to avoid the whole "manufacturers send cherry picked units" and "manufacturers swap out components for cheaper ones after getting tested," problems also addressed in that Tech Quickie video.

I'd also really like to see the results of the FSP Hydro Ti PRO 1000W PSU, as it appears to be one of the best PSUs in the entire Cybenetics database atm... I'd like to see if it's really worth the price (assuming they ever come back in stock  )

[Spotty]

Another thing you could add to the test suite is operating temperature testing with the environmental chamber. The EVGA N1 you tested is rated for only 25°C operating temperature (compared to 40°C for the seasonic TX-650). Some PSUs might pass the test suite at room temp (around 20°C) but fail in higher temperatures. You could test those manufacturers claims and it would be another point of data to sort the weak from the strong. 

 

Also testing at both 120V and 240V. Some PSUs are 220-240V input only so you'd need to include it in your test anyway if you wanted to test those power supplies. I imagine that is already planned to be included in the final test suite for the labs site but maybe just wasn't overly relevant to this particular video (or the EVGA N1 and thermaltake Smart died before you could test at 240V lol). 

[Ubersonic]

Probably popular because, it's a ThermalFake so people probably assume it's a cheaply engineered clone of a Seasonic PSU or something, so should be good for the money lol.

[chunkychalice]

Electrical engineer here!

 

I'm actually quite shocked that the power factor of most units is so close to 1... I would never have guessed that they cared that much. But I'm also quite shocked that the power factor on the EVGA 400 N1 is *so brutally bad*, especially when all the others are so close to unity. I really have to think that that unit must be defective and wonder if the LTT team tried a second one. Even with no power factor correction, I'd still expect a switched-mode power supply to have a pf better than 0.7 or so. Less than 0.5 at 50% load is w i l d.

[Spotty]

3 minutes ago, chunkychalice said:

Electrical engineer here!

 

I'm actually quite shocked that the power factor of most units is so close to 1... I would never have guessed that they cared that much. But I'm also quite shocked that the power factor on the EVGA 400 N1 is *so brutally bad*, especially when all the others are so close to unity. I really have to think that that unit must be defective and wonder if the LTT team tried a second one. Even with no power factor correction, I'd still expect a switched-mode power supply to have a pf better than 0.7 or so. Less than 0.5 at 50% load is w i l d.

The EVGA N1 lacks power factor correction. The power factor numbers they got are likely correct. It's not a defective unit, they're just atrociously bad.

[XNOR]

2 hours ago, FlyingPotato_is_taken said:

Nice

 

 

Where can I apply to work for sugary daddy Linus?

 

For those who don't know:

The R&S is one of the hottest scopes on the market.

The R&S one the left was the hottest thing for a while and still an absolute dream to use.

That thing on the right is a Lecroy with 12 bits@10GS/s (1 GHz bandwidth)!

If there is one scope missing is the Tek MSO-5.

Aren't Lecroys rebranded Siglents nowadays? They're pretty decent for oriental oscilloscopes, but they're not quite on par with R&S and Keysight. Tektronix used to belong to that list, but they've been on a downslide.

[Spotty]

@AlexTheGreatish In the video you guys say that the Thermaltake Smart is in Tier C on the tier list, however the model in tier C and that is highlighted in the video shows the "Thermaltake Smart BM2" which is a completely different power supply than the one you tested. The Thermaltake Smart 80+ unit which you tested is actually located in Tier D on the tier list on the forum which hasn't been updated in 2 years and Tier F on their most recently updated list on their own website. You made the point that they were wrong about ranking the EVGA BR Tier C but you compared it to the Thermaltake Smart which you mistakenly identified as also being ranked Tier C not Tier F. Considering the results in your video I would say that the rankings of the PSU tier list seem reasonable.

 

This is the Thermaltake Smart BM2 (Tier C according to the tier list). https://www.thermaltake.com.au/smart-bm2-750w-tt-premium-edition.html

Which is definitely not the same as the Thermaltake Smart 80+ which you tested https://www.thermaltake.com/smart-600w.html

 

The people behind the PSU tier list have been trying to establish themselves as their own independent media outlet publishing news and reviews. Please consider the impact your video might have in discrediting them entirely unfairly because of your mistake. I hope you issue a correction and consider editing that portion out of the video.

[XNOR]

6 minutes ago, Spotty said:

@AlexTheGreatish In the video you guys say that the Thermaltake Smart is in Tier C on the tier list, however the model in tier C and that is highlighted in the video shows the "Thermaltake Smart BM2" which is a completely different power supply than the one you tested. The Thermaltake Smart 80+ unit which you tested is actually located in Tier D on the tier list on the forum which hasn't been updated in 2 years and Tier F on their most recently updated list on their own website. You made the point that they were wrong about ranking the EVGA BR Tier C but you compared it to the Thermaltake Smart which you mistakenly identified as also being ranked Tier C not Tier F. Considering the results in your video I would say that the rankings of the PSU tier list seem reasonable.

 

This is the Thermaltake Smart BM2 (Tier C according to the tier list). https://www.thermaltake.com.au/smart-bm2-750w-tt-premium-edition.html

Which is definitely not the same as the Thermaltake Smart 80+ which you tested https://www.thermaltake.com/smart-600w.html

 

The people behind the PSU tier list have been trying to establish themselves as their own independent media outlet publishing news and reviews. Please consider the impact your video might have in discrediting them entirely unfairly because of your mistake. I hope you issue a correction and consider editing that portion out of the video.

LTT pinned someone else's comment saying exactly this below the video.

 

Quote

Worth noting that the regular Thermaltake Smart you tested is in the F-tier of the PSU tier list, not in the C-tier like the Smart BM2 or 450BR, so it is at least rated properly according to its performance on that tier list.

 

[Spotty]

3 minutes ago, XNOR said:

LTT pinned someone else's comment saying exactly this below the video.

That comment isn't pinned (it's just high up on the list of comments due to likes) and liking the comment of somebody who pointed out their mistake is not the same as issuing a correction. A correction should be done in the same medium that the mistake was made and actually make effort to inform the audience of the mistake - not buried in a way that it is difficult to find.

 

LTT does have a pinned comment on the video issuing a correction, but it's not related to the mistake they made about the tier list.

Quote

Correction: 5:37 - The location of the 80 Plus Certification Levels that are shown on the Efficiency Sweep 115V graph actually correspond to the 230V 80 Plus Certification numbers. Additionally, since our setup isn’t finalized, the numbers in this video should be considered in a bubble and shouldn’t be interpreted to confirm or refute 80 Plus or any other specification (Labs will get there Soon).

[XNOR]

11 minutes ago, Spotty said:

That comment isn't pinned (it's just high up on the list of comments due to likes) and liking the comment of somebody who pointed out their mistake is not the same as issuing a correction. A correction should be done in the same medium that the mistake was made and actually make effort to inform the audience of the mistake - not buried in a way that it is difficult to find.

 

LTT does have a pinned comment on the video issuing a correction, but it's not related to the mistake they made about the tier list.

A comment liked by the video creator usually brings it to the top, or very near it, even when it's much less popular than other comments. It is there by virtue of the like/pin/whatever you want to call it. It's a decent way of improving visibility. Note that editing a message removes the like/pin to prevent abuse, but it seems @pierregonzalez580 edited their message before the like.

 

The correction could be added to LTT's own pinned message for better visibility and for the sake of being as clear as possible? Sure. Calling the second top level comment hard to find seems a stretch, though.

 

[LMGcommunity]

48 minutes ago, Spotty said:

@AlexTheGreatish In the video you guys say that the Thermaltake Smart is in Tier C on the tier list, however the model in tier C and that is highlighted in the video shows the "Thermaltake Smart BM2" which is a completely different power supply than the one you tested. The Thermaltake Smart 80+ unit which you tested is actually located in Tier D on the tier list on the forum which hasn't been updated in 2 years and Tier F on their most recently updated list on their own website. You made the point that they were wrong about ranking the EVGA BR Tier C but you compared it to the Thermaltake Smart which you mistakenly identified as also being ranked Tier C not Tier F. Considering the results in your video I would say that the rankings of the PSU tier list seem reasonable.

added it into the pinned comment

[socrguy143]

Just hopping in here hoping somebody will see and take to heart. I generally like the video overall as it highlights the importance of the tier lists and actually shows they were correct. This power supply is garbage and should not be put in any computer. However, if you're going to use your massive platform to subtly (not really) attack a community that has dedicated a lot of time doing what you're just now starting to do, you better be right. In the first minute of the video, which is for sure the most watched portion of the video, you made a massive mistake. YOU, not the community, put this garbage power supply in a higher tier than it deserved, and the entire video is tainted by that. People who are less informed will watch this video, and disregard the tier list in the future because "Linus said". Given how big your reach is, you have made it more difficult in the future for flippers, and possibly even full size companies, to justify to buyers why they need to pay more for a higher quality power supply in their rigs.

 

As for the more informed audience who does understand the differences in the power supplies, you make fools of yourselves when you demonstrate an inability to read a tier list. If you want to be taken seriously by the whole community and not only cater to the uninformed, a bit more attention to detail would be advised, especially when putting a community on blast. Otherwise, you'll just turn yourselves into the clowns among the pc world. The last thing I'd assume you want is for one of your videos to be released and the pc community comes together to say "oh boy, another LTT lab video. Wonder how many corrections will be issued this time?"

[Spotty]

21 minutes ago, XNOR said:

A comment liked by the video creator usually brings it to the top, or very near it, even when it's much less popular than other comments. It is there by virtue of the like/pin/whatever you want to call it. It's a decent way of improving visibility. Note that editing a message removes the like/pin to prevent abuse, but it seems @pierregonzalez580 edited their message before the like.

 

The correction could be added to LTT's own pinned message for better visibility and for the sake of being as clear as possible? Sure. Calling the second top level comment hard to find seems a stretch, though.

A users comment pointing out a mistake is not a correction, no matter how easy it is to find. If your local news station ran a story on the 6 o'clock news saying you were a murderer and then the news station liked a tweet saying you weren't a murderer that would not be the same as the news station issuing a correction, even if the tweet went viral.

 

There's actual journalistic standards and ethics to follow when issuing corrections. Generally when it comes to broadcast media it's accepted that corrections are issued in the same form that the error was originally made. For example if the mistake was on the 6 o'clock news the correction should also be made on the 6 o'clock news; not as a Facebook post where it is less visible. It's basic ethical practice so that media outlets don't make false or erroneous statements in broad reaching media forms and then silently issue corrections where most people won't see it. If LTT makes a false claim during a video that gets millions of views it's not good enough to like some random person's comment in the comment section. In my personal opinion pinning a comment to the video isn't good enough either (especially in cases where the error may discredit, defame, or damage somebody else) as most people who watch the video aren't going to read the comment section, though unfortunately it seems that is the norm for issuing corrections on Youtube videos.

I would like to see LTT start issuing corrections during WAN show. Even if pinning a comment is treated as the norm on Youtube I would like to see LMG go above and beyond. People have complained about misinformation or mistakes in LMG videos from time to time. I understand that mistakes are going to happen but even if those are entirely innocent mistakes LMG should still make a fair effort to correct them. It should only take a few seconds to issue the corrections on WAN show and the writers can just add them in to the WAN show doc throughout the week (in combination with issuing a correction by pinned comment on the original video). All you would need to say is "I'd like to issue a correction for <video> earlier in the week. In that video we incorrectly stated <this>, however the correct thing is <this>. We apologise for the error".

 

 

I do have some concerns about PSU testing from the labs in future if they weren't able to correctly identify the model of power supply they were testing. It's common for PSU brands to use similar names or even reuse the same names for different models. For example the CP-9020056 and CP-9020235 are different power supplies despite both being called "Corsair RM850". If you test them and the CP-9020056 fails testing but the CP-9020235 passes testing then it's important that you correctly identify which is which and properly inform your audience. I'm assuming it will be a bit easier to convey this information on a written labs page but please keep this in mind when doing labs testing and publishing your reports.

 

This might also be a good idea for something for the labs team to look in to and directly compare a bunch of different PSUs that have the same/similar name. "Can you trust that PSU recommendation?" "You've done your research and you've read the reviews... but wait! Are you actually getting the power supply you think you are or has the manufacturer pulled a fast one on you?... It's not uncommon for power supply brands to use the same name for different power supplies. Does that old review that was published a few years ago praising this power supply for its great performance still hold true on the new model with the same name? Or is the new model a piece of junk that is riding the wave of good reviews it didn't earn? We put 10 models with deceptively similar model names to the test head to head to find out. <sponsor> Despite having the same model name these are different power supplies and the reviews for the old units cannot be applied to the new units or vice versa. This can make shopping for a power supply very confusing if you don't know what you're looking at and those recommendations you see for <example> might not be true for <example>...".

Would also be good to teardown the power supplies to show the differences between the models. Some might be the same but have a minor revision change such as a different model fan but others might have completely different designs.

[XNOR]

30 minutes ago, Spotty said:

A users comment pointing out a mistake is not a correction, no matter how easy it is to find. If your local news station ran a story on the 6 o'clock news saying you were a murderer and then the news station liked a tweet saying you weren't a murderer that would not be the same as the news station issuing a correction, even if the tweet went viral.

 

There's actual journalistic standards and ethics to follow when issuing corrections. Generally when it comes to broadcast media it's accepted that corrections are issued in the same form that the error was originally made. For example if the mistake was on the 6 o'clock news the correction should also be made on the 6 o'clock news; not as a Facebook post where it is less visible. It's basic ethical practice so that media outlets don't make false or erroneous statements in broad reaching media forms and then silently issue corrections where most people won't see it. If LTT makes a false claim during a video that gets millions of views it's not good enough to like some random person's comment in the comment section. In my personal opinion pinning a comment to the video isn't good enough either (especially in cases where the error may discredit, defame, or damage somebody else) as most people who watch the video aren't going to read the comment section, though unfortunately it seems that is the norm for issuing corrections on Youtube videos.

I would like to see LTT start issuing corrections during WAN show. Even if pinning a comment is treated as the norm on Youtube I would like to see LMG go above and beyond. People have complained about misinformation or mistakes in LMG videos from time to time. I understand that mistakes are going to happen but even if those are entirely innocent mistakes LMG should still make a fair effort to correct them. It should only take a few seconds to issue the corrections on WAN show and the writers can just add them in to the WAN show doc throughout the week (in combination with issuing a correction by pinned comment on the original video). All you would need to say is "I'd like to issue a correction for <video> earlier in the week. In that video we incorrectly stated <this>, however the correct thing is <this>. We apologise for the error".

 

 

I do have some concerns about PSU testing from the labs in future if they weren't able to correctly identify the model of power supply they were testing. It's common for PSU brands to use similar names or even reuse the same names for different models. For example the CP-9020056 and CP-9020235 are different power supplies despite both being called "Corsair RM850". If you test them and the CP-9020056 fails testing but the CP-9020235 passes testing then it's important that you correctly identify which is which and properly inform your audience. I'm assuming it will be a bit easier to convey this information on a written labs page but please keep this in mind when doing labs testing and publishing your reports.

 

This might also be a good idea for something for the labs team to look in to and directly compare a bunch of different PSUs that have the same/similar name. "Can you trust that PSU recommendation?" "You've done your research and you've read the reviews... but wait! Are you actually getting the power supply you think you are or has the manufacturer pulled a fast one on you?... It's not uncommon for power supply brands to use the same name for different power supplies. Does that old review that was published a few years ago praising this power supply for its great performance still hold true on the new model with the same name? Or is the new model a piece of junk that is riding the wave of good reviews it didn't earn? We put 10 models with deceptively similar model names to the test head to head to find out. <sponsor> Despite having the same model name these are different power supplies and the reviews for the old units cannot be applied to the new units or vice versa. This can make shopping for a power supply very confusing if you don't know what you're looking at and those recommendations you see for <example> might not be true for <example>...".

Would also be good to teardown the power supplies to show the differences between the models. Some might be the same but have a minor revision change such as a different model fan but others might have completely different designs.

Wow, that's a lot of text. I see the correction was added to LTT's pinned message in the meanwhile, so everyone is happy.

 

I'm not sure what also discussing corrections in the WAN show help, as that's both a much more informal format, and has nowhere near the same viewership. I don't see how an even more obscure response adds anything of value.

 

The ethics of corrections on the real world are rather hit and miss, and usually entail a tiny correction hidden in the corner of page 15 after getting a front page story wrong. LTT definitely makes mistakes and has its biases, but seems to at least strive to admit mistakes and correct them as soon as they're aware of them. That's honestly what I'd reasonably expect, but unfortunately so many media outlets seem to think differently and take nowhere near the same level of responsibility.

[Make it werk]

Where is this power supply tier list you reference? I'm very curious to see it, but I. I'm having a hard time finding it with how many topics are on this form.

[socrguy143]

1 minute ago, Make it werk said:

Where is this power supply tier list you reference? I'm very curious to see it, but I. I'm having a hard time finding it with how many topics are on this form.

https://cultists.network/140/psu-tier-list/

[Spotty]

12 minutes ago, XNOR said:

I'm not sure what also discussing corrections in the WAN show help, as that's both a much more informal format, and has nowhere near the same viewership. I don't see how an even more obscure response adds anything of value.

 

The ethics of corrections on the real world are rather hit and miss, and usually entail a tiny correction hidden in the corner of page 15 after getting a front page story wrong. LTT definitely makes mistakes and has its biases, but seems to at least strive to admit mistakes and correct them as soon as they're aware of them. That's honestly what I'd reasonably expect, but unfortunately so many media outlets seem to think differently and take nowhere near the same level of responsibility.

The WAN show regularly gets 500k-1.5M views which is more views than a comment in the Youtube comment section receives. It's much more in line with the viewership of regular upload videos. As I mentioned corrections should be done in the same form that the error was made. Due to the way LTT produces their content it isn't sensible to expect they include corrections in the following days normal upload video, and due to the way Youtube works I definitely do not expect them to upload a dedicated video just to issue a correction. I think the WAN show provides a very good opportunity for them to make a fair effort at correcting any errors made throughout the week in addition to the comment they pin on to videos.

 

Another reason for issuing corrections through the WAN show is that on mobile view the comments section are hidden by default and you need to open the comment section to view them. People watching the video might not see those comments at all unless they are actively looking for the comments.

Spoiler

 

 

An additional benefit of having Linus provide any necessary corrections from the week prior on WAN show is that it adds additional incentive for them to ensure their videos are free from errors prior to publishing. Right now the culture seems to be "Publish the video and quickly move on to the next video. There's no time to review it. If there's any mistakes we'll just add a comment later".

 

These are the pinned comment corrections on LTT videos from the past few weeks. Some of these mistakes really should have been caught before publish like the number of cores a CPU they're reviewing has, number of PCIe lanes a GPU has, or the correct name of a product. Mistakes do happen but I think they can do better.

Spoiler

I haven't watched all of those videos, I cannot say how serious any of those mistakes were. I'm only highlighting that it appears to be a common issue

[BlueChinchillaEatingDorito]

9 hours ago, da na said:

I have the 430w Smart PSU in a small system of mine, it's served me rather well I will say. Going on 5 years.

I have an OCZ PSU running ModXStream Pro my 12700 system right now. A PSU people would probably say is a bomb. Still running fine after being in service for 10 years first powering an 1100T, then FX6300, and now this.

[nhand42]

My strategy has always been to buy the biggest and most expensive power supply from a brand that I recognize. PSU for $50? I'll budget $500. My reasoning being a good power supply will last minimum 10 years, will outlive any of the components it powers, and there's almost zero chance of it frying my boards or burning down my house.

 

Sounds like an expensive luxury but if you only buy it once and you can reuse it in several computers, it actually makes more financial sense than buying budget components several times. Same as with Noctua coolers I'm still using the same DH-15 cooler I bought in 2015.

 

* That's 500 Aussie dollarydoos. That's how much it costs for a Seasonic Prime Titanium down under.

[nhand42]

9 hours ago, FlyingPotato_is_taken said:

The R&S is one of the hottest scopes on the market.

I can only dream of one day being able to afford Rohde & Schwarz anything. That thing makes Rolls Royce look like an off-brand. They've been the gold standard for decades.

[Fred Hardy]

My 2 cents:

• Please test with 110V and 220V.(for us in europe  )
• Repeat the test on multiple samples of a model, if possible with very different serial number. I undestand that you can't test hundreds of samples but as much as needed to have an idea of the quality control of the model.

Thanks for your work!

[SmileyGG]

Do anyone tested Be Quiet! System Power 10 (650W)? I'm going to buy this PSU and I want to know if itf good or not

[Mark_nz]

12 hours ago, chunkychalice said:

I'm actually quite shocked that the power factor of most units is so close to 1...

I've been surprised too about how efficient modern day switchmode power supplies are.

(In my experience with signage and server power supplies.)

 

If @LMGcommunity is able to, server power supplies would be interesting for efficiency standpoint. I use them for projects because they sell cheaply secondhand (crypto crash).

and just because LMG may be able too..... find a cheap medical device with a PC inside and test that PSU in it. Compare 'medical grade' to a standard consumer PSU .

 

Another point to show is power losses into heat and the impact on a system. E.g. if the PC case air intake or air outtake is through the PSU.

Heck, even how the power efficiency changes with air flow restrictions. Like components within a case or the PC case being too close to a wall.