[OLD] PSU Tier List 3.0 (Legacy)

[LukeSavenije]

1 minute ago, Azims said:

CX is better than CXM confirmed by JonnyGuru.
[media]https://youtu.be/QJ_VV3UjEBY?t=1196[/media]

i'm aware

 

it's fixed in my viewable revision, but lienus hasn't kept up

[Blademaster91]

1 hour ago, LukeSavenije said:

i understand what you mean here, but you forget two things i've noted earlier about the s12ii/m12ii and many other older units

• a complete lack of undervoltage protection
• out of spec possible if you load the 12v too much while loading the 5v too little

 

the overloading issues are on the focus and focus+ line

No uvp is bad I agree, but how much is too much on the 12v? I also understand the point of replacing it as if the person can afford a new GPU, they should have the budget for a $60 PSU. but i'm saying why replace it in a realistic sense of most people aren't going to overload the 12v rail.

And I was referring to the tier list 3.0 thread in which most of the Seasonic units got put in the "do not buy" it's gonna set your house on fire tier, even though the overloading issue really isn't that much of an issue as again the average gamer with a single GPU isn't going to overload their 650w PSU or even use half of the 650w rating.

1 hour ago, 5x5 said:

S12II is a product of it's time. Was good in 2014, not so much in 2019

It's certainly out of date, but not the worst thing ever and still better than some low end new stuff.

[LukeSavenije]

3 minutes ago, Blademaster91 said:

how much is too much on the 12v?

5% more (12.6v) according to atx spec. but it's the 5v that goes out of spec when loading up (NOT overloading) the 12v and not the 5v. as the same 5% is required for 5v, this would mean it's allowed to have up to 5.25v, but PSUs based off this platform have been able to show 5.4v under certain loads. 

 

this is why it doesn't have c6/c7 or haswell certifications

[Flying Sausages]

Hello

 

Any info about this psu? Apevia ATX-BT700W Beast 700W ATX Gaming Power Supply, Supports Dual/Quad Core CPUs, SLI, Crossfire, Haswell

https://www.amazon.com/Apevia-ATX-BT700W-Supports-Crossfire-Haswell/dp/B01IFTQVPA/ref=sr_1_21?crid=39KD78K5Z8GUE&keywords=750w+power+supply&qid=1569601012&sprefix=750w%2Caps%2C200&sr=8-21

 

I can predict it is a bad psu based on the price.

[LukeSavenije]

47 minutes ago, OlympicAssEater said:

I can predict it is a bad psu based on the price

and i don't have much evidence to what it exactly is, but i know that isn't a 700w psu

[BiG StroOnZ]

Nice interview with JonnyGuru:

 

 

[LukeSavenije]

6 minutes ago, BiG StroOnZ said:

Nice interview with JonnyGuru:

yeah... I've seen it

 

it sure is a good video

[FALC0N]

18 hours ago, Blademaster91 said:

No uvp is bad I agree, but how much is too much on the 12v? I also understand the point of replacing it as if the person can afford a new GPU, they should have the budget for a $60 PSU. but i'm saying why replace it in a realistic sense of most people aren't going to overload the 12v rail.

And I was referring to the tier list 3.0 thread in which most of the Seasonic units got put in the "do not buy" it's gonna set your house on fire tier, even though the overloading issue really isn't that much of an issue as again the average gamer with a single GPU isn't going to overload their 650w PSU or even use half of the 650w rating.

It's certainly out of date, but not the worst thing ever and still better than some low end new stuff.

 

18 hours ago, LukeSavenije said:

5% more (12.6v) according to atx spec. but it's the 5v that goes out of spec when loading up (NOT overloading) the 12v and not the 5v. as the same 5% is required for 5v, this would mean it's allowed to have up to 5.25v, but PSUs based off this platform have been able to show 5.4v under certain loads. 

 

this is why it doesn't have c6/c7 or haswell certifications

 

Everything LukeSavenije says here is either untrue or out of context.  

 

The S12II does have undervoltage protection.  Myself and another poster have pointed this out to him previously and even linked to reviews confirming it.  He has himself referenced reviews showing working UVP on S12II derivatives.  So why he keeps repeating this is a bit of a mystery.

 

The S12II NEVER hits 5.4 volts on the 5v rail under any know load condition.  He was told this by someone else.  That person was wrong and so it he.  Again, I have previously informed him of this, but he keeps repeating it anyways.

 

Finally, the one out of spec load he does correctly describe occurs only on a maximum minor rail crossload with next to no 12v load. This is NOT a real world load scenario.  It's like putting a car on a dyno and simulating 7000 rpm at 5mph.

 

 

[LukeSavenije]

1 hour ago, FALC0N said:

The S12II does have undervoltage protection

i stand corrected on that, I've done a deeper dive into the ic (will put in a spoiler to let you check yourself), it does have uvp, but only on the minor rails. with the lack of ocp there, it doesn't have my trust. the protection on the 5v and 3.3v are set rather low, 1,8V for 3,3V and 3,3V for 5V.

 

1 hour ago, FALC0N said:

He was told this by someone else

i don't work with single sources here. i have this confirmed by @Ave with a model o, @Stefan Payne, ebasler from tech-review.de, @PSUGuru and others, so not the least.

 

1 hour ago, FALC0N said:

Finally, the one out of spec load he does correctly describe occurs only on a maximum minor rail crossload with next to no 12v load.

the other way around, a higher load on 12v with a low load on 5v. this is an issue with a lot of group regulated units, and I've seen some doing a lot worse (a lepa unit at hardwareluxx did around 7v)

 

this is with a modern system more likely, as the 12v gets widely used by many components.

 

@jonnyGURU's website used to have a s12ii review with crossloads (i recall it being the corsair vx), but that one got lost in the server moving.

[seon123]

8 minutes ago, LukeSavenije said:

@jonnyGURU's website used to have a s12ii review with crossloads (i recall it being the corsair vx), but that one got lost in the server moving

The reviews of the VX aren't lost at all. You're just too lazy to actually look them up on the website. 

http://www.jonnyguru.com/blog/2007/09/25/corsair-vx550w-power-supply/2/

http://www.jonnyguru.com/blog/2007/07/31/corsair-vx450w-power-supply/2/

[OrionFOTL]

1 minute ago, seon123 said:

The reviews of the VX aren't lost at all. You're just too lazy to actually look them up on the website. 

http://www.jonnyguru.com/blog/2007/09/25/corsair-vx550w-power-supply/2/

http://www.jonnyguru.com/blog/2007/07/31/corsair-vx450w-power-supply/2/

Also the Seasonic platform in the VX450 has nothing in common with what they're talking about (S12II Bronze), so I have no idea what's the relevance of bringing VX up. 

[LukeSavenije]

6 minutes ago, seon123 said:

The reviews of the VX aren't lost at all. You're just too lazy to actually look them up on the website.

then i guess I've missed them in my search

 

4 minutes ago, OrionFOTL said:

Also the Seasonic platform in the VX450 has nothing in common with what they're talking about

then realhardtechx reported it wrong, they list the 450 as a seasonic s12ii

[OrionFOTL]

10 minutes ago, LukeSavenije said:

then realhardtechx reported it wrong, they list the 450 as a seasonic s12ii

That's correct though, VX450 is S12II. How is it relevant to discussing S12II Bronze? 

 

Are you not aware that Seasonic had an S12II series before they made S12II Bronze, at that they're not the same?

Didn't something tick you off in the VX review, like the fact it doesn't look anything like the platform you're discussing, or that it's from 2007, when S12II Bronze was made in 2009?

[Rexper]

2 hours ago, FALC0N said:

The S12II does have undervoltage protection.

Not that I'm on anyone's side here...

The S12ii and M12ii uses the HY510N as seen by this review.

This IC does support UVP on the minor rails (3.3V and 5V) but not 12V.

The PSU lacks OCP on every rail.

And has no OTP.

 

2 hours ago, FALC0N said:

The S12II NEVER hits 5.4 volts on the 5v rail under any know load condition

As per the review I linked, the 12V rail drops to 11.633V during a crossload test. Though in spec, this put strain on modern demanding components as the 12V rail deviate +-242mV as the workload deviates.

 

2 hours ago, FALC0N said:

This is NOT a real world load scenario.

The Haswell C6/C7 sleep states require a sort of crossload that this unit and any other group regulated power supply cannot handle.

 

54 minutes ago, LukeSavenije said:

1,8V for 3,3V and 3,3V for 5V.

Evidence? Last I checked no one has tested the protections on the S12ii/m12ii. 

[LukeSavenije]

7 hours ago, Rexper said:

Evidence?

https://www.computerbase.de/2017-07/cooler-master-cougar-xfx-zalman-netzteil-test/3/#abschnitt_schutzschaltungen

[Mr. Budget]

What about the Molex 650W 12V ATX unit? I couldn't find it on either this or the legacy list.

[LukeSavenije]

5 minutes ago, Mr. Budget said:

Molex 650W 12V ATX unit?

ehm... model number?

[Mr. Budget]

7 minutes ago, LukeSavenije said:

ehm... model number?

Couldn't get the model number on either of the two I was wondering about and the Moxel went out of stock so what about this: (y'all don't allow shortened links so apologies.) https://www.ebay.com/itm/NEW-650w-Genuine-Dell-Power-Supply-PowerEdge-1800-TJ785-GD323-U2406-PS-5651-1/202698331714?_trkparms=ispr%3D1&hash=item2f31c31642:g:Ur8AAOSwoW1cgzR1&enc=AQAEAAACMBPxNw%2BVj6nta7CKEs3N0qWQDIF2THanPl%2FroXqSO5qlCUy9gHMZsSbQ%2BV%2B5xhIgQ%2FN%2FIiThuy%2BuXpNr8IjzZMticGAE5usr2nNPFayfB9%2FOrLacDpOTs5VdoFT1G5qUX5Y1jfJ82LUrGeNf8ShKOPGKviC7T6DzSOThwlx5rvZPOpsk39J0Nq%2BPW%2FOMWy4goEdHJKWfZiPqD4p66xZV089h93EYx5j67NgyLtKDuLty9aPngGuHclThWjUpDGtIabnEZlsF9Ahr2bixLbylUAY7TdIZX9RJ%2BI7DT7LIP5m0CWvXjbW%2BR5peQ4P%2B3rq2dtfpTw76MOdSb%2BV56YRaiPjKNWHuptwmaoHbV16mjn5QmZU7e6xF%2F7hGZ91pOBHNydCGHNPEo9b7yAsFAXCKZDd6koNN0nBPXiKHZgqXWhV0et7zMQ0sowQEFkiybINvnQTf51xWh9mQYp7V%2FPZ%2B1SWscCV%2BcOvaiOknpZBmQPh%2BAxyGMCslAEuztthzD2g5hrysXE%2BqfrHu6HwC37IsH4PEI9FVLO1UtWE3jpBBKSUsMImsWl1I3%2FHwlU%2FYGCYmaHKuN%2FdSqwkwSQ347XTIQOkoMfvASZQ2C4Ydxm%2BRz2vI%2FvYFNyBmM26H1nKz742TySixg8ExsomVxomyPfOYuubxwc%2BYQtmg2Akmlck7Cg7MNSoW15X4TnFg8BVz90vOXgBhEV%2FM4nagyE8BfCP7GoKXn7J6%2Bb5PemGxWo4UAJyX&checksum=2026983317146a61c0896487410fa699603cecc1fd1e

[LukeSavenije]

14 minutes ago, Mr. Budget said:

y'all don't allow shortened links so apologie

i mean... you could've put it in a spoiler

 

but that's an oem psu, we don't tier those

[Guest]

1 hour ago, Mr. Budget said:

What about the Molex 650W 12V ATX unit? I couldn't find it on either this or the legacy list.

Where did you get "Molex" from?

 

Any way.. That Dell PSU isn't even following ATX standards.  It has a +3VSB instead of a +5VSB.  Why?  Who the heck knows.

 

[Mr. Budget]

20 minutes ago, jonnyGURU said:

Where did you get "Molex" from?

Molex as in the company... If you mean website I found it on Amazon aswell as eBay and Newegg.

[Guest]

1 hour ago, Mr. Budget said:

Molex as in the company... If you mean website I found it on Amazon aswell as eBay and Newegg.

I know who Molex is.

 

But I don't know why you called that PSU a Molex PSU. 

 

[Mr. Budget]

1 hour ago, jonnyGURU said:

I know who Molex is.

 

But I don't know why you called that PSU a Molex PSU. 

 

I clarified. I didn't find the exact Molex PSU so I put up a different one.

[Guest]

17 minutes ago, Mr. Budget said:

I clarified. I didn't find the exact Molex PSU so I put up a different one.

There is no Molex PSU.  Molex is a manufactuer of connectors:

 

https://www.molex.com/molex/home

 

Anyhoo...  That Dell PSU is not standard and therefore wouldn't be seen on a tier list for aftermarket PSUs for the DIY market.

 

[Mr. Budget]

3 minutes ago, jonnyGURU said:

There is no Molex PSU.  Molex is a manufactuer of connectors:

 

https://www.molex.com/molex/home

 

Anyhoo...  That Dell PSU is not standard and therefore wouldn't be seen on a tier list for aftermarket PSUs for the DIY market.

 

Molex has made PSUs...