Discussion PSU recommendations and power supply discussion thread - Tom's hardware

Page 25 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Well, they sort of do. Higher load means higher temperature means lower expected lifespan.

You do notice that I didn't say anything like capacitors don't care that you are using 750w on a 550w unit, right? LOL. I specifically said they don't care that you are using only 250w on a 550w unit. And, even then they really wouldn't because the protections would have kicked in long before that and shut the unit down, unless of course it was one of those units at the bottom of the first post on this thread. Then, who knows.
 
are you arguing capacitor degredation? because that was also debunked years ago. there is a liferating at a certain temperature sure, but thats at a temp a PSU usually doesnt run at.

Yes, I am arguing it. It exists. It is a thing. I've seen motherboards that were boxed for years, never run, and years later when they were needed to be swapped in to a system with a board failure they could not be used because the capacitors were degraded and either leaking or bulging and the unit couldn't be used. I've also seen this with power supplies. And actually, I 've seen this with capacitors that sat in the top drawer of my tool box, not even installed in anything or ever having been part of any device. So yes, I am. I don't care what anybody says, no matter who it is, it happens. It is an actual thing.

It might not ALWAYS happen and it CERTAINLY happens more if it's a cheap part to begin with, but to try to say it is fake, or made up, or doesn't exist, is pure nonsense.
 
  • Like
Reactions: Aeacus
Do these people have any real experience or credentials that make them trustworthy sources, other than compiling a questionable tier list?

No, they don't. And I've seen that list. Several versions of it. And it makes the list we had here look like a Ribeye compared to a hot dog that has been sitting in the sun for a week, despite the fact that Dottorent's list had a few minor short comings. That list on LTT would be laughable if it wasn't so purely pathetic. Last time I looked there were units on the top tier that had scored like 7.5 on GJ testing or failed other testing outright. Pathetic.
 
Apr 4, 2019
59
18
35
Last time I looked there were units on the top tier that had scored like 7.5 on GJ testing or failed other testing outright. Pathetic.

and they were placed the way they do for a reason. if you got sources that contradict the given methodoligy at their placement, please send them. otherwise you will just be seen as just another "muh precious unit is at x"

and i encourage you to do so.

edit: also people read more than the score number.

to stick with the folks on the JG forum than anybody you might find on LTT.

does this include the entire staff of the PSU tierlist? because they have been far more informative and open than here.
 
Last edited:
Honestly, I couldn't care enough to be bothered to do that. In truth, I wouldn't care if they put an Logisys 250w unit on their top tier. At all.

Perhaps you should just hang out there if you like them so much better than the members here. We get drifters from there all the time. They don't normally stay long. We are noncommittal either way. If you prefer to stay, stay, but at least TRY to act like the people you are talking to who have been working on systems since before there were systems to work on, are not stupid idiots.
 

Rogue Leader

It's a trap!
Moderator
Yes, I am arguing it. It exists. It is a thing. I've seen motherboards that were boxed for years, never run, and years later when they were needed to be swapped in to a system with a board failure they could not be used because the capacitors were degraded and either leaking or bulging and the unit couldn't be used. I've also seen this with power supplies. And actually, I 've seen this with capacitors that sat in the top drawer of my tool box, not even installed in anything or ever having been part of any device. So yes, I am. I don't care what anybody says, no matter who it is, it happens. It is an actual thing.

It might not ALWAYS happen and it CERTAINLY happens more if it's a cheap part to begin with, but to try to say it is fake, or made up, or doesn't exist, is pure nonsense.

Side note about this. I'm into retro console gaming quite a bit, and one of the more expensive units out there, the NEC Turbo Duo, you can many times find still new in box (for a LOT OF MONEY). However most times these new boxed never touched units when you plug them in, don't work. Why? The caps fail. Now admittedly they are 25 years old, but they also have never had a single drop of electricity through them.

But basically if you're selling a Turbo Duo that hasn't had its caps replaced the value is SIGNIFICANTLY lower than one that has. Because without them even if it turns on, its only a matter of a short time before its a doorstop.
 
  • Like
Reactions: Darkbreeze

Rogue Leader

It's a trap!
Moderator
Caps have a shelflife, and its usually much shorter than 10 years.

Caps in use can last much longer than caps on a shelf.

They do, but they also when in use while lasting longer, have a rating of anywhere from 1000 to 10000 hours. 1000 hours is not a very long time. This rating can be affected by temperatures as well (so in theory a 1000 hour cap run cool could last say 2000, or more who knows).
 
Apr 4, 2019
59
18
35
They do, but they also when in use while lasting longer, have a rating of anywhere from 1000 to 10000 hours. 1000 hours is not a very long time. This rating can be affected by temperatures as well (so in theory a 1000 hour cap run cool could last say 2000, or more who knows).

Usually the rating is at 85* or 105*. And if you go below that, the life expectancy goes up very quickly. Assuming no other variables change
 
All of which, contrary to what you said earlier, is capacitor degradation. Degradation, for ANYTHING that can fail whether it is capacitors or the engine in your car, is what happens PRIOR to failure. Obviously, if a capacitor that has never been part of a powered device fails, then it fails because it degraded. It didn't fail because it "haz a sad" that day.
 
  • Like
Reactions: TJ Hooker

TJ Hooker

Titan
Ambassador
You do notice that I didn't say anything like capacitors don't care that you are using 750w on a 550w unit, right? LOL. I specifically said they don't care that you are using only 250w on a 550w unit. And, even then they really wouldn't because the protections would have kicked in long before that and shut the unit down, unless of course it was one of those units at the bottom of the first post on this thread. Then, who knows.
Huh? When I said "higher load", I didn't mean "higher than the PSUs rated wattage". I meant that if you have two identical 550W PSUs and one is regularly loaded to say 450W while the other is only loaded to 250W, the former will be running hotter due to the higher load, and therefore experience faster capacitor ageing (although not necessarily enough to be significant). I.e. the capacitor can "care" about the particular load that was being placed on the PSU, in terms of the capacitor's lifespan.
 
Last edited:
Apr 4, 2019
59
18
35
So you're saying that caps DO have a lifespan and degradation... because thats what the last page of argument was all about.

im saying they have a lifespan. unshure about degredation (havent heard from jonny on that yet), but at least on that part if it is a thing, its not a concern in PSUs.
 

Rogue Leader

It's a trap!
Moderator
im saying they have a lifespan. unshure about degredation (havent heard from jonny on that yet), but at least on that part if it is a thing, its not a concern in PSUs.

Disagree.

Capacitor degradation is nearly 100% the reasons PSUs fail over time. I don't know where you could possibly have gotten the idea otherwise, but thats flat out wrong. I hope @jonnyguru can reply to this to give his insights.
 

Aeacus

Titan
Ambassador
if you have any suggestions to make the list better, im all ears to passing it on.

but it has methodoligy laid out on the table and its based on reviews. if there are units that do not have reviews, they may get preliminary placements if some info can be gathered from internal shots or otherwise. those placements are usually conservative and do not happen very often.
That LTT PSU tier list is flawed beyond repair. And it is not based on reviews. It's biased and based on personal preference.

Few examples:

1. Tier D - Potentially dangerous, but only in specific situations
Seasonic S12ii - M12ii - M12ii Evo - Focus FX - Focus Plus FX

In what world those Seasonic units are "Potentially dangerous"???

Yes, S12II series (80+ Bronze), alongside M12II EVO series is old platform, released some 9 year ago. However, S12II-520 (80+ Bronze), despite is group-regulation, it is the best group-regulated PSU ever made. S12II (80+ Bronze) is tried, tested and proven to be reliable. Ever wondered why Seasonic still produces and sells S12II (80+ Bronze) series PSUs? In 3rd world markets, S12II (80+ Bronze) is often the best PSU to choose from, alongside no-name PSUs that are available there.

Also, there is a nice fact about M12II EVO series which you don't seem to know. M12II-520 EVO and M12II-620 EVO are group-regulated, just like their fully-wired S12II counterparts. However, M12II-750 EVO and M12II-850 EVO aren't group-regulated, they are instead DC-DC, making them better than same series lower wattage units. But i guess you don't care about it. Or don't want to acknowledge it since it doesn't fit the agenda. I've seen several instances in that list where only one PSU from the lineup is looked at and then entire lineup is classified based on how one, ONE PSU, performs. You can't classify the entire lineup based on one PSU performance.

History lesson:
Seasonic S12, S12II (80+ White), M12, M12II are all discontinued PSUs, making me ask why there is M12II in the list? Successor of those 4 PSU lineups are: S12II (80+ Bronze) and M12II EVO series. I don't see any differentiation between S12II (80+ White) and S12II (80+ Bronze) in that list.

And what's with the Focus+ lineup? Just because Seasonic was brave enough to release a statement (source) where there could be issues with Focus+ Gold PSUs based on few user reports, isn't enough to completely blacklist the whole Focus+ Gold lineup, including Focus+ Platinum lineup.
If Flextronics would be brave enough and release similar statement for their platform used in Corsair AX760i PSU, would it get it's Tier S status ripped off and slapped into Tier D? I don't think so. However, more on that below.

Moreover, why are Seasonic Focus (Gold) series in the Tier D as well? Just because they share the similar name of Seasonic Focus+ (Gold/Platinum) series? Since i didn't see any sources/references/reviews in that LTT list which warrants it's position in Tier D.


2. Tier C - Recommended for entry level desktops, low profile HTPCs, Office desktops, preferably GPUs with no external PCI-e power connection.
Corsair CX 2012 - CXM 2012 - VS 2017

Corsair VS series, the known system killer, with it's 3 years of warranty being better than e.g Seasonic Focus+ (Platinum) with 10 years of warranty?
Someone really hates Seasonic and loves Corsair.


3. Tier A - Recommended for high end gaming systems and workstations with single GPU setup
Seasonic Focus GX - Focus GM - Focus PX - Focus SGX - Prime Platinum - Prime Titanium - X - XP - XP2 - XP3 - Snow Silent- Prime Fanless - X Fanless

What's with the "single GPU setup"? Do you not know that there are 750W, 850W and 1kW PSUs that are more than capable running 2, 3 and 4 GPU setups?

Also, i see Seasonic PRIME (Platinum/Titanium) series but not PRIME (Gold) and PRIME Ultra (Gold/Platinum/Titanium) series. How come? Is it because PRIME Ultra 750 Titanium scored 9.7 on JG review (link), with the same 9.7 score from JG as Corsair AX1600i (link) and once added, it needs to be in Tier S, alongside AXi, which isn't allowed?


4. Tier S - Recommended if you're looking for some of the best units brought out in the recent years, generally come at a premium
Corsair AXi 2013 - AXi 2014 - AXi 2018

There is no such lineup as AXi 2013. What there is, is: AX760i, AX860i and AX1200i, released in 2012. Also, AX760i and AX860i are discontinued PSUs at current date. Some digging and found the possible reason why:
When the Corsair AXi series of power supplies came out in 2012 they featured a single rail design. Corsair switched to a multiple rail design for the power supply series in 2013 (previously unknown to us). This is obviously a significant difference in the design of the power supply. We also learned that the earlier single rail power supplies did not have OCP enabled by default.
source: https://www.legitreviews.com/intel-x99-motherboard-goes-up-in-smoke-for-reasons-unknown_150008

With that public statement, Corsair AX760i, AX860i and AX1200i (2012), without OCP enabled, max they are suited is for Tier B+ but for the public fact that they have fried systems, Tier D is well suited for them.


These are just few examples pointed out in the fundamentally flawed LTT PSU tier list.
 
Apr 4, 2019
59
18
35
Since i didn't see any sources/references/reviews in that LTT list which warrants it's position in Tier D.

i suggest reading the placing methodoligy then

Corsair VS series, the known system killer,

do you have pointers for this claim? because its a "eh" unit, but nothing is seriously wrong with it.

Corsair AXi 2013 - AXi 2014 - AXi 2018

There is no such lineup as AXi 2013. What there is, is: AX760i, AX860i and AX1200i, released in 2012. Also, AX760i and AX860i are discontinued PSUs at current date. Some digging and found the possible reason why:

thanks, something to look at, ill pass it on.

What's with the "single GPU setup"? Do you not know that there are 750W, 850W and 1kW PSUs that are more than capable running 2, 3 and 4 GPU setups?

its meant as a generlization for newbies to see what they are looking for, ive brought it up, but at this time other things have taken priority.


However, M12II-750 EVO and M12II-850 EVO aren't group-regulated, they are instead DC-DC, making them better than same series lower wattage units

that one seems to have gone lost in a revision at some point, previous placement was in Tier B (old ranking, which makes it either Tier B or B+) thanks for notifying as that unit was included previously.

And it is not based on reviews.

it is, its why there is linked sources and units rarely if ever get a placement before propper sources are in place.

It's biased and based on personal preference.

only thing that could be biased is the placement methodoligy, tho the methodoligy is not made by 1 person alone, and the people involved dont have any resentment towards certain brands or manufacturers.

In what world those Seasonic units are "Potentially dangerous"???

i suggest reading the entire thing as the previous focus fx units had serious ripple issues under certain load conditions. since that is a requirement for Tier D, thats the reason for its placements. its been covered multiple times before.
 
Apr 4, 2019
59
18
35
Which was fixed with capacitors in the cables. Now the ripple is servicable.

untill reviews are out, nothing can be 100%, but the Focus GX/GM units have been given a temporary placement in Tier A.

note that stock of the fx is still sold alongside the GX in many stores.
 
i suggest reading the placing methodoligy then

i suggest reading the entire thing as the previous focus fx units had serious ripple issues under certain load conditions. since that is a requirement for Tier D, thats the reason for its placements. its been covered multiple times before.

These are two textbook examples of why that list is garbage. The methodology you describe is worthless and exists only to give you an excuse for being wrong. And the alleged "serious ripple issue" with the Focus only occurred in a single Chinese test at 900w during a 150w overload. Nobody competent is going to describe that as a "serious issue". And the list is littered with nonsense like that.