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

Aeacus · Jan 12, 2018

Onus · Jan 12, 2018

I think this thread will be useful. The tier lists are fine for "My PC has "these parts," and is used for [office|gaming|content creation|whatever]; what PSUs will work for me?" It won't give the best suggestion, but it will not fail to give good suggestions. Then, if they want to know Why? they can come here.

Darkbreeze · Jan 12, 2018

Yes, I agree. I can't help but feel like there is never "too much" information available to those looking for it. There is only ever "good" and "bad" information, and so long as you are leaning towards the "good" side of it, it is going to be of use and helpful to somebody. More options/information is almost always better. Lack of/not enough is almost always bad or at least frustrating.

Darkbreeze · Jan 12, 2018

Ok, so you can call this a retraction, or an olive branch, or simply recognizing the truth for what it is (Although it still does not convince me of the quality or value of the CXm550w unit since there are still no actual real world reviews of that unit) but I'm here to say that I see and agree with the fact that the PSU reviews at Hardware insights is probably tragically flawed. After checking about four other models reviewed there and comparing those same models to reviews from other well respected reviewers, I found that NONE of the reviews there I looked at gave even a passing grade to units we KNOW are good and that other reviews show wildly different results of.

So, Hardware insights is off the "nice" list and on to the "naughty" one. Not that I had bothered looking at them for anything other than that one review previously anyhow, but that's enough for me to be convinced that the testimony against that particular site is fairly convincing. Something is seriously wrong with how they are doing things there.

Onus · Jan 12, 2018

HardwareInsights reviewed an InWin PSU at my request. I think they might differ because their sources for units differ; aren't they in AUS or NZ? This calls to mind Coolermaster, which removed some of their questionable units from the US market, but indicated (to Gabriel Torres) that they were just going to sell them in other parts of the world.
That's more of the dishonesty that has me saying "friends don't let friends buy Coolermaster."

Darkbreeze · Jan 12, 2018

Actually, I think they are a Czech based site, from what I read on one of the threads over at JG. I'm not sure that being in a different part of the world can explain away having failing grades for power supplies that multiple other well respected sites have passed with flying colors. I was willing to entertain the idea that on that ONE unit, they might have hit upon a real problem, but not with five different units, the results of which four are not consistent with other reviews. I gotta scratch them off the list.

FWIW, I am almost entirely in agreement on the Cooler Master front. That doesn't mean I won't include relevant reviews for them if/when I come across some worth adding, but I would certainly not ever purchase one for myself or recommend them to anybody when there are a lot of other as good or better units at almost, if not every, price point.

Onus · Jan 12, 2018

I just wonder if the units that H_I has reviewed are internally identical to those sold in the more demanding US market, despite having the same name.

turkey3_scratch · Jan 12, 2018

Some of those Coolermaster Toughpower RGB units (whatever they're called) have been super well priced. Plus the Vs are usually good in other countries.

<Mod edit - "Toughpower and Toughpower Grand" are Thermaltake series, not Cooler Master>

Rexper · Jan 12, 2018

G-series
Seasonic SSR-360GP @ JonnyGuru:
Seasonic SSR-360GP @ PCPerspective:

Those units use the platform GP which only belong to the SSR-360GP, SSP-300GT, SSP-350GT and KRPW SS350W/90+ according to orionpsudb.com

The rest of the G-series use a different platform, which makes those reviews a bit irrelevant for 'G-Series'. Maybe rename to 360w G, GP platform or similar

Aeacus · Jan 12, 2018

Rexper :

According to the Seasonic, SSP-360GP belongs in the G-series, alongside with SSR-450RM, SSR-550RM, SSR-650RM and SSR-750RM.
Proof: https://seasonic.com/g-series

Seasonic SSP-300GT and SSP-350GT are part of a SSP-xxxGT/RT (OEM) series which are industrial PSUs and not intended for consumer use as the SSP-360GP is.

According to the orionpsudb.com platform index (LINK), Sesonic G platform is also used in the: Seasonic G, Seasonic S12G, Antec TruePower Classic, Antec EDGE 550-750, Corsair HX650, Fractal Design Edison M, XFX TS Gold and XFX XTR PSUs.

If you'd arrange the PSU reviews by the platform then it would make everything more complicated. But it's up to Darkbreeze to decide if he wants to arrange the PSU reviews by the PSU series or by the platforms used inside the PSUs.

Rexper · Jan 12, 2018

It belongs to the G series, though it's layout and likely performance will be different because it's a different platform. You can't apply the performance of one PSU which has a whole different platforms to the rest of the PSUs in that series. It's just misleading. Just use a review of a PSU which uses the platform the majority of G series units use, http://www.jonnyguru.com/modules.php?name=NDReviews&op=Story&reid=348

Darkbreeze · Jan 12, 2018

You're getting WAY too (Choice word) far into the realm of nit picky-ness now. If it's part of the series, it's part of the series. Period. There are PLENTY and MANY power supplies that are in the same series but use completely different or significantly altered platforms. It doesn't matter. If the naming scheme says that's the series it's in, then that's how it's going to be named, accurately, by the model you'd find on the box (Or something close to it) if you bought the damn thing. I am 100% NOT going to go through the process of trying to figure out every single unit by platform and group them that way. Technically speaking, I'm only loosely grouping these units anyhow.

Loosely, as in, MOSTLY going from lower capacity to higher capacity within a brand and MOSTLY trying to keep bronzes with bronzes, golds with golds, platinums with platinums, etc. Also, SOMEWHAT older to newer, but not in any definitive way. Hell, the RealHardTechX database list has their actual list lined up by series but if you click on ANY of them, it simply takes you to ALL of them. Click on Platinum series, you go to the database page but are in the bronze listings. So, whatever. So long as the reviews are listed and are somewhat easily found on the page, I'm not getting into that kind of nit picky bs. At least not this early on. Maybe when it comes time to fine tune things a bit later.

I mean this is what, three, four days along at most? I'm lucky to have even gotten half of these put up by now the way I feel. I feel like a bag of smashed (@$$h!!!s) to be honest, so for now I'm just gonna do what I can and it can get sorted further as it goes.

You know, actually, for a bunch of guys who intially thought this was a big waste of my time, sure does seem like it must be important or interesting enough that you want to be a part of it, which is a good thing I think, but let's just keep in mind that Rome didn't happen over night and all that. I mean, look at ALL other types of PSU resources whether they are Tier lists, review databases, platform databases, OEM databases or just regular old model roundups. I promise you none of those happened overnight. Most of them, some at least that I know of for sure like the Who's who in power supplies OEM database, took MONTHS from start to finish, so this will get there too.

It's already more today than it was yesterday, and tomorrow it will be more than it was today.

Darkbreeze · Jan 12, 2018

turkey3_scratch :

Yes, absolutely regional availability sometimes plays a role. I've seen more than one occasion when a CM unit, even one that's not terribly good, was about the best option available. So in cases like that, clearly better is better even if better doesn't mean "good". Plus, they DO absolutely have SOME good units (They've got a lot of junk too) but it's rare that those units are in a price range where there are not much better units for the same price, or an eqivalent unit for significantly less.

For the record, Toughpower and Toughpower grand are Thermaltake PSU series, not Cooler master.

Still, much as CM has fallen flat with me, based on personal experiences with their products over the years AND as Onus has mentioned, with some of their shady practices, there clearly will be some occasions where one of their products will simply be the best available choice and that just is what it is. I would personally, even against my better judgement in some cases, much prefer to recommend a Corsair product over a CM product any gven day of the week. Usually though this ends up being a case of saying "You COULD go with this, and it would be better than some of the options available to you, but I and your system would both be happier if you did THIS instead", even if it's a bit less convenient or slightly more costly.

goldstone77 · Jan 14, 2018

OP is having issues I think are probably related to the power supply. Does anyone have any information on these PSUs? Also, if you wish to take a look at the thread and comment there that's great too.

Sentey 450w (using now) is old and came with the Pentium G630 PC. Eurocase 500w come with Core i5 PC.
Core i5 was using Eurocase PSU before Mobo stopped working.

http://www.tomshardware.com/answers/id-3617675/cpu-freezes-randomly.html#20596248

Darkbreeze · Jan 16, 2018

See, this is one of those "things" that makes it rather difficult to actually get a feel for whether or not a unit is any good. Either they are simply getting completely different quality units to review, or there is simply not enough consistency across the variety of different methods and equipment being employed for testing.

That, or as has been mentioned in the past, OW is simply afraid to accurately score a Corsair power supply negatively due to the obvious affiliation his overlord has with them.

OW says the TX750m is superb. Fantastic. Mythic levels of everything. Nothing bad about the unit at all except maybe being priced a little bit high and having semi-modularity.

http://www.jonnyguru.com/modules.php?name=NDReviews&op=Story6&reid=528

While Aris on the other hand says, "meh, it could be a lot better". It has poor hold up time. Poor +12v ripple control under full load. Terrible fan profile. Bad transient response on the 3.3v rail.

http://www.tomshardware.com/reviews/corsair-tx750m-psu,5062-11.html

So those are WILDLY different reviews of the same unit. Then in the same review OW goes on to say how great the EVGA B3 unit is and how it's this units biggest competitor, and we all KNOW what Aris found on all the B3 units he tested. Something is seriously wrong SOMEWHERE in this equation. Either somebody is sucking at their job or somebody is actually being that paid shill we all seem to get accused of being from time to time.

turkey3_scratch · Jan 16, 2018

Darkbreeze :

I think you are overthinking it. Under OW's testing conditions both those units performed excellently. OW doesn't test the fan profile, doesn't test transient response, doesn't test holdup time. So he has no way to critique those things on the Corsair TXM PSUs. Likewise, OW does not test protections, so there is no way he can critique the protections not working on the EVGA B3.

This is a misunderstanding that caused misled articles like this to be released https://www.techpowerup.com/237767/evga-caught-sending-golden-samples-of-the-supernova-b3-to-jonnyguru

EVGA never sent a "golden" sample to Jonnyguru, it's just that Jonnyguru's testing methodology is basically the same as it was nearly a decade ago so the majority of flaws that are present in power supplies you won't see in a Jonnyguru review.

Darkbreeze · Jan 16, 2018

Well, then they should step it up, or step off. I doubt Aris would bother testing those factors if they weren't important, and if they ARE, then SOMEBODY should be figuring out a way to get OW or Taz the necessary equipment so he too can bring his reviews up to modern standards.

These are the two primarily looked to venues for trustworthy reviews and if the reviews are not accurate by way of omission, then that's just as bad as being inaccurate by way of flawed data like Hardware Insights.

Darkbreeze · Jan 16, 2018

Also, you're wrong, at least in part, because OW said the units have good ripple control, and Aris says they do not. I know damn well OW tests for ripple.

turkey3_scratch · Jan 16, 2018

Jon Gerow has no association with Jonnyguru.com anymore and hasn't for years.

OW can't get anymore equipment. Jonnyguru, sadly, does not churn in a lot of money. OW makes barely any money doing those reviews, and between him and Tazz they're not earning much money. Investing in more equipment would be too costly. Not to mention how much OW is going through with his father's cancer

But stepping out of the game seems unwise to me. Not everybody needs to test everything.

Plus even if OW did have the equipment to test more stuff, it'd be far too time consuming then. He wouldn't have the time to write that much stuff for a review and test that much stuff. And I don't feel like the site would make much more money anyway; what keeps people going there is probably the simplicity of the reviews. Start making the reviews more extensive and then there will be one every other week instead of every week. And that'll actually cause them to lose money.

The ripple difference could just be contributed to some unexpected variance or difference between the two samples as well as the differences in load. 100% at JG may not be the same load combination as 100% on Toms. Not to mention different equipment. And different mains being tested on. There's a ton of things that could have an effect.

Darkbreeze · Jan 16, 2018

Maybe he needs to market his skills to a DIFFERENT company then, one that WILL pay him according to his capabilities so that he's not practically working for free. Sorry, but I think there are probably places he could do those reviews for that WOULD offer him a bit more compensation. And some investments in additional equipment. If I choose to work for grease monkey when I'm capable of being a Ford line tech, that's my own fault.

I can't complain about Grease monkey only paying me minimum wage if I'm not willing to go take on the work required to make what I'm worth. How does Jon not have anything to do with JG anymore? Did he sell off his interest in it or something? Seems funny his avatar still says site founder and he's a full on administrator on the site if he has no relationship with it anymore. Who owns it now then?

I think that's incorrect on the ripple. Shouldn't matter. Ripple at 500w output should be the same no matter whether it's a load tester, graphics card or whole system pulling the 500w from the socket. Certainly different equiptment might show SOME variance in measurements, but to go from "mythic" to "bad", no f-ing way that can happen unless something is terribly different or wrong.

turkey3_scratch · Jan 16, 2018

Darkbreeze :

It wasn't even "bad" per say in the Tomshardware review, only about 50mv whereas it was around 25 in the JG review.

Neither site is lying or falsifying data. It's just what happened. You'll find if you look at lots of power supply reviews of the same power supply you'll very often find differences in ripple measurements like this across the different sites.

Or another example is to look at HardOCP's transient response testing. Units always perform way worse on there than on Tomshardware.

turkey3_scratch · Jan 16, 2018

Also, I'd highly recommend you visit the Jonnyguru forum if you have such concerns with the review and post those concerns in the thread on that particular review.

Darkbreeze · Jan 16, 2018

They don't like me over there. I don't go there anymore. I've tried to reach out several times with requests for specific information and have basically been told go away, you're not wanted here. Even by email. So F those guys IMO. I still like the reviews they do, don't get me wrong, I just wish it were possible for there to be more consistency in results between sites. Eh, whatever though.

goldstone77 · Jan 16, 2018

Guyz didn't someone just get baited and switched not long ago? EVGA sent some higher quality PSU to get reviewed, and when the retail PSU were test there was a shocking difference in quality. I'm sure this isn't the first time a company has done this type of thing, and why it's important to have multiple review sources.
https://www.techpowerup.com/237767/evga-caught-sending-golden-samples-of-the-supernova-b3-to-jonnyguru

An Overclock.net member, "shilka", posted late last month over the failure of power protection in EVGA's SuperNova B3 power supplies, specifically the 450 and 850 W models. This adds to the concerns first raised by Tom's Hardware, one of whose review units created fireworks. Normally, for a product that has been on the market for some time, as the B3 series has, one could chalk up such incidents to a faulty batch. But it appears that might not be the case, considering JonnyGuru also reviewed EVGA's B3 and did not encounter any fire hazards.

Rexper · Jan 16, 2018

But it appears that might not be the case, considering JonnyGuru also reviewed EVGA's B3 and did not encounter any fire hazards.

... That's because reviewers at Jonnyguru doesn't test OPP (or any protections for that matter. That doesn't mean the flaw didn't exist in the sample they tested)

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

Titan

Titan

Retired Mod

Retired Mod

Titan

Retired Mod

Titan

Expert

Respectable

Titan

Respectable

Retired Mod

Retired Mod

Distinguished

Retired Mod

Expert

Retired Mod

Retired Mod

Expert

Retired Mod

Expert

Expert

Retired Mod

Distinguished

Respectable

Share this page