We don't seem very interested in PSU reviews, this review has no comments after three days lol. A PC enthusiast 'should' be interested in this important component for all the reasons Sir Hilbert has stated in previous PSU reviews so feeling sorry for PSU's, I thought I'd comment. I'd never heard of FSP Hydro before but a search reveals that they are a well established Taiwanese 'mid position' PSU manufacturer. Hilbert, has the introduction of ATX 3 caused you to change your long established postion that a PSU should be double your maximum total system power usage? I know that ATX 3 PSU's have to provide a transient demand of twice their wattage so perhaps that changes your advice? I've just had my gaming computer sunstantially upgraded with an MSI Suprim X 4090 and an Asus Thor II Platinum 1200w PSU now installed in a new Corsair 5000T case (the courier damaged my CM H500 Mesh case beyond repair). I looked at ATX 3 PSU's but I don't purchase the 1st gen of any product and I upgraded my Asus Thor mark 1 Platinum 850w to the mark 2 1200w version (others might hold different views of 1st gen products). I don't understand your comments about the maximum power of 500-550w with an i9 13900 + 4090 as my previous set-up with a i7 10700K + 3080 OC displayed 580w on the 850w PSU's OLED display while Metro Exodus Enhanced with RT, no DLSS, Ultra settings and the new 1200w Asus PSU shows 680w while playing the same game with the 4090. Same i7 10700 processor (standard clock, I don't use the optional OC), same mobo & memory, same Asus PG35VQ 35" UW with 200Hz, HDR1000, Gsynch Ultimate, VA panel, ESS Sabre dac. So 680w is just over half my Asus Thor II Platinum's power rating and unless ATX 3 is the reason you now recommend a 1000w PSU then I can only say I'm glad I went with your previous PSU advice. There's no need to pay the Asus/Seasonic 'tax' (I really like the power display & RGB) but there are other much better known brands than FSP Hydro PSU who make PSU's at similar prices (the Corsair Shift RM1000X ATX 3 is £208, be quiet Pure Power M ATX 3 is £160, prices as at 26 February 2023).

Hi @ChisChas, I'm the author of this test - in this particular review, you can find info "A PC like the one used in this test (i9 13900K + Geforce RTX 3080) rarely exceeds 500-550 W of power draw.". You can't find info about the power consumption of a pair with RTX 4090. This review was prepared earlier than the BQ Pure Power M (which I also checked -> https://www.guru3d.com/articles-pages/bequiet-pure-power-12-m-850w-atx30-psu-review,1.html, and it earned the "Recommended" award, which is more than "Approved" for the FSP) In this BQ test, you can see -> "A PC like the one used in this test (i9 13900K + Geforce RTX 4090) rarely exceeds 650 W power draw. ". Does that differ much from the 680W you're stating? Also I have an Inno3D card, which has 450W power limit. It's always better not to be on the edge of PSU capability (especially since the fan is usually louder at 75%+ of load). Also, you buy it for a couple of years, probably longer than the CPU/GPU. 10-12 years of warranty also show that there shouldn't be a problem after extended use.

@ChisChas It is not that that kind of reviews aren't needed. They are needed. It is just mostly enthusiasts forum/portal and at such as, we largely know HOW PSU IMPORTANT IS. This particular PSU is good. Is efficient. Has all features. IF I would be at need of buying one at the moment, I surely would consider this PSU. And for that, NO COMMENT IS NEEDED. AND as is known, and stated above, having more power available means that PSU working at lower load, heats less and thus uses (itself) less energy. And ofc there is noise factor, but far less important.

Captain_Hook:

Hi @ChisChas, I'm the author of this test - in this particular review, you can find info "A PC like the one used in this test (i9 13900K + Geforce RTX 3080) rarely exceeds 500-550 W of power draw.". You can't find info about the power consumption of a pair with RTX 4090. This review was prepared earlier than the BQ Pure Power M (which I also checked -> https://www.guru3d.com/articles-pages/bequiet-pure-power-12-m-850w-atx30-psu-review,1.html, and it earned the "Recommended" award, which is more than "Approved" for the FSP) In this BQ test, you can see -> "A PC like the one used in this test (i9 13900K + Geforce RTX 4090) rarely exceeds 650 W power draw. ". Does that differ much from the 680W you're stating? Also I have an Inno3D card, which has 450W power limit. It's always better not to be on the edge of PSU capability (especially since the fan is usually louder at 75%+ of load). Also, you buy it for a couple of years, probably longer than the CPU/GPU. 10-12 years of warranty also show that there shouldn't be a problem after extended use.

The 'unloved' PSU has some comments (now) lol. I don't have an i9 processor. How much more wattage does an i9 13900 uses v my i7 10700K (when playing a demanding game like Metro Exodus Enhanced at max qual settings on a 3K screen)? Yes, I'm aware that some games use a processor more than others before anyone thinks I haven't been paying attention in Sir Hilbert's classes for the last 10 years (I've been a member of this excellent forum since 2013). Standard Intel clocks in both cases (I have an i7 10700K overclock option in BIOS but don't use it, at the moment). And my Suprim X 4090 has a 520w power limit and the Asus Thor II Platinum 1200w comes with a 600w power cable (one of its many appeals & why I was prepared to pay the Asus 'tax'). Don't all the quality PSU's come with 10 year warranties nowadays? And personally I don't want to call upon a warranty, I'd rather buy a PSU that has a very low failure rate so I don't have the hassle of implementing that 10-12 year warranty. And my personal choice is to pay the extra for a Platinum or Titanium PSU especially at the level of a 4090 system. I wouldn't have bought any other 4090 than an MSI Suprim X, Asus TUF OC or Asus Strix OC (I had an Asus Strix OC 3080 but couldn't get supply of either Asus 4090 at my PC builder). My money, my choice, other's money, their choice. And yes, I've studied hard in Sir Hilbert's classes and have learnt about PSU fan speeds & db's. I'm subscribed to some 40 tech websites including guru3d and I'm hugely appreciative of the tech information these 40 websites have brought to me. The most specialist monitor website enabled me to set-up my Asus PG35VQ monitor so it performs better than out of the box.

GREGIX:

@ChisChas It is not that that kind of reviews aren't needed. They are needed. It is just mostly enthusiasts forum/portal and at such as, we largely know HOW PSU IMPORTANT IS. This particular PSU is good. Is efficient. Has all features. IF I would be at need of buying one at the moment, I surely would consider this PSU. And for that, NO COMMENT IS NEEDED. AND as is known, and stated above, having more power available means that PSU working at lower load, heats less and thus uses (itself) less energy. And ofc there is noise factor, but far less important.

'No comment is needed'? None of the comments on the guru3d reviews are 'needed' so I've no idea what valid point you're trying to make? We leave a comment because we wish to and none of us require anyone else's permission to leave a comment, do we? All reviews of major gaming components are needed and the guru3d reviews help us to make better choices. After being a member for only 10 years, I didn't realise this was a 'mostly enthusiasts' forum (couldn't resist teasing you). And do you really think that I haven't built up a reasonably high level of general tech/gaming component knowledge after 20 years of gaming? Your last paragraph assumes that I've apparently haven't even learnt the basics after two decades, being a member of guru3d for 10 years and subscribed to some 40 tech/gaming websites. As Sir Hilbert might remember, four gaming computers ago I had a Cosmos case with triple GTX580's, three radiators, Corsair 1200w PSU etc. My builder & I had great fun choosing the components for this amazing system, nowadays the custom cards are so well cooled that you don't 'need' a full water cooled loop.