Corsair RM750i Power Supply Review

Not open for further replies.


Sep 30, 2010

Thanks for the thorough review.

The RM750i does beat the G2 750 at $10 more, which is a good trade for a mpore quite operation, and is $20 cheaper than the Snow Silent 750, but how does it compare to the $10 more expensive P2 750?


Contributing Editor
Feb 5, 2015
I am sorry but I haven't reviewed the P2 750 model yet. However since it is Platinum it should be compared with the HX750i.


Sep 30, 2010
No I'm comparing it based on price.

Since the gold rated RM750i and the platinum P2 750 are $10 apart, excluding the rebate, and as Jonny (The Jonny??) said, a few %s of efficiency isn't as important to me as something as tangible as emitting noise.


Dec 24, 2014
This seems like the droids/PSU I've been looking for. That or maybe the 850i. Also, I like the white lettering - that way it will match my case no matter what lighting I have inside.
I'm confused, there's no other product that compares to the corsair rm750i except the evga g2 750 and the seasonic snow silent 750 which is a higher category and price. Yet isn't that the ss-750km3 sitting at $10 cheaper?

Cwt may be improving but it's not ss quality. Hard to face the two off when they're so close in price or the seasonic is cheaper. I fail to see how it's a better option than the competition. A better option than their own lineup maybe, but that's not saying a ton.

I suppose it's true it comes with corsair link, though with psu's already this quiet not sure the need is there to control the fan beyond active variable control built in and silent mode at low draw. I've never needed to monitor my psu temp or control the fan speed even on a plain old active fan design. Comes off as a bit gimmicky to try and create added value. Don't get me wrong, this would be nice on an otherwise loud psu with no fan control and no silent operation mode at all but in the face of being quiet and having variable/silent fan operation it's a little redundant.

Idk, like I said it's nice to see they're trying to improve quality in some areas but they don't even come in under the competition. As of right now they're more expensive than the competition. Personally I'll stick with seasonic.


Jan 27, 2003
If any PSU companies read this, I want to see molex discontinued, at least on the non-modular power supplies. At least offer some power supplies that don't have molex.


Nov 29, 2010
It's good to see that all the caps a Japanese on this one. I like Corsair PSUs, but the last gen RM series had some lower quality caps in there, which quite frankly shouldn't have been in an enthusiast PSU at that price point in the first place.

How do you know? ;-)

You can also monitor voltages and load and calculate efficiency. So it does a lot more than what you're stating.
Wasn't cwt who manufactures all the cx/vs/rm lines? When's the last time seasonic has had a tier3 unit? Comparing a company with a track record of weak designs over one that's been at the top of the game for well over a decade leads to doubts. When the proven track record comes out cheaper, it doesn't take a masters degree to sort that one out.

Granted it's not a complete list and the tiers are rough tiers but many based on the rec's of jg and others. Someone else's list, so it's not cherry picked either. The th psu tier list.

27 tier 1 units
14 are seasonic, 3 are cwt.

36 tier 2 units
11 seasonic, 2 cwt.

43 tier 3 units, no seasonics fell into this low of a category while 10 of them were cwt.

15 tier 4 units, barely a grade above "don't bother" populated by 4 cwt's. Again, seasonic had nothing this low quality.

It's not to say cwt can't come up with something quality but lets face it, their track record speaks for themselves. Personally I wouldn't use a corsair unit if it was given to me. That's just my personal opinion, they're not reliable enough. It's not even as if comparing this latest unit and saying well look, at it's price and performance there's nothing like it. The article started to say that until it was mentioned that there are two others at least that easily compete at performance and price.

It's halfway through 2015 and the big sales pitch is 'using all japanese caps' - something quality motherboards and psu's have been doing for years. In that sense I'm glad they're finally catching up, but the ss750km was released in 2013 I think it was? I'm not hateful towards them or anything but anyone can see cwt's quality has definitely been all over the place. There's no reason for it, the cost savings isn't even there. I can't recall how many times I've seen potential system builds going with a lower tier unit likely not using all japanese caps that was cwt based and at the same price they can easily get a higher quality ss or evga (superflower) unit. I won't say it directly impacts the performance of this model psu, that would be foolish - but cwt/corsair hasn't exactly had a track record of quality/value either.

You mention that the software allows other features that makes it 'do more' like calculate efficiency. So if I calculate it and see how efficient it is - and it's not efficient, is there a setting to increase efficiency? That's what I mean by 'gimmicky' features. A hardware monitor can check things like cpu temp, something that can be problematic and/or resolved by a higher fan profile, something the user can alter and react to. Efficiency is hardwired into a psu, so if efficiency isn't ideal then what? It's already supposed to be tested and proven 'efficient' from the factory and through tests done on the units to confirm the specs. I can monitor voltages of any psu with programs a lot of people already use to check thermals, things like hwinfo64. It might allow someone to see if a unit is going defective, though I think that goes back to quality. Users haven't had to worry about quality units for years, decades, all the sudden it's an issue.

I suppose on their lower tier models it would be helpful if it had a warning indicator before it blew and took other components with it. :)

You've probably seen my participation in that "TH PSU TIER LIST" thread. My opinion is that whole list is bullshit, therefore your reference to it is also bullshit. Pardon my bluntness. I tried to update it with a list of my own, but it's too much of a time suck and, quite frankly, if you have to refer to a tier list instead of a proper PSU review, you're just lazy.

Seasonic is a smaller factory that doesn't do any custom platforms. Asking them for any kind of customization of existing product is like asking someone to cut off their own penis. The good news for the consumer is consistency. The bad news for the industry is a lack of differentiation between competitor's products and you end up commoditizing your own product because you can only really compete on price. CWT has many lines of varying capability. If you have a good engineering team, you can have them build whatever you want... or you can just pick something from their existing line up (which is generally "lower tier" like VS and CX). But they've also done very good work, like Corsair's HX, Thermaltake's Toughpower Grand Platinum or Toughpower DPS Platinum and Titanium., Lepa's Platinum 1700W or the entire Enermax product line (using Enermax's own designs) for the past two or three years. I can guarantee you that none of those products come off the same like as a Corsair VS or CX.

Personally, I'm not sure where Seasonic got such a "high tier". Sure, they're consistantly good quality and almost always get good reviews, but their RMA rate certainly isn't any lower than most of the other CM's out there. For example: Corsair's AX line is still made by Seasonic... and the RMA rate is just as high on them than any of Corsair's CWT built products.

Consequential data, so I suggest taking it in with a grain of salt, but you might find this interesting:

Based on what data? I guarantee that if you were to poll Amazon, Newegg, Scan, etc., they're not going to tell you that Corsair is RMA'd more than any other major, enthusiast brand.

Corsair usually uses Japanese caps (look at the old VX or TX), but there was a different product manager for CS and RM and he didn't specify and AVL for caps. AX, AXi, RMi, HXi... all use Japanese caps. VS, CS and CX don't, but they compete at a lower price, entry level price point that Seasonic doesn't even try to compete in. An SS750KM is NOT in the same price range as a CX, so at least compare apples to apples.

Of course you can't "adjust efficiency". I suspect you're joking... But efficiency can change with different loads. But aside from seeing your PSU's efficiency at different loads..... You have no interest in knowing how much power you're using when idle? When gaming? When benchmarking? I see that as another level of hardware enthusiast; one that likes to be educated in actual power usage/requirements of particular PC hardware and then, in turn, can educate others on precisely how much power they may need.

And you must be joking about monitoring voltages with programs like AIDA, HW Info, etc. It's a KNOWN ISSUE how inaccurate those programs are. Not because of bad programming, but because of how and where they get their data. They're not getting it from the PSU itself. I have a Skylake build down in the lab that HW Info tells me has +7.4V on the +12V. I have a DMM shoved into one of the spare PCIe cables and it's reading +12.3V. Really freakin' accurate there. And certainly no third party software is going to tell you the actual load on any of those rails, never mind total wattage used.

You have your opinions, I have mine. If you can't be bothered to make a list of your own you're not that dedicated quite frankly. Talk about lazy lmao. Yes I was joking about adjusting for efficiency, it's gimmicky crud to add to their problematic software. How many times have people reported in the forums 'help, my corsair link isn't working right, this doesn't work, that doesn't work'. Inventions of solutions to problems which don't exist. If I really want to know my mb voltages I'll check using a voltmeter at the voltage checkpoints on my mobo. Why does my mobo have those? Because I didn't settle for cheap junk. Trying to convince people that software from corsair is anymore accurate than other 3rd party software you listed as junk is laughable. Anyone knows if you want accurate info you use a voltmeter, period. You should know that right? That defeats your own embracing of the included 'value' and reiterates my point that it's in fact a 'gimmick'.

No I really don't care to know as an average user how much power I'm using. As an enthusiast, there are tools for that. Again, hardware trumps software so the right tool for the right job. Not some playskool cruddy software gimmick. The typical pc even when gaming draws as much as maybe 2-3, maybe 4 lightbulbs. I've had kitchen chandeliers that drew more power. If someone's penny pinching their power bill that hard, a gaming pc is the least of their concerns. Of course as before if for other purposes, testing and so forth it's time to break out the tools and put up the toys.

I didn't say the ss750km was in the same range/tier/quality or comparison to a cx, I specifically said regarding this rm750. Way to read though. I see you're just as lazy reading posts as you are about your take on other things. And if you'd happened to pay attention to the article, it said the ss was priced higher when in fact the gigantic link to newegg said $10 cheaper.

Here's a screeny in case you missed it. I've seen how thorough you are. ;)

If you're trying to tell someone how much power they may need based on bundled software with a single psu at a certain efficiency rating by a single manufacturer, you've already missed the boat. Extensive testing and experience will answer that better, not "my gold rated psu with proven efficiency at this watt rating with this component is this, so there's your answer". Just like trying to tell someone what overclock to use because "my cpu" does x ghz at this or that. Is there mobo the same? Their chip identical, with same oc potential? Is their psu the same? Ram? Bios version? Probably not. Can I get them in the ballpark? Sure. Can someone who's read up on it and seen real world results from various users get them within the same ballpark? Yep.

To each their own, no one is going to change the opinion of the other but I think you'd be hard pressed to find fault in seasonic's track record for being one of the top industry standards to which others are compared. There's a reason for that. Corsair does have a few decent models but they're few and far between showing they have great inconsistency. Obviously they use different oems and don't actually create their own, but they choose what to put their label on so it's no excuse. Each company is responsible for the reputation they earn, good, bad or otherwise. No one forced them to flood the market with mediocre hardware.

I'm happy for them they have an additional piece of hardware to add to the selective few that can be called 'quality' in their lineup. As with any company, they can eventually overcome their past history but it's a cut-throat business with a lot of competion, takes awhile to get the bad taste out of ppl's mouths and they've no one to blame but themselves.
Yep. No one here is going to change the other person's opinion. I can see that.

You can call me lazy, but you can see what work I did try to do for a tier list if you bothered to read towards the end of that thread. And it is kind of ironic that you call the person that started a whole website dedicated to reviewing PSUs "lazy" just because he doesn't want to categorize a PSU as "good", "kind of good" and "not quite the best" simply based on the OEM.

Then again, just the fact that you think quoting someone as saying "I have a Gold PSU blah blah blah" tells me that you have ZERO idea of who I am. But that's ok. My ego is in check. 80 Plus isn't an indication of quality and they only test efficiency at 20%, 50% and 100% and I have plenty of people asking "what about at 5 to 10% where my computer is running 90% of the time. "Can someone who's read up on it and seen real world results from various users get them within the same ballpark?" For hardware like CPUs and GPUs, sure. But end users don't have the kind of equipment required to accurately measure power consumption, efficiency, etc. So your example holds no water. Thanks to reviewers like Aris and Jeremy @ and the rest, we can answer those questions... but you can choose to live in ignorance because that is your choice.

I'll just leave this here so people can see what I have to say and what you have to say and I'll be perfectly satisfied with that.
You'll sit there and bash a list and call people lazy then admit you can't be bothered to do so yourself. So yea, lazy. If you'd read what I said, I didn't say anything about 80 plus being quality. I said that simply using the included software packaged with this psu gives the home user no relevant point of reference to suggest power usage to others as you stated it might. That merely having an (for instance) 80 plus gold xyz model paired with (fill in the blank specific setup, mobo, cpu, gpu etc) means beans. That won't tell someone how efficient to expect their psu to run unless it's the same model with the same hardware. At best it will get them ballpark which other references will do. Having a readout doesn't make it exact science for the purposes of informing others as you suggested - "one that likes to be educated in actual power usage/requirements of particular PC hardware and then, in turn, can educate others on precisely how much power they may need." I believe that's what you stated.

Either one of two scenarios, either you can argue that using the rm750i with included software to monitor its efficiency at load is exactly equatable to whichever psu others may have and there's no tolerance differences between different 750w psu models based on design and components used. In which case anyone can inform said user based off a chart that says 'this gpu need xyz psu watt rating' and leave it a blanket statement. Or you can argue that from one manufacturer to another there are differences, handle power delivery better/worse - aka if a user has an antec or if they have a cooler master unit comparing those to the readings off the rm750i. How does that work? It has severe flaws regardless of the scenario.

The ONLY time it would be of any relevance is if someone were running a similar system to yours (cpu power draw was similar, same amount of ssd or hdd, same gpu) - but then of course if they had the same power supply they could just look at their own software meter. So again, the case for the 'value' of the software is a moot point. Using the psu with the software as a reference point to advise others, how does that work when you have a 750w unit and they have a 550? When you have a gtx 980 and they have an r9 285? When their psu is a bronze rated (doesn't correlate to a gold rated in efficiency), is a lesser quality unit that may struggle to push the full 550w it's rated for and begins to flake out at 375w or 400w load. It makes no sense, anymore than trying to tell someone what overclock settings to use on their amd fx based off my i5 oc settings.

It had nothing to do with your ego until you made it about that so that's on you. If you misread what I clearly said, again that's on you. The fact you spun off yapping about 80 plus not being a designation of quality says you were out to lunch when you read my post as I've never said any such thing. You argue what about the person asking 'what's the efficiency of my pc running at 5-10%' - how would they even begin to know this? In order to know your pc was running at 10% power consumption you'd have to know the rigs max power draw under load to begin with and the value of the draw that it's 'usually' running at (for them, since this is about solving the riddle that applies to their specific concerns). Only once you have those two values can you determine what percentage you're running at. So for the end user who doesn't have that equipment as most don't, they wouldn't even be asking an accurate question in the first place.

Let's pretend I complained to my mechanic my car was running at only 80% of its fuel efficiency. Even if I knew the rated fuel economy which is handily posted on the sale sticker for that model (which solves one of the unknowns, in the case of the psu the user wouldn't even have that much since each individual system has slightly different load draw) - how would I know what my current fuel economy is? If I lacked the knowledge to assert my current actual fuel economy, the mechanic's first question would be "well how can you know you're running at xyz percentage of its rating?" Or is it pure guesstimating in which case you could spout out any figure in the 'ballpark'?

I don't see any evidence for someone to be concerned with something as trivial as power efficiency when they don't even know what their load factor is. They might ask 'how efficient is my power supply the majority of the time I'm using it', but if you don't know what that percentage is I bet they don't. And if you claim to know I call b.s. You don't know how often their system is idling, being used for gaming, what they consider 'moderate' tasks so there's no way to even claim that their 'avg' usage is 5-10% load. My system stays on 24/7 so its average isn't the same as someone who fires up their rig strictly to game for 3 out of 4hrs then shuts it off. It's all variable with too many puzzle pieces missing to suggest any amount of accuracy, best case scenario again is random ballpark. You and your reviewers can test the exact same model psu as a user has, but unless you know what their average usage is without making baseless assumptions that it's some arbitrary 5-10%, it matters little. You give them an efficiency rating for 5-10% and come to found out their avg is 20%, 80% of the time. Well there went the test results.

I can grab make believe figures out of thin air and attempt to give them credibility too, doesn't make me a guru. Just shows I'm good at making assumptions. If you have people coming to you for answers with false figures (uneducated or flawed because they lack the testing abilities as you say), saying well what about xyz percentage use xyz of the time - how do you expect to give them a meaningful answer? You might provide an answer that applies to the question but without consideration that the question was flawed. I tell someone I weigh 150lbs, what size shirt do I need? They tell me a medium, so I order a medium - oops, I'm 200lbs. Shirt doesn't fit. Same principle. Am I saying it's your fault if you're given faulty numbers or assumptions to work with? No. If you don't factor that inherent error into the overall equation and think you're supplying people with meaningful data based on a highly likely flawed data set, then yea you're not helping them much. Satisfying an answer to 'a scenario', not 'their scenario'. Isn't that really what 80 plus does to begin with? Highlights a few basic load points and gives a rough idea of efficiency at those loads?

I have nothing else to say, I know who you are and hopefully you and your ego can forgive me for not being impressed. My condolences to both of you. Trust me when I say this, you're definitely the right 'tool' for the job. lmao.


Mar 23, 2010
I've got a RM750i in both my computers, and I've got no complaints. Everything about the RMi series oozes of quality, right down to the soft cables.

I found it particularly interesting to see exactly how much juice my computers use during idle/stress. My CPU's are 4790K and 3770K, which are the top CPU's of their generations (of the LGA 1150/1155). Apart from the motherboards, CPU's and graphics cards, both systems are identical.

The 4790K system's power consumption sits at 50-55W idle and around 300W while I'm gaming, so 750W is obviously overkill. This is something I've always been curious about for years, and now I finally get to see the actual numbers, not to mention the conversion efficiency.

The 3770K system sits at about 100W idle and I can't remember how much it was while gaming. Considering the fact that these systems are only one generation apart, it's impressive to see the idle power consumption nearly cut in half on the 4790K.
The gpu power consumption would make a difference both idle and under load, but in terms of the general difference between the two cpus power consumption is easy to locate.,11.html

The much lower power consumption of the 4790k system at idle was published over a year ago. Similar figures are available depending on the gpu in question and can help people determine which power supply they need before they purchase it. It might be fun to see the figures if you opted to buy the rm750i but these figures are something people should consider prior to purchasing a power supply to fit their needs, not after it's bought and installed. Both were widely available and known with a 2min google search with no need for bundled software alongside the psu. Motherboard, cpu and gpu (especially cpu/gpu) will have the biggest factor on power consumption differences. By contrast, hdd's and fans take very little power. Those two factors are enough to make the two systems compared nothing alike really unless the different gpus happen to have the same power draw.

It would be surprising if the software reported figures widely different from system load tests already performed using those components. In other words, it shouldn't really come as a surprise, instead confirming what was already known to be true. If people enjoy or get a kick out of confirming existing data there's nothing wrong with that but I don't think it qualifies it as a major tool or anything.
Not open for further replies.