Question Worried about my PSU

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Nov 23, 2022
43
2
35
Hello, i currently have an FSP Hydro G 750w PSU..

im worried about it becouse i bought it used.. and im not sure about the PSU cables.. becouse it came with an EVGA box and lots of cables.. i’ve used it for about 4 months without any problems.. no wierd sounds no nothing.. i have no clue about the past of the PSU..


should i change it?
My only option would be an NZXT C650W.. the thing im worried about here is that it may not be enough for my system.. i would like to know if 650w is okay for my system..

i also experience a lot of GPU coil whine.

here are the specs.

AMD Ryzen 7 5800x Stock.
AMD Radeon RX 6650 XT
ASrock B550 Pro4
AQIRYS 240mm AIO
2 Sata SSD’s (Samsung 870 500gb, 256 Gigabyte UD PRO)
2 Nvme’s (Samsung 980 500gb, kingston NV2)
About 6 RGB Fans.
2 Monitors
Logitech Keyboard and Mouse.
Razer Microphone.
Bluetooth adaptor..
And other small stuff..

And should i change it?
Or just keep the FSP Hydro G PsU?
 
Yeah, Memtest86 is a good starting point to rule out MOST physical or configuration problems with memory, but it's not foolproof. I've seen ten or fifteen cases where the memory passed Memtest86 with flying colors and they still had issues and replacement of the memory resolved it and that was BEFORE Ryzen even existed as a platform. So you have that PLUS the fact that Ryzen really simply does not like some memory. Corsair Vengeance sticks are a prime example and a lot of it is simply due to the fact that it is a more budget oriented series and Corsair tends to make unannounced changes to the ICs that get used from one production run to the next. I've seen a lot of issues with Ryzen platforms using Kingston/Hyper X Fury kits, and just Kingston memory in general much the same as with the Corsair Vengeance kits.

Add to that the fact that you have two separate kits, and it makes absolutely no difference if they are the same kit model or not, and you effectively double the chances of having problems with it.

For a little more expanded explanation of why that is, please read section two titled "Mixed memory (Or the odd man out)" at the following link and the single post at the second link.



https://forums.tomshardware.com/threads/amd-ram-compatibility.3210050/#post-19785792
 
I bought a used 800 watt power supply two years ago. It is still working without any problems. I don't see the point in replacing it if it works properly. Especially if you have been using it for 4 months without any problems, there is most likely nothing wrong with it or the cables.
 
I bought a used 800 watt power supply two years ago. It is still working without any problems. I don't see the point in replacing it if it works properly. Especially if you have been using it for 4 months without any problems, there is most likely nothing wrong with it or the cables.
That is 100% not even remotely accurate. Not in any way, shape or form. Besides which, your sample of one literally has nothing whatsoever to do with another persons experience or outcome. Even if you both had the same power supply model, which you don't, and even if they were both brand new out of the box, you still would be inclined to have entirely different experiences either in the short or long term. Either unit could fail, either prematurely or eventually, due to age, while the other continues working fine. It's no different than with people. You can take any two people, feed them the same diet, same exercise, same environmental conditions, etc., and one gets cancer while the other is perfectly healthy.

It also matters IMMENSELY what the condition of the hardware it's being used with is like AND in fact WHAT hardware is being used with it, not to mention the specific model of the power supply in question. "800 watt power supply" literally means nothing, at all, in any conversation where you are discussing the merits of any given power supply. Neither does it's "80 plus" certification. The ONLY thing that matters is the internal quality and component selection (Assuming the unit has sufficient capacity and can actually sustain that capacity, which is a big assumption with any unit that isn't specifically known to be of at least moderately good quality) which can only be determined by those of us on this end of things by looking at units that have been reviewed with a complete teardown and associated testing by a reputable reviewer using appropriate equipment. Or, to some degree, also units that are known to use the same platform as another unit that HAS been tested and reviewed in such a way, although even units with the same platform that are different models might have wildly different performance results and some internal components might be of different quality, so it's usually best to only rely on actual reviews of a specific model.

The idea that "because it worked for 4 months" has some bearing on whether or not it might be faulty, is, though, is the part that bothers me the most. The fact that a thing was working, in NO way, has anything to do AT ALL with whether it might or might not be the problem. In fact, in 99% of cases, a thing was ALWAYS working fine, before it wasn't, so that is just completely irrelevant to any conversation of whether a thing might be to blame or not.

I will however concede that if the cables themselves were the problem, THAT you'd have known about immediately. Likely there would have been magic smoke in abundance if the cables were not compatible, for any power supply. And the point of replacing a power supply when you are having problems is that if you've got a good reason to suspect the unit, or if you've eliminated everything else as being probable, what you are generally left with is the power supply and motherboard and while it's often a toss up which of those it might be, you can pretty well hedge your bets towards the winning guess if one of them is significantly older than the other. Either way it's a roll of the dice but when you don't know the history of the power supply but you DO know that it's been in service for a fairly long time, it's almost always a no brainer to lean that direction if for no other reason than because it's a high probability type component AND has significant miles on it.

Never make assumptions when it comes to power supplies. A PSU will break you of those assumption making tendencies more often than not. You've been lucky. Most are not.