EVGA SuperNOVA 650 G3 650W Modular 80+ Gold PSU £89.99 - cclonline
247°

EVGA SuperNOVA 650 G3 650W Modular 80+ Gold PSU £89.99 - cclonline

27
27 Comments

Hot+ Still £100 on amazon

Very hot. Thanks.

What warranted the price jump from the G2 to G3? Mine was less than £70, is this really £20 better? If you're not using custom cables, may as well get a semi modular GQ. Even the 850W is less than this.

This is generally not worth the money over a PSU half the price, you won't notice any difference in use.

Danze19841 h, 30 m ago

What warranted the price jump from the G2 to G3? Mine was less than £70, …What warranted the price jump from the G2 to G3? Mine was less than £70, is this really £20 better? If you're not using custom cables, may as well get a semi modular GQ. Even the 850W is less than this.


The G3 is marginally more efficient and slightly smaller. Just pick whichever is cheapest at the time of purchase.

And, in response to the last comment, do not buy a cheap PSU unless you only use your PC for browsing, etc, and don't mind if it fails in a few years. Good PSUs are expensive because they hold up under heavy loads and are usually covered by a long warranty.

Original Poster

Danze19841 h, 38 m ago

What warranted the price jump from the G2 to G3? Mine was less than £70, …What warranted the price jump from the G2 to G3? Mine was less than £70, is this really £20 better? If you're not using custom cables, may as well get a semi modular GQ. Even the 850W is less than this.


G2 is £85 now

Nice find Op. Heat.

Recently replaced a 6 months old Corsair 750W PSU with one of these because the PC kept rebooting when playing DOOM or running 3DMark (but worked 100% ok doing anything else)... popped in the EVGA and everything has been rock solid ever since.

So yeah, cheap-o PSUs aren't always super great...

EndlessWaves1 h, 49 m ago

This is generally not worth the money over a PSU half the price, you won't …This is generally not worth the money over a PSU half the price, you won't notice any difference in use.


No. Just no.

You won't notice any difference...until you do. Unfortunately, the difference you finally notice could well be a cheapo PSU frying something much more expensive in your PC.

Your PSU is the one thing you shouldn't cheap out on. A bad PSU can quite easily damage other parts in your PC, whilst a good PSU will last for many years and is one of the most transferable parts between PC builds.

I'm not saying this specific PSU is the one people should buy, but saying that you won't notice a difference between any high quality PSU and a cheapo one is misleading nonsense.

matt10110129 m ago

saying that you won't notice a difference between any high quality PSU and …saying that you won't notice a difference between any high quality PSU and a cheapo one is misleading nonsense.


Prove it. Statistics on failures, tests on the protection functions, whatever. I've seen absolutely no evidence to believe that a £45 PSU is any more likely to cause problems than a £90 one.

I'm not saying buy the absolute cheapest PSU you can find from a dodgy seller on ebay, I'm saying that this is a luxury product with little benefit over the mainstream.

It's like cars. If you searched the world for the cheapest car then you would find it was pretty poor in a crash test. That doesn't automatically mean you should buy the most expensive car possible and if you look at crash test results you'll find once you get past £10-15K the differences between models no longer correspond to how much you spend.

Decent price thanks OP! I need to replace my ageing corsair PSU & I'm torn between the EVGA & Seasonic Focus Plus units...

EndlessWaves2 h, 50 m ago

This is generally not worth the money over a PSU half the price, you won't …This is generally not worth the money over a PSU half the price, you won't notice any difference in use.



How about you link a '45 quid PSU' that you think is good and we can see just how many complaints it has, statistics, etc.

Nate149213 m ago

How about you link a '45 quid PSU' that you think is good and we can see …How about you link a '45 quid PSU' that you think is good and we can see just how many complaints it has, statistics, etc.


Yoyotech currently have the XFX XT 600W for £40
yoyotech.co.uk/com…htm

I bought a fully modular EVGA 650GS 80+ Gold for around £65 a few years ago. Does anyone know the reason why the price has gone up for modular PSUs?

(People actually giving a crap about cable management nowadays etc?????)

All tech has gone up a fair old bit after Brexit. Also - another vote for the "don't cheap out on PSUs" from me. Some of the branded bronze units have a place, but are often much more aggressive on the fan speed. I'd rather pay the extra little bit for a good gold unit with nice fans that shut off at low loads.
Edited by: "mikedigitales" 27th Nov

Excellent PSU, well worth it over cheap alteratives IMO. Great deal.

EndlessWaves5 h, 44 m ago

Yoyotech currently have the XFX XT 600W for …Yoyotech currently have the XFX XT 600W for £40https://www.yoyotech.co.uk/components/power-supplies/600w-xfx-xt-series-psu-bronze-wired.htm



orionpsudb.com/new…red


So, not only is this a non-modular PSU, it's a 600W PSU that is grossly mislabeled *AND* eats 5% more power (again, AT least!)


So, you're buying, what, is ostensibly a 550W PSU, non-modular PSU with lower power ratings.


If you run anywhere near capacity on that PSU, let's say 400W max draw, you are losing 5% of that compared to the Gold rated PSUs. That's 20W just disappearing into the air.


That's actually a lot of money over 3 years, so you pay less at the till, but when you fill it up, it chugs the energy.

Let's be generous and say you run it at 200W at 6 hours a day, 5 days a week, for 3 years.


That's 9 quid a year. Times 3 years, 27 quid.



So, you'r 40 quid PSU, used LIGHTLY, costs 67 quid very quickly.


Ramp that figure up, say you are on your computer 6 hours a day, but it's mostly gaming or heavy work and lets say 7 days a week.

That's over double the yearly cost. Putting this 40 quid PSU at 100 quid operating cost (Energy+Unit) over 3 years.


A good PSU can last you 10 years. Just do the math and see how much you can save with your usage.


Not to mention... It's a GOOD PSU!

Nate149236 m ago

http://www.orionpsudb.com/news/the-mystery-of-xfx-budget-xt-line-solved-oem-and-platform-discoveredSo, not only is this a non-modular PSU, it's a 600W PSU that is grossly mislabeled *AND* eats 5% more power (again, AT least!)


That article was clearly written right after the first english language batch launched on the US market, do you have any evidence that any of the mis-labelled units were actually sold in the UK?

A quick check on XFX's website shows correct label, so clearly it's not 'grossly mislabelled' right now.
32586095-xFlR1.jpg
Nate149236 m ago

.If you run anywhere near capacity on that PSU, let's say 400W max draw, ….If you run anywhere near capacity on that PSU, let's say 400W max draw, you are losing 5% of that compared to the Gold rated PSUs. That's 20W just disappearing into the air.That's actually a lot of money over 3 years, so you pay less at the till, but when you fill it up, it chugs the energy.Let's be generous and say you run it at 200W at 6 hours a day, 5 days a week, for 3 years.That's 9 quid a year. Times 3 years, 27 quid.So, you'r 40 quid PSU, used LIGHTLY, costs 67 quid very quickly.


The difference between bronze and gold certification is 4%, not 5%. 92% efficient for Gold, 88% for Bronze at 50% load. At 200W computer draw that's 227W vs. 217W mains draw if we use the 50% figures (there are no 33% ones).

10W for 6 hours a day, 5 days a week is 10w * 6h * 5d * 52w = 15,600Wh a year. At around 12p per thousand watts for a decent electricity deal that's £1.87 a year extra over a £90 PSU. I'm not sure how you got £9.

200W is the typical i7 & GTX 1070ti going full pelt in a demanding game, and 6 hours a day of that certainly can't be considered light use for the majority of people.
tomshardware.com/rev…tml


And yes, modular is nice for the final stage of the build, but makes no difference otherwise.


So your saying people should pay £50 extra for £10 worth of electricity savings and something that's going to be noticeable for 0.2% percent of the time you own the product (3 days building every 5 years of ownership).


Yes, there will be niches where a PSU like this makes sense but for most people it's just wasted money.

Nate149256 m ago

http://www.orionpsudb.com/news/the-mystery-of-xfx-budget-xt-line-solved-oem-and-platform-discoveredSo, not only is this a non-modular PSU, it's a 600W PSU that is grossly mislabeled *AND* eats 5% more power (again, AT least!)So, you're buying, what, is ostensibly a 550W PSU, non-modular PSU with lower power ratings.If you run anywhere near capacity on that PSU, let's say 400W max draw, you are losing 5% of that compared to the Gold rated PSUs. That's 20W just disappearing into the air.That's actually a lot of money over 3 years, so you pay less at the till, but when you fill it up, it chugs the energy.Let's be generous and say you run it at 200W at 6 hours a day, 5 days a week, for 3 years.That's 9 quid a year. Times 3 years, 27 quid.So, you'r 40 quid PSU, used LIGHTLY, costs 67 quid very quickly.Ramp that figure up, say you are on your computer 6 hours a day, but it's mostly gaming or heavy work and lets say 7 days a week.That's over double the yearly cost. Putting this 40 quid PSU at 100 quid operating cost (Energy+Unit) over 3 years.A good PSU can last you 10 years. Just do the math and see how much you can save with your usage.Not to mention... It's a GOOD PSU!


Couldn't have said it any better! Anyone who thinks an unbranded, half the price psu is just as good or reliable as a branded, gold rated psu's are delusional. When cheapo psu fries under heavy load it will most likely take the mobo and graphics card with it!

Each to their own but I know I'd rather pay extra for a decent psu that will last me 5+ years and won't blow up on demanding load. And I won't even bother going into better fan quality, Nippon capacitors and a relaxed fan profile for the silent builds.

EndlessWaves11 h, 25 m ago

That article was clearly written right after the first english language …That article was clearly written right after the first english language batch launched on the US market, do you have any evidence that any of the mis-labelled units were actually sold in the UK?A quick check on XFX's website shows correct label, so clearly it's not 'grossly mislabelled' right now.[Image] The difference between bronze and gold certification is 4%, not 5%. 92% efficient for Gold, 88% for Bronze at 50% load. At 200W computer draw that's 227W vs. 217W mains draw if we use the 50% figures (there are no 33% ones). 10W for 6 hours a day, 5 days a week is 10w * 6h * 5d * 52w = 15,600Wh a year. At around 12p per thousand watts for a decent electricity deal that's £1.87 a year extra over a £90 PSU. I'm not sure how you got £9.200W is the typical i7 & GTX 1070ti going full pelt in a demanding game, and 6 hours a day of that certainly can't be considered light use for the majority of people. http://www.tomshardware.com/reviews/nvidia-geforce-gtx-1070-ti-8gb,5311-16.htmlAnd yes, modular is nice for the final stage of the build, but makes no difference otherwise.So your saying people should pay £50 extra for £10 worth of electricity savings and something that's going to be noticeable for 0.2% percent of the time you own the product (3 days building every 5 years of ownership). Yes, there will be niches where a PSU like this makes sense but for most people it's just wasted money.


1) The difference between Bronze and Gold is as simple as you suggest.

en.wikipedia.org/wik…lus

115v I N-R is 85% and 90%. 230v IR is 85% and 92%. 203v INR is 88% and 92%.

Add them up and average, it is 5.3% better. Using 5% is fine.

Your math is simply wrong.

Feel free to try a tool to assist in your calculation.

ukpower.co.uk/too…ity

I did use a more reasonable price per electricity based on the national average in the UK.


blog.comparemysolar.co.uk/ele…es/

But even if you use .12, you get more than your figure. You are missing some number in your calc.

Anyway...

Nate14921 h, 7 m ago

1) The difference between Bronze and Gold is as simple as you …1) The difference between Bronze and Gold is as simple as you suggest.https://en.wikipedia.org/wiki/80_Plus115v I N-R is 85% and 90%. 230v IR is 85% and 92%. 203v INR is 88% and 92%.Add them up and average, it is 5.3% better. Using 5% is fine.


Why would I be interested in the efficient in a 110v or industrial context though? In this discussion only 230v EU non-redundant is what matters as that's what is used by everyone reading this.
Nate14921 h, 7 m ago

.Your math is simply wrong.Feel free to try a tool to assist in your ….Your math is simply wrong.Feel free to try a tool to assist in your calculation.https://www.ukpower.co.uk/tools/running_costs_electricityI did use a more reasonable price per electricity based on the national average in the UK.http://blog.comparemysolar.co.uk/electricity-price-per-kwh-comparison-of-big-six-energy-companies/But even if you use .12, you get more than your figure. You are missing some number in your calc.Anyway...


Care to point out where it's wrong?

I get the same results with that calculator. The average month is 30.4 days long, and using it 5 days out of every 7 is 21.7 days. 6 hours a day is 130 hours a month. Putting in 130 hours a month on a 10W difference yields an average monthly cost of 17p, or £2.04 a year. Very similar to my original figure and nowhere near £9.

I'm assuming the cheapest package here, this is HUKD so again in this context it's likely most people are close to that. The average only adds 25p per year.

EndlessWaves32 m ago

Why would I be interested in the efficient in a 110v or industrial context …Why would I be interested in the efficient in a 110v or industrial context though? In this discussion only 230v EU non-redundant is what matters as that's what is used by everyone reading this. Care to point out where it's wrong?I get the same results with that calculator. The average month is 30.4 days long, and using it 5 days out of every 7 is 21.7 days. 6 hours a day is 130 hours a month. Putting in 130 hours a month on a 10W difference yields an average monthly cost of 17p, or £2.04 a year. Very similar to my original figure and nowhere near £9.I'm assuming the cheapest package here, this is HUKD so again in this context it's likely most people are close to that. The average only adds 25p per year.



Perhaps my numbers are wrong then, I will look more later!

One other thing to consider is that running PSUs at 50% is 'peak efficiency' as well, so if you need to run 400W of power, the best PSU to do so is an 800W. So running this (essentially 550W) PSU would at 400W would lead to worse power efficiency. Runninga 650W PSU at 400W would be a little better, probably just a % or two.

Nate149242 m ago

Perhaps my numbers are wrong then, I will look more later!One other thing …Perhaps my numbers are wrong then, I will look more later!One other thing to consider is that running PSUs at 50% is 'peak efficiency' as well, so if you need to run 400W of power, the best PSU to do so is an 800W. So running this (essentially 550W) PSU would at 400W would lead to worse power efficiency. Runninga 650W PSU at 400W would be a little better, probably just a % or two.


Peak efficiency varies depending on the design and can be anywhere between 50% and 80%. Most PSUs on the market have very similar efficiency at 50% and 60%, which would mean a 650W PSU was just as efficient as an 800W one in that example.

Given there's often a big drop off in efficiency at lower power draws a 650W PSU would often be more efficient overall than an 800W PSU. Certainly in percentage terms, actual kwh would depend on the relative models and use of the computer.

But as mentioned above, 400W is a huge power draw, most systems are well below that.

EndlessWaves2 h, 14 m ago

Peak efficiency varies depending on the design and can be anywhere between …Peak efficiency varies depending on the design and can be anywhere between 50% and 80%. Most PSUs on the market have very similar efficiency at 50% and 60%, which would mean a 650W PSU was just as efficient as an 800W one in that example.Given there's often a big drop off in efficiency at lower power draws a 650W PSU would often be more efficient overall than an 800W PSU. Certainly in percentage terms, actual kwh would depend on the relative models and use of the computer.But as mentioned above, 400W is a huge power draw, most systems are well below that.



Not really at all.

400W power draw is quite common actually, and to suggest a 650W PSU would 'often be more efficient' than an 800W PSU is generically wrong.

jonnyguru.com/mod…500

jonnyguru.com/mod…494


Have a look at Jonny's site, he does a TON of testing and you can see that almost every single 'model equivalent' PSU has the higher W models performing better.

Maybe this XFX is an ok PSU, but the generic advice of 'just get the cheaper PSU' is so far from good advice I would say it borders on dangerous.


Apart from the efficiency aspects, simply have more stable 12/5/3.3v rails makes your hardware last longer.

Buying a 'good' PSU means they often include 5-7-10 year warranties and they actually last for multiple builds, while a 'bad' or 'cheap' PSU has 2-3 year Warranties and isn't built to last.

The XFX from above is a 3 year Warranty. The EVGA SuperNOVA from this post is 7 years.

That by itself should be enough to convince a potential shopper.

Also, modular PSUs are *absolutely awesome* for disassembly and adjustment as you can leave all your tiddied cables in place while you remove the PSU.

Nate14921 h, 4 m ago

Maybe this XFX is an ok PSU, but the generic advice of 'just get the …Maybe this XFX is an ok PSU, but the generic advice of 'just get the cheaper PSU' is so far from good advice I would say it borders on dangerous.


Again, evidence is needed that a mainstream PSU is in any way dangerous or more likely to cause problems. Nobody's even attempted to provide it yet.
Nate14921 h, 4 m ago

Apart from the efficiency aspects, simply have more stable 12/5/3.3v rails …Apart from the efficiency aspects, simply have more stable 12/5/3.3v rails makes your hardware last longer.Buying a 'good' PSU means they often include 5-7-10 year warranties and they actually last for multiple builds, while a 'bad' or 'cheap' PSU has 2-3 year Warranties and isn't built to last.The XFX from above is a 3 year Warranty.


It's a nice idea, but it's never worked out in practice for me.

My first PSU purchase was an Antec 350W that died near the end of the first system's life. The next system had a Tagan 480W that is still going but lacks PCI-E connectors. So I needed a new PSU for the system after that, which was an 82% efficient Enermax 525W. That one survived an (indirect) lighting strike but was no good for my current PC because it was too large for my new case.

In over fifteen years I've never been able to reuse a PSU. That sort of gamble may be worth a fiver, but I wouldn't call the odds particularly favourable.

EndlessWaves29 m ago

Again, evidence is needed that a mainstream PSU is in any way dangerous or …Again, evidence is needed that a mainstream PSU is in any way dangerous or more likely to cause problems. Nobody's even attempted to provide it yet. It's a nice idea, but it's never worked out in practice for me.My first PSU purchase was an Antec 350W that died near the end of the first system's life. The next system had a Tagan 480W that is still going but lacks PCI-E connectors. So I needed a new PSU for the system after that, which was an 82% efficient Enermax 525W. That one survived an (indirect) lighting strike but was no good for my current PC because it was too large for my new case.In over fifteen years I've never been able to reuse a PSU. That sort of gamble may be worth a fiver, but I wouldn't call the odds particularly favourable.



I mean, do you believe those were good PSUs? Those, to me, sound like very run of the mill PSUs followed by one that got hit by lightning.

Was it under Warranty? Because you could get it warrantied, you know, like this one.

Also, are you seriously saying you don't believe that cheaper, mainstream, PSUs have a higher chance of failure?

Do yourself the favor and google it, like, just for a few minutes and check just how many horror stories abound.

There is so much information, I can't even begin to parse it for you.

I'm not saying the XFX XT is cheap ( I think it's a rebranded Seasonic, which is a pretty decent make)... But the generic advice of 'cheap is fine' is not good advice.

Nate149235 m ago

I mean, do you believe those were good PSUs? Those, to me, sound like very …I mean, do you believe those were good PSUs? Those, to me, sound like very run of the mill PSUs followed by one that got hit by lightning.Was it under Warranty? Because you could get it warrantied, you know, like this one.Also, are you seriously saying you don't believe that cheaper, mainstream, PSUs have a higher chance of failure?Do yourself the favor and google it, like, just for a few minutes and check just how many horror stories abound.There is so much information, I can't even begin to parse it for you.I'm not saying the XFX XT is cheap ( I think it's a rebranded Seasonic, which is a pretty decent make)... But the generic advice of 'cheap is fine' is not good advice.


Of course there are more failure reports, they sell in greater numbers. They're probably not equally durable, but I do believe they're sufficiently durable that the vast majority will replace them due to obsolescence rather than failure.

As for the Antec 350W This was back in the days before anyone had heard of seasonic, corsair, EVGA or anyone else who is popular today. Every power supply was a grey box and the top of the range 9800Pro graphics card I had at the time drew about 60W. I doubt it was still under warranty but I can't even remember which model it was.

Again, if you're building an ultra-reliable system then maybe it's worth it, but this is aimed at the market that's overclocking CPUs, using non-ECC memory and other stuff that'll likely impact reliability far more.

EndlessWaves9 h, 42 m ago

Of course there are more failure reports, they sell in greater numbers. …Of course there are more failure reports, they sell in greater numbers. They're probably not equally durable, but I do believe they're sufficiently durable that the vast majority will replace them due to obsolescence rather than failure.As for the Antec 350W This was back in the days before anyone had heard of seasonic, corsair, EVGA or anyone else who is popular today. Every power supply was a grey box and the top of the range 9800Pro graphics card I had at the time drew about 60W. I doubt it was still under warranty but I can't even remember which model it was.Again, if you're building an ultra-reliable system then maybe it's worth it, but this is aimed at the market that's overclocking CPUs, using non-ECC memory and other stuff that'll likely impact reliability far more.


I think you are greatly misrepresenting PSUs with a lack of proper knowledge, basing it off your limited experience of your own PSUs and a refusal to consider the benefits of higher quality units.

This isn't about 'ultra-reliable' systems.

I'll tell you what, you can recommend, to your friends, that they cheap out on the PSU. Built 20+ PCs for people. See how things work out in the medium to long run.

I can tell you from a sample of around 20-30 PCs constructed, 100+ PCs serviced, and countless discussions on top of those without first hand experience... The PSU is the worst part of the PC to save 20-30 bucks on.

I also can't believe you would suggest a budget PSU for someone who is looking to Overclock a system. That's actually even a worse situation to be in.

When you are pushing the limits of your hardware, unstable rails are the *worst* thing you can subject your motherboard/CPU/GPU to.
Post a comment
Avatar
@
    Text