1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Build Advice Should you buy a more efficient power supply?

Discussion in 'Hardware' started by Gareth Halfacree, 24 Jan 2014.

  1. Gareth Halfacree

    Gareth Halfacree WIIGII! Staff Administrator Super Moderator Moderator

    Joined:
    4 Dec 2007
    Posts:
    12,187
    Likes Received:
    1,683
    I wrote this for a discussion in another sub-forum, but decided to double the reward for my effort by plonking it 'ere as advice for anyone who's ever wondered whether it's worth spending the extra cash on buying a more efficient PSU when speccing up a build.

    THE NUMBERS IN FULL
    Using several assumptions, I built three models: a gamer who plays two hours a day, a pro-gamer who plays eight hours a day, and a Folder or miner who has the system fully loaded 24 hours a day. Each of these users is building a new rig, the specifications of which are given below. In all three cases, it's a completely new rig: no existing parts are being used.

    THE RIG
    Radeon R9 290(X) TDP: 300W
    Intel Core i7 4770K TDP: 84W
    Motherboard, Fans and So Forth: 10W
    Non-Green Hard Drive: 8W active, 3W idle (based on a 2011 Ars Technica post)
    Total maximum system power draw: 402W.

    ASSUMPTIONS
    When gaming, the GPU is 100% loaded and the processor 60% loaded (two cores versus all four cores, plus overhead), while the hard drive is mostly idle for a total power draw of 363.4W rounded down to 363W for simplicity's sake.
    When participating in distributed computing projects like Folding@Home or Litecoin mining, both CPU and GPU are 100% loaded, while the hard drive is mostly idle for a total power draw of 398W.
    Electricity currently costs on average 15.32p per kilowatt hour (KWh), based on figures from the Energy Saving Trust. From the same page, generating each KWh of electricity causes 0.517kg of carbon dioxide to be emitted into the atmosphere.
    The cost of electricity is rising at 7 per cent annually, based on an average of the most recent price rises listed on USwitch.
    The PSUs in question have a five-year warranty, and thus five-year worst-case lifespan. All calculations, therefore, are based over a five-year period.
    The two PSUs under comparison are both 80 PLUS Titanuim rated, one at 450W and one at 900W. As a result, at the system's peak load the 450W offers 91 per cent efficiency, and the 900W offers 96 per cent efficiency. Buying the 900W PSU costs £50 more than the 450W PSU.

    With that in mind, let's run the numbers.

    The Gamer
    The gamer works in an office all day, during which time his or her PC at home is powered off. On average, the gamer manages to get in around two hours of gaming every day - some days there's no gaming at all, but on a weekend it might be an eight-hour marathon. At all other times, the computer is switched off or in an extremely low power mode.

    Result of Simulation:
    Over a five-year period, paying the extra £50 for the 900W PSU will have cost the user £37. In other words, this use-case makes no financial sense. Environmentally, however, the move will have reduced the environmental impact of the PC by preventing the emission of 7.84kg of carbon dioxide into the atmosphere.

    The Pro-Gamer
    The pro-gamer works at gaming all day. Eight hours a day, seven days a week he or she is hammering the system, honing skills and pwning the opposition. Outside the 'office hours,' the PC is switched off or in an extremely low power mode.

    Result of Simulation:
    Over a five-year period, paying the extra £50 for the 900W PSU will have saved the user £3. Not much, but it is a saving. Environmentally, the move will have reduced the environmental impact of the PC by preventing the emission of 31.36Kg of carbon dioxide into the atmosphere.

    The Folder
    This user has their system forming part of a distributed computing cluster. Perhaps they're running Folding@Home or BOINC for scientific research, or renting their system out as a renderfarm, cracking passwords and generating rainbow tables, or perhaps they're trying to mint the latest cryptocurrency. Whatever the reason, the system is at full load - CPU and GPU - all day, every day. Hey, on the plus side: at least their room is nice and warm.

    Result of Simulation:
    Over a five-year period, paying the extra £50 for the 900W PSU will have saved the user £126. Hey, that's enough to buy a replacement PSU! Environmentally, the move will have reduced the environmental impact of the PC by preventing the emission of 103.17Kg of carbon dioxide into the atmosphere.

    CONCLUSIONS
    If you only load your PC by a couple of hours a day, don't bother speccing it with a PSU capable of delivering double your wattage requirement. You'll never recoup your investment, and the environmental impact is minimal. If you're a pro-gamer, it could be worth doing - especially as you'll be able to claim the cash spent on the PSU as a business expense against tax, something I didn't take into account in my calculations. If you're a folder, though, absolutely go for maximum efficiency - it has a real-world environmental benefit and gives you the cash you'd need to replace said PSU once it's out of warranty. Win-win!
     
    Last edited: 24 Jan 2014
    Guest-16 likes this.
  2. Xaine

    Xaine Death... destroyer of brain cells

    Joined:
    19 Sep 2012
    Posts:
    371
    Likes Received:
    8
    Extremely helpful thank you. Axi860 it is!
     
  3. sandys

    sandys Well-Known Member

    Joined:
    26 Mar 2006
    Posts:
    3,050
    Likes Received:
    86
    Some of your assumptions are off a single case fan is 3-4w, 12v@~0.3A, you will probably have more than one, including the fan on your CPU cooler, a motherboard and RAM consumption does not come for free etc. would be about 30-40w there on a reasonably featured board, less on a uATX setup. Though I guess you could argue you wouldn't hit TDPs in some cases, in other case TDPs aren't max power.

    Intel TDP used to be based on typical usage and not the maximum the processor could pull in worst case scenario which was higher than TDP, don't know if that is still the case.

    A quick look an they still define it as below which would imply it is still not a max.

    Just looking at this site it would seem most gamers over clock so you can throw these TDPs out the window anyway :D
     
  4. Gareth Halfacree

    Gareth Halfacree WIIGII! Staff Administrator Super Moderator Moderator

    Joined:
    4 Dec 2007
    Posts:
    12,187
    Likes Received:
    1,683
    The original thread was suggesting I was mad for thinking spending £50 more on an efficient power supply would ever pay for itself, so I used absolute worst-case figures. Taking your increased figures into account, and thus increasing the overall system power draw in all three scenarios by 30W, you're looking at the following changes: the Gamer would lose £36 instead of £37; the Pro Gamer would save £8 instead of £3; and the Folder would save £139 instead of £126.
     
  5. Parge

    Parge the worst Super Moderator

    Joined:
    16 Jul 2010
    Posts:
    12,864
    Likes Received:
    552
    Great work. Very interesting indeed. Also, makes me feel better about having just bought a 450W SFX PSU for my rig (it is Gold Rated)
     
  6. Margo Baggins

    Margo Baggins I'm good at Soldering Super Moderator

    Joined:
    28 May 2010
    Posts:
    5,607
    Likes Received:
    242
    Makes me feel slightly happier about having a 1kw gold psu in my server at home that often is only pulling about 200w.
     
  7. GeorgeStorm

    GeorgeStorm Aggressive PC Builder

    Joined:
    16 Dec 2008
    Posts:
    6,050
    Likes Received:
    237
    I'm with Parge, my 450W Gold rated PSU may not be as bad when it comes to energy consumption/cost etc as I worried it was :)
     
  8. Gareth Halfacree

    Gareth Halfacree WIIGII! Staff Administrator Super Moderator Moderator

    Joined:
    4 Dec 2007
    Posts:
    12,187
    Likes Received:
    1,683
    80 PLUS Gold at 20% load requires 88% efficiency - which is the same as 100% load. In other words, you're running at exactly the same efficiency as if you'd bought a 200W PSU (except, of course, you've got headroom for peak draw being considerably higher than that.) For maximum efficiency without changing from 80 PLUS Gold, you should have a 400W PSU which would run at 92% efficiency for a 200W load. Of course, if you've got enough hardware in your server for instances where you'd exceed 400W, then that wouldn't be possible without releasing the magic smoke.
     
  9. AlienwareAndy

    AlienwareAndy New Member

    Joined:
    7 Dec 2009
    Posts:
    3,421
    Likes Received:
    70
    Good article Gareth.

    The whole power argument has long annoyed me tbh. People go around spouting so called figures yet don't understand the actual costs involved, which usually turn out to be pennies rather than pounds.
     
  10. Yslen

    Yslen Lord of the Twenty-Seventh Circle

    Joined:
    3 Mar 2010
    Posts:
    1,961
    Likes Received:
    48
    Wouldn't a more valid comparison be between two PSUs of equivalent power but different efficiency?

    For example, the £59 Corsair TX550M vs the £85 be quiet! BN198 Straight Power E9

    Both have roughly the same power capabilities, but the Corsair is 85% efficient and the be quiet! is 93.2%, on paper. That's a much bigger gap than the two PSUs you compare, but with half the price difference.

    Nice work, anyway, it made a good read.
     
  11. Gareth Halfacree

    Gareth Halfacree WIIGII! Staff Administrator Super Moderator Moderator

    Joined:
    4 Dec 2007
    Posts:
    12,187
    Likes Received:
    1,683
    No, because the original thread was specifically about the gains that come from running at 50% load compared to 100% load - and thus that was the question I was trying to answer. In the real world, anything that improves the efficiency of the power supply by 5% and costs less than £50 - whether it's switching to 900W from 450W, or Titanium from Bronze, or a combination of both - will result in the savings (or otherwise) shown.
     
  12. MSHunter

    MSHunter Well-Known Member

    Joined:
    24 Apr 2009
    Posts:
    2,461
    Likes Received:
    55
    Well I went to a 850 when 400 would have been enough for 2 reasons, size and noise. the model I got doesn't turn on fan under 50% load :D and it was one of the few PSUs that fit in my case.

    TBH: Mostly got a large PSU so no matter what I do in the future I do not need to buy a new one.
     
    Last edited: 27 Jan 2014
  13. Shirty

    Shirty Time travelling rogue Super Moderator

    Joined:
    18 Apr 1982
    Posts:
    12,019
    Likes Received:
    1,385
    It is a good write up Gareth. What I think is equally as interesting is how people calculate their loads prior to purchasing a PSU.

    For example, if I were using my rig for benching/mining/folding 24/7 I would want a PSU which was as efficient as possible at the load level. It wouldn't bother me what the percentage load on the PSU was, just the efficiency at said load.

    But I'm not that guy - I game a bit, casually. I still consider myself an enthusiast in the sense that I have a self-built reasonably priced ITX PC capable of HD gaming, but I'm definitely not hardcore.

    I have an in-line power meter with the extension powering my sig rig, monitor & speakers plugged in. The exact contents of my rig which can draw power from the socket are as follows:
    • Dell U2311H
    • Creative T6060 5.1 Speakers
    • Logitech G500
    • Corsair K70
    • Bitfenix Recon
    • Corsair SP120 x 1
    • Corsair AF140 x 1
    • 2 x 1TB 7200rpm HDD
    • 1 x SSD
    • EVGA GTX770 SC
    • Asrock Z77E-ITX
    • 16GB Corsair Vengeance (2 x 8GB)
    • Corsair H55 (wearing the SP120 above)
    • i5 2500K (usually at 4.2 but happy up to 5GHz)
    These are all plugged in to a Corsair CX500M - not the greatest PSU in history by any means, but it cost me £40 new with 3 years warranty, and it's more than suitable for my needs.

    Now here's the thing - as I said above, I don't bench, fold or mine. I play games at max settings for a few hours a day on average. I have never managed to get my rig to pull more than 330W from the wall socket when gaming.

    Case in point, when I run the Metro: Last Light benchmark at the settings detailed in this thread I get an average of 55fps for a total draw of 321W, although in reality it varies between 200 - 310W. For the whole rig, screen and speakers, playing one of the most graphically intensive games there is at basically max settings at 1080p.

    If I turn my speakers (at a sensible volume) and monitor off I lose around 45W from that total bringing me down to 276W from the wall. Time for fun facts:
    • The PSU fan is spinning at ~800rpm at this load (inaudible to all intents and purposes)
    • The current-gen CX500M is around 87% efficient at 50% load, based on a couple of reviews I've seen
    • I'm pulling 276W from the wall to supply 240W to my rig, playing a maxed out graphically demanding game
    • The same workload using an SF-1000F-14MP would pull around 262W to produce the same amount of power

    The bottom line for me was that whilst I could have spent more money for extra warranty and slight efficiency savings, the reality is that it would have been money wasted based on my usage patterns. When my warranty expires in 3 years I could buy another and still have change from what an AX760 would have cost me. I have also concluded that for a single card gaming rig, anything over 500W is likely to be overkill. I'm sure that's the general consensus anyway, but I do see an awful lot of 700-1000W PSUs in signatures here and elsewhere which are completely wasted based upon the other components listed. Not that I'm questioning people's choices at all, but the less knowledgeable system builder might well believe the hype that more is better.

    Relevant article.

    TL;DR I was jealous of Gareth's wall of text so I wrote one too.
     
  14. Gareth Halfacree

    Gareth Halfacree WIIGII! Staff Administrator Super Moderator Moderator

    Joined:
    4 Dec 2007
    Posts:
    12,187
    Likes Received:
    1,683
    \o/
     
  15. Shirty

    Shirty Time travelling rogue Super Moderator

    Joined:
    18 Apr 1982
    Posts:
    12,019
    Likes Received:
    1,385
    ส็็็็็็็็็็็็็็็็็็็_(ツ)_ส้้้้้้้้้้้้้้้้้้้้

    <3 UNICODE
     
  16. Margo Baggins

    Margo Baggins I'm good at Soldering Super Moderator

    Joined:
    28 May 2010
    Posts:
    5,607
    Likes Received:
    242
    I agree with that - the only reason I have a 1kw psu in my server is because I went down to 650 on my main rig which left me with a psu spare. My server is mostly cobbled together out of old bits so it's the nature of the beast. 650 on my main rig is probably still overkill - but i saw it on the marketplace here for what I thought was a bargain price, as is the driving force behind most of my purchases. I rarely ever buy anything new, and I buy things if I see them for a good price, kind of dictates my upgrades and the frequency of them ;)
     
  17. Shirty

    Shirty Time travelling rogue Super Moderator

    Joined:
    18 Apr 1982
    Posts:
    12,019
    Likes Received:
    1,385
    Oh fo' sho' - if you get a deal on a big PSU then it's a no-brainer!
     
  18. Yslen

    Yslen Lord of the Twenty-Seventh Circle

    Joined:
    3 Mar 2010
    Posts:
    1,961
    Likes Received:
    48
    Okay, fair enough!
     
  19. true_gamer

    true_gamer Well-Known Member

    Joined:
    4 May 2009
    Posts:
    6,357
    Likes Received:
    987
    Great read Gareth! ;)
     
  20. Yadda

    Yadda Well-Known Member

    Joined:
    25 Jul 2003
    Posts:
    3,217
    Likes Received:
    49
    Interesting read, good stuff. Unfortunately I, like many tend to overspec the PSU in my builds to "leave some headroom" for future potentially power-hungry upgrades, multiple graphics cards or even to last the next couple of complete system upgrades.

    At least that was my thinking when I bought my PSU (in sig) nearly 5 years ago - at the time I had a single 5770 and a C2D not C2Q. I think I may have misjudged things slightly. :D
     

Share This Page