Will Gigabyte Radeon HD 6850 run on 400W PSU?

Aug 6, 2011
Post New Reply
  1. Hello. I have bough a 400w psu a month ago. I believe that I have made a mistake buying such a low wattage psu. Anyway. I really need to upgrade my Graphics card. My question is will Gigabyte Radeon HD 6850 run on my 400w psu?

    CPU: Intel E7500 2.92Ghz
    Motherboard: Intel DG33BU
    RAM: 2x2gb Kingston 800
    Fans: 2 case fans 80mm and 120mm
    HDD: 2x 7,200RPM
    Graphics Card: MSI ATI Radeon 2600 HD Pro 256mb

    My psu specs are here:

    Thank you.
  2. Leeky

    Leeky TS Evangelist Posts: 3,797   +116

    I think you should be OK as long as the CPU is not overclocked and therefore drawing more power.
  3. hellokitty[hk]

    hellokitty[hk] Hello, nice to meet you! Posts: 3,435   +145

    AFAIK, that brand is rather cheap.
    Maybe, you could try.
  4. rockyjohns

    rockyjohns TS Rookie Posts: 80

    I would try it. While it is not a PSU I would select, you already have it and this review shows a system with an HD 6850 running at 260w peak load, which provides adequate reserve, and your PSU has two 12v legs with 18a each and APCF.

    I selected the review because it is for the GTX 560 (note without the Ti suffiix which is a faster and more expensive card) which you also might want to consider.

    Edit: Opps - left out the link

    MIXEDD TS Member Topic Starter Posts: 38

    Thanks for replies. rockyjohns I can't understand what have you wrote. Can you explain a little bit more further please?
  6. Mizzou

    Mizzou TS Enthusiast Posts: 823

    The HD 6850 power requirements are fairly modest so it might work with your current power supply. As Leeky noted, you wouldn't want to overclock anything in your system as this would quickly put you over the top.

    MIXEDD TS Member Topic Starter Posts: 38

    Thank you for your help. I do not want to overclock anything. :) I think I will buy this card in a few days maybe and I'll have to wait a few days to it arrive. :) Thanks again everyone!
  8. Mizzou

    Mizzou TS Enthusiast Posts: 823

    Let us know how it works out :)
  9. rockyjohns

    rockyjohns TS Rookie Posts: 80

    Sorry I left out the link in my post above (but now have edited it in). Here it is:

    The benchmarks in the article show a test system with HD 6850 requiring 260 watts at peak load - see the second chart on the page linked above. So a good 400w PSU should provide ample power.

    The power coming out of the PSU to your components is split into 3 different levels - 3.3v, 5v, and 12v. The graphics card, and the CPU, use 12v power. In the "older" days, video cards did not need much power and most of the power was allocated to the 5v circuit. But newer or better PSUs provide a higher allocation to the 12v circuit because of the increased requirements for video card. So besides considering total power, when upgrading grading video cards, one must also evaluate if it provides enough power for the 12 volt circuit. Your PSU provides good power allocation to the 12v cicuit.

    For more information explaining power requirements for video cards, and other helpful information about upgrading, check out this site:

    On the link above, the PSU part is about half way down the page.

    I said I would not buy the brand of PSU you have as there are better brands with better quality builds including better components and longer warranties. One item many low priced brands leave out is Active Power Factor Correction - which I and many consider a basic protection in any PSU. Your has that.

    So while I would buy a different PSU if I were building a new system, I would not pay the added cost if I already had yours since it does provide ample total power, has a good allocation to the 12v circuit, and has Active PFC.
  10. hellokitty[hk]

    hellokitty[hk] Hello, nice to meet you! Posts: 3,435   +145

    Someone do clarify, but that seems like a ridiculous way of judging your power needs.
  11. MIXEDD

    MIXEDD TS Member Topic Starter Posts: 38

    I have ordered the card yesterday so it should come after 3-4 days. I'll let you know guys how i'ts working. :) Thanks for replies.
  12. rockyjohns

    rockyjohns TS Rookie Posts: 80

    Are you saying that using real, actual results from a reputable benchmark review showing total requirements for a system under peak load is a ridiculous way? If so,. why?
  13. MIXEDD

    MIXEDD TS Member Topic Starter Posts: 38

    Good people. I have got the graphics card. IT'S AWESOME! All games work at max settings. The card works very silent and it's not overheating. I believe that everything is fine. Thank you! :)
  14. Mizzou

    Mizzou TS Enthusiast Posts: 823

    Good news, glad to hear that it worked out for you :)
  15. rockyjohns

    rockyjohns TS Rookie Posts: 80

    That was a quick 3 to 4 days Mixedd - glad it was shorter rather than longer.
  16. hellokitty[hk]

    hellokitty[hk] Hello, nice to meet you! Posts: 3,435   +145

    Huh? Oh, not sure really. I don't know.
    Most of those 260w are taken from the 12v rail though, so you'd have to go measure which power supplies will be able to handle that load from their 12v rail only, since that's usually not going to be given. That'd be difficult. You might as well be normal and not think like that.
  17. rockyjohns

    rockyjohns TS Rookie Posts: 80

    This is the most intelligent statement you have made. At least you got one thing right.

    As I reported before, his PSU has two 12v legs with 18a each. The power supply chart for the PSU shows it has 360w of 12v power. Gee that is more than the 260 watt TOTAL SYSTEM USAGE AT PEAK POWER on the benchmark. Took all of about 10 seconds to click on the link provided by MIXED to find that. For most PSUs it is usually NOT DIFFICULT to find.

    And there you go embarrassing yourself, and violating board rules, with another childish personal attack.
  18. Leeky

    Leeky TS Evangelist Posts: 3,797   +116


    Your method assumes the test system is identical to the OP's system. That in my honest opinion is a bad way of calculating things. Sure its fine if your just paying attention to the total peak power consumption of a GPU(s) in Watts, or its peak current in Amps, but its not a good way of calculating the whole power consumption system wide.

    I find a much more effective way is find out the peak W/A readings of a GPU, and then factor in the overhead for the system, plus a percentage of "headroom".

    Without a dedicated test kit, or hardware reviews identical to the system your trying to calculate, I draw on the knowledge learnt here and elsewhere online.

    It is also worth noting that cheap PSU's can also deliver very poor power under load. It is also often said that the cheaper PSU's peak wattage rates are conservative at best, and really not reliable when pushed hard either.

    We all have slightly different ways of coming to the same conclusions. Thats not to say one is worse than the other (assuming the same end result is achieved) - they're just different.

    I don't see his response being any worse than the second sentence of your last reply, and your further comments to be honest.
  19. hellokitty[hk]

    hellokitty[hk] Hello, nice to meet you! Posts: 3,435   +145

    I think it's too much to depend on manufacturers providing accurate specifications'd really have to go out and test it yourself.
    It's not a childisn attack; I've never seen such a way of guesstimating power before :confused:.
    My apologies.
  20. rockyjohns

    rockyjohns TS Rookie Posts: 80

    No it only assumes that the Op's system's power requirements will not exceed it. Since test system is configured with high end components to ensure none will bottleneck the card being tested, they will usually consume more power than the average system If Op's includes additional power hungry components they need to be identified, But this is no different than your assumption of adding in some standard overhad. And I too allow a factor for headroom.

    I don't understand your comment. Please explain what you mean. I am looking at the peak power consumption of the whole system and reporting peak load usage for the whole system.

    Systems don't have to be "identical". I already answered that above but I think relying on actual computations made by experts is more reliable than anecdotal guestimates by one or two users.

    I fully agree which is why I never recommend cheap PSUs and already stated that I would not buy the one OP has.

    What did you mean above then when you said "it's not a good way of calculating the whole power consumption system wide"

    The difference is he called my method "ridiculous" which it clearly is not. He even admitted after the charge that he could not say why it was ridiculous. But that did not stop him from turning right around and in the same post making another personal attack and saying that I should try being "normal". Two lame and bogus personal attacks I chose to respond - with facts. His statement "Huh? Oh, not sure really. I don't know." was the most intelligent thing he wrote. And it is a fact that he was embarrassing himself, and violating board rules, with another childish personal attack. Frankly I am surprised that you never condemned his abuse and instead chose to attack the victim when I choose to defend myself.
  21. Leeky

    Leeky TS Evangelist Posts: 3,797   +116

    It makes more sense when explained like that, for sure. Like I said, we all go about it in different ways, and as long as the end result is the same (or at the very least correct) its fine.

    Thats kind of answered by your comments in the first quote in your reply to me.

    But for the sake of clarity; I did not feel the way you calculated the load using a test system was representative of the OP's system.

    Not they don't, but using a test system is not representative of the OP's system. I agree with what your saying though, but there are plenty of places online I use to check GPU load, like here for instance.

    The Corsair unit is in the same category as the OP's original choice, if not worse. I'm being critical of your component choice, not making a personal insult. That is well within the forum rules.

    Well the test system will differ from another users computer. For example, the number of discs, lighting, fans, etc.

    The most accurate way without question is to attach it to a power meter and find out the wattage and current draw. This however is beyond the scope of the average user.

    For two simple reasons:

    1. You retaliated in this and another thread. Your both back and forth between each other, and to be perfectly frank with you, you are responding back at him, making you both as bad as each other. If you feel something is offensive report it, then a member of the team can deal with it appropriately.
    2. While HK's choice of words were questionable, he was not personal insulting you in a manner I deem to be agressive, or personally insulting. It definitely could have been worded better, but he wasn't swearing, and I saw no derogatory comments made of an offensive nature. This however is as always, open to personal interpretation, and if you feel it warrants it, report the posts and I'll let another member of the team deal with it.

    HK has also apologised for offending you, so I consider it dealt with.

    Lets be very clear; I did not attack you. I mearly gave my opinion based on what I read, based on my experiences. The fact it appears one-sided is unintentional, assuming you even feel that's the case? I assume that is how you view it given the fact you think I'm "attacking" you.

    I simply gave my opinions on your comments which is again, well within forum rules. If you don't like it, so be it, but to suggest I am attacking you is far fetched at best. I'm allowed to give my opinion just like any other member as long as I adhere to the forum rules.
  22. hellokitty[hk]

    hellokitty[hk] Hello, nice to meet you! Posts: 3,435   +145 bad :rolleyes:
  23. rockyjohns

    rockyjohns TS Rookie Posts: 80

    Yes let us be very clear. He did not simply disagree with what I had suggested, he said it was "ridiculous". When I responded very civilly and asked him to explain, he could not explain or support his comment, but instead said "You might as well be normal and not think like that." A clear personal insult and abusive statement.

    I can't see how you can say telling someone that they "might as well be normal" is not "personally insulting" or of a "derogatory nature".

    The first time, the "ridiculous" comment, I choose to let it go and responded with objective reason, simply asking him to explain. Since he had no defense or justification for his statement, he responded with another attack. The second time, when he made the "normal" comment, chose not to continue being a silent victim of his bogus attacks. And since reason did not work, I chose to respond in the language he understood.

    I think there is a major difference between someone who initiates a totally unprovoked attack and someone who is defending themself from it, and who, having tried reason, speaks the language of the initiator to get his message accross.

    You did not criticize either of his statements, but chose to attack mine - and you did attack when you said my comments were as bad as his. You said "I don't see his response being any worse than the second sentence of your last reply, and your further comments to be honest."

    The situations were totally different. He wrongly labeled my good suggestion as "ridiculous" and then made it even more personal by saying, after I had only asked him to explain the first comment, that I "might as well be normal" - esstentially stating that I was not normal. Clearly neither of those were intelligent statements - thus my comment in response to his attacks. Yes I was beiing abusive too at that point - which I felt justified in responding to hm. I am surprised you cannot see the difference.

    And my further comment that you object to was "And there you go embarrassing yourself, and violating board rules, with another childish personal attack." That was a very accurate statement and justified to point out to him how he was behaving.

    I do not wish to belabor the point - but apparently my first explanation did not seem to be understood. You may have a right to express your opinion, but that does not mean it is not an attack when you say I behaved like he did.

    And just to be clear, I never objected to your disagreeing with me on he technical issues regarding the OP's system , nor HK either, I have only taken offense at his abusive words and your then, after never objecting to his words, taking me to task for defending myself.
  24. Leeky

    Leeky TS Evangelist Posts: 3,797   +116

    You should have reported the posts you were unhappy with, and allowed us to deal with it. Once you've responded it makes it much more difficult to intervene.

    I stand by my original comments, you were retaliating back, so if you really want to push this, you'll both need punishing for it. Personally I'm far happy just leaving it as it is. HK apologised to you for his actions, so I consider that settled. Whether you choose to accept it or not is none of my concern.

    I'm really not going to get drawn into an argument because your unhappy with the way I worded my comments to you. Your not going to get an apology from me for giving my opinion, or for suggesting you were acting just as bad in some of your replies. Those are facts that everyone can see, and given that you did retaliate, pushing this will only mean that both of you are punished.

    Had you followed forum rules and reported the post the outcome might have been different, but you chose not to, so we are were we are now because of it.

    If you have anything else to say on the matter take it to PM, we have interfered with this thread and taken it off topic enough already.

Similar Topics

Add New Comment

You need to be a member to leave a comment. Join thousands of tech enthusiasts and participate.
TechSpot Account You may also...