1. Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

NVIDIA GeForce GTX 480 1536 MB Fermi

Discussion in 'Reviews' started by W1zzard, Mar 19, 2010.

  1. pantherx12

    pantherx12 New Member

    Joined:
    Jan 2, 2009
    Messages:
    9,714 (4.16/day)
    Thanks Received:
    1,699
    Location:
    ENGLAND-LAND-LAND
    Why would google go off diggs man?

    having a high digg rating means its high up in digg results XD


    Google is based on traffic and frequency of search terms etc.
     
    Last edited: Mar 29, 2010
  2. Robbaz

    Robbaz New Member

    Joined:
    Jul 12, 2006
    Messages:
    28 (0.01/day)
    Thanks Received:
    4
    Location:
    Norrland, Sweden.
    SNiiPE_DoGG says thanks.
  3. hertz9753

    hertz9753

    Joined:
    Sep 5, 2009
    Messages:
    1,100 (0.53/day)
    Thanks Received:
    484
    Location:
    Bismarck, ND
  4. @RaXxaa@ New Member

    Joined:
    Jun 29, 2009
    Messages:
    473 (0.22/day)
    Thanks Received:
    20
    Location:
    Pakistan & US
    Six months later and it can barely out perform a 5870 and not a 5890.
    The way its meant to be!
     
  5. OnBoard

    OnBoard New Member

    Joined:
    Sep 16, 2006
    Messages:
    3,044 (0.96/day)
    Thanks Received:
    379
    Location:
    Finland
    You sure went on about how to cool this. Really easy, Accelero Xtreme GTX 480 does it with ease. It will be the same as GTX 280, but with moved contact point. Maybe they'll add a pipe just to be cool.

    45C idle and 75C load my prediction with Accelero :p

    edit: forgot, not all the 320W has to be cooled, just core. That number has memory+voltage regulation parts in it.
     
    Last edited: Mar 29, 2010
  6. Delta6326

    Delta6326

    Joined:
    May 21, 2008
    Messages:
    3,874 (1.51/day)
    Thanks Received:
    682
    Location:
    Iowa, USA
  7. pantherx12

    pantherx12 New Member

    Joined:
    Jan 2, 2009
    Messages:
    9,714 (4.16/day)
    Thanks Received:
    1,699
    Location:
    ENGLAND-LAND-LAND
    Probably needs a lot of video ram.
    That and it being a Nvidia optimised game wouldn't help XD
     
  8. btarunr

    btarunr Editor & Senior Moderator Staff Member

    Joined:
    Oct 9, 2007
    Messages:
    29,766 (10.67/day)
    Thanks Received:
    14,221
    Location:
    Hyderabad, India
    You have great difficulty in understanding simple things. Given, hypothetically this cooler would ace on Radeon HD 5870, giving better cooling (although in reality it won't even fit), but that doesn't make it the "best cooler", or even a good one at that. If it were the best cooler, it would at least work "good" on the GPU it is designed to cool. Since the GPU is ending up >90C for even basic gaming, this cooler isn't doing the job it's meant for. Hence it is not the "best cooler" by a long shot.

    Don't use the "bigger/smaller" argument, we're talking "best" here. So don't throw semantics at me. Size and weight won't count in the argument. Radeon HD 5970's cooler is able to cool way more components (combined TDP higher), and ending up with lower temperatures. Hence it's better than GTX 480's cooler, if not "the best" cooler.

    Even otherwise I'm suspecting that Fermi actually weighs higher than Radeon HD 5970, by the looks of it. I'm searching for a review with such weight measurements. That would nail your "but it's bigger" argument.
     
  9. imperialreign

    imperialreign New Member

    Joined:
    Jul 19, 2007
    Messages:
    7,043 (2.45/day)
    Thanks Received:
    909
    Location:
    Sector ZZ₉ Plural Z Alpha
    Regarding the 4870x2, I can attest to that . . . I have two of those beasts spooning in my setup, all 4 GPUs OCed, and none of the 4 GPUs stay that hot under short- moderate workloads . . . only if they're working their tales off for a couple of hours straight (only gaming) do they settle in at 90C+

    But, regarding their coolers, the 4870x2 setup is actually pathetic and still cools better, IMHO - just for comparison:

    [​IMG]

    (inside the fan shroud, with fan removed)
    [​IMG]


    . . . and it somehow manages to keep two RV700s, 2GB of GDDR5, the PCIE chip, and all other nuclear PCB components happy . . .


    What do I see wrong with the 480s cooler? It appears to be aluminum fins + HP - copper would've been a more efficient choice . . . although heavier, copper is more willing to give up heat than aluminum is. Also, the annodized coating negatively affects cooling efficiency as well . . . is it a major impact? Not typically; but when dealing with components that become so hot so quickly, and dealing with an item designed specifically to cool an extremelly hot component that must operate in an extremelly enclosed space, plus a rather poor choice of metal to begin with, I'm sure it's probably doing more harm than good. The cooler could also make do with more fins, too . . .

    Either way, it really looks more like the cooler was designed more for show than go . . . it looks pretty and all, but it's simply not cut out for the work it must endure.
     
    TheMailMan78 and DaedalusHelios say thanks.
  10. pantherx12

    pantherx12 New Member

    Joined:
    Jan 2, 2009
    Messages:
    9,714 (4.16/day)
    Thanks Received:
    1,699
    Location:
    ENGLAND-LAND-LAND
    God BTA its like you can't even read, or read what you want to read.

    when did I say it was the "best cooler"

    Or did I say it was probably the "best stock cooler"


    Thought I'd point something very important out, I am only disagree with you implying ( and then later confirming) that you think the cooler is BAD. That is all.
    The cooler is actually good for a stock cooler and if it was fitted to any other cards people would be ranting and raving. That would make it infact a GOOD stock cooler.
    See what I mean now?


    Also BTA it looks like the 480 uses 55mm mounting holes to me, thus the cooler would fit on a 5870 ;)

    Or hell switch it round and stick a 5870 cooler on the 480 and watch the card fry.


    Regardless of how right you think you are, you are wrong.

    As by your logic, metric shit tons of after market coolers are now bad coolers thanks to Fermi : [


    Also Bta for the last time, the 480 gpu by its self outputs MASSIVE amounts of thermal energy, easily nearly as much as everything the 5970 has to cool. ( but with less room for cooling!)

    God damn what a round a bout!





    Non-nonsensical order to this post sorry folks! I've not slept XD


    Oh by the way, guy above me copper is less willing to give up heat. thought I'd clear that up. Its Brilliant at soaking up heat though : ]

    also

    "they're working their tales off for a couple of hours straight (only gaming) do they settle in at 90C+"

    You realise to get temperatures in reviews they often use bench marking programs that really stress the gpu right? thus its under 100% load constantly.
     
    Last edited: Mar 29, 2010
  11. btarunr

    btarunr Editor & Senior Moderator Staff Member

    Joined:
    Oct 9, 2007
    Messages:
    29,766 (10.67/day)
    Thanks Received:
    14,221
    Location:
    Hyderabad, India
    Of course we're talking about stock coolers, isn't that a given? Otherwise wouldn't I pull out some Arctic Cooling Accelero Xtreme numbers? Smell the coffee.
     
  12. imperialreign

    imperialreign New Member

    Joined:
    Jul 19, 2007
    Messages:
    7,043 (2.45/day)
    Thanks Received:
    909
    Location:
    Sector ZZ₉ Plural Z Alpha

    No.


    Copper is more thermally conductive, and therefore dissipates heat better over longer periods of time from the component it's trying to cool.

    What makes it generally confusing is that if you were to remove an aluminum or copper HS from a component, alum would return to room temp faster than copper - copper is a denser metal, so it will retain more heat during cool-down . . .

    Also because of it's density, copper it able to absorb more heat from the component, whereas aluminum can't - if you have sufficient airflow across the cooler, copper will outperform aluminum every time.

    Dissipation of heat is simply the opposite of absorption . . . that's where the fan comes in, to help with this process. The copper is dissipating more heat than alum, as it's absorbing more heat than alum.

    They could've gone with a hybrid style cooler (copper base/HP + aluminum fins), which would give better cooling efficiency than the current setup, while not being as heavy as a full-blown copper cooler . . . but copper still reguarly out-performs the vast majority of hybrid coolers as well.
     
  13. pantherx12

    pantherx12 New Member

    Joined:
    Jan 2, 2009
    Messages:
    9,714 (4.16/day)
    Thanks Received:
    1,699
    Location:
    ENGLAND-LAND-LAND
    Not really given the discussion we've been having :laugh:


    Anyway fact of the matter is its not a bad GPU cooler.

    Stick it on any other card ( and it would fit) and it would perform better then what ever stock cooling is already on.

    ( single gpu)




    Also poster above me, mind if I reply in the morning I'm tired as hell ha ha

    To tired to run around finding info etc.

    But for a quick example check out reviews of the true 120 ex and the copper version. The difference would be greater if copper could dissipate heat as fast as it could absorb heat.
     
  14. btarunr

    btarunr Editor & Senior Moderator Staff Member

    Joined:
    Oct 9, 2007
    Messages:
    29,766 (10.67/day)
    Thanks Received:
    14,221
    Location:
    Hyderabad, India
    No it's a given, use commonsense. If I were talking about "the best" cooler, I'd be talking about Kingpin LN2 evaporators. I was talking about stock coolers all this while, it's like you're the one who can't read or make things out in contexts.
     
  15. pantherx12

    pantherx12 New Member

    Joined:
    Jan 2, 2009
    Messages:
    9,714 (4.16/day)
    Thanks Received:
    1,699
    Location:
    ENGLAND-LAND-LAND


    Common sense goes out the window when someone disagrees with science and logic :laugh:
     
  16. btarunr

    btarunr Editor & Senior Moderator Staff Member

    Joined:
    Oct 9, 2007
    Messages:
    29,766 (10.67/day)
    Thanks Received:
    14,221
    Location:
    Hyderabad, India
    And yours is critically flawed.
     
  17. pantherx12

    pantherx12 New Member

    Joined:
    Jan 2, 2009
    Messages:
    9,714 (4.16/day)
    Thanks Received:
    1,699
    Location:
    ENGLAND-LAND-LAND

    Not by a long shot.

    In fact my reasoning and logic have been described as " off the scale" before now.

    :laugh:


    Seriously stop trying to insult me and get back on topic.

    If this cooler was on a 5870 would it be considered a good stock cooler? ( a copper plate would be added to the bottom of the cooler to simulate a IHS)
     
  18. imperialreign

    imperialreign New Member

    Joined:
    Jul 19, 2007
    Messages:
    7,043 (2.45/day)
    Thanks Received:
    909
    Location:
    Sector ZZ₉ Plural Z Alpha


    You can replay in the morn, no problem . . . but it still won't change the fact that copper is a better cooling solution than aluminum.

    It's simply one of the basiscs of physics - energy in must equal energy out.


    :toast:
     
  19. pantherx12

    pantherx12 New Member

    Joined:
    Jan 2, 2009
    Messages:
    9,714 (4.16/day)
    Thanks Received:
    1,699
    Location:
    ENGLAND-LAND-LAND

    I'm not debating that at all, how ever the rate at which something looses is thermal energy is important.

    I need Alex p :laugh: He's better at explaining it then me, but check out all the thermal properties of copper you'll find some interesting numbers : ]
     
  20. DaedalusHelios

    DaedalusHelios

    Joined:
    Feb 21, 2008
    Messages:
    4,965 (1.87/day)
    Thanks Received:
    826
    Location:
    Greensboro, NC, USA
    Dear pantherx12 and btarunr,

    I regret to inform you both that you have been in disagreement through a
    simple miscommunication. Btarunr has been saying the cooler is bad because
    it barely handles its intended application. Pantherx12 you are judging the
    cooler by itself and not by the single action of cooling Fermi.

    PS. You are both right but arguing different things.
     
    dadi_oh and El_Mayo say thanks.
  21. pantherx12

    pantherx12 New Member

    Joined:
    Jan 2, 2009
    Messages:
    9,714 (4.16/day)
    Thanks Received:
    1,699
    Location:
    ENGLAND-LAND-LAND

    I figured that out quite a few posts back :laugh:

    Cheers though, hopefully it clears things up for other people. :toast:
     
  22. btarunr

    btarunr Editor & Senior Moderator Staff Member

    Joined:
    Oct 9, 2007
    Messages:
    29,766 (10.67/day)
    Thanks Received:
    14,221
    Location:
    Hyderabad, India
    Leave others to judge that, trolls judge themselves. There are simple arguments presented to you.
    • There is a stock cooler (Radeon HD 5970) that handles way more components, and keeps them cooler. Doesn't that make it better than the one on the GTX 480?
    • Who cares if it's longer/bigger/wider? We're talking about which is better, and the one on the HD 5970 is handling a higher thermal load, and ending up cooler.
    • A GeForce GTX 480 cooler won't fit a Radeon HD 5870, just as the HD 5970 cooler won't fit it, so we're left to hypothesize, and the HD 5970 will do a better job since it's on half the thermal load.
    • A good VGA cooler keeps temperatures down at acceptable levels, the one on the GTX 480 doesn't. It's worse than the HD 5970's cooler in this aspect. The immediate repercussions of GPUs running at 90C falls on overclocking potential. For example, look at EVGA's product lineup. Look at how the out of the box overclocking is unusually low by NVIDIA's/EVGA's standards, and how for a mere 7% higher clock speeds (FTW variant), EVGA immediately jumps into water-cooling (there's no air-cooled FTW variant).

    Among us, you're the one who started with "You're highlighting how little you know about cooling here BTA", and later "can't you read". If I took that as an attempt to insult, I would probably have banned you. When I reply with "maybe you're the one who can't", it's better you not take it as an insult. If you do, you shouldn't have "can't you read" me first.

    Probably, but it's not. Stock cooler is something that's designed to fit the needs of something (a CPU/GPU it's designed for). Since this cooler is unable to give the GPU it's designed for more comfortable temperatures, it's not a good cooler.
     
    El_Mayo says thanks.
  23. imperialreign

    imperialreign New Member

    Joined:
    Jul 19, 2007
    Messages:
    7,043 (2.45/day)
    Thanks Received:
    909
    Location:
    Sector ZZ₉ Plural Z Alpha

    Exactly, and the rate at which a metal gives off thermal energy is equivalent to the rate at which it absorbs thermal energy.

    Cu absorbs larger amounts of thermal energy over a given period of time than Al is capable of . . . if you were to have both an Al and Cu HS at room temperature, and apply an equivalent amount of heat to them, you would find that the Al HS will warm up faster than Cu will . . . the converse is true as well, remove that source of heat, and Al will return to room temperature faster than Cu.

    Does this mean that Cu is a less-efficient cooler material than Al? No, quite the contrary.

    The density of the metal has a lot to do with the amount of thermal energy the metal is absorbing. Cu is absorbing more thermal energy than Al, and when applied to cooling a component, this means that Cu is absorbing more thermal energy being given off by the component, and therefore releasing more thermal energy as well.

    This is basic physics, man.
     
  24. btarunr

    btarunr Editor & Senior Moderator Staff Member

    Joined:
    Oct 9, 2007
    Messages:
    29,766 (10.67/day)
    Thanks Received:
    14,221
    Location:
    Hyderabad, India
    I'm arguing on his pane. Not only am I saying that "it barely handles its intended application", but also "there's another stock cooler that handles higher thermal load, and stays cooler, so is better".
     
  25. Grings

    Grings

    Joined:
    Nov 16, 2006
    Messages:
    2,303 (0.74/day)
    Thanks Received:
    187
    Location:
    Blighty
    Has anyone tested how hot fermi gets when manually setting the fan to 100%?

    I really dont think a t-rad or accelero extreme style cooler would work very well here, with heat levels like this you NEED to be exhausting heat out of the rear... i wouldnt be surprised if aftermarket coolers for this card all have big ass ducting systems to get more heat out the case

    I wonder if we'll see more of those cases with 'wind tunnels' for the graphics cards soon?
     

Currently Active Users Viewing This Thread: 1 (0 members and 1 guest)

Share This Page