[H]ard Does New Heat / Noise Test (real world) for 480

[Apocalypse23]

It's really puzzling and funny to me that after many reviews of Fermi, the folks are still trying to hear and believe what they want to (aka Fermi runs cool and quiet...), when clearly that's not the case.

Fermi runs hot and loud, period. Kyle from HardOCP probably was getting annoyed by fans asking him about reevaluating his decision so much that he went ahead and tested the cards again, the cards are not a success imo.

 

[ronnn]

what do you recommend? I don't mind spending a few $$$ for some quiet. I have an ssd now, too, so quieter case fans could be nice!

I just got a Silverstone Fortress FT01 case with a corsair hx650 ps. I know expensive, but both were less than a 5850 and nice quality. Anyways with the old sunbeam fan controller is very quiet with stock fans. The nice thing about a fan controller is you can crank it up on a hot when gaming.

http://ncix.com/products/index.php?s...facture=Noctua These fans are very good and often go on sale.

 

[apoppin]

I had a 2900 Pro @ 800MHz, never thought it was that loud. When I'd launch a game you could clearly hear the fan spin up, but it wasn't too loud in my opinion.

i had one too. When i set up my test, you can hear the XT at full bore and then you decide.

As it stands right now - with me halfway thru the OC'ing tests (i am almost done with HD 5870) - "features" ASIDE!

i think the GTX 480 is a very good raw performance value for $500 if you are not worried about the 250W TDP and associated issues. It is the fastest single GPU. (period)
- it is also on beta drivers and there is room for improvement

i think HD 5870 with a decent overclock is an exceptional value for $400 or so. It's bang for buck is awesome and it is relatively cool running and a mature process on mature drivers.

It's really puzzling and funny to me that after many reviews of Fermi, the folks are still trying to hear and believe what they want to (aka Fermi runs cool and quiet...),

WHO is saying that?

i also notice that you also pick and choose what you want to believe, exaggerate it and then pass it on as fact. Why is that?

 

[ronnn]

Concur, Apoppin tends to not suffer dumb asses well and they end up misinterpreting that as being a fan boy.

Liking the guy doesn't make him smart or unbiased or give him good ears. Will be hard to beat Kyle's review on noise. Those vids were a brilliant idea and until I actually hear a 480 in action - no amount of reviews without audio will change my mind.

 

[Apocalypse23]

i had one too. When i set up my test, you can hear the XT at full bore and then you decide.

As it stands right now - with me halfway thru the OC'ing tests (i am almost done with HD 5870) - "features" ASIDE!

i think the GTX 480 is a very good raw performance value for $500 if you are not worried about the 250W TDP and associated issues. It is the fastest single GPU. (period)
- it is also on beta drivers and there is room for improvement

i think HD 5870 with a decent overclock is an exceptional value for $400 or so. It's bang for buck is awesome and it is relatively cool running and a mature process on mature drivers.



WHO is saying that?

i also notice that you also pick and choose what you want to believe, exaggerate it and then pass it on as fact. Why is that?

I don't know about that, guess it's just me?

 

[apoppin]

Seriously?

LOL

I have never seen anyone tout extra heat put out by any electronic equipment, outside of a space heater/furnace/etc, as being a FEATURE.

Who the heck said it is a "feature" ?

i don't like the extra heat in Summer; but i don't whine about it
- i turn the cooling up to compensate for the fact that i like the higher performance from my core i7 at 3.8 GHz and from everything else. i put with the extra heat because i demand extra performance.

But then, MOST of what i do on my PC is on my notebook PC; the gaming rig is for gaming and benching. i compensate.

I don't know about that, guess it's just me?

Probably not

 

[Rebel44]

I really like Kyles reviews because they are so close to real world situations.

This test was done with GPU in case with very good cooling, GPU was stressed with games (not Furmark or any other benchmark) and sound was measured from reasonable distance from PC.

For me Fermi is failure - its 10-15% faster than 5870 but it eat much more power and its louder so I wouldnt pay for it more than for 5870.

 

[Madcatatlas]

Much the same for me, i might end up getting one anyway tbh, it depends on what gameengine and features are put into Mechwarrior 5.
And lets not forget that as a singlegpu the 480 has pretty descent minimum framerates aswell, though it depends on the title etc

However, i agree, i dont like the heat and i wouldnt call it a feature, neither do i enjoy the extra draw on my power bill as a sideeffect and i wouldnt call it whining either if someone brought the point up.
Its kinda a repeat of 480s weak points, power needed, heat created and soundlevels above what the competition offers.

The really bad part is, will i be able to buy one..

 

[blanketyblank]

I'd prefer you test metro 2033 since Crysis is getting outdated.

btw remember to let us know your ambients since you are saying it's cold there it would be nice to know what to expect when the temp heats up.

 

[MrK6]

Did you miss the smilie?
- the fun part was that KyleB *banned* a long-time forum member for *daring* to question his methods

That wasn't what happened. That "long-time forum member" was an idiot and should have been banned awhile ago. It's a good thing someone finally did. And since he was personally insulting the forum owner who's taking the time to answer everyone's specific questions... well, that's just ignorant.

Anyway, interested in seeing the results.

 

[Ben90]

The GTX 480 is hot. I know many enthusiasts are not going to be satisfied with a video card that idles at 73C (and even hotter in most SLI setups) and runs 93C under pretty much any kind of load. There have to be long term implications to this and I would just suggest that you buy from a board builder that has a rock solid warranty and a reputation for great customer service if Fermi is for you.

I have an ATI 9800 pro in an old Thinkpad that idles around 80 and loads into the 110s. The CPU core has been operating above 100*C almost 24/7 since I got this early 2003. Temps are not as big of a concern as people make them out to be. Anyone with experience in engineering microprocessors can concur.

 

[ronnn]

i also notice that you also pick and choose what you want to believe, exaggerate it and then pass it on as fact. Why is that?

Someone who Kyle B banned at his own forum, suggested the pine wood in his house was echoing the sound and exaggerating the louder sounds

So why pretend the banning was for disagreeing with Kyle?

And why go after Kyle?

 

[pmv]

Seriously?

LOL

I have never seen anyone tout extra heat put out by any electronic equipment, outside of a space heater/furnace/etc, as being a FEATURE.

Doesn't make sense to me. Electricity is a very inefficient and expensive way to produce heat, compared to other fuels. If you need heat, turn the gas up or put more coal on the fire. Using an electric heater (one _insiide_ your PC case at that!) is a very eccentric way to do it. Unless, of course, you don't have any other fuel supply, in which case, fair enough.

 

[RussianSensation]

http://ncix.com/products/index.php?s...facture=Noctua These fans are very good and often go on sale.

Scythe Gentle Typhoon is my pick for $5 less.
http://www.ncix.com/products/?sku=35...facture=Scythe

You can always get Yate Loons. Those are awesome at $5.

 

[RussianSensation]

i had one too. When i set up my test, you can hear the XT at full bore and then you decide.

Just want to make sure you are testing at 100% gpu load not 100% fan speed :awe: I don't think anyone is interested what the sound level at 100% fan speed is.

 

[apoppin]

So why pretend the banning was for disagreeing with Kyle?

And why go after Kyle?

Who is going after Kyle?
-He does things his way. That is fine.

Everyone has different impressions of the same card. it is good to hear from everyone's PoV including someone as experienced as KyleB. Undoubtedly having 6 months with a quiet card like HD 5870 and then putting in a GTX 480 is a bit of a surprise. If you come from a 4870-X2, it is not so shocking.

Have you actually read that thread over at HardOCP forum? It was for disagreeing with Kyle - the guy who was banned *questioned* Kyle's methods .. he was joking yet Kyle took offense. Which is Kyle's right; it is his forum and he is free to ban his forum guests for a perceived insult.

Just want to make sure you are testing at 100% gpu load not 100% fan speed I don't think anyone is interested what the sound level at 100% fan speed is.

*EXACTLY!*

Now.. what *short* benchmark (from a real game) will start out "slow" and then ramp up the GPU to 100% relatively quickly? i want to use something familiar - something that is 100% identical and repeatable since i am testing several cards. Short because i don't want to bore the hell out of everyone.

 

[RussianSensation]

*EXACTLY!*

Now.. what *short* benchmark (from a real game) will start out "slow" and then ramp up the GPU to 100% relatively quickly? i want to use something familiar - something that is 100% identical and repeatable since i am testing several cards. Short because i don't want to bore the hell out of everyone.

Why don't you use Unigine Heaven v2 bench? at 8AA/16AF it will bring every single card to its knees and it loads them fully almost right away like no other. It's identical every time too! Although it's not a game, I think it's about as intense as any game will be, but a lot more realistic than FurMark.

 

[Zebo]

Living in a softwood house, it deadens sound. Pine is even softer, especially if it has turned blue from the beetle.

Not really sound is block by mass. Look up "mass loading."

This is why studios are solid concrete blocks and 1' sand floors. Go underground sometime. I worked at Nevada Test Site in some of the bunkers there and you couldn't hear bombs at air force range just a miles away. Pretty awesome.

Incidentally all my PC cases are lined with 1/4" asphalt substrate to block most sounds.

 

[Attic]

Apoppin, I'd prefer if you did real world game tests for your noise and thermal? tests.

The synthetic benchmarks and Furmark i've never given much consideration to because they aren't games. I understand their place however. Crysis is always good and BFBC2 should be 90-100% GPU usage.

I've grown to really like HardOcp's review methods, it's obvious those guys are gamers and are looking to inform gamers in meaningful ways about gamings potentials and pitfalls with current tech. I've seen Kyle handle a few of the board trolls with solid reserve, if he banned someone I dobut it was for anything truely minor. His noise evaluation is great thing IMO, best so far and he went through the trouble of using a $300 mic for his tests.

 

[BFG10K]

As to noise, what all the tech site are pointing out is GTX 480 at *MAX load*
- it never gets there in gaming .. how can that be "real world"

I just fired up Quake 3 and I was able to get a 99% GPU load at the settings I play the game at, and that game is 11 years old. I did the same thing with Stalker SoC, again using the settings I actually play the game at.

Such a scenario is very much real world, and you don’t need Furmark to do it. If you always play games at their highest playable settings (like I do) then you’ll often get a 95% (or higher) GPU load.

This is why the noise is a big deal to me. With the way I configure my games and graphics card, I’ll almost certainly cop the worst case scenario from the GTX480.

 

[apoppin]

I just fired up Quake 3 and I was able to get a 99% GPU load at the settings I play the game at, and that game is 11 years old. I did the same thing with Stalker SoC, again using the settings I actually play the game at.

Such a scenario is very much real world, and you don’t need Furmark to do it. If you always play games at their highest playable settings (like I do) then you’ll often get a 95% (or higher) GPU load.

This is why the noise is a big deal to me. With the way I configure my games and graphics card, I’ll almost certainly cop the worst case scenario from the GTX480.

Here is what i am looking for:

A *short* timedemo or benchmark from a game that starts out relatively light on the graphics card and then ramps up GPU usage to 100%

i am thinking the GPU demo from Crysis will do .. although it is longer than i like

I've grown to really like HardOcp's review methods, it's obvious those guys are gamers and are looking to inform gamers in meaningful ways about gamings potentials and pitfalls with current tech. I've seen Kyle handle a few of the board trolls with solid reserve, if he banned someone I dobut it was for anything truely minor. His noise evaluation is great thing IMO, best so far and he went through the trouble of using a $300 mic for his tests.

i am there also and i appreciate the work they do for gamers.

Sadly i cannot find my AKG mic; a nice one, but will make do with something different.

 

[cbn]

Here is what i am looking for:

A *short* timedemo or benchmark from a game that starts out relatively light on the graphics card and then ramps up GPU usage to 100%

i am thinking the GPU demo from Crysis will do .. although it is longer than i like

Why does the demo need to be short?

 

[apoppin]

Why does the demo need to be short?

who wants to listen to a long audio file?
- especially if a short one will do as well?

Let me see if i got this right:

i fire up a game demo and let the automatic fan profile take care of the fan RPMs as we listen to it as the GPU loads to 100%
--rinse and repeat with each video card with the testing parameters exactly the same ... i'll also include case temps.

Do we need a long demo to make the test valid?

 

[Chumster]

who wants to listen to a long audio file?

[snip]

Do we need a long demo to make the test valid?

You wouldn't have to record the entirety of the test - just the last 2 minutes (?) or so. You said you wanted this to be as 'real world' as possible. I'm not sure about you, but I don't fire up BFBC2 for only 2 minutes at a time.

 

[Axon]

I was hoping they could cook an egg on it