AT fermi review is up!

[Piano Man]

Man, did you hear the dbs these things put off under load? Unacceptable. I thought my 5850 was way too loud. These things are friggin' leaf blowers. Gonna have to break out the water cooling if you want any peace of mind when playing games with these space heaters.

Well, now we know why ATI isn't going to lower prices. Sucks for consumers.

 

[Maximilian]

Can you guys say dustbuster? =/

 

[allthatisman]

The Firing Squad has a nice review that you can actually read... good graphs as well. I have never seen so many people this juiced, and subsequently let down by a videocard release lol.

http://www.firingsquad.com/hardware/nvidia_geforce_gtx_480_470_performance/

Moral of the story, its faster than a 5870 on average of about 7-15% give or take... but both are still a hair slower than my 4870x2... I guess it's staying around for a while...

 

[dguy6789]

I'm impressed. It's faster than I thought it was going to be. Hopefully it will be available in sufficient enough quantities so that the price won't be inflated way beyond msrp.

 

[RussianSensation]

http://hardocp.com/article/2010/03/26/nvidia_fermi_gtx_470_480_sli_review

Kyle's review is up. His numbers look like the 470= 5850 and 480 = 5870 pretty much but with MUCH higher power draw.

Ya based on his review, GTX470 is overpriced and inefficient compared to 5850 and doesn't produce any better gameplay:

"Starting from the bottom up, we would say the least relevant video card is the GeForce GTX 470. In all our gameplay testing today, not once did the GeForce GTX 470 provide a superior gameplay experience compared to the Radeon HD 5850, even in Metro 2033. In fact, performance was very close between both video cards, and in some cases the Radeon HD 5850 proved to provide faster framerates. Looking strictly at performance, these video cards are equal. Looking at power, the GeForce GTX 470 consumes a good bit more power than the Radeon HD 5850, while also producing higher thermals."

with GTX480 vs 5870, again more or less of the same:

"We are seeing some trades back and forth; in some games the GTX 480 is slightly faster, in other games the HD 5870 is slightly faster. There wasn’t anything that really stood out, other than Metro 2033. Once again, we have to factor in cost, power, and heat. The GeForce GTX 480 consumes more power than the HD 5870, produces more heat, and costs more money (which might also include the cost of a new PSU). It doesn’t seem like the best value with all this factored in. The only thing that "blew us away" was the heat coming out of the video card and the sound of the fan. If you have not taken a look and a listen to the videos on the previous page, you should.".... "We see no reason to purchase a GeForce GTX 470."



-

Here is TweakTown Review of GTX470:
http://www.tweaktown.com/articles/3203/nvidia_geforce_gtx_470_video_card_tested/index17.html

-- The high heat level isn't because the fan is spinning quietly either. Noise at 74 DB at load....ouch. Performance about 5850's level.
-- Power consumption at load is 114W more than 5850! and just 2W lower than 5970.

 

[Lonyo]

Well the first UK prices I've seen are an utter joke.

£220 HD5850.
£310 HD5870

£310 GTX 470
£430 GTX480

£500 HD5970

ATI don't need to drop their UK prices based on this sort of crap.

 

[konakona]

I am not so concerned about power (TX850 should be able to handle it no sweat) but noise is one thing I really can't stand.

Just keep in mind that the GTX 470 is still going to be louder/hotter than any 5800 series card

and this

The 5850 is the bargain hunter’s card

That's pretty much all I had to know, glad I got my 5850. Maybe some aftermarket cooling can help the NV cards with noise. Too bad AT review doesn't have anything on overclocking.

 

[DefRef]

What single-card solution is SLIed 8800GT 512s comparable to? I was thinking about putting a single GTX 275 in my rig, but everything I saw said it would be a wash performance-wise. Looking at the Battleforge chart, a single 8800GT gives 23.1 fps, the GTX 275 is 45, and the GTX 480 does 85.5. Does that mean the 480 would be twice as fast as my current setup? Thanks.

 

[sxr7171]

To say it is 10-15% faster is meaningless. What matters is if it can make certain games not playable on the 5870 playable given a set of quality settings. The other thing that is relevant is if it will allow you some quality settings that the 5870 won't allow at playable frame rates.

I couldn't give a rat's ass if my game which runs at 120fps on the 5870 runs at 138fps on the 480.

 

[Lonyo]

To say it is 10-15% faster is meaningless. What matters is if it can make certain games not playable on the 5870 playable given a set of quality settings. The other thing that is relevant is if it will allow you some quality settings that the 5870 won't allow at playable frame rates.

I couldn't give a rat's ass if my game which runs at 120fps on the 5870 runs at 138fps on the 480.

The main problem is that the higher the resolution, the lower the gap.
It's more like 15% at 1680x1050, 10% at 1920x1200 and 5% at 2560x1600.
That's not good when you consider the game gets less playable, and the chances of someone running a high resolution are considerable considering how much these cards cost.

10~15% faster including "low" resolutions that these cards aren't really meant for isn't good considering how much more they cost.
That's the most shocking thing. Higher res -> smaller gap. What the hell is going on there.

 

[lopri]

What is the official TDP of the 480? The thing consumes 120W+ more than 5870 under load.

 

[SlowSpyder]

Holy crap... I just watched the GTX480 video on Hardocp.com. It just keeps geting louder, and louder, and louder...

FX5800 anyone?

 

[SlowSpyder]

What is the official TDP of the 480? The thing consumes 120W+ more than 5870 under load.

I've been told that no one cares about that, it doesn't matter. Hardware enthusiasts don't care about the efficiency of NASCAR engines, or something like that. :/

 

[Lonyo]

What is the official TDP of the 480? The thing consumes 120W+ more than 5870 under load.

250W TDP for GTX480, 188W for the HD5870.

62W difference in TDP, 100W difference in system load under gaming conditions (consider potential difference due to efficiency of the PSU)

 

[cbn]

Yes. That's the only thing I've done @HardOCP and my goodness.. It's insane. I'm normally not too picky on load power consumption of enthusiast parts, but that's simply not acceptable. And the test was performed on open bench so it's scary to think what'll happen inside a case.

How does it compare to other cards in the 290 watt TDP range?

I need to listen to more sound tests. I wish he had HD5970 in there for comparison.

 

[Compddd]

Chizow on EVGA forums says the GTX 480 keeps up with the 5970 sometimes?

 

[pmv]

Well the first UK prices I've seen are an utter joke.

£220 HD5850.
£310 HD5870

£310 GTX 470
£430 GTX480

£500 HD5970

ATI don't need to drop their UK prices based on this sort of crap.

Kind of baffling why anyone would buy the gtx 470 at that price difference from the 5850. Why does the card even exist? What's the point of it?

The 480 looks almost as pointless, though I guess those who don't care how much it costs for a few extra fps might be able to justify it to themselves.

 

[allthatisman]

To say it is 10-15% faster is meaningless. What matters is if it can make certain games not playable on the 5870 playable given a set of quality settings. The other thing that is relevant is if it will allow you some quality settings that the 5870 won't allow at playable frame rates.

I couldn't give a rat's ass if my game which runs at 120fps on the 5870 runs at 138fps on the 480.

Agreed... but numbers are an easy way for people to wrap their heads around results, and that's what were all after isn't it? And to be fair, at this point... aside from Metro 2033 and Crysis, what other powerhouse games have come out in the last couple years that have brought even my 4870x2 to it's kness??... the answer is none. My unqualified theory is because of the role consoles play in PC gaming design of late, and also the hardware seems to have caught up... at least on the GPU side anyway...

 

[SlowSpyder]

Chizow on EVGA forums says the GTX 480 keeps up with the 5970 sometimes?

I saw one bench where it did. The GTX480 is definitely not on the 5970 level as far as performance in general.

According to Hardocp, the GTX480 uses more power than a 5870 CF setup... I thought that was noteworthy.

 

[cbn]

250W TDP for GTX480, 188W for the HD5870.

62W difference in TDP, 100W difference in system load under gaming conditions (consider potential difference due to efficiency of the PSU)

How is the PSU going to affect these numbers?

 

[Lonyo]

I've been told that no one cares about that, it doesn't matter. Hardware enthusiasts don't care about the efficiency of NASCAR engines, or something like that. :/

Not to mention my 400w PSU means I am not an enthusiast and shouldn't complain about power.

Lets ignore the fact that a 400w PSU could run an HD5850 or HD5870. They aren't enthusiast cards.

In fact, the only single GPU cards it couldn't run are all made by NV, hmm...

 

[Compddd]

I'll prolly get a GTX 480 so I can play with physX I guess

 

[cbn]

Kind of baffling why anyone would buy the gtx 470 at that price difference from the 5850. Why does the card even exist? What's the point of it?

The 480 looks almost as pointless, though I guess those who don't care how much it costs for a few extra fps might be able to justify it to themselves.

For someone that uses H20 and doesn't like Multi-GPU, overclocking Fermi might be really interesting.

Could water improve efficiency at high clocks in this scenario?

 

[Lonyo]

How is the PSU going to affect these numbers?

Different load means different level of efficiency.
If anything it will make the GTX480 look more favourable.
If you assume the efficiency is 70% at the HD5870 load then you get real power draw (rather than from the wall) of ~225w
If you assume it goes up to 80% at the 420w GTX480 load, then draw from the wall would be ~336w.

Instead of the 102w difference, it would be 111w.

I have no idea what the efficiency figures are for that PSU (since AT haven't finished their review yet...), but it's an example that as power load goes up, so does efficiency, slightly altering the differences in actual load of the cards.

 

[cbn]

Not to mention my 400w PSU means I am not an enthusiast and shouldn't complain about power.

Lets ignore the fact that a 400w PSU could run an HD5850 or HD5870. They aren't enthusiast cards.

In fact, the only single GPU cards it couldn't run are all made by NV, hmm...

In this graph we have a 40nm single GPU card using more power than a 55nm dual GPU card?