It's starting to show (Fermi) (NEW official MSRP's)

Dev1ant said:
73529807.jpg

gtx4802.jpg


Well, there is 1920x1200
Click to expand...

damn, if those are accurate, It looks like I'm better off sticking with my 295 for the forseable future

Hopefully they release a 495 late this year/early next year
 
dr.stevil said:
damn, if those are accurate, It looks like I'm better off sticking with my 295 for the forseable future

Hopefully they release a 495 late this year/early next year
Click to expand...

Holy cow! I didn't realize Metro 2033 was that demanding...
 
Derfman said:
Holy cow! I didn't realize Metro 2033 was that demanding...
Click to expand...
It's not. It's one setting - Advanced DoF, and it pretty much cuts FPS in half for an extremely minor IQ benefit. Most reviewers seem to neglect to actually play the game they are testing and fail to notice this.

Thanks for posting those charts Dev1ant, very interesting.
 
So, have you got it yet? ;)

Btw. why did you send me this PM?

Don't mistake laughter for tears... unless it's laughing so hard at you fangirls that I'm crying from doing so that much! Your little post about Charlie being right is hilarious, he's been wrong on virtually everything for years.You saying "he's right on Fermi, it's slow and hot and horrid and OMG IT KICKED MY DOG!!!!" is just pathetically hilarious.

I bid you adieu, young child!
Click to expand...
 
GoldenTiger said:
I wouldn't know, since it's around 20% faster avg. and 30% minimum fps, at 1920x1200 with 4x AA in most benches. Cute bait, though...

http://www.hardwarecanucks.com/foru...iews/30297-nvidia-geforce-gtx-480-review.html
2560x1600 is stated to be a driver issue on the BFBC2 page for a couple of games, quoted direct from nVidia. Hence why I call out 1920x1200.
Click to expand...
Yeah, I'd cherry pick benches too to make sure I didn't look like I pissed away my money either.
The average across reviews is 5-15%, and is reduced close to 0 once both cards are overclocked. You can try to justify it all you want to yourself, don't waste our time.
 
Mr. K6 said:
Yeah, I'd cherry pick benches too to make sure I didn't look like I pissed away my money either.
The average across reviews is 5-15%, and is reduced close to 0 once both cards are overclocked. You can try to justify it all you want to yourself, don't waste our time.
Click to expand...

Who's trying to convince who against the facts now? Looks like you: the reviews show 15-20% average gains across all of them, and 20-30% minimum FPS gains at normal gaming settings, period. You can cherry-pick situations where it doesn't, but that's not the norm. I don't give a crud what you think of my spending, rofl, not like we're in grade school here... I just think it's sad when people completely misread and fail to comprehend data right in front of their noses.
 
GoldenTiger said:
I just think it's sad when people completely misread and fail to comprehend data right in front of their noses.
Click to expand...
Wow... Seriously?
To wrap things up, let’s start with the obvious: NVIDIA has reclaimed their crown – they have the fastest single-GPU card. The GTX 480 is between 10 and 15% faster than the Radeon 5870 depending on the resolution, giving it a comfortable lead over AMD’s best single-GPU card.

Meanwhile let’s talk about the other factors: price, power, and noise. At $500 the GTX 480 is the world’s fastest single-GPU card, but it’s not a value proposition. The price gap between it and the Radeon 5870 is well above the current performance gap, but this has always been true about the high-end. Bigger than price though is the tradeoff for going with the GTX 480 and its much bigger GPU – it’s hotter, it’s noisier, and it’s more power hungry, all for 10-15% more performance. If you need the fastest thing you can get then the choice is clear, otherwise you’ll have some thinking to decide what you want and what you’re willing to live with in return.
Click to expand...
http://www.anandtech.com/video/showdoc.aspx?i=3783&p=20

The GeForce GTX 480 is more relevant in the market but it hasn’t exactly come out of the gate wowing us with performance either. There are some games where it is faster than the Radeon HD 5870, and there are some games where it is even with the Radeon HD 5870. Factor in the cost and power, and include the ability to run Eyefinity on a single GPU, the Radeon HD 5870, to us, seems like the better value for the gamer right now.
Click to expand...
http://www.hardocp.com/article/2010/03/26/nvidia_fermi_gtx_470_480_sli_review/8

Yeah, I'll just go back to "ignoring" all those "awesome" reviews showing it 20-30% faster. :rolleyes:
 
CrimandEvil said:
Wow... Seriously?

http://www.anandtech.com/video/showdoc.aspx?i=3783&p=20


http://www.hardocp.com/article/2010/03/26/nvidia_fermi_gtx_470_480_sli_review/8

Yeah, I'll just go back to "ignoring" all those "awesome" reviews showing it 20-30% faster. :rolleyes:
Click to expand...
Beat me to it, but exactly what I was going to say.

EDIT: Let me also direct everyone's attention to this summary:
480vs5870final.png

Note that this group includes extremely varied reviews. Some didn't even run the 58xx series on the 10.3a's, which give a substantial performance boost. Take this as the best case scenario for a GTX480.
 
Last edited by a moderator:
Mr. K6 said:
Beat me to it, but exactly what I was going to say. Like I said, GoldenTiger, parrot whatever fanboy crap you want in the mirror, don't waste our time.

EDIT: Let me also direct everyone's attention to this summary:
[*IMG]http://img638.imageshack.us/img638/1360/480vs5870final.png[/IMG]
Note that this group includes extremely varied reviews. Some didn't even run the 58xx series on the 10.3a's, which give a substantial performance boost. Take this as the best case scenario for a GTX480.
Click to expand...

Yeah, I kinda posted that already from XS here:

http://hardforum.com/showthread.php?t=1506377

Yesterday.
Since then, it's been updated here: http://www.xtremesystems.org/forums/showpost.php?p=4309768&postcount=238 .


Note how it is holding a solid 30% min. fps lead at 1920x1200 4x aa, and still with buggy drivers managing 15% average over all games across all reviews, in many cases spiking higher to 30+% gains on avg. and far higher on the min.? Yeah. That's not even taking into account *minimum* fps gains which are far more important and show more substantial gains such as:

Anandtech regarding Crysis said:
"We’ve also gone ahead and recorded the minimum framerates for Crysis, as in our testing we’ve found the minimums to be very reliable. And in doing so, we have some data even more interesting than the averages. The GTX 400 series completely tramples the 5000 series when it comes to minimum framerates, far more than we would have expected." (13% faster average fps, versus 33% faster minimum fps)
Click to expand...

HardwareCanucks regarding DX11 Aliens vs. Predator said:
Things really start out on an interesting note for the GTX 480 in this game since it struggles a bit with AA disabled but when we bump up the image quality with 4xAA, it simply destroys the HD 5870 and even comes close to the HD 5970 in some cases. While performance without AA may be on the low side, the real story here is the minimum framerates at every IQ setting and here the GTX 480 can claim absolute dominance.
Click to expand...

So, call it "fanboy crap" all you want, but it doesn't change the hard facts staring you in the face. Unless you're insinuating that every review site on the 'net that used newer ATI drivers & a GTX 480 with a fast CPU (i7 920 oc'd/better) are fanboys of nVidia too?
 
Last edited:
Goldentiger, I hope you get your 480s and enjoy them as much as you want because you are obviously in love with them. Nobody is going to change your mind and you aren't going to convert anybody here. I appreciate you had a bad experience with the 5870 and want an alternative. I think it's irrefutable that the average performance increase over the 5870 is about 15% corresponding to the 16% price difference and you said so yourself in the other thread that you started. That basically means that currently, the 5870 and Fermi about about equal on performance per dollar (excluding your electricity bill :), but maybe you'll save on your heating bill) We don't need to argue about that.

Let us know how the sound levels are when you get one.

I'd wager the 5870 has much move overclocking headroom though, due to the low TDP and low heat. If AMD improves their cooler, adds more VRM cooling, bumps up the power a notch, and brings out the 2GB model at around the $500 price point as a direct competition to the GTX 480, that model may come out a winner as well but for now, if you want to pay 16% more for 15% performance, that makes perfectly fine sense to me.
 
Last edited:
Spare-Flair said:
Goldentiger, I hope you get your 480s and enjoy them as much as you want because you are obviously in love with them. Nobody is going to change your mind and you aren't going to convert anybody here. I appreciate you had a bad experience with the 5870 and want an alternative. :)

Let us know how the sound levels are when you get one.
Click to expand...

I don't particularly hate the 5870 or anything... I just find it irritating to an extent that people seem unable to read charts that are straight in front of them with number sets in hard writing: not subjective, but actual, objective numbers sitting plainly. .I'm not out to convert anyone. A basic search could probably find posts of me saying the same things in regard to 5870 vs. GTX 280 as well,just as excitedly but in ATI's favor on that one especially over on the XS forum ;). I'm solely trying to point out performance deltas here, not complain about the 5870's drivers, and haven't been for quite a few posts :). Hardware is a hobby to me, and I have fun playing with the latest & greatest.

I'd wager the 5870 has much move overclocking headroom though, due to the low TDP and low heat. If AMD improves their cooler, adds more VRM cooling, bumps up the power a notch, and brings out the 2GB model at around the $500 price point as a direct competition to the GTX 480, that model may come out a winner as well but for now, if you want to pay 16% more for 15% performance, that makes perfectly fine sense to me.
Click to expand...

The couple of reviews that oc'd the GTX 480 showed pretty good #'s (800mhz core, 4000-4100mhz ram) but they're nvidia-selected cards, so I'm more interested in retail card results on that front. 5870 seems to stably sit at ~925mhz for most people for 24/7 gaming, vs. 850mhz stock on the core. I'm not so convinced that in 1920x1200 4x AA tests the RAM capacity is making a difference, really, though.
 
Last edited:
The hell is up with pcsuperstore charging my credit card for something they won't even ship out for a few weeks at least. oh and they actually charged it, went from a $1.00 pre auth transaction, then to the full amount pending and now it's a completed transaction.
 
Mr. K6 said:
EDIT: Let me also direct everyone's attention to this summary:
(long summary omitted)
Click to expand...
Mr K6, did you build this table? You did a good job collecting the numbers from all the sites. The math you use to compute the averages, however, are not quite right. You used arithmetic mean (add n numbers and divide by n) when you should have used the geometric mean (multiply n ratios and take the nth root). Here is a short example to illustrate the differences:

Card A: 60 FPS in Game A, 30 FPS in Game B
Card B: 30 FPS in Game A, 60 FPS in Game B

It is obvious that neither card has any advantage over the other. Here is the speed up of Card A over Card B:

Game A: +100%
Game B: -50%

Here is what you get by computing the arithmetic mean (as was done in your table) of the two speed-ups:

Arithmetic mean = (100%+(-50%))/2 = 50%/2 = 25%

This result is obviously wrong: Card A is not 25% faster than Card B. Here is what you get by using geometric mean:

Geometric mean = (200%*50%) ^ (1/2) = 100% ^ (1/2) = 100%

Instead of using the speedup (100% and -50%), here we must use the speed ratio (200% and 50%) in the computation. And we get the expected result: Card A is not any faster than Card B (performance of Card A is 100% of the performance of Card B).

I know it is a lot of work, but maybe you can redo the table for us?
 
I haven't posted for a long time but I just had to log in here to make a comment. I'm tires of seeing these AMD / ATI fanboys bashing on the new fermi line, it's pathetic.

Some of the BIGGEST draws of nvidia cards at this point is the PhysX, 3D Vision, Cuda, etc. support. I could care less if you ATI guys can run three monitors with just one card, it's meaningless to me. I personally care about 3D Vision, I think it's fantastic and it's one of the coolest investments I've made in the gaming realm for quite some time. I also love physics in my games, and at this point only nvidia cards can utilize PhysX.

While I won't be running two 480s, I am really looking forward to getting more frames out of 3D once my card arrives. That's why I'm upgrading to the 480 from my current 280.
 
You must log in or register to reply here.
Back
Top