GeForce GTX 480 and GeForce GTX 470

$499 for the 448 shaders, im sure of it 99%.

Nvidia might as well shift their entire production line to Tesla then, because they aren't going too find many suckers who will pay $100-150 more for roughly the same performance as the 5870.

The 512sp version should be around 500-525 if they want a shot at competing. Otherwise myself and most others will be trying out the 5000 series or waiting on the 6000 series. I might even rebuild my computer around an i7 with the money I save.
 
I STILL don't understand ATI's logic of undercutting Nvidia so hard on the 4XXX series. Yes, they gained market share, but man did they loose out on profits. They could have easily made 3 or 4 times thier profit margin on those cards. I'd happly trade 1/2 my sales for 3 times the margin.
Because they want much more market share, which the combination of 4xxx and 5xxx cards has given them. There are a lot of rumours created by NVidia fanboys and NVidia itself, and perpetuated by first-time card/PC buyers who don't know better, that NVidia is inherently superior, drivers are better, etc. I'm sure you know what I'm talking about. If AMD can get a significantly greater market share, then they get a significantly increased pool of people in online forums like this one, who will stand up and say, 'hey, I use an AMD card and I'm perfectly happy with the performance and the drivers work very well'. Then in the long term, the perception against AMD starts swinging the other way, and people start buying AMD cards because they perform better, they run cooler, quieter, whatever, rather than all this fabricated nonsense about superior drivers.
 
Because they want much more market share, which the combination of 4xxx and 5xxx cards has given them. There are a lot of rumours created by NVidia fanboys and NVidia itself, and perpetuated by first-time card/PC buyers who don't know better, that NVidia is inherently superior, drivers are better, etc. I'm sure you know what I'm talking about. If AMD can get a significantly greater market share, then they get a significantly increased pool of people in online forums like this one, who will stand up and say, 'hey, I use an AMD card and I'm perfectly happy with the performance and the drivers work very well'. Then in the long term, the perception against AMD starts swinging the other way, and people start buying AMD cards because they perform better, they run cooler, quieter, whatever, rather than all this fabricated nonsense about superior drivers.

I agree with you on most parts. But I've always found ATI's Catalyst a bit sluggish to use and unstable at times. Of course Nvidia's forceware isn't perfect. Even then, I still can't help prefer nvidia drivers. But that's just me of course.
 
Because they want much more market share, which the combination of 4xxx and 5xxx cards has given them. There are a lot of rumours created by NVidia fanboys and NVidia itself, and perpetuated by first-time card/PC buyers who don't know better, that NVidia is inherently superior, drivers are better, etc. I'm sure you know what I'm talking about. If AMD can get a significantly greater market share, then they get a significantly increased pool of people in online forums like this one, who will stand up and say, 'hey, I use an AMD card and I'm perfectly happy with the performance and the drivers work very well'. Then in the long term, the perception against AMD starts swinging the other way, and people start buying AMD cards because they perform better, they run cooler, quieter, whatever, rather than all this fabricated nonsense about superior drivers.

roflmao
 

Why? hehe. He's kind of right but he forgot ATI's other strategy.

Average Joe consumer walks into store. Consumer sees Wii for $199.99. Consumer sees ps3 for $599.99. Consumer leaves with Wii.

Consumer walks into store. Consumer sees ATI with low price tag, comparatively and both boxes say '1GB' or '512MB'. Huh, must be same speed but ATI is cheaper by ___ dollars. Consumer leaves with ATI.

Nvidia's counter-strategy; Next week consumer walks into store. Consumer sees ATI with low price tag. Consumer sees nVidia box says '640mb' or '768mb' and looks at ATI's box saying 512MB. 'Ohh, its more expensive but I get 1.5x as much. Ohhh that makes sense'. Consumer might leave with nVidia.

ATi's current counter-strategy; Average Joe customer walks into store. Consumer sees ATI with low price tag. Consumer notices ATI box says DirectX 11. 'Ohh, 1GB just like the nVidia but it says DirectX11 on the box. That must be better than the nVidia box that says Directx 10 with is also 1GB? and its cheaper? Sweet'. Consumer leaves with ATI.

Really, that's the way most average-joes who buy an aftermarket video card shop. The other method is 'call my friend Bob and/or Bob's son who knows about PCs and ask what is best'. Honestly, if you can give good price:performance ratios, especially in the low-budget ranges, you'll get a lot of business as while nVidia might have the 8800 Ultra or the top of the high-end range. When Bob goes to recommend his friend a video card at the $100.00 range, he doesn't say buy an nVidia cause they have the 8800 Ultra unless he's an nVidia supporter. Hopefully, he would say well at $100.00, its kind of hands down the best model is the _____ and list the model with the best performance:price ratio. If he's an ATI fan, he'll probably mention the ATI card no matter what. The larger you can grow your market share, when you grow it by providing a equal or better product at cheaper costs, the faster your market share should grown on its own from reputation/word of mouth/reviews/recommendations from 'tech-guys'.
 
No "average joe" I know buys individual components. Not to rain on your reasoning, which I think is fairly sound, but in my experience most 'average joes' don't even know what a graphics card is.

I agree with you on most parts. But I've always found ATI's Catalyst a bit sluggish to use and unstable at times. Of course Nvidia's forceware isn't perfect. Even then, I still can't help prefer nvidia drivers. But that's just me of course.
I've never found Catalyst as anything other than working perfectly fine. Of course, I don't use it much, but when I do open it, it's pretty quick, and pretty responsive.
 
No "average joe" I know buys individual components. Not to rain on your reasoning, which I think is fairly sound, but in my experience most 'average joes' don't even know what a graphics card is.

i was an average joe 7years ago... walked into bestbuy and got me a geforce3 ti200. cuz it had better "specs" knew nothing of benchmarks back then till i bought the card and started benchmarking
 
I know several people that believe ATI cards don't work with TWIMTBP titles :)
 
No "average joe" I know buys individual components. Not to rain on your reasoning, which I think is fairly sound, but in my experience most 'average joes' don't even know what a graphics card is.


I've never found Catalyst as anything other than working perfectly fine. Of course, I don't use it much, but when I do open it, it's pretty quick, and pretty responsive.

I've seen a hundred times where people who know sqaut about computers will go out and get some new "awesome" game (usually Sims 3 or WoW) only to find out their computer can't run it because Intel graphics can barely handle Solitaire. They will run out and buy a new video card at the recommendation of whatever techie they know (even if that's just the Best Buy drone). Sure, they're not going to be buying $500 video cards.. yet.. but they will be following the recommendations of someone who might prefer one manufacturer over the other. If AMD can get a good solid reputation going, then there's more of a chance that the Best Buy drones and techie friends are going to be pushing AMD cards to their customers/friends.
 
Why? hehe. He's kind of right but he forgot ATI's other strategy.

Average Joe consumer walks into store. Consumer sees Wii for $199.99. Consumer sees ps3 for $599.99. Consumer leaves with Wii.

Consumer walks into store. Consumer sees ATI with low price tag, comparatively and both boxes say '1GB' or '512MB'. Huh, must be same speed but ATI is cheaper by ___ dollars. Consumer leaves with ATI.

Nvidia's counter-strategy; Next week consumer walks into store. Consumer sees ATI with low price tag. Consumer sees nVidia box says '640mb' or '768mb' and looks at ATI's box saying 512MB. 'Ohh, its more expensive but I get 1.5x as much. Ohhh that makes sense'. Consumer might leave with nVidia.

ATi's current counter-strategy; Average Joe customer walks into store. Consumer sees ATI with low price tag. Consumer notices ATI box says DirectX 11. 'Ohh, 1GB just like the nVidia but it says DirectX11 on the box. That must be better than the nVidia box that says Directx 10 with is also 1GB? and its cheaper? Sweet'. Consumer leaves with ATI.

Really, that's the way most average-joes who buy an aftermarket video card shop. The other method is 'call my friend Bob and/or Bob's son who knows about PCs and ask what is best'. Honestly, if you can give good price:performance ratios, especially in the low-budget ranges, you'll get a lot of business as while nVidia might have the 8800 Ultra or the top of the high-end range. When Bob goes to recommend his friend a video card at the $100.00 range, he doesn't say buy an nVidia cause they have the 8800 Ultra unless he's an nVidia supporter. Hopefully, he would say well at $100.00, its kind of hands down the best model is the _____ and list the model with the best performance:price ratio. If he's an ATI fan, he'll probably mention the ATI card no matter what. The larger you can grow your market share, when you grow it by providing a equal or better product at cheaper costs, the faster your market share should grown on its own from reputation/word of mouth/reviews/recommendations from 'tech-guys'.


So sad, but so true. I'm building custom rigs for the Folding Project and my #1 question isn't about the graphics card type or performance, but how much RAM is on the card.

"Dude, sweet machine. how much graphic ram is it?"

"I wanna get a 1gb card. what the $?"

"I dont want that graphics card in it now. I have an uberfast 8800gt 1Gigibyte card now" (He doesn't want the GTX 260 card because it doesn't have 1GB...lol)

On average, about 9 out of 10 people who call me about the custom rigs have no clue. They just see the numbers. If I was unethical, I could just sell them a $40 9400GT 1GB card. :rolleyes:
 
Nvidia might as well shift their entire production line to Tesla then, because they aren't going too find many suckers who will pay $100-150 more for roughly the same performance as the 5870.

The 512sp version should be around 500-525 if they want a shot at competing. Otherwise myself and most others will be trying out the 5000 series or waiting on the 6000 series. I might even rebuild my computer around an i7 with the money I save.

the 448sp version will be a much faster card than the 5870 for games that use heavy tessellation effects.

even with normal games it should be at least 25% faster than the 5870.
 
the 448sp version will be a much faster card than the 5870 for games that use heavy tessellation effects.

even with normal games it should be at least 25% faster than the 5870.


you really have no idea if that is true or not, since NV is using their shaders to do tessellation, there could be a significant performance hit to other functions that the shaders normally perform.

when you say it should, doe it mean it should according to you? or facts that are somehow known to you and not everyone else :p?

I think it's just hopefulness without anything to say otherwise
 
you really have no idea if that is true or not, since NV is using their shaders to do tessellation, there could be a significant performance hit to other functions that the shaders normally perform.

So is AMD. Do people really still not know how DX11 tessellation works?
 
the 448sp version will be a much faster card than the 5870 for games that use heavy tessellation effects.

even with normal games it should be at least 25% faster than the 5870.

I agree, from the look of the architecture Nvidia will have a tessellation advantage. On the other hand, by the time games start using tessellation, AMD will have had their 6000 series out for awhile, with Nvidia pushing out its 500 series.

As to the 470 being 25% faster than the 5870, I will be surprised to see that. I expect around a 15% edge for the 470. Also, even if it was 25%, $100-150 extra for that is ridiculous. I might as well just go ahead and buy a 5970 at that point.
 
I agree, from the look of the architecture Nvidia will have a tessellation advantage. On the other hand, by the time games start using tessellation, AMD will have had their 6000 series out for awhile, with Nvidia pushing out its 500 series.

As to the 470 being 25% faster than the 5870, I will be surprised to see that. I expect around a 15% edge for the 470. Also, even if it was 25%, $100-150 extra for that is ridiculous. I might as well just go ahead and buy a 5970 at that point.

dont forget, nvidia is already one step ahead of amd with this new dx11 tech. They are already ahead of ati even if amd releases the 6000 series it wont matter cuz nvidia's Fermi design can easily be upgraded for few more generations in no time since it already exists.
 
dont forget, nvidia is already one step ahead of amd with this new dx11 tech. They are already ahead of ati even if amd releases the 6000 series it wont matter cuz nvidia's Fermi design can easily be upgraded for few more generations in no time since it already exists.

They will have to rely on die shrinks to increase scale. Meanwhile, we will have to see what AMD/ATI does with its 6000 series, which is supposed to be a new architecture.

The whole thing is though, at present, Nvidia cannot afford to price itself out of the market, which they themselves have acknowledged when talking about the 200 series and how it was too expensive. We will just have to see if they follow through.

I expect ATI will probably drop the 5870 down to $350-375 to undercut Nvidia and steal more of the market. If they drop the 470 at $500, they better drop the price within a month, or myself and thousands of other loyal Nvidia fans will be going AMD.
 
They will have to rely on die shrinks to increase scale. Meanwhile, we will have to see what AMD/ATI does with its 6000 series, which is supposed to be a new architecture.

The whole thing is though, at present, Nvidia cannot afford to price itself out of the market, which they themselves have acknowledged when talking about the 200 series and how it was too expensive. We will just have to see if they follow through.

I expect ATI will probably drop the 5870 down to $350-375 to undercut Nvidia and steal more of the market. If they drop the 470 at $500, they better drop the price within a month, or myself and thousands of other loyal Nvidia fans will be going AMD.

I can tell you for a fact if the 470 is priced at 500.00 and the 480 is 600.00 or over I might go with some 5870's or I might skip this gen all together and wait for refreshes from both camps. Simple reason is I am not hurting for performance at all and this upgrade to fermi is a want not a need. I also need 2 or whatever I purchase since 1 of either card wouldn't be an upgrade in performance. That being said I will not be bent over a barrel and handled with no lube just for the sake of my desire to upgrade for kicks, no no no. Sorry Nvidia. :mad:
 
1. Programmable Hull Shader - determines LOD and converts HOS representation from one format to another if necessary.
2. Tessellator - splits the higher order patch into triangles based on LOD computed above.
3. Programmable Domain Shader - creates and positions the new vertices using Hull Shader and Tessellator output and doing lookups into a displacement map.

The Hull and Domain shaders run on the shader core on both Nvidia and AMD hardware. The Tessellator is dedicated logic on both Nvidia and AMD hardware.
 
Wow, excellent article. Anand really has the right connections. Doesn't it deserve its own thread though?
 
Nice article and good read! Very glad for ATi...will be buying next gen. I still haven't forgiven NV for the FX series, but I bought their later cards bc they were simply better. Keep it up ATi! :cool:
 
Nice article and good read! Very glad for ATi...will be buying next gen. I still haven't forgiven NV for the FX series, but I bought their later cards bc they were simply better. Keep it up ATi! :cool:

I find this statement a little odd. Due to your cheer leading for ATI, I take it you've forgiven them for R600? Have you forgiven Intel for Netburst? Will you forgive AMD for K10? Every company is going to have a bum product from time to time. These things happen and holding a personal grudge against a company over a lackluster product is pretty childish.
 
I think I was running a X800 Pro during the FX release. Got it when my 9700 Pro died after much service. :(
 
I find this statement a little odd. Due to your cheer leading for ATI, I take it you've forgiven them for R600? Have you forgiven Intel for Netburst? Will you forgive AMD for K10? Every company is going to have a bum product from time to time. These things happen and holding a personal grudge against a company over a lackluster product is pretty childish.

Good reply there:D
 
I think I was running a X800 Pro during the FX release. Got it when my 9700 Pro died after much service. :(


Didn't go for X800 tho I wanted too. I had a 9500 Pro softmodded from 4 pipes to 8 to make it a virtual 9700 Pro. I tried an FX 5900 at it was meh for UT2004...no better than then my 9500 Pro. I was new to hardcore PC gaming at the time, so that card rocked for the money. It cemented my fanboi leanings for ATi, tho my Ti 4200 was a close second.:D
 
yea FX ran against the 9x00 line from ATi which just kicked the shit out of the FX5x00 cards
Xx00 line was when the 6x00 line from NV came out
 
that was also right at the time we went to pciex from agp 8x, furthering the confusion. now i cant remember if i went from an agp 6800gt to a pciex x850xt or from an agp x850 to a pciex 6800gt. i think it was the latter because i had 6800gt sli, and i dont think sli ran on agp. no i am sure it didnt run on agp. there were no multi agp slot boards, except for 1 or 2 off the wall boards there were not meant for 2 graphics cards.
 
I find this statement a little odd. Due to your cheer leading for ATI, I take it you've forgiven them for R600? Have you forgiven Intel for Netburst? Will you forgive AMD for K10? Every company is going to have a bum product from time to time. These things happen and holding a personal grudge against a company over a lackluster product is pretty childish.

That's why they make ice cream in vanilla, chocolate and strawberry. :D
 
Guys, I don't remember where I saw it, but is it already known what connections Fermi will have?
Will that be two DVI and one hdmi?
 
Guys, I don't remember where I saw it, but is it already known what connections Fermi will have?
Will that be two DVI and one hdmi?

I don't know for sure, but it will prolly be the standard 2 x DVI-d and 1 x S-video. They may go the 1 x DVI-d and 1 x HDMI route though. Not much point in them putting 2 x DVI-d and 1x HDMI on one card as long as they include the DVI to HDMI adapter that most cards from both companies have been shipping with for a while. Fermi is going to be hot supposedly, they prolly need as much room as possible for the fan exhaust.
 
There are other options, you know...

Nvidia is allowing these new SLI modes on existing cards, so you could get two MUCH CHEAPER Nvidia cards and run Nvidia's triple-monitor solution that way. Getting two cheaper Fermi-based cards will be a possibility at some point in the future as well.

There's also the Matrox TripleHead2Go (and the recently released Mview), which split one DVI or DisplayPort connection into 3. This would allow you to run three monitors with a single GTX470 or 480. This also works on existing Nvidia hardware.

There's also SoftTH, which would let you use a single GTX470 or 480, with a secondary card like an 8400GS, to drive triple-monitor. This also works on existing Nvidia hardware.

Or you could just buy a single 5850 ...
 
I wonder if NV will support one monitor per card in triple SLI for Nfinity? That would be a high horsepower config for sure, if they support it.
 
I wonder if NV will support one monitor per card in triple SLI for Nfinity? That would be a high horsepower config for sure, if they support it.

Would be slow as hell. There would be wwwaaayyyy to much latency involved getting all 3 cards to sync up, especially since 3D surround is being done in software.
 
Back
Top