• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

"Fermi" GF100 performance numbers are out

Status
Not open for further replies.

DeadSkull

Supreme [H]ardness
Joined
Jun 7, 2008
Messages
4,482
I am surprised no one posted this yet.

http://www.guildwarsguru.com/forum/rahjas-freeeeeeee-t10420384.html
http://www.guildwars2guru.com/forum/rahjas-freeeeeeeeeee-t2238.html
http://www.overclock.net/8085586-post389.html

GF100 outperforms ATi's 5870 by 46% on average
GF100 outperforms ATi's 5970 by 8% on average

The GF100 gets 148 fps in DiRT2
The GF100 gets 73 fps in Crysis2
The GF100 gets 82 fps in AvP3

Here are a few more bits of FPS data for you. These were collected at 1920x1200 with 4x MSAA (no SSAA this time) and 16xAF. All settings are on high (perhaps not extreme/highest though, it isn't clear from the included documentation)

Crysis Warhead: 58 FPS
Left For Dead 2: 143 FPS (Judging from this result, this appears to be CPU limited at this point)
Hawx: 127 FPS
World of Warcraft: 46 FPS (this result seems off to me, I assume this is a driver issue)
Fallout 3: 72 FPS (clearly a driver acceleration issue here again)

Numbers look insane but no word on the chip that was used for this testing; if it had all the 512 SPs enabled and running at full speed.
 
The intresting thing is the date, 1st of March, if this is total BS or not it won't be long before we know. Damn, just as Kyle had me psyched about getting a set of 5870s! Oh well back to my original plan, one 5770 for now and wait.:mad:
 
Heres the thing that strikes laughter into me, What if these numbers in the end come up true when fermi launches? I also understand that all that has been said so far is a bunch of data or words with no official source or trusted I should say.

Guess we will have to wait and see :p
 
The guy does claim to be an ex nvidia engineer who worked on Fermi. So...who knows.
 
Last edited by a moderator:
This is hot BS.

Not necessarily. Don't be surprised if these numbers are too far off. This is almost what nVidia HAS to do to regain momentum and its not like that haven't produced amazing stuff when they were down, just like ATI/AMD.

Honestly, I don't get people who boo hoo these stories, BS or not. Would you rather nVidia produce a really crappy product so that AMD can sell 5000s at high prices and rest on its laurels?

Didn't think so!:D
 
Not necessarily. Don't be surprised if these numbers are too far off. This is almost what nVidia HAS to do to regain momentum and its not like that haven't produced amazing stuff when they were down, just like ATI/AMD.

Honestly, I don't get people who boo hoo these stories, BS or not. Would you rather nVidia produce a really crappy product so that AMD can sell 5000s at high prices and rest on its laurels?

Didn't think so!:D

I want nvidia to perfect cold fusion, but lets be realistic.
 
I want nvidia to perfect cold fusion, but lets be realistic.

I don't think that these number are unrealistic, they may be pure BS, who knows. Like I said, much less than this would not be good for nVidia so I see it as a "this better be kind of true or else..." situation for them. Successful companies like nVidia have a way of pulling rabbits out of their hats when you least expect.
 
I don't think that these number are unrealistic, they may be pure BS, who knows. Like I said, much less than this would not be good for nVidia so I see it as a "this better be kind of true or else..." situation for them. Successful companies like nVidia have a way of pulling rabbits out of their hats when you least expect.

I think they are unrealistic in the same way that aliens are unrealistic. Neither exist from what we can tell. We haven't even seen a card yet, much less performance numbers on two games that aren't out yet.

Why would you wait? Why not wait until ATi's 6000 series then? ATi clearly has the superior product at this point, and Nvidia has no product to put that superiority in doubt.

If Nvidia had anything convincingly better at this point, don't you think they'd be showing it? At the very least, to try and stop people from buying ATi cards so close to Fermi's supposed release?
 
I've already stated that Fermi has outperformed the 5970 by 20% in a few games already. This is really old news. And this is coming from a 5870 owner.
 
Why would you wait? Why not wait until ATi's 6000 series then? ATi clearly has the superior product at this point, and Nvidia has no product to put that superiority in doubt.

Why? Because these are the last video cards ever! Better make sure they are good ones.:D
 
I've already stated that Fermi has outperformed the 5970 by 20% in a few games already. This is really old news. And this is coming from a 5870 owner.

Troll poster opinions don't count.

Regarding Fermi, I hope that these numbers are true even though the story is outlandish. Faster Nvidia means ATI price drops.
 
I've already stated that Fermi has outperformed the 5970 by 20% in a few games already. This is really old news. And this is coming from a 5870 owner.


lol believe you.. or believe a forum post on guildwars.. even though both are 100% BS id believe some guildwars forum post over anything you have to say...



until nvidia has something physically out its all a load of crap.. though i wish they would hurry the eff up so i could get a 5870 for under 400 dollars damnit!
 
I want pics, screenshots, validations, not the blatherings of some forum post out there with no evidence.

But if the numbers are true...I'd better start saving...
 
I think they are unrealistic in the same way that aliens are unrealistic. Neither exist from what we can tell. We haven't even seen a card yet, much less performance numbers on two games that aren't out yet.

Why would you wait? Why not wait until ATi's 6000 series then? ATi clearly has the superior product at this point, and Nvidia has no product to put that superiority in doubt.

If Nvidia had anything convincingly better at this point, don't you think they'd be showing it? At the very least, to try and stop people from buying ATi cards so close to Fermi's supposed release?

Unrealistic is not the same as unconfirmed. These numbers are realistic just unconfirmed.

Why wait? Did you see the date? 1st week of March. I GUARANTEE you if I pull the trigger on three 5870s right now that these numbers are LOW! That's just my luck!

The only thing I'm really missing out on is Eyefinity and I pretty much will be building a new rig when the i7-980x launches (i7-980x, 1TB SSD, 24 GB RAM, and the 3 best single GPU cards at the time) so I would like the rig to be nice for at least a day!:D
 
Not necessarily. Don't be surprised if these numbers are too far off. This is almost what nVidia HAS to do to regain momentum and its not like that haven't produced amazing stuff when they were down, just like ATI/AMD.

Honestly, I don't get people who boo hoo these stories, BS or not. Would you rather nVidia produce a really crappy product so that AMD can sell 5000s at high prices and rest on its laurels?

Didn't think so!:D

Except they started developing this card right after they finished releasing the GT200. That means they had no idea when they were building the architecture the 5870 would be anything or they would be late to market or they would have manufactoring problems. Yet according to you they instead of designing a super powerful chip from the begining just decided at the last min to delay it again for another 3 months and during that time double it's power when they struggle to do that over an 18 month cycle. Yeah I'd believe that. :rolleyes:
 
Except they started developing this card right after they finished releasing the GT200. That means they had no idea when they were building the architecture the 5870 would be anything or they would be late to market or they would have manufactoring problems. Yet according to you they instead of designing a super powerful chip from the begining just decided at the last min to delay it again for another 3 months and during that time double it's power when they struggle to do that over an 18 month cycle. Yeah I'd believe that. :rolleyes:

Huh? I never said they did anything. I simply said that I don't think that these numbers are unrealistic because it's almost what the market is expecting.
 
When this is released, fact will remain, I called it. :cool:

Just like your 1-6 month prediction...

I predict Fermi will come out sometime between tomorrow and December 21, 2012, and will perform between 100% slower than a 5870 up to 500% faster than a 5870.

Looks like I called it too...
 
I like how they have tested on Crysis "2" when the game its not even out :D
 
Just like your 1-6 month prediction...

I predict Fermi will come out sometime between tomorrow and December 21, 2012, and will perform between 100% slower than a 5870 up to 500% faster than a 5870.

Looks like I called it too...

WOAH!!! Amazing! You need to help the White House estimate future events!!!
 
When this is released, fact will remain, I called it. :cool:

Well, I saw about 500 news sites saying the same thing long before you even had heard of the name Fermi... I guess they called it ;). It's only a range the size of the broad side of a barn!
 
e0083r.jpg
 
Fergie will be the fastest card ever, no piece of technology will ever surpass it.
 
Well, I saw about 500 news sites saying the same thing long before you even had heard of the name Fermi... I guess they called it ;). It's only a range the size of the broad side of a barn!

Not to mention this is the same guy who didn't know the actual name of the card not long ago.

That said, I don't think the numbers are too unrealistic. Like someone else said, this is about where Fermi needs to be for nVidia to regain some much lost respect. While we don't have any official numbers on price, I think there is little doubt that it's going to be a pricey GPU based on it's size and all the issues that came along with it. nVidia knows where the performance needs to be based on the cost, and that could be why it's taking so long for Fermi to become "ready" as they try to squeeze every last bit of performance they can. I would also expect some disappointing figures for those who enjoy overclocking their cards, atleast with the first couple production batches.
 
Not to mention this is the same guy who didn't know the actual name of the card not long ago.

That said, I don't think the numbers are too unrealistic. Like someone else said, this is about where Fermi needs to be for nVidia to regain some much lost respect. While we don't have any official numbers on price, I think there is little doubt that it's going to be a pricey GPU based on it's size and all the issues that came along with it. nVidia knows where the performance needs to be based on the cost, and that could be why it's taking so long for Fermi to become "ready" as they try to squeeze every last bit of performance they can. I would also expect some disappointing figures for those who enjoy overclocking their cards, atleast with the first couple production batches.

Give the man a cookie! :D This is EXACTLY what I was saying and think. Remember, nVidia has been making claims about how awesome Fermi is AND they are running 6 months behind. If Fermi doesn't come close to this level of performance they would be better off probably waiting until it is, and I think you hit the nail on the head, they working to get it there.

And on top of that, if they have no answer "Eyefinity" they could see a lot of gamers slip away even with the performance crown.
 
Not to mention this is the same guy who didn't know the actual name of the card not long ago.

That said, I don't think the numbers are too unrealistic. Like someone else said, this is about where Fermi needs to be for nVidia to regain some much lost respect. While we don't have any official numbers on price, I think there is little doubt that it's going to be a pricey GPU based on it's size and all the issues that came along with it. nVidia knows where the performance needs to be based on the cost, and that could be why it's taking so long for Fermi to become "ready" as they try to squeeze every last bit of performance they can. I would also expect some disappointing figures for those who enjoy overclocking their cards, atleast with the first couple production batches.

Agree with this 100%... pretty much what I was thinking.
 
Except they started developing this card right after they finished releasing the GT200. That means they had no idea when they were building the architecture the 5870 would be anything or they would be late to market or they would have manufactoring problems. Yet according to you they instead of designing a super powerful chip from the begining just decided at the last min to delay it again for another 3 months and during that time double it's power when they struggle to do that over an 18 month cycle. Yeah I'd believe that. :rolleyes:


odds are they started designing these cards before the GT200.. given the fact that it took almost 4 years of designing the architecture for the ATI HD3k and HD4k series cards.. so id say they pretty much started the designing of the fermi architecture around 9 series release..
 
Nice numbers, need these cards out in the hands of reviewers though. Either way, I will likely be purchasing the GF100 when it is released. :)
 
Status
Not open for further replies.
Back
Top