-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
Fermi March?
- Thread starter straxusii
- Start date
Associate
- Joined
- 10 Nov 2009
- Posts
- 441
- Location
- Southampton
looks like Fermi will be relased around same time as 58XX card are
nope didnt need fixing. 58XX have had to softess launch ever
looks like Fermi will be relased around same time as 58XX card are
i could have sworn that i have had one for over 2.5 months....
Caporegime
- Joined
- 18 Sep 2009
- Posts
- 30,561
- Location
- Dormanstown.
Me too.
fixed it for u again..
lol. I'm waiting for a 5970 to come down a bit to buy on as well as when someone can stock them. but love how ATI's low stocks is nvidia fault also?
Caporegime
- Joined
- 18 Oct 2002
- Posts
- 33,188
I really wish people would stop talking about how fermi wasn't really designed for gaming, it was flat out. It has almost nothing over a 5870 that makes it a gpgpu monster, almost none of the die offers things ATi don't, nor things Nvidia/ATi offered last gen.
Their architecture is almost identical to the last gen, bar a small amount of die space to increase its double precision power, each shaders cluster, module, whatever is incredibly similar to last gen. THe massive majority of the die is shaders, rops, memory controller and the like, you're talking 1-2% of the die space adding minor extra functionality. They've only marketed it as such so far because they've been WELL below clock speed targets, gpgpu that doesn't matter, 10% faster will be paid for by people who need them, 10% faster than their last card would be awful in the graphics side of the market. You can advertise an underclocked non final gpgpu, not a gpu, its that simple. They've been talking about the only thing it can do, even Nvidia aren't dumb enough to design a card for 2% of their business at the expense of 98% of their business and 99.9% of their profits.
Its a gpu , the gpgpu market was worth an entire $80 in REVENUE, not profit last year for Nvidia, its maybe, if lucky a few percent of their business, they can't offset the graphics side cost due to gpgpu profits, there aren't any.
They've made a gpu first and foremost, and added as many tiny things as they can to bump up gpgpu performance, and realistically thats only the double precision power.
iT being late is bad for us all, and the gpgpu market isn't going to pull them out of any trouble they have, its a non existant market that they are spending as much in marketing and software developement in than they can make in profit.
As for the date, its been widely reported for a month that the March date is what they told their less favoured customers, their more important customers who they have to tell the truth to, have been told May, and supposedly the March date told to guys who are less important has slipped to April now.
Their architecture is almost identical to the last gen, bar a small amount of die space to increase its double precision power, each shaders cluster, module, whatever is incredibly similar to last gen. THe massive majority of the die is shaders, rops, memory controller and the like, you're talking 1-2% of the die space adding minor extra functionality. They've only marketed it as such so far because they've been WELL below clock speed targets, gpgpu that doesn't matter, 10% faster will be paid for by people who need them, 10% faster than their last card would be awful in the graphics side of the market. You can advertise an underclocked non final gpgpu, not a gpu, its that simple. They've been talking about the only thing it can do, even Nvidia aren't dumb enough to design a card for 2% of their business at the expense of 98% of their business and 99.9% of their profits.
Its a gpu , the gpgpu market was worth an entire $80 in REVENUE, not profit last year for Nvidia, its maybe, if lucky a few percent of their business, they can't offset the graphics side cost due to gpgpu profits, there aren't any.
They've made a gpu first and foremost, and added as many tiny things as they can to bump up gpgpu performance, and realistically thats only the double precision power.
iT being late is bad for us all, and the gpgpu market isn't going to pull them out of any trouble they have, its a non existant market that they are spending as much in marketing and software developement in than they can make in profit.
As for the date, its been widely reported for a month that the March date is what they told their less favoured customers, their more important customers who they have to tell the truth to, have been told May, and supposedly the March date told to guys who are less important has slipped to April now.
Last edited:
I quite enjoy being a member of the elusive 5xxx series club. We should get t-shirts made.
Caporegime
- Joined
- 8 Jul 2003
- Posts
- 30,080
- Location
- In a house
