• We should now be fully online following an overnight outage. Apologies for any inconvenience, we do not expect there to be any further issues.

Nvidia GF110?

Toggle sidebar Toggle sidebar
T

tviceman

Diamond Member
Mar 25, 2008
6,734
514
126
www.facebook.com
The rumor of Nvidia making a GF110 part is not new. However, this sounds like a heresay article. It's vague and doesn't even give a time frame, and the math of a 48 shader cluster doesn't divide into 512 shaders so the part can't just be a GF104 extrapolated to 512 shaders - it has to be a different grouping.

EDIT: Ooops I had to reread that. If they are regrouping to match the gt200 sets then that would be a big bump in ROP's. Are the polymorph engines grouped as 1 per cluster? If so, that would also boost the tessellation power. If true, it could be a monster of a performer.
 
Last edited:
Gloomy

Gloomy

Golden Member
Oct 12, 2010
1,469
21
81
What's most interesting is that they mentioned making the existing chips more efficient. I'd be most interested in 480SLI after something like that happens (unless AMD's CF is up to par).

I'll skip the GF110 if Nvidia makes another attempt at bringing the surface of the sun to my house ;)
 
MentalIlness

MentalIlness

Platinum Member
Nov 22, 2009
2,383
11
76
If this chip is real, I have a feeling it will be like the 8800 Ultra was. Sky high price and rare.
 
S

SlowSpyder

Lifer
Jan 12, 2005
17,305
1,002
126
If I am reading this correctly, Nvidia is calling the 512SP part they announced in January for the GF100 as a new part? And/or they are adding shaders or other parts of the GPU from the GTX2x0 family? Fermi is that bad (for gaming) that they are going backwards? I don't know that I buy what that article is saying.

At any rate, it's nice to hear some Nivdia rumors. Everything points to the Radeon 6xxx dominating for a while, so it's good to see Nvidia starting to stir. Things have been sort of quiet on the Green side it seems.
 
O

OCGuy

Lifer
Jul 12, 2000
27,224
37
91
From the current rumours, it sounds like GF100 and it's derivatives already compete with 6XXX.
 
Keysplayr

Keysplayr

Elite Member
Jan 16, 2003
21,219
55
91
SlowSpyder said:
If I am reading this correctly, Nvidia is calling the 512SP part they announced in January for the GF100 as a new part? And/or they are adding shaders or other parts of the GPU from the GTX2x0 family? Fermi is that bad (for gaming) that they are going backwards? I don't know that I buy what that article is saying.

At any rate, it's nice to hear some Nivdia rumors. Everything points to the Radeon 6xxx dominating for a while, so it's good to see Nvidia starting to stir. Things have been sort of quiet on the Green side it seems.
Click to expand...

You asked me before if I really believed Nvidia has an answer to the 6xxx series.
Now I ask you, how can you not think this to be true? Why wouldn't they?
Fermi was late, but you guys can't seem to get your mind around the fact that everything else AROUND Fermi doesn't automagically become late as well. Just consider the hypothesis as if Fermi launched on the same day that the 5xxx series did. That should help you understand the time frame lapse that some of you seem to be experiencing.
 
D

dangerman1337

Senior member
Sep 16, 2010
385
45
91
I woudln't be surprised if the GF110 is based upon the GF104 (332nm i think?) but 50% larger with some further tweaks getting out much more perfomance per watt which could add up altogether with a possible 50% boost. It could be called the GTX 580/570, then a fully unlocked GF104 (or modified even more) and released as the GTX/S 560, not sure about the lower end though.

Though i think 512 bit bus is BS due to the fact that GF100 is already hot and adding extra memmory bus is going to make it larger and hotter. But with a upscaled and modified GF104 we could be looking at 576 shaders and 96 TMUs but with a 384 mem bit bus (GF100 has 64 and was a downgrade in TMUs compared with the GT200/b).
 
Last edited:
L

Lonyo

Lifer
Aug 10, 2002
21,938
6
81
Keysplayr said:
You asked me before if I really believed Nvidia has an answer to the 6xxx series.
Now I ask you, how can you not think this to be true? Why wouldn't they?
Fermi was late, but you guys can't seem to get your mind around the fact that everything else AROUND Fermi doesn't automagically become late as well. Just consider the hypothesis as if Fermi launched on the same day that the 5xxx series did. That should help you understand the time frame lapse that some of you seem to be experiencing.
Click to expand...

Both ATI and NV said at the beginning of the year they would aim to refresh, for ATI it was "second half" and for NV it was "late 2010".
NV also said (before GF100 launched) that other products weren't delayed due to that delay.

One problem with all this is that NV parts have been "delayed" compared to where early rumours situated their release, so it might not be too surprising if the Fermi refresh also slipped a bit (maybe to 2011).

Part of the problem I have with the idea of NV suddenly massively increasing performance though, is the fact that they don't have much chance to make a bigger die, so they will have to work by making their architecture more efficient in terms of 'speed per transistor'.

ATI seem to have managed that, but we knew they had an often 'inefficient' architecture in many ways, and significant weakness in many areas of DX11, so it will be interesting to see how much of an increase NV can get without increasing die sizes (since their is already very large, and they might not want to go too much bigger).
One thing in their 'favour' is that their current shaders seem much weaker in many situations than the GT200 shaders (e.g. GTS250 with 128 shaders beats 192 shader GTS450 in many situations), so maybe they will be able to make some tweaks ot make it a more effective architecture, but who knows.
 
netxzero64

netxzero64

Senior member
May 16, 2009
538
0
71
Lonyo said:
Both ATI and NV said at the beginning of the year they would aim to refresh, for ATI it was "second half" and for NV it was "late 2010".
NV also said (before GF100 launched) that other products weren't delayed due to that delay.

One problem with all this is that NV parts have been "delayed" compared to where early rumours situated their release, so it might not be too surprising if the Fermi refresh also slipped a bit (maybe to 2011).

Part of the problem I have with the idea of NV suddenly massively increasing performance though, is the fact that they don't have much chance to make a bigger die, so they will have to work by making their architecture more efficient in terms of 'speed per transistor'.

ATI seem to have managed that, but we knew they had an often 'inefficient' architecture in many ways, and significant weakness in many areas of DX11, so it will be interesting to see how much of an increase NV can get without increasing die sizes (since their is already very large, and they might not want to go too much bigger).
One thing in their 'favour' is that their current shaders seem much weaker in many situations than the GT200 shaders (e.g. GTS250 with 128 shaders beats 192 shader GTS450 in many situations), so maybe they will be able to make some tweaks ot make it a more effective architecture, but who knows.
Click to expand...
it just gave us an idea on what may come next, nothing is certain until nvidia itself has spoken something but the idea stirs things a bit..

the nuclear powered GF100 chips is just too much for a gamer level of person, it will be good to see an improved and more efficient fermi chips though =)
 
D

dangerman1337

Senior member
Sep 16, 2010
385
45
91
netxzero64 said:
it just gave us an idea on what may come next, nothing is certain until nvidia itself has spoken something but the idea stirs things a bit..

the nuclear powered GF100 chips is just too much for a gamer level of person, it will be good to see an improved and more efficient fermi chips though =)
Click to expand...

IIRC the GF100 were more designed for CUDA/HPC stuff while the GF104 was more game orientated, if the GF104's die size is 332mm2, then its likely a 50% die increase with possibly futher optimization/alteration could increase the perfomance by a considerable amount to counter cayman since cayman from the looks of it is going to be somewhat near 2900XT size (less though and hopefully has 384 bit bus) though i bet that tje GF110 has a good chance of outperfoming it if the GF110's arch is improved over GF104/100 and has a good amount TMUs, shaders and clocks but we'll see.
 
S

SlowSpyder

Lifer
Jan 12, 2005
17,305
1,002
126
Keysplayr said:
You asked me before if I really believed Nvidia has an answer to the 6xxx series.
Now I ask you, how can you not think this to be true? Why wouldn't they?
Fermi was late, but you guys can't seem to get your mind around the fact that everything else AROUND Fermi doesn't automagically become late as well. Just consider the hypothesis as if Fermi launched on the same day that the 5xxx series did. That should help you understand the time frame lapse that some of you seem to be experiencing.
Click to expand...


If I remember correctly, I asked if you honestly believed Nvidia had a 'KO' for the Radeon 6xxx parts, that is how I remember it being phrased.

Here is my thought. Nvidia launched a big, power hungry chip on 40nm with plans to move to 32nm. Those plans had to be cancelled, so they have to work with 40nm, but they already pushed some of their limits. They can't go bigger, AMD can. They can't go more power hungry, AMD can. And, Fermi is a brand new architecture, I doubt they had another brand new architecture in the works for a 'mid-life kicker'... on the other hand AMD is introducing a new architecture (maybe not totally new, but cerainly revised and very likely improved).

So where does Nvidia go? I am sure they are not sitting still, but I do wonder how they are going to counter. I can see a 512SP Fermi at 750MHz - 800 MHz on a respin or if 40nm has matured for them enough. But will it be enough? If AMD's $250 card is only 240mm2 (I think that's what the rumor is) and is matching the GTX470 and 5870, you have to wonder what their 'big gun' single GPU will do with another 33% - 50% transistor count (just a number I pulled out of my rear, I have no idea how big or small Cayman will be, but we know it will be more and it will be faster). Not to mention their new dual GPU part, Nvidia hasn't even matched the 5970 yet.

I don't know how things will go, but my best guess is this will be sort of like with the 5870 launched, AMD will be unmatched for at least a while. I guess we'll have to wait and see what Nvidia does.

If you have something you can tell us that you know about a product to counter AMD's 6xxx, please share. :)
 
T

tijag

Member
Apr 7, 2005
83
1
71
OCGuy said:
From the current rumours, it sounds like GF100 and it's derivatives already compete with 6XXX.
Click to expand...

This isn't really true. AMD's 230mm2 sized part will compete with nvidia's 360mm2 sized part in price and performance.

If nvidia comes out with something new that has better performance, amd has all kinds of room to lower prices and remain competitive.

Cayman XT is probably just a bit bigger than GF104. My guess is that the performance between the 2 won't even be close.
 
RussianSensation

RussianSensation

Elite Member
Sep 5, 2003
19,458
765
126
netxzero64 said:
the nuclear powered GF100 chips is just too much for a gamer level of person, it will be good to see an improved and more efficient fermi chips though =)
Click to expand...

You realize your 875 overclocked GTX460 consumes MORE power than a GTX470 (GF100), and you are only 30-40 Watts away from a GTX480??? :whiste: I do agree that they need to improve performance/watt so that they can make faster chips.

As far as this rumour goes, it's too hard to digest it right now. They could slap 2 x GF104s together to come up with a 512-bit card. Or the rumour can be talking about a new larger GF104 chip. The "renaming" of GF1xx to GTX580.....
 
Last edited:
J

Janooo

Golden Member
Aug 22, 2005
1,067
13
81
Gloomy said:
What's most interesting is that they mentioned making the existing chips more efficient. I'd be most interested in 480SLI after something like that happens (unless AMD's CF is up to par).

I'll skip the GF110 if Nvidia makes another attempt at bringing the surface of the sun to my house ;)
Click to expand...
Watch out. Some people here would try to persuade you that you got it wrong. The performance is more important than power consumption. :)
 
You must log in or register to reply here.

TRENDING THREADS

COMPANY
RESOURCES
FOLLOW
Top Bottom