January 2009: GT 100m series announced:
http://www.notebookreview.com/default.asp?newsID=4785
June 2009: GT 200m series announced:
http://hothardware.com/News/NVIDIA-Announces-New-GeForce-200M-Series-GPUs/
December 2009: GT 300m series annouced:
http://www.notebookcheck.net/Newsentry.153+M5a2c3b7fb78.0.html
Should we expect Mobile Fermi around June 2010?
What do you think and why?
-
H.A.L. 9000 Occam's Chainsaw
Nope, because it just won't be efficient enough to use in a notebook in that time frame, if it even ever makes it into desktops.
-
-
H.A.L. 9000 Occam's Chainsaw
-
you get GT 100m-GT200m- GT300m and so on but did you actually get any significant performance increase? Nah.
-
If I not mistaken, we havent got the true GT200 architecture based mobile GPU yet right?
-
masterchef341 The guy from The Notebook
the 100m, 200m, and 300m series are not all actually updates.
the 100m series is the same as the 9000m series (renamed to 100m, for kicks). the 200m is based on the newer stuff, the desktop 200 series. the 300m is the same as the 200m series (renamed, again, just for kicks).
the next set of cards is going to be the 400 series, which is also known as fermi. june is a bit optimistic in my opinion. maybe august. -
masterchef341 The guy from The Notebook
i think the 200m and 300m series are actually based on the desktop GT200 series, not the g92 / 9000 parts.
however, they are crippled compared to their desktop parts- the number of cores and clock speeds haven't gone up much from the 9000 series, but the feature set matches the GT200 parts. so, unfortunately, the performance isn't leagues better. -
H.A.L. 9000 Occam's Chainsaw
I can confirm that my G210m is a 40nm GT218 core... so the GT architecture is in today's notebooks. They are just a lot more efficient than the 9000 series, power wise and performance wise.
-
Nothing has happened worth noting since the 8000 series was released, just been minimal overclocks and rebadging.
-
-
We already have GTS 360M. GTX 380M will continue the pattern, with a bigger slice of SP's than any other 40nm GT200 part thus far. As the information hasn't yet been disclosed, we don't know the true specs, aside the fact it will be GT200 based, 40nm, and DX10.1. DX11 is being reserved for Nvidia's 400 series.
-
Next round for Nvidia has to be rebadged 200 series. As said before, FERMI GF100 barely runs on desktop alone. Seen a few reviews where special case, and cooling system with upgraded power source were needed.
GF100 is sucking over 250 watts of power. It's the most inefficient computer hardware probably ever made for consumer use. It's downright pathetic. -
I doubt we are going to get mobile ferni till 2011.... They can't even make the desktop one by June... and not to mention excessive power consumption... I doubt there gonna be any mobile directx 11 parts from NVDIA till 2011...good thing for me that i'm buying ATI...NVDIA nver again..
-
I think it will be 380M announced at Cebit and available somewhre in June. Maybe it will be 256bit with GDDR5. And then Fermi will come... end 2010, beginning 2011.
-
-
Another interesting point of note: the GTX 260M, 280M, and 285M are 55nm fabricated while the rest are 40nm. -
GTX 2xxM isn't based off G92b, it is G92b.
Semantics aside, Nvidia is admittedly digging their own grave. This isn't exactly helping us as consumers, as Ati is having a runaway success with their 5000 series, and we will unfortunately never see prices as low as we had with the 4000 series. -
If they can graft GDDR5 and 10.1 onto the 92b then it's absolutely possible to add DX11 features. Up the shader count a bit, keep the DDR5 and you have something you already know works @ 40nm.
Course right now Nvidia is spending all their time and $ on Fermi. So sadly we may not see DX11 mobile chips for a long time. As always, hoping I'm wrong. -
Alexrose1uk Music, Media, Game
Also nice post in that other thread, I was going to respond to your points but I've lost the thread(I cant remember which thread it was in and it wasnt the ones I looked in haha).
Either way, dont expect a mobile Fermi any time soon, they're having enough trouble with yields and the desktop release, meaning it is a long way from laptop consumption.
The next chip will likely be GT200 based, how powerful and fast it'll be though is anyones guess until we see the figures. -
You are kidding right? If ATi could do GDDR5 256 bit don't you think they would have done it? With Nvidia's miserable record of power consumption and temperature management in mobile form, how do you think they will be able to release a GDDR5 256 bit that doesn't crash any laptop that uses it?
The crappy Go7700 that was made to be more energy efficient and run cooler than the 7800 still ran at 90C. The HD5870 OVERCLOCKED hasn't reached over 76C for me. Nvidia should just get out of the mobile market until they figure out how to stop making GPUs that fry, literally.
-
Nevermind, ignore this post.
-
Might not be fun to see 1 company dominate the Mobile market...
-
Howitzer225 Death Company Dreadnought
If one does indeed come out, I'll bet it will be a thoroughly nerfed Fermi GPU.
-
-
No, what you on about, i want more G92b core! MOAAAW!
But seriously, futures looking grim for Nvid in the mobile sector, hope they do something to pull themselves through!
Personally i still love voodoo 3dfxhaha.
-
All I know is that the HD5870 is incredible in the G73.
I have overclocked the GPU to 800 Core/1100 Memory and I have yet to exceed 76C.
And not only that I don't hear the fans screaming, it remains quiet. I play NFS Shift on my lap on my bed when I'm bored and it doesn't get hot underneath.
When I browse the web with FireFox with my previous G1 with an Nvidia Go7700, Nvidia GPU ran so hot I thought it would scorch my balls and be hospitalized. Not kidding. -
-
-
SoundOf1HandClapping Was once a Forge
-
There isn't too much to talk about here, because right now not even nVidia can keep the story straight with desktop Fermi cards. Let alone notebook cards, of which there is no official information or even half-way decent speculation to be found anywhere.
nVidia dropped the ball this round, plain and simple. All you can hope for is that it does not kill nVidia, and that Fermi's introduction into the market will actually be a competitive one. -
You would think that. But laptops using the Go8400 had a recall regardless of the laptop brand because the 8400 was running too hot and bricking itself. So yeah, Nvidia has been failing with mobile GPUs for a while.
A mobile GPU should not require someone to buy a $3,000 Clevo that is 3 inches thick and weighing over 10 lbs to have good cooling. That's inexcusable.
-
Sager NP8690
15", $1569
7.38 LBs with Battery Pack
14.75" (w) x 10" (d) x 1.65" ~ 2.0" (h)
Asus G73
17", $1549
8.5lbs with 8-cell battery
16.33"x12.59"x1.8" (W x D x H) -
This reminds me of those times when the FX cards came out on the market...
-
Of course, there are many other factors... If only business were that simple. -
My thinking is several points on why they'd add DX11 to the 3xx and rebadge it as 4xx.
- It fits their style, add a couple features and add a number series.
- We all know Fermi won't make it to the mobile world for years (die shrinks) but there are design elements that could be reused.
- Tesselation is actually a rather simple item, as it is not a very complex Tesselator. Note that a derivative of the feature has been on AMD chips for years.
- Tesselation and Shader 5.1 are features which add serious performance and visual appeal when well used. That means the competition could wave that in Nvidia's face for how long? A year? That won't sit well with them.
- 40nm and GDDR5 were attained rather quietly, not a big fanfare aside from a 3xx series announcement.
- Displayport and finer power gradations could be added for next to no silicon space.
- The 3xx series is already a compact chip on which to base these improvements, so size is likely to stay compact.
There are more thoughts but that's the list for now -
Nice try.
Although I think you are the one trolling. Ironic.
Should we expect Mobile Fermi around June 2010? (It's always been 6 months interval between GT100m->GT200m->GT300m)
Discussion in 'Gaming (Software and Graphics Cards)' started by lgmcben, Feb 21, 2010.