[FUD] GT300 is codenamed Fermi

[Kai Robinson]

I see the personal insults have started flying.

I'm an AMD fanboy am i?

Well that would explain why i've owned a GeForce 3 Ti, GeForce 6600GT, GeForce 7600GT and a GeForce 8800GT then, would it?

My basis for the 5870 choice at the moment, is 1) performance far better than both of my 4850's with a single gpu solution (and therefore fewer driver issues) 2) driver performance is due to improve 3) the price is appealing

Currently, idle power is far lower than any current offering than i can see which can only translate into lower temperatures. As for my 4850CF setup generating noise, they're both sat at 56-58*c and 35% fan speed. They're pretty quiet and definitely quieter than my evga 8800GT 512mb, which, at idle, would sound loud enough to irritate my ears.

If you *read* my posts, you'd see that i choose cards on more than just the sticker basis. And as for calling me a retard - who the hell do you think you are?

[Nightcover]

They're pretty quiet and definitely quieter than my evga 8800GT 512mb, which, at idle, would sound loud enough to irritate my ears.

If it's a reference cooler, I don't know what you are talking about. The 8800GT cannot be heard at around 30% and it never goes above 49% at load. (at least mine don't.)

[Bodkin]

Yeah, my refernece 8800gt were great, no noice noticible over watercooling pump

[flopper]

If one guy tells you his foot has a knife in it and it bleeds, are you telling him he dont have a knife in his foot and it dosnt bleed?
I love videocards, especially new ones, and they all even when same manufacture are different, some fans has pitching noise, some people have really good hearing for subsonic sounds.
Its basically impossible to use ones own preference to a card and how it affect people around it, kai tells people his 8800GT sounds worse than his 4850 crossfire, I rather belive him any day than people who dont sleep at his house and room.


didnt a lot of 8800GT broke down due to cheap third manufacture parts?

[saaya]

I'm an AMD fanboy am i?
Well that would explain why i've owned a GeForce 3 Ti, GeForce 6600GT, GeForce 7600GT and a GeForce 8800GT then, would it?

ahhh come on admit it, you only bought those to trash them and rma them over and over, trying to hurt nvidia as much as you can ^^

seriously, can we get back on topic?
what does 8800s fanspeed and noise have to do with gt300?
i dont see any connection at all

so the latest launch date rumor i heard is november 27?
that means 58 days to go for gt300... supposedly...
yet still no hint of a leak...

[Bodkin]

November 27, sounds like they will try and get at lest one batch out for Christmas. Any source on that Saaya?

[trinibwoy]

Fuad claims Jen Hsun will discuss Fermi at today's keynote at 1pm. Maybe even give a demo.

http://www.fudzilla.com/content/view/15741/1/

We expect that Nvidia CEO Jensen Huang will demonstrate the card at the GTC keynote that should start at 1.00 PM Pacific time or roughly 10 PM Central European time.

Theo posts specs: http://www.brightsideofnews.com/news...6gb-gddr5.aspx

3.0 billion transistors
40nm TSMC
384-bit memory interface
512 shader cores [renamed into CUDA Cores]
32 CUDA cores per Shader Cluster
1MB L1 cache memory [divided into 16KB Cache - Shared Memory]
768KB L2 unified cache memory
Up to 6GB GDDR5 memory
Half Speed IEEE 754 Double Precision

[BenchZowner]

AFAIK:

512 "SPs"
384 Bit
40nm
GDDR5
~245Gbps
Lots of architectural changes
Native C++ execution
Codename: GF100

[RPGWiZaRD]

Theo posts specs: http://www.brightsideofnews.com/news...6gb-gddr5.aspx

How I wish this was accurate! There's much sense in going into such direction at a point like this though, as it really seems the traditional GPU won't live that many more years in it's current state and NVIDIA might start preparing towards a more hybrid CPU/GPU solution now. Of course the first generations won't be very fast excecuting CPU tasks and still get easily dominated by highend CPUs but it would be a start.

[kazuyakun]

http://www.anandtech.com/video/showdoc.aspx?i=3651

Love how Nvidia's engineers reply to the reason for the set-back, "because designing GPUs this big is "ing hard". Wonder if they'll be heat issues like the 280GTX.

[Solus Corvus]

Native C++ execution

What's this supposed to mean? C++ is a compiled language not an interpreted language. So are they supplying a C++ compiler that makes native G300 binary code or developed a way to interpret uncompiled C++ code on the fly?

[SimBy]

Then timing is just as valid, because while Fermi currently exists on paper, it's not a product yet. Fermi is late. Clock speeds, configurations and price points have yet to be finalized. NVIDIA just recently got working chips back and it's going to be at least two months before I see the first samples. Widespread availability won't be until at least Q1 2010.

I asked two people at NVIDIA why Fermi is late; NVIDIA's VP of Product Marketing, Ujesh Desai and NVIDIA's VP of GPU Engineering, Jonah Alben. Ujesh responded: because designing GPUs this big is "ing hard".

http://www.anandtech.com/video/showdoc.aspx?i=3651

EDIT: sorry already posted

[BenchZowner]

AFAIK:

512 "SPs"
384 Bit
40nm
GDDR5
~245Gbps
Lots of architectural changes
Native C++ execution
Codename: GF100

http://www.nvidia.com/content/PDF/fe...Whitepaper.pdf

What's this supposed to mean? C++ is a compiled language not an interpreted language. So are they supplying a C++ compiler that makes native G300 binary code or developed a way to interpret uncompiled C++ code on the fly?

http://www.nvidia.com/content/PDF/fe...Whitepaper.pdf

[Solus Corvus]

http://www.nvidia.com/content/PDF/fe...Whitepaper.pdf

Thanks. That clears it up.

[Farinorco]

What's this supposed to mean? C++ is a compiled language not an interpreted language. So are they supplying a C++ compiler that makes native G300 binary code or developed a way to interpret uncompiled C++ code on the fly?

Yep, I was wondering about exactly the same thing...

EDIT: Hell, I didn't see the BenchZowner's link. I'll read it now...

RE-EDIT: So a few changes to the compute instruction set to allow compiling programs in C++ to that IS.

[xbrian88]

HwMania - Visualizza messaggio singolo - [Thread Ufficiale] nVIDIA Fermi