Nvidia GT200-successor on 22th October?

[Hornet331]

Dude... that's not right actually.
Unless you expect the people who have sources to expose their sources...
What are you expecting ?
Recording a tel conversation or a video recording of a meeting and posting it in public ?

There are two kinds of people...
People who are attention whores, who'd post anything just to get some attention or hits ( like some sites that I'm not going to mention now ).
And... people who have sources and get some real info, and sometimes leak some info because of their excitement and "will" or "love/passion" on helping the PC enthusiasts, etc.

I can easily name some people that are trustworthy and have brought some real info in the past and of course didn't provide what you call "proof" back then.

no offense,

Friendly,
Bill

yes but those people usually dont make money out off this "informations."

[zerazax]

yes but those people usually dont make money out off this "informations."

Agreed

We need to start holding people accountable for spreading rumors then not backing them up if products dont show up - sometimes, i see questionable rumor spreading out there that is worse than Fud or Inq - borderline shilling for people

[T_Flight]

People like you are what I'd call trustworthy, because when you post things they have some basis in fact, and you post things that are truthful. When you do that you don't need any proof, because you have a reputation of posting factual info and people take you at your word.

These other sites have proven just the opposite. If they had official sources, then they've broke NDA's a long time ago and used them up. So what's left? Fabrication and guesses. This stuff about "insider info" and "We cannot expose our source" is just a cover for they don't have any, and are guessing.

If they had a source which in this example would be nVidia or somebody who worked with nVidia and was allowed to release info they'd have no trouble saying "official Souces at nVidia have allowed us to release this latest info", and it would be to their advantage and nVidia's to do so, but that's not what happened. They made up some tall tale about a new GPU refresh that was coming out, and even made up a date which we now know for a fact was a complete lie.

See? It's just like this thread. I held back and just waited this one out. I pretty much figured there was no refresh, because it was so soon after release of the last new gen of GPU's, but I waited to proove it. It's now been proven beyond any doubt. This forum is littered with threads like this and lost in them are the corrections from people with actual first hand info that *can* say something. These threads always have one or two places in common...Fud, or that Inq. I could promise people the world and be right 50% of the time. All I;d have to do is contradict myself 100% of time, and I'd be right 50% and would draw millions to a site and make all kinds of money. There's only one problem...I would'nt want to ruin my word to do it.

There is a BIG difference between something that has real actual info and can be trusted, and places that just make stuff up out of the clear blue. There was no reason at all to make some stuff like this up. What would possess somebody to do stuff like this? Do they not know the damage they are causing? Do they not know the trouble they are causing people?

[Sly Fox]

We need to start holding people accountable for spreading rumors then not backing them up if products dont show up - sometimes, i see questionable rumor spreading out there that is worse than Fud or Inq - borderline shilling for people

Wouldn't that scare off legit people from posting anything out of fear for possibly being wrong in the end? I mean, launch dates and whatnot do get pushed back I'm sure.

I definitely agree certain people's words should be read with a great deal of skepticism. Not going to name names, but I think most people can figure out roughly the type of people

In any case, I think it's good that there are well informed people (like benchZowner) who from time to time post accurate information, and I'd rather not risk losing that.

[Mk]

Word is going around that NVIDIA's 55nm version of the GT200 GPU will arrive next month in the form of replacements for the GeForce GTX 260 and GeForce GTX 280 cards.

We heard that they might be out by November, but this is not something we can confirm at the moment. There should be two products, one to replace GT280 and one to replace GT260.

From the original prognosis this new card might actually end up marginally faster than Radeon 4870 and 4870 X2, but it won’t leave them in the dust which means that Nvidia will have to wait for 40 nm chips in 2009 to turn the situation around.

http://www.dvhardware.net/article30735.html

[Beefy22]

modify date in thread title? ...close thread?

My vote -> Close thread

Just out of mild curiosity, is there any knew information regarding GT200b's core/shader clk, also is it still expected to be a mere shrink of GT200 and nothing more?

[NapalmV5]

so whats going on.. nothing?

no new GTX290 ?

[adamsleath]

who knows?

i would have thought a shrinky would decrease costs...but maybe they waiting for revenues from existing models before committing to next gen production...
total speculation; maybe there's something closely under wraps.

or else it's just sit on current product line-up and milk it for what it's worth.

[wrangler]

If they have the ability to produce these chips on 55nm process (would think that they would have that ability by now) then economically, they absolutely positively must. There is no doubt that they will. Just when.

I also have to ask (apologies for being ignorant if they have) but have they produced any chip @ 55nm? There was also a rumor of 55nm 9800GT's with Hybrid power.....or all Hybrid power 9800GT's would be 55nm.......something to that effect. Don't recall seeing any of those.

[Mk]

found this new pic on the net

[Nedjo]

wow those beast are HUGE! How much does it cost to produce them in acceptable yield?

[clonez]

the number is at around 120$ afaik, per chip (the gt200, shrink shouldnt be much cheaper)

and you dont get acceptable yealds with that size

[NapalmV5]

guys thats pic of G80

[Hornet331]

found this new pic on the net

more like 1 year old if the IHS is correct. That picture shows a G80-300 aka 8800gtx compared to the G92 65nm shrink.

[largon]

^
That's a 65nm G200 chip/package on a G80GTX.

[Gunslinger]

I also have to ask (apologies for being ignorant if they have) but have they produced any chip @ 55nm?

I was under the impression that the 9800 GTX+ and 9800 GT are 55nm parts.

[fellix_bg]

That pic above shows an uncapped G80 (8800GTX board?) next to a misterious GPU from NV with dimensions of 22-by-22 mm (484 mm²), so it's anything else but a 65nm GT200 (577 mm²) -- click!

[alexio]

Wow, Nvidia really has some problems. I was told by an inside source that the 55nm variants were planned for the beginning of Q4 (I reported this a few months back). I'll check if I can get a bit more information for you guys.

[fellix_bg]

Yep! Here is a truck load of defective batches:

[alexio]

Yep! Here is a truck load of defective batches:

[...pic...]

Lol @ that floppy disc.

But is this picture real?

[largon]

:o
Someone has had flippin fun decapping all those chips...

[Gunslinger]

The label on the box appears to say P/N G200-E8-A1

[wrangler]

I was under the impression that the 9800 GTX+ and 9800 GT are 55nm parts.

Well, my Hybrid power Evga 9800GT came today and GPU-Z reports it as 65nm.

[adamsleath]

i believe some 9800gt's were to be 55nm...?

and some not.

[BenchZowner]

Well, my Hybrid power Evga 9800GT came today and GPU-Z reports it as 65nm.

GPU-z cannot detect if your core is 65nm or 55nm.
Only way to find out is by looking at the core itself with your own eyes and measure