Those using GPGPU type functionality will - and they will have no problem shifting stock off on joe public who knows no better - don't underestimate how many people will go out and buy 2x 295 for quad SLI even now thinking its still the top dog.
-
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.
Nvidia works on second generation Fermi
- Thread starter Dark Sun
- Start date
Those using GPGPU type functionality will - and they will have no problem shifting stock off on joe public who knows no better - don't underestimate how many people will go out and buy 2x 295 for quad SLI even now thinking its still the top dog.
Second generation Fermi is coming in late 2010, just like the first generation Fermi was coming in 2009. I'll believe it when I see it. My prediction with this is they'll keep pushing the date back slowly and we'll be in the same situation the same time next year.
Last edited:
I feel really sorry for these people!!!
But he said the delay would be March
I think he was just trying make up the word count TBH. No one from Nvidia is going to say we are delaying the launch from March to March.
I think they should sit tight on the GF100 until they can build them at 32nm, if they cant get sub 200 watt and beat the 5000 range at 40nm that is.
The whole sept, nov, christmas launch was banded about to try and stiffle 5 series sales and I'm not even sure if anyone at nVidia actually confirmed those dates (tho they have publicly confirmed end of Q1 2010) most of the people giving 2009 dates were staff on certain pro-nVidia websites.
I'm pretty sure it was nVidia themselves who said the 2009 launch, to then have to admit it was delayed.
Caporegime
- Joined
- 18 Oct 2002
- Posts
- 33,188
NO, it hasn't and to constantly so something so ridiculously untrue that NO ONE at NVIDIA backs up your story is an entire joke.
THe article is saying MAINSTREAM FERMI is not delayed, not the high end part, GF100/Fermi the release in March is not the Mainstream Fermi, they mean a half sized 5770-type part. They are saying that's on time, largely because Nvidia don't release their mainstream/midrange part within a month or two of high end, they are always 6 months later with it, right now its on time because they aren't late, when they miss its release, it will be late. Midrange parts that have no profit will be pointless, yields won't match the 5770 while being circa 50% bigger.
Considering Nvidia, the CEO, and everyone there made a huge big launch saying there would be volume availability on a specific date in November, for one lone Nvidia guy on a forum to say, its not late, is ridiculous. Nvidia themselves today don't say "we aren't late, we always planned on a March release" because its untrue. They also admitted delays and lateness in their recent "architecture" release. The fact that the industry didn't expect it till March and hasn't for a long time, doesn't mean its not late, they just expected Nvidia to screw up again and again. Nvidia hoped to release at A1 silicon, than prayed for a A2 release, and then prayed on a A3 release, which they might get.
Nvidia and AMD are both working on anything up to 3 generations ahead ALWAYS, so working on their next generation is no surprise and never will be.
Anyway, TSMC won't have a decent 28nm yield by the end of the year, if they can make anything launchable this year I'd be shocked, they've switched from gate first and a far better quality process to a gate last version, which introduces a bunch of problems but is cheaper and easier to get going, but much worse. Half of the issues with their 40nm process are being carried over to the 28nm process. IF thats late, Nvidia are late even if their chip is completely ready, when TSMC are finally ready, we've got to pray to the process gods that Nvidia don't need 3 spins to get it right, if they do it will be VERY late again.
Soldato
- Joined
- 11 May 2006
- Posts
- 5,786
I think it's probably best to just take a break from the forums and enjoy playing games on the card you currently have, untill nvidia actually release something.
It's getting tiring, same crappy threads every day, spewing the same old rubbish out again and again. It wasn't even this bad prior to ATI releasing the HD2900XT.
It's getting tiring, same crappy threads every day, spewing the same old rubbish out again and again. It wasn't even this bad prior to ATI releasing the HD2900XT.
What this basically says though surely is "were going to release first gen fermi please buy it and ignore our info on the arrival of second gen fermi a few months later". Sorry but this is laughable you don't start advertising your next gen till you have the first one out the door and selling. If i was waiting for fermi1 i think i would now bin that and wait for fermi2 as it doesn't seem to be that far behind.
This all confirms that my gut feeling on original fermi was right and they have problems that they cannot sort out so start the hype on next gen. Nvidia for the love of god will you stop talking and release something hell anything because hot air is all well and good but we want cards and competition.
This all confirms that my gut feeling on original fermi was right and they have problems that they cannot sort out so start the hype on next gen. Nvidia for the love of god will you stop talking and release something hell anything because hot air is all well and good but we want cards and competition.
There are roadmaps from way back showing end of Q1 2010... theres also plenty of evidence of the progression nVidia intended and what happened there... the 200 series 40nm refresh was delayed and eventually killed off except the low end parts... the only reason nVidia made any kinda motions about a release previously was to try and head off ATI sales.
Associate
- Joined
- 8 Oct 2008
- Posts
- 1,313
Where is this advert you talk of ?
The original plan back in September was to re-brand DX10.1 gt200 cards, and milk them through 2010. I remember giving you some stick about it because you kept calling DX10.1 useless after Nvidia said they would never use it.
I don't think any of this story is from Nvidia TBH. I cant see why Nvidia would talk about second version of the GF100, They would just call the next card.
I'm pretty sure someone from Nvidia confirmed a pre Christmas re-release too.
DX10.1 is pretty useless really 
sure it gives ATI cards a bit of an edge - but most of the decent features from it aren't really useable to any great deal with the hardware out before now (on either side)... the only reason the GT240, etc. have it is because they were cut down from a failed attempt at a high performance line that would have had the power to utilise it to some extent... the low end cards themselves don't benefit from it... but I guess nVidia still see it as having some worth as a sales point.
sure it gives ATI cards a bit of an edge - but most of the decent features from it aren't really useable to any great deal with the hardware out before now (on either side)... the only reason the GT240, etc. have it is because they were cut down from a failed attempt at a high performance line that would have had the power to utilise it to some extent... the low end cards themselves don't benefit from it... but I guess nVidia still see it as having some worth as a sales point.DX10.1 is pretty useless really
It is now we have DX11 ! The savour of PC gaming.
Its taken six years and three OS's to move on from DX9c because of pigheadedness in the PC industry
I'm not having a pop at Nvidia for the "Cough DX10.1 incident Cough", so don't you Nvidia nuts start going mental, I'm just saying its a sad state of affairs.
I feel the big companys should get together and set some proper rules in place, specificly aimed at pushing the PC gaming industry in the right direction for the benefit of the of the customer.
yes definatly...theres no source for this article he doesn't even say that he heard it from nvidia.
just pure guesswork on fud's part.