Fermi 'slips into January'

[Bluescreendeath]

I have to admit I started to question some of his later articles, putting them down mostly as 'wishful thinking'.

But tbh...those 1.7% yield figures are starting to look pretty plausible right? At least you can believe it now that TSMC are back below 40%. Take a scratch-built chip that big, with Nvidia's lack of expertise on so many fronts and you can quickly come to the conclusion that 1.7% might even be generous.

http://www.semiaccurate.com/2009/09/15/nvidia-gt300-yeilds-under-2/

It really doesn't look so far off now does it?

But that original figure stems from a complete mistranslation. The article was originally based wrong information.

Just because some silicon industry in Taiwan went to hell and now the yields suck doesn't make Charlie's articles correct. Even if the yields are below 2%, it was due to unforeseen consequences that had nothing to do with Charlie's predictions.

 

[jennyh]

No he explained in a later article that the information was actually correct. There was a mistranslation in place but in that same article he mentioned he had numerous sources all coming to the same conclusion.

I'm willing to believe in lucky coincidences but only so far. I think it's a lot more likely that Charlie has (ie is paying for) good insider information. On this one he's been more than half right, a lot more - while everybody else is just guessing and getting it horribly wrong.

 

[jennyh]

I mean honestly, if somebody had to come out from TSMC and say 'Fermi yields are below 5%', I really don't think anybody would die of shock, do you?

 

[L1qu1d]

I hear Intels quick to retaliate tho
http://i645.photobucket.com/albums/uu171/jaydeejohn/2r5uo2c.png

That alone belongs in the New Yorker, nothing could be more accurate.

If BFG or EVGA start making ATI cards, I'm prob going to grab the 5890 I've been impressed with the new gen dual GPU cards. The 4870 X2 still had dips in min frames but the 295 GTX was very impressive, and acted like 2 cards in sli. I'm sure the 5890 will do the same (I'm assuming the 5890 is going to be the dual GPU or is the 5980? OR w.e the dual GPU is😛)

To be safe I'll just say 5870 X2😛

 

[L1qu1d]

Yeah here's hopping for the OpenCL client ASAFP (the F stands for folding, not FU.... ), it's sad the amount of lost potential there, not from a BS marketing standpoint, but just from the spinning your wheels in neutral standpoint. It's not like if there was an OpenCL client today ATi could possibly sell more cards with their current shortages, but unfortunately alot of science is being left in the double calculation. At work we still can't use the new cards cause no one is an OpenCL guru yet, and I'm not working in that department anymore, so I'm not busting my huump learning F'in code, when I can play with shiny toys (currently getting ~20Mb/s on my wireless connection ). I have to say ATi dev team needs more guys, but I bet there's a hiring freeze there too. I just wish we were already a year into the future when people were more adept than we are now. It's all promise and so little to really work with right now, and no one has spare cash to hire young students yet to become the future experts. So much wasted potential. 🙁
Funny thing speaking of Charlie's article(s), did you read Theo's @ BSoN about Batmangate, really a good read, even if you don't belive whatever part of it, and interesting attempt at balance in a very contentious issue.

Oh, and congrats on having the first silver badge (next to your username) I've seen I'm still waiting for a bronze one

I'd like to thank the Academy, and my Parents Hannah and Barbera. [:thegreatgrapeape] [:thegreatgrapeape:3] [:thegreatgrapeape]

I'm going for the GOLD Who do I have to sleep with? 😀

 

[jennyh]

As if on cue...

http://www.semiaccurate.com/2009/11/09/tsmc-slips-their-32nm-process/

Charlie reporting TSMC's 32nm has even worse issues than their 40nm. That doesn't really take a huge leap of faith however.

 

[Harrisson]

For example, if Nostradamus says that the world will end due to nuclear war, but it actually ends due to a giant meteorite....the result is the same, but the reasons are wrong. So Nostradamus is still incorrect.

Accurately predicting the end result without accurately predicting the cause...

Besides, saying Charlie makes accurate predictions is just cherry picking the articles that are correct while ignoring the articles that are totally wrong. It's not exactly hard to foresee Fermi's delays given the heaps of trouble Nvidia has been going through...

In most cases Charlie get both end results and its causes right, at least at the core, you could go anal about some details only. And since I read a lot of his articles, I dont have to cherry pick anything to say he was right very often passed year, while YOU would have to cherry pick to find where he was wrong 😉 Oh, and to say "It's not exactly hard to foresee Fermi's delays" is easy post factum, try that like a year before it happens and see majority hunt for your head till everyone finally sees it as well

 

[Harrisson]

But that original figure stems from a complete mistranslation. The article was originally based wrong information.

Just because some silicon industry in Taiwan went to hell and now the yields suck doesn't make Charlie's articles correct. Even if the yields are below 2%, it was due to unforeseen consequences that had nothing to do with Charlie's predictions.

Thats Fudo version, but as it appears Charlie was probably right, again, and Kyle from HardOCP confirmed it as well. It was quite amusing to read HardOCP thread about it, where majority were bashing and mocking him, and even after Kyle said he heard from inside sources the same info, STILL posters were telling how Charlie got this from his ass, mistranslation or w/e

 

[TheGreatGrapeApe]

Charlie is sometimes right...but in the times he is right, he is right for the wrong reasons. So that still makes most of his arguments still totally wrong...

I understand what you're trying to say, but I totally disagree with this being applied to Charlie.

He said they would be late, because it's very BIG (check), very [overly?] Complex (check), thus take longer to finish designing (check), then due to the size and complexity and the already known poor yield of 40nm for both ATi and nV will likely mean very low initial yields with few working chips (check), likely requiring a respin to fix (check), qnd which may even require a third spin (check), which even if everything goes well will mean the chip gets pushed into 2010 (check).
This was all in APRIL with a follow-up in May, and meanwhile every other pundit out ther said he was way off and just bitter, and up until last week (Wed) Fuad et al were still believing there was a chance that this thing would ship for November.

So really, who's right here? The guy who took an edjucated guess and got alot of it right if not 100% precise on dates, or the people getting fed PR by nV who's changed his time frame 3 times sofar?

Remember Charlie's as locked-out of nV as any man can bem yet did far better than those who get fed leaks by them. So that makes them more 'right' because their arguments aren't as pointed with personal opinion, just diluted nV PR?

I want nV to launch something soon, it's not healthy to have these issues for either company (also low 40nm yields [for both] and excess demand for desktop DX11 GPUs means laptop chips are pushed out further ) but I'm also not a fan of the recent FUD towards the current product to try and distract from the reality that the actual competition isn't here right now or soon.

Fermi may solve world hunger... but not this year.

 

[randomizer]

At least you can believe it now that TSMC are back below 40%.

I'm willing to bet that number is padded to keep the shareholders happy too.

 

[TheGreatGrapeApe]

For example, if Nostradamus says that the world will end due to nuclear war, but it actually ends due to a giant meteorite....the result is the same, but the reasons are wrong. So Nostradamus is still incorrect.

Accurately predicting the end result without accurately predicting the cause...

Which is still more accurate than those saying that 'everything will fine, nothing is going to happen and the world will keep going on an on, the other guy is crazy'.

Your argument would hold some weight if anyone were anywhere near more accurate on the subject.

Also explain to me where he majorly missed the boat in that April article compared to those written by everyone else even so recent as the day before Fermi's 'Launch'?

It's easy to criticize him when his speculation is way off, but in this case it's equally hard not to give him credit for calling this playoff before pre-season started.

 

[jennyh]

I do feel Charlie deserves a lot of credit on this one lol.

Sure he might just have got exceptionally lucky, but tbh if Nvidia high command had believed him they probably wouldn't have ended up with such an abortion of a gpu.

 

[Bluescreendeath]

Which is still more accurate than those saying that 'everything will fine, nothing is going to happen and the world will keep going on an on, the other guy is crazy'.
Your argument would hold some weight if anyone were anywhere near more accurate on the subject.
Also explain to me where he majorly missed the boat in that April article compared to those written by everyone else even so recent as the day before Fermi's 'Launch'?
It's easy to criticize him when his speculation is way off, but in this case it's equally hard not to give him credit for calling this playoff before pre-season started.

I have a feeling that he would've said Nvidia is going down regardless of whether Nvidia really was in deep sh*t or not. It just worked out in his favor that it was true.

 

[jennyh]

BSD you have to admit he has been pretty much 100% on this one from the start.

I agree a lot of it could be and probably is lucky, but even at that he has been incredibly accurate all the way on this. To me that points to pretty good insider info, he just knows too much even with his hatred of nvidia.

 

[Bluescreendeath]

For this particular issue then perhaps that is true.

 

[JAYDEEJOHN]

Dont know how edjmacated this guess is, but going by chipsets, ATIs seem to be fine for ocing and regular clocks, whereas nVidias have been known to run hot the last few gens, unstable etc.
So, maybe the GF goodness does come into play here vs TSMC
Id heard before launch of the 5xxx series the x2 solutions wouldnt be out before the end of the year. Maybe its wrong, and maybe ATI knew they wouldnt need it, and Fermi was a ways off, who knows for sure, and the naming either?

 

[TheGreatGrapeApe]

I have a feeling that he would've said Nvidia is going down regardless of whether Nvidia really was in deep sh*t or not. It just worked out in his favor that it was true.

Perhaps, but he didn't do the FanATic / nVidiot typical YOU SUCK, he went into detail, and like I pointed out above alot of those details came to be true.

Alot of the Fermi supporters and ATi detractors didn't provide anywhere near the substance in their ANALysis, and turned out wrong with just the basics.

So I'll give Charlie props for being far more specific than others, and also in this particular case being far more accurate than others.

Overall he may still be hitting .200, but on that one hit hit a Grand Slam, and you gotta at least admit that even if you think everything else he writes is BS. 😉

 

[turboflame]

I wrote a nice little poem for you guys:

1.7

Wood Screws

Q1 2010

Bad cartoons

It's over, Nvidia is finished

 

[L1qu1d]

God forbid unless you want to see a monopolized market

 

[EXT64]

They had better not be finished or decide to leave. I was kind of feeling going NVidia again after all these years (last was a 6800), just like my next build is going to be Intel (my first since the P4 days), but not if this mess persists. I think I'll just stick with my 4850 for now.

 

[noob2222]

Nvidia confirms it. http://www.theinquirer.net/inquirer/news/1561640/nvidia-fermi-delayed

Seems to me Nvidia is doing what AMD did back in the athlon xp/64 days, putting all their eggs in one basket. They should have been working on 2 projects instead of just trying to make fermi work. Update the 200 series so you can have cash flow, instead they EOL the 200s.

AMD did just the opposite this time around, they are working on their gpgpu, but pushed out a new faster updated 4000 series.