Nvidia Q1 Financial Results

[SirPauly]

A key to sensationalism, to me, is throw out many alternatives so one may spin that they're correct.

 

[Zanovar]

Regarding charlie if u spread loads of BS over and over again u ought to be correct sometime right?

To be fair,i think he has his insiders.and he gets alot right.lots wrong too

 

[KompuKare]

Charlie also had some incomprehensible, self-contradictory crap about how Kepler was fast only because of PhysX.

Probably one of the most useful anti-NV diatribes of his was bumpgate, which ended up being very real and very expensive for NV to settle out of court.

Well, he's a sensationalist journalist and for some reason Nvidia must have rubbed him up wrong at some stage but on bumpgate he was pretty much 100% spot on. While the settlement might have cost them 'upto' $250 million a lot of that ended up with OEMs who may or not have sorted out their customers (I've seen at least two HP DVs laptop who had all the bumpgate signs but who according to HP were not affected: this despite the serial numbers above and below them were on the recall - HP's reputation rightly suffered a lot with those DV laptops). But those who took part in the class action lawsuit cannot have been happy: high end laptops replaced by essentially netbooks.

And parts I've personally had go bad on me (2x 8800GT, nForce 7150) were not covered by the settlement anyhow. The only thing know about those parts is

1) they have been going bad more than one would expect and

2) they were on that list of part number / revisions which Charlie mentioned in his 'Nvidia 55nm parts are bad too' article.

 

[Jaydip]

To be fair,i think he has his insiders.and he gets alot right.lots wrong too

The problem is how can u trust a man like him?
If he says Kepler architecture is probably better than the scientist itself how can u be sure unless u actually see the reviews?:biggrin:
Kepler had dedicated PPU remember.

 

[Zanovar]

The problem is how can u trust a man like him?
If he says Kepler architecture is probably better than the scientist itself how can u be sure unless u actually see the reviews?:biggrin:
Kepler had dedicated PPU remember.

Personally i wouldnt trust the dude as far as i could throw him,lot of peeps seem to do though.

 

[Olikan]

Charlie also had some incomprehensible, self-contradictory crap about how Kepler was fast only because of PhysX.

please...

Sources tell SemiAccurate that the ‘big secret’ lurking in the Kepler chips are optimisations for physics calculations. Some are calling this PhysX block a dedicated chunk of hardware, but more sources have been saying that it is simply shaders, optimisations, and likely a dedicated few new ops. In short, marketing may say it is, but under the heat spreader, it is simply shaders and optimisations.

where charlie was wrong here?

 

[blastingcap]

please...

what's wrong here?

Why are you selectively quoting? Here's the entire Charlie rant: http://semiaccurate.com/2012/02/01/physics-hardware-makes-keplergk104-fast/

I stand by my comment. If you want to defend Charlie's rant go ahead. I won't argue with you, I will just laugh at you instead, and so will everyone else who reads Charlie's article and compares it to reality.

 

[RussianSensation]

where charlie was wrong here?

I am too lazy to look for a quote but I specifically remember Charlie saying at the last minute that GK104 is unbalanced, that it wins in PhysX titles and sometimes loses to Pitcairn. Basically, NV was going to try to make Kepler unique by integrating out of this world PhysX functionality and use those games to show how superior it is. But before Kepler launched, he changed his tune to state that Kepler will win in the major performance metrics that count against HD7970 (performance/watt, performance, power, etc.).

EDIT: From Blastincap's link, he says GK104 will go against Pitcairn, maybe end up a little faster.

 

[tviceman]

Charlie is an idiot who is on a lifelong personal crusade against Nvidia. Nvidia's financials, compared to what they did a year ago, aren't spectacular. But during Nvidia's previous conference call, they said they were expecting gross margin to be around 48% (they beat that) and they also expected revenue to be lower than what it was. Given that they had similar margins and revenue as they did a year ago but made less money, to me it looks like they are sitting on a large inventory of chips that have yet to be shipped for revenue.

 

[Olikan]

Why are you selectively quoting? Here's the entire Charlie rant: http://semiaccurate.com/2012/02/01/physics-hardware-makes-keplergk104-fast/

I stand by my comment. If you want to defend Charlie's rant go ahead. I won't argue with you, I will just laugh at you instead, and so will everyone else who reads Charlie's article and compares it to reality.

dude, i am not defending charlie atitude, he is indeed an idiot... just saying that he was very accurate, and the "PhysX block" was from someone else...

EDIT: From Blastincap's link, he says GK104 will go against Pitcairn, maybe end up a little faster

long time. On some games, GK104 will handily trounce a 7970, on others, it will probably lose to a Pitcairn. Does this mean it won’t actually do what is promised? No, it will. Is this a problem? Depends

...this one? well, if we count gpgpu....he stays correct

 

[thilanliyan]

I agree but i really don't understand all these terminology very much.I skipped most of the "finance and business accounting" classes during engineering May be a refresher course is in order

Lol me too. I got a mercy pass for my Engineering Economics course haha

 

[Jionix]

Charlie was also the one to report that Kepler would beat Tahiti on "every metric", months before anyone even had a clue, and everyone said he was just trolling because he hates Nvidia.

It's not like he is making all this up. Charlie has his sources --- sometimes he gets it right, sometimes he doesn't. When you dabble in the "unofficial" news like he does, you are never going to be 100% right, because if you were, then it really wouldn't be unofficial and someone would start to ask questions..

 

[sontin]

Charlie was also the one to report that Kepler would beat Tahiti on "every metric", months before anyone even had a clue, and everyone said he was just trolling because he hates Nvidia.

After we saw the performance of Tahiti this was so easy to predict. Even the GTX560ti has a 70+% perf/watt improvement over the GTX285 that it was more diffecult for nVidia to miss this "target".

 

[Jionix]

Um, do you know what "Every metric" actually means? Not just performance... Die size, power consumption, temperature.. Every metric! He was spot on. You can't spin it any other way..

 

[ShintaiDK]

Charlie was also the one to report that Kepler would beat Tahiti on "every metric", months before anyone even had a clue, and everyone said he was just trolling because he hates Nvidia.

It's not like he is making all this up. Charlie has his sources --- sometimes he gets it right, sometimes he doesn't. When you dabble in the "unofficial" news like he does, you are never going to be 100% right, because if you were, then it really wouldn't be unofficial and someone would start to ask questions..

Well thats the fun part of Charlie. He makes as many articles he needs to "be right". Then he can pick the one he needs. Because in another article it would barely compete with Pitcairn.

I can also make alot of articles about the same subject. Then take whatever of them is right and say. See I told you so, I´m always right!

 

[KompuKare]

Well SA is a rumour site... But the nature of unofficial news (in this case a retailer /card AIB / techsite) is that what someone like Charlie gets to see / hear about with NDA is mostly prototypes. Although clock may not accurate for pure fud, in the case of GK104, I think we can safely assume that the prototypes were not as highly clocked / overclocked as the release. So, that makes rumour sites not much good for rumours...

 

[SirPauly]

Um, do you know what "Every metric" actually means? Not just performance... Die size, power consumption, temperature.. Every metric! He was spot on. You can't spin it any other way..

He downplayed that to some levels -- especially performance and said AMD wouldn't reduce pricing.

 

[DeeJayeS]

Charlie is good at trolling, as evidenced by the existence of this thread

Don't hate on him--ignore him instead.

 

[Olikan]

Well thats the fun part of Charlie. He makes as many articles he needs to "be right". Then he can pick the one he needs. Because in another article it would barely compete with Pitcairn.

I can also make alot of articles about the same subject. Then take whatever of them is right and say. See I told you so, I´m always right!

right....Please, write about the next HSA features that will be found in the 8xxx series....

we are saying that he is very accurate, not always accurate...

i mean, i think that only charlie itself belives that nvidia have yet to sell 10k kepler, world wide

 

[krumme]

All this public labelling Charlie an idiot, and at the same time you are all reading his articles makes you seem like small men. There is a huge difference from critizizing the work, to caling him an idiot. As you dont know s..t about the man, the labelling only says something about the person calling names, witch mean yourself.