• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Was Charlie Right?

Brackle

Old Timer
2FA
Joined
Jun 19, 2003
Messages
9,101
http://www.semiaccurate.com/2010/03/29/why-nvidia-hacked-gtx480/

I have to give him credit where its due. Alot of what he mentions is accurate. It's just the way he presents it which is VERY biased IMO.

Not sure how accurate the Chip Manufacturing part is...but Hey he was Semi-Accurate about Fermi so far. I mean shit he said there would be no 512 shader part, but XMAN25 and everyone else said he was wrong......

What do you guys think?
 
He's right, but more of a "lucky right".

If fermi were released 4 months sooner it would be completely in charge. See, the longer it wasn't released, the longer speculation and anti nvidia spew got to surface, it caught on and now only the extremely nvidia loyal have held out for it.

Xman was just wishful thinking.
Posted via [H] Mobile Device
 
He's right, but more of a "lucky right".

If fermi were released 4 months sooner it would be completely in charge. See, the longer it wasn't released, the longer speculation and anti nvidia spew got to surface, it caught on and now only the extremely nvidia loyal have held out for it.

Xman was just wishful thinking.
Posted via [H] Mobile Device

You cant get that lucky. He was definitly getting insider Info....Cause be was damn close to being right about Fermi period...

But the way he presented it was TOTALLY biased....
 
Charlie had some of the facts right and corroborated by the Anand, but he did claim 300 TDP when it turned out to be 250W. Fermi could very well have been undergoing testing at 300 TDP to determine how much performance could have been pushed out of the chip and Nvidia could have lowered it to 250 months later after the Semiaccurate report when it reached retail however.

I'm guessing it was not exactly easy for Nvidia to muffle reports of low yields (the AIBs had to have known regarding supply issues) or hot cards and power requirements and Charlie used these base points to get an approximate idea of the situation which was written in a very biased way.
 
Charlie had some of the facts right and corroborated by the Anand, but he did claim 300 TDP when it turned out to be 250W. Fermi could very well have been undergoing testing at 300 TDP to determine how much performance could have been pushed out of the chip and Nvidia could have lowered it to 250 months later after the Semiaccurate report when it reached retail however.

I'm guessing it was not exactly easy for Nvidia to muffle reports of low yields (the AIBs had to have known regarding supply issues) or hot cards and power requirements and Charlie used these base points to get an approximate idea of the situation which was written in a very biased way.


Yea it was, Also if you read the article, from what Charlie says (snicker) they were suppose to be 295 TDP, but at the last minute they were able to get it down to 250 TDP.....of course this is what he says.

And of course its writtin VERY biased like you said
 
Charlie was substantially correct.

Fermi would also be well received if they had used this extra time to get an extra 20% of performance for maybe 25% more in price. Instead the performance is neither poor nor a great improvement, but the premium remains, the time was lost, and the heat is a concern.
 
The time has come from Charlie being mocked by some... to becoming recognized as da man.

Nvidia owes Charlie a written apology. He owned them. And all the haters.
 
Yeah, he was right. From what I've heard, his sources are actually moles. If there's one thing that has amused me throughout Fermi's postponement, it's how much the expectations from the green loyalists were so utterly violated. No problem for me, as price/performance ratio usually dictate my decisions but I'd like to thank him for ensuring my purchases were solid.
 
He's right, but more of a "lucky right".

If fermi were released 4 months sooner it would be completely in charge. See, the longer it wasn't released, the longer speculation and anti nvidia spew got to surface, it caught on and now only the extremely nvidia loyal have held out for it.

Xman was just wishful thinking.
We have no proof of this, nor can we test it. All we can go off is this one instance, and I have to say, Charlie has been remarkably accurate throughout the whole debacle.

Charlie had some of the facts right and corroborated by the Anand, but he did claim 300 TDP when it turned out to be 250W. Fermi could very well have been undergoing testing at 300 TDP to determine how much performance could have been pushed out of the chip and Nvidia could have lowered it to 250 months later after the Semiaccurate report when it reached retail however.
But the thing is, that seeing as the cards actually draw closer to 300W than 250W, part of me wonders whether that change was purely for marketing purposes.
 
Seemed like did a pretty decent job on this. The only thing of course he couldn't have predicted was the price, but that's definitely understandable as that as that could be any arbitrary number
 
Yes, hate him or not he's definitely got some inside sources at TSMC and probably not Nvidia. He's always right about the manufacturing stuff. Another thing to remember: alot of the stuff could also have been 100% right at the time of writing, and last minute changes were done by Nvidia.
I remember lurking on here before fermi and reading all of Xman and some other tool's (forgot the name blue man or something) posts about having inside info. They were even worse than Charlie good stuff :).
 
I think its obvious he was right.

Remember we are judging what he wrote on the past based on a cut down geforce 100 core. It should have had a 512 cuda cores. The ones shipping have less. Who knows what would have happened if they released a part with that amount of cores and the target clock speeds Charlie was hearing for the part. It could have been well over 300w and as we know the 512 core chip is not producable because there is no part avalible.
 
After all this I'm seriously worried we have another 8800m on our hands. I'm still interested in going nvidia for the next rig, but I'm refusing to touch the first revision of these puppies. Charlie was right, but only in 1 or 2 sentences of his page long tirades. The rest was boring conjecture no one cared about.
 
I like the way this guy writes, the way he phrases things drip with venom while pretending to be descriptive.

"Yields were still far below 20 percent for both the full 512 shader version and the 448 shader part combined. For comparison, Nvidia's G200 chip, which became the GTX280 family of GPU parts, had a yield of 62.5 percent, give or take a little, and that yield was considered so low that it was almost not worth launching. Given a sub-20 percent yield, to call the Fermi or GF100 or GTX4x0 line of GPU chips unmanufacturable is being overly kind."



PS - Need a history lesson, did Nvidia slaughter his family or something?
 
To be honest it was always fairly unlikely that someone would have made such specific predictions on the basis of nothing.

I don't understand why ATI vs Nvidia is so tribal. Ultimately a graphics card is just a piece of machinery, and neither company gives a damn about you or your loyalty or lack of it, or is even aware of it either way. It's just about getting the best value out of their offerings.
 
Stopped clock, etc etc. :)

I hate the way the guy writes, I hate his obvious bias over the years, but you have to give credit when credit is due. He wasn't right about everything, but small details and small FUD that turned out to be wrong or exaggerated don't matter much when the big message was so on target.
 
Stopped clock, etc etc. :)

I hate the way the guy writes, I hate his obvious bias over the years, but you have to give credit when credit is due. He wasn't right about everything, but small details and small FUD that turned out to be wrong or exaggerated don't matter much when the big message was so on target.

Yea thats something people cannot deny. His OBVIOUS bias against Nvidia.
 
I don't think this was luck at all, it was very educated guessing of course, so he turned out right. Small details are hard to get gotcha points with because the parts aren't 512 core 750MHz parts. Could have easily been a 300W TDP part if Nvidia pushed the product they designed it to be.

Remember it was the very last minute that Nvidia finalized the spec to get it producable within the proper thermal envelope. He certainly divulges a lot more info about the chip making industry than a lot of websites put in to their pieces, as he had schedules and processes all well accounted for.

Other rumor sites were just regurgitating marketing info without actually trying to find out what was happening with the process.

I wish he could present his stories in a more neutral manner but the Nvidia spite just makes me laugh.
 
I like how all these people who "hate" the guy and "hate" his writing so much are obviously reading it anyway.

I hate how everyone whines about "Charlie" all the time, can I start that bandwagon? I wouldn't even know who the guy is (and still don't care) if not for all the wailing and gnashing of teeth about him.
 
More right than wrong especially when you consider how many months ago a lot of this was written. What we see released today was in no way what was planned.

Regardless of what he has written before or writes after. Regarding fermi, he was correct and all those who sneered at him and claimed nvidia would prove him wrong will have to eat humble pie.

Where are all those posters who slated him anyway?
 
He will only be useful to me if and when he closes his eyes and draws upon his reservoir of hatred for nVidia to will the 5870 to catch up to the GTX 480.
 
He obviously has a source that is close to GPU manufacturing. He was right a lot about the physical attributes of the chip, but was very much hit and miss when it came to the architecture and how it would ultimately perform. I can't wait till the next Nvidia release so we can go through this all over again.
 
I think this guy is funny but he definitely seems like the type of person who if you presented him with facts proving him wrong, he'd find ways to be right. Thankfully I don't care if he rips Nvidia because they deserve everything they get.
 
I don't care about his bias, you're either right or wrong and he was mostly right.

And it's always fun ripping on Nvidia. I've owned more Nvidia cards than I have ATi and I'll rip on them any chance I get.
 
I think the worst thing he did was keep claiming it had 2% yields at first, when that was revealed a long time ago to be a translation error... yet he still kept referring to it.

I think he over-speculated and was too quick to post articles with less-than-credible sources.
 
and in the end he was still mostly right, as said above, your either right or wrong on and on Fermi he was right, i love seeing how even though he was right, people are still trying to dis-credit him....
 
He's an ass. Even if he is SEMI accurate, even if some of the stuff he spewed it correct, that doesn't make up for the fact that he is a Ati troll plain and simple. Simple as pie.
 
the 300wat TDP was probably correct with 512 cores the GTX480 still draws alot of power

one thing he was wrong about were the spec's of the cores, but even those could have changed last minute
 
lol some just refuse to admit that someone is right cause they are one sided........
 
I like the way this guy writes, the way he phrases things drip with venom while pretending to be descriptive.

PS - Need a history lesson, did Nvidia slaughter his family or something?
I think it has less to do with a grudge of any sort and more to do with Charlie just doing what he has to do to make a living. Charlie's venom is just standard shock jock/tabloid journalism. He is pulling the usual Howard Stern or Weekly World News lines of saying outrage, insulting, venemous, etc, things just to get people to pay attention and get clicks to his articles.

Was he right this time? Yeah, absolutely. He does wrap his venom around enough kernels of truth to not just be total "Bat Boy marries dolphin girl" trash. If Nvidia ever comes out with the best gfx card on the market that is simultaneously half the price of any competition and does your taxes for free, will Charlie find a way to be absolutely venomous and derogatory about it? Yeah, absolutely. If Charlie didn't act like Charlie he wouldn't have a job. Tabloid writers don't make their money by being fair and unbiased.
 
I think it has less to do with a grudge of any sort and more to do with Charlie just doing what he has to do to make a living. Charlie's venom is just standard shock jock/tabloid journalism. He is pulling the usual Howard Stern or Weekly World News lines of saying outrage, insulting, venemous, etc, things just to get people to pay attention and get clicks to his articles.

Was he right this time? Yeah, absolutely. He does wrap his venom around enough kernels of truth to not just be total "Bat Boy marries dolphin girl" trash. If Nvidia ever comes out with the best gfx card on the market that is simultaneously half the price of any competition and does your taxes for free, will Charlie find a way to be absolutely venomous and derogatory about it? Yeah, absolutely. If Charlie didn't act like Charlie he wouldn't have a job. Tabloid writers don't make their money by being fair and unbiased.

Problem is, that's how you get hurt :D. I would not be surprised to see this guy get punched square in the mouth just for being the kind of guy that he is. He reminds me of Perez Hilton, who did get punched a few times for being the dick that he is.
 
It's impossible to be 100% right about a product being changed every month.
I wannna see someone else have the guts to report rumors as accurate as Charlie did.

Most sites are very careful about reporting the news Charlie does, that why you barely saw any other sites confirm what Charlie said.
Even if he was 50% right, he had more info than any other site on the net and wasn't afraid to report it.
 
He's an ass. Even if he is SEMI accurate, even if some of the stuff he spewed it correct, that doesn't make up for the fact that he is a Ati troll plain and simple. Simple as pie.

He was right though.

Exactly, but some people just will find any reason to deny that.

lol some just refuse to admit that someone is right cause they are one sided........

yup, this is about someone being right that everyone claimed as a troll and couldn't be trusted,
 
You cant get that lucky. He was definitly getting insider Info....Cause be was damn close to being right about Fermi period...

But the way he presented it was TOTALLY biased....

you have answered your own post. yes he was right, he just painted it in the worst of all possible lights. Fermi is not a viable product in itself (I am withholding judgment for now on the product line, look how many incarnations of the G80 chip we have seen) and nvidia has really screwed themselves and their AIB here. I can't see any other way to look at it.


The other shoe here is that while charlie was always poking Nvidia with a sharp stick (maybe even inserting it in strange places) they kept HANDING him sticks to be hit with. please, lying about mobile chips, all they had to do was come clean the first time and it would have been done. That stick went up their ass. Wood screws? why? No wonder Charlie has had a field day here. I really wish Nvidia management would take a look events over the last couple of years and reassess their methods of deal with issues.
 
I like the way this guy writes, the way he phrases things drip with venom while pretending to be descriptive.

"Yields were still far below 20 percent for both the full 512 shader version and the 448 shader part combined. For comparison, Nvidia's G200 chip, which became the GTX280 family of GPU parts, had a yield of 62.5 percent, give or take a little, and that yield was considered so low that it was almost not worth launching. Given a sub-20 percent yield, to call the Fermi or GF100 or GTX4x0 line of GPU chips unmanufacturable is being overly kind."



PS - Need a history lesson, did Nvidia slaughter his family or something?

that is somewhat clouded. apparently there was and issue that Charlie broke the NDA or that Nvidia was trying to enforce something of a bullshit one (I consider either or both very possible) and what your seeing the result of simple pissing contest getting out of hand. Nvidia appearntly thought they could ignore him and dismiss him but Charlie (love him or hate him) is an EXCELLENT asshole. that combined with Nvidia being stupid about things have produced this very entertaining drama that we have seen as Charlie vs Nvidia. :D
 
I think the worst thing he did was keep claiming it had 2% yields at first, when that was revealed a long time ago to be a translation error... yet he still kept referring to it.

I think he over-speculated and was too quick to post articles with less-than-credible sources.

that number was not (at first at any rate) a translator. the so called hot lots (IE the first production run) simply did not go well. as I understand it that was the number he was always referring to. it was true, in a way
 
He was right in some things, wrong in others. But what opened him up to "looking" like he was right more so was nvidia's shadyness and lies. They gave him everything he wanted and needed.
 
Back
Top