• 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.

Fermi A3 Silicon...

NKD said:
man I love reading your posts, are you really telling me the hd 4800 series wasn't competitive, oh shit as far as I remember that is the only reason nvidia dropped prices, hmm, lets see I am pretty damn right about that, may be you can go back and check or ask the forum members here, an honest person would agree, gtx 285 went from 600 to 450 and gtx 260 went to 300 from 500, is that because ati didn't have a product to compete, and thanks to ati hd 4800 series I was able grab a gtx 285 for less than for hundred a few months after launch, seriously I am just being real and stating facts here, you should start charging nvidia for their defense.
Click to expand...

I'm sorry, but can you read english ?

Here's a quote from the post you quoted me from:

Silus said:
So ATI finally competes with them since the release of the HD 4800s
Click to expand...

And what you take of that is that I said the HD 4800s was not competitive ? Do you not grasp simple english where "competitive" does not equal "not competitive" ?
 
Wotan said:
I should have been more clear as to what he was right on the money about
Click to expand...

Yes you should. Including that link doesn't help you either...

Wotan said:
I meant his timeframe for release. The performance numbers I haven't got a clue about to be honest. I have no idea how he is working his math

His Fermi timeline has been pretty damn good thus far. He has a pretty good idea of how long things take in terms of production.

From his article on May 14th http://www.theinquirer.net/inquirer/news/1137331/a-look-nvidia-gt300-architecture

He has a pretty decent understanding of Nvidia's situation at the moment and even way back in May.
Click to expand...

Have you ever heard that even a broken clock is on time, twice a day ? Getting one thing right, doesn't make the dozens of other things he got wrong, right. I already pointed quite a few in this thread, which obviously he never admitted he was wrong.
And he has no understanding of anything, but parroting what others tell him (often AMD's PR). If you like him so much, you probably should see the ass he makes of himself in B3D forums, with people that do know what they are talking about, telling him how things actually work in the real world. He even recently confirmed that he does't even fact check what those people tell him. But that's not surprising...if it's something that can be twisted to fit his agenda of "NVIDIA doom and gloom" he posts it and obviously his "fans" plaster his links all over the place.

WotanAlso said:
release[/URL] put DPP at 630 GFlops whereas other sites like http://www.techreport.com/articles.x/17815/4 put DPP somewhere around 870 GFlops. The Tesla may be clocked lower for whatever reason, but you get the idea.
Click to expand...

Again the usual mistake. Tesla is not clocked lower "for whatever reason". Teslas exist for the HPC market and they undergo a much more rigorous certification process, which makes them be clocked in a much more conservative way, when compared to GeForces.

Wotan said:
If someone came and said that GF100 was going to underperform at this point I wouldn't be surprised and considering that Charlie does seem to have pretty accurate information regarding Fermi as of late (tapeouts, yields, memory controller probs) I'd take his information to be more accurate than other sources like Faud who was saying that we'd see Fermi in late Novemeber up until a few weeks ago.

It's not the Ten Commandments and it could easily be incorrect, but at this point it's very plausible.
Click to expand...

Memory controller probs ? There are no confirmed memory controller probs. He seems to have good insider info on tapeouts I'll give you that. As for yields, they are of public domain, with TSMC itself stating them. And they affect both NVIDIA and ATI.
 
Silus said:
I'm sorry, but can you read english ?

Here's a quote from the post you quoted me from:



And what you take of that is that I said the HD 4800s was not competitive ? Do you not grasp simple english where "competitive" does not equal "not competitive" ?
Click to expand...

you said ati competes with them since the release of 4800 series, competes the them finally? you mean gtx 200 series with the hd 5000 series, if I misunderstood you I am sorry but to me it sounded like hd 5000 competes with gtx 200 series since the release of hd 4800 series, so it basically means that hd 4800 series was nothing and hd 5000 is true competition for gtx 200 series, hd 5000 is not a competition for gtx 200 series it is way ahead in feature and performance.
 
Wotan said:
Don't forget it was the 280, not the 285 that they were pricing at 600.

A great article on the 4800 series: http://www.anandtech.com/video/showdoc.aspx?i=3469&p=1

Also, GTX 280/260 were not competitive parts (not for NV anyway) they were designed to be $400 and $600 parts respectively. The 55nm refresh couldn't have come soon enough. Also, the lack of 5800 cards recently has played in NV's favor by allowing them to keep prices high on their GT200b parts and not have to sell at a loss/cost.
Click to expand...

First, how weren't they competitive ? Are we going through NV's costs to produce them ? What do you care ? Competitive means that they compete with the...er...competition and they did, so who cares what it costs to NVIDIA. What matters for the consumer is the price they pay for the hardware and last time I checked, both the HD 4870 and GTX 260 (which competed directly) cost about the same.

Second, NVIDIA doesn't keep prices high on their GT200b cards because of not selling at a loss. There was a shortage of 55 nm orders, because both AMD and NVIDIA are focusing on their product portfolio @ 40 nm and thus do not want to have excess of 55 nm parts.

http://channel.hexus.net/content/item.php?item=20566
http://www.digitimes.com/news/a20091005PB204.html

Fact checking is a good thing you know ?
 
NKD said:
you said ati competes with them since the release of 4800 series, competes the them finally? you mean gtx 200 series with the hd 5000 series, if I misunderstood you I am sorry but to me it sounded like hd 5000 competes with gtx 200 series since the release of hd 4800 series, so it basically means that hd 4800 series was nothing and hd 5000 is true competition for gtx 200 series, hd 5000 is not a competition for gtx 200 series it is way ahead in feature and performance.
Click to expand...

How the hell did you figure that from my post ? I didn't even mention GTX 200s or HD 5000s in my post and that's what you took from it ??? :eek:

I quite clearly replied to someone that said that NVIDIA quarter after quarter, didn't have a competitive product, when in fact, up to one year and a half ago, NVIDIA had absolutely no competition and dominated the market. ATI only started being a worthy competitor again, with the release of the HD 4800s, which is what I said in my post.

With the HD 5000s they are obviously ahead of NVIDIA.
 
This is for the usual suspects who bashed Charlie at the start of the thread.
Charlie is far more credible than any of the upset nobodies bashing him. Period.
He has proven to have real sources inside the industry. If he exaggerates to generate hits to his site that is his prerogative. If that somehow hurts your ego, you need to get over yourself, no one is forcing you to read his site or comment on anything he says. The fact that people feel the need to bash Charlie and do Nvidia's damage control PR work for free speaks how effective his tactics are. I don't see anyone complaining constantly about Fuad. If people bash only Charlie and not Fuad that makes them hypocrites.

On this topic. Let's see if the same people who bashed AMD for having low supply of RV870 on the first couple of months after launch call out Nvidia for their upcoming limited quantity launch of Fermi. Something tells me we won't hear a word from them.
 
Last edited:
Once again Charlie's name pop up. Well, I don't really care what that guy have to say, he can bash or praise nVidia all he wants. Doesn't change the fact that these are all just guesses and predictions, which is a hit or miss thing, sometimes he's lucky, other times he makes a fool of himself. Anyone who make their buying decisions base on rumors and predictions are making a huge mistake.

I agree with Kyle, I don't think nVidia's next gen card will be out in the market by February. We'll see. What I'm really interested in are the benchmarks when its released and tested by reputable sites like hardcop.com here. That will be the decisive factor in a person's buying decision. (I saw some user mentioning that charlie made some estimation of the performance base on clock comparison, which IMO is hardly accurate at all even for a wild estimation)

I also agree that currently there's very little incentive to buy a new flagship card. For my next build, I'll probably be looking for a cheap GTX 275 as a temporary solution, which I will use for physicsX thing later on. If Fermi is a disappointment, I think I'll be just fine with a GTX 275 anyway and just put in my current noisy 9800GTX for physx :p . (not that I have anything against ATI, but I'd love to have physx soon and I run folding@ home too)
 
What he says is soooo irrelevant and yet look at this thread. Says it all. Spin spin spin for nvidia. He does have better sources than any of us here.
 
Lorien said:
This is for the usual suspects who bashed Charlie at the start of the thread.
Charlie is far more credible than any of the upset nobodies bashing him. Period.
He has proven to have real sources inside the industry. If he exaggerates to generate hits to his site that is his prerogative. If that somehow hurts your ego, you need to get over yourself, no one is forcing you to read his site or comment on anything he says. The fact that people feel the need to bash Charlie and do Nvidia's damage control PR work for free speaks how effective his tactics are. I don't see anyone complaining constantly about Fuad. If people bash only Charlie and not Fuad that makes them hypocrites.

On this topic. Let's see if the same people who bashed AMD for having low supply of RV870 on the first couple of months after launch call out Nvidia for their upcoming limited quantity launch of Fermi. Something tells me we won't hear a word from them.
Click to expand...

Only recently has he ben RIGHT about anyhting related to Nvidia and that has only been about tapeouts and dates regaurding those. Prior to this, anything concerning Nvidia has has been more wrong than right. Even the boy who cries wolf continously eventually gets it right on occassion. His track record is shit. He said R600 would thrash G80, just the oppisite, said R670 would retake the crown, nope didn't happen. Said R770/700(refferring to BOTH not 1) would retake performance crown. Only R700 did until GT200b arrived. Claimed GTX275/295 were canned, then continued to claim GTX295 would never see light of day becuase it was impossible to do with do monster chips. Then whensaid It wouldn't take back performance crown. All this was before G300 tapeout reports on issues concerning it and 40nm. So get your facts stright about his credibility when it comes to Nvidia.

As far as Fuad goes, most people give him the "grain of salt" attitude that ATI lovers seem to not give Charlie.
 
Silus said:
Have you ever heard that even a broken clock is on time, twice a day ? Getting one thing right, doesn't make the dozens of other things he got wrong, right. I already pointed quite a few in this thread, which obviously he never admitted he was wrong.
.............
But that's not surprising...if it's something that can be twisted to fit his agenda of "NVIDIA doom and gloom" he posts it and obviously his "fans" plaster his links all over the place.
Click to expand...

It's not one thing it's many things and you just decided to ignore it. Also, Nvidia IS in dire straits at the moment. Their chipset business is all but dead, IGPs are going to die soon as well along with the chipset. Their market share and their reserve cash can only last so long if they don't have competitive products.


Silus said:
Again the usual mistake. Tesla is not clocked lower "for whatever reason". Teslas exist for the HPC market and they undergo a much more rigorous certification process, which makes them be clocked in a much more conservative way, when compared to GeForces.
Click to expand...

Just from comparing the GTX280 and the Tesla C1060 it doesn't look like there's a clock difference between them except for memory speed. Why would they need to clock the new Tesla lower? Hopefully Fermi's thermals aren't f'ing insane.

Silus said:
Memory controller probs ? There are no confirmed memory controller probs. He seems to have good insider info on tapeouts I'll give you that. As for yields, they are of public domain, with TSMC itself stating them. And they affect both NVIDIA and ATI.
Click to expand...

Yield was referring to the number of functional chips that NV was getting from their earlier steppings. Pretty sure ATI was receiving back more than a handful of functional chips at the time. :)

Silus said:
First, how weren't they competitive ? Are we going through NV's costs to produce them ? What do you care ? Competitive means that they compete with the...er...competition and they did, so who cares what it costs to NVIDIA. What matters for the consumer is the price they pay for the hardware and last time I checked, both the HD 4870 and GTX 260 (which competed directly) cost about the same.

Second, NVIDIA doesn't keep prices high on their GT200b cards because of not selling at a loss. There was a shortage of 55 nm orders, because both AMD and NVIDIA are focusing on their product portfolio @ 40 nm and thus do not want to have excess of 55 nm parts.

http://channel.hexus.net/content/item.php?item=20566
http://www.digitimes.com/news/a20091005PB204.html

Fact checking is a good thing you know ?
Click to expand...

Wotan said:
Also, GTX 280/260 were not competitive parts (not for NV anyway)
Click to expand...

When your part costs significantly more to make than your competitor's at the same price point your part is NOT competitive. Not to mention right now you can buy a 4890 for less than a GTX 260. That's pretty pathetic.

Yes, I know about the shortage, but for NV it's also helpful that ATI was having problems with 5800 supply. For a while there was nothing that ATI had to offer in the higher price ranges. Even now there is an enormous gap between the 4890 ($200) and the 5850 ($300) that needs to be filled and is currently being filled by NV with the 275.
 
Last edited:
XMAN245 said:
As far as Fuad goes, most people give him the "grain of salt" attitude that ATI lovers seem to not give Charlie.
Click to expand...

How about because Fuad is even less reliable? I haven't seen Fuad say anything accurate or on-time, basically, ever. He said Fermi was fine months ago, then changed his tune a month later... long after Charlie - and so far, still no Fermi

As for the rest of your post, Charlie has rarely been right about performance comparisons across cards, and here's the kicker - he doesn't appear to have good sources with ATI. Any time he tries to bring ATI into the mix, he's usually wrong. An example - he had no clue whether RV770 had 800SP's, when other sources were saying it was.

While he loves making every Nvidia situation dire, should we forget that he talked about the death of their chipset business before it was public? Fact of the matter is, his track record on NVIDIA BUSINESS ONLY is far more than just "a broken clock is right twice a day." It's when he tries bringing in comparisons to other things he doesn't have insider info on, he starts dreaming up all sorts of scenarios :rolleyes:
 
something interest more is this..

http://translate.google.com/transla...bir_kez_daha_gundemde-16991.htm&sl=auto&tl=en

bheg.jpg


I kind of doubt if this is even true though, since 40nm is just beginning to get mature, how in the world they start making 28nm...
 
edifyingkarma said:
Yep, good ole, completely unbiased Charlie.



It is completely valid to not trust a word this man writes because anyone with that level of brand worship and brand hate should not be taken seriously. A rational person would report on facts not based on the whole "my team rules, your teams drools" mentality. He is an embarrassment to IT journalism and should be fired. Even the doomsayers make accurate predictions once in a while, that doesn't make a single word that comes out of their mouths credible.
Click to expand...

um, for the record if charlie is giving GOOD news about a Nvidia product its because he didn't have a choice. we can probably take it to the bank here
 
ZerazaX said:
How about because Fuad is even less reliable? I haven't seen Fuad say anything accurate or on-time, basically, ever. He said Fermi was fine months ago, then changed his tune a month later... long after Charlie - and so far, still no Fermi

As for the rest of your post, Charlie has rarely been right about performance comparisons across cards, and here's the kicker - he doesn't appear to have good sources with ATI. Any time he tries to bring ATI into the mix, he's usually wrong. An example - he had no clue whether RV770 had 800SP's, when other sources were saying it was.

While he loves making every Nvidia situation dire, should we forget that he talked about the death of their chipset business before it was public? Fact of the matter is, his track record on NVIDIA BUSINESS ONLY is far more than just "a broken clock is right twice a day." It's when he tries bringing in comparisons to other things he doesn't have insider info on, he starts dreaming up all sorts of scenarios :rolleyes:
Click to expand...

They aren't out yet. Their chipset business is on hold until the issue with intel is resolved. They currently have chipsets ready for TSMC to make but are holding off making them until the licencing issues are resolved. Again, Charlie was wrong, but keep trying to make his Nvidia Business affairs look good.
 
If they're legit, those numbers are pretty impressive. But the AMD GPUs, especially the 5970, will only improve over time, so having a comparison now is a bit premature. Wait till Fermi actually comes out, and do the comparison with the contemporary drivers then.

The real question is, will those numbers be good enough to justify the price that NVidia wants to sell them at?
 
Mr. Pedantic said:
If they're legit, those numbers are pretty impressive. But the AMD GPUs, especially the 5970, will only improve over time, so having a comparison now is a bit premature. Wait till Fermi actually comes out, and do the comparison with the contemporary drivers then.

The real question is, will those numbers be good enough to justify the price that NVidia wants to sell them at?
Click to expand...

Most likely bogus. Anyone can make such slides. As you mentioned, ATI drivers are not mature yet and I would think the same would be for Fermi. I doubt Nvidia have performance numbers ready to be "leaked" yet. We're talking about a card that will come in Feb/March/April sometime... Too early to launch performance numbers yet.
 
Mr. Pedantic said:
If they're legit, those numbers are pretty impressive. But the AMD GPUs, especially the 5970, will only improve over time, so having a comparison now is a bit premature. Wait till Fermi actually comes out, and do the comparison with the contemporary drivers then.

The real question is, will those numbers be good enough to justify the price that NVidia wants to sell them at?
Click to expand...

For their competition its not really looking good. AMD is just not gonna sit on their hands, they're most likely working on the refresh right now. Just what happens when you're late to the party 5+ months.
 
For their competition its not really looking good. AMD is just not gonna sit on their hands, they're most likely working on the refresh right now. Just what happens when you're late to the party 5+ months.
Click to expand...
Typically, they're both working on something, all the time. It just depends on when their work is ready to release. But yeah, I'm looking forward to the 5870 refresh and Fermi at roughly the same time; should be good for prices...
 
Silus' step-by-step workflow:

- Continuously browse [H], Techreport, and potentially others, while watching for nVidia threads / news posts.
- If (nVidia thread detected), activate "lol Charlie" irresistible defense.
- if (personal credibility attacked), attempt to convince that he's unbiased.
- Always: attempt to change topic to "where's x AMD rumor? huh?"
- Return to initial step and repeat.

Just let it go, man. ;)
I think most of us see you for who you are, but learn to take every piece of news, nV or AMD in stride. It would be much better to return evidence of your own and do your own research with respect to the truth of reportings as opposed to simply activating your defensive mode.

Also, before you ask, yes, I'm watching you. In fact, even if I don't want to, you still manage to show up everytime there's an nV posting.
 
Tamlin_WSGF said:
Most likely bogus. Anyone can make such slides. As you mentioned, ATI drivers are not mature yet and I would think the same would be for Fermi. I doubt Nvidia have performance numbers ready to be "leaked" yet. We're talking about a card that will come in Feb/March/April sometime... Too early to launch performance numbers yet.
Click to expand...

It would be in the ballpark of what I would expect to see from Fermi, though. Especially with the oncoming refresh from AMD it would need some breathing room so to speak to keep up with what will be 'current gen' by then.

Also, we know that A2 silicon has been manufactured and tested (engineering samples). It wouldn't be improbable that some benchmarks were run with it. Which would also mean that the A3 revision is probably going to beat these benchmarks, if they're real :p

Ah, I so love pre-release speculation :D
 
Mr. Pedantic said:
Typically, they're both working on something, all the time. It just depends on when their work is ready to release. But yeah, I'm looking forward to the 5870 refresh and Fermi at roughly the same time; should be good for prices...
Click to expand...

Exactly and I think thats what some nvidia fans in this thread seem to think. 30% more performance than 5870 great job Nvidia however may be not enough cuz of what I said earlier being late. Their parade will be short lived and rained on.
 
Cyrilix said:
Silus' step-by-step workflow:

- Continuously browse [H], Techreport, and potentially others, while watching for nVidia threads / news posts.
- If (nVidia thread detected), activate "lol Charlie" irresistible defense.
- if (personal credibility attacked), attempt to convince that he's unbiased.
- Always: attempt to change topic to "where's x AMD rumor? huh?"
- Return to initial step and repeat.
Click to expand...

lol I guess I wasn't the only one that noticed this pattern.
The moment I saw a Charlie article thread 6 pages long I knew Silus would be fueling it with multi-quote defensive mode.

I think this fanboy reeking thread can RIP now.
 
Haha, im relatively new to this forum and am trying to get reads as to the bias' and wow, thank you Cyrilix, i was thinking the exact same thing but wasn't completely sure. I can rest now.

Anyways, the general consensus thus far is this: Charlie's timetables on Fermi are fairly accurate. Everything else is BS.

Put it to rest now. As for the tables posted on the GT3xx numbers, there's another thread for that, but i'll save you the trouble of reading. They're fake.
 
You must log in or register to reply here.
Back
Top