Printable View
[S|A]
http://www.semiaccurate.com/2010/03/...artners-cebit/
"Most of the questions were answered a bit ago, but lets recap. The benchmarks are on par with the ATI HD5850 for the GTX470 and about 5 percent above the HD5870 for the GTX480, but both Nvidia cards run much hotter. Clocks are 625/1250MHz, way short of the intended 750/1500, and wattage is through the roof."
That's what he had said in the previous article.
+ classical NDA's terms ... Bad nVidia !!!!! Naughty nVidia !!!! :ROTF:Quote:
Originally Posted by annihilat0r
What is nVidia really trying to achieve by doing this? They are basically pissing off partners, while the specs will have to be made public sooner of later anyway... Then again, if it could prevent people from buying ATI GPUs right now, because it isn't worth the wait, and for them to wait for Fermi, they might actually be able to sell to those people by having a good marketing campaign and similar performance. Then again, it could also be that fermi is extremely powerful, and they want to take ATI by surprise. But then again, why leak those farcry2 and heaven benches, where it has a huge lead over the HD5870? Those could be cherrypicked benches to prevent people from buying ATI...
Argh, this is way too confusing. I'll just sit tight and wait and see. I already have 2x 5870, so I don't have any reason to buy Fermi anyways. I'll just enjoy this setup (and maybe add a third 5870 if I feel like it :p:), and wait for the next generation. We're already getting close to ATI's HD6xxx series. Speaking of which, would be fun if ATI released info about the 6-series as soon as Fermi lands, to make to waiters wait some more :p: I've learned my lesson a long time ago. Stop waiting for hardware, and just enjoy what's readily available.
...anyone actually believes Charlie? ...really?
"wattage is through the roof." for Chuckles means a 2A bump in +12V requirement over GT200, for 2.3x the perf in some apps. 5% more power, double the performance.
Quote:
Originally Posted by Gilgamesh
Yes. And even if he's completely wrong now everyone will soon forget and the cycle will repeat as it has so many times in the past.Quote:
Originally Posted by Amorphous
fixedQuote:
Originally Posted by trinibwoy
Except when it comes to Charlie and Nvidia your "fix" isn't relevant at all.
somebody told me but i think he was wrong :DQuote:
Originally Posted by annihilat0r
technical analysis? :confused:Quote:
Originally Posted by trinibwoy
i said that imo doom3 sucked, hows that a technical analysis? :D
and whooooo fully dynamic shadows... so what?
you apparently liked it, i didnt... it looked cool in a few moments but that was it...
back then everybody was betting on hl2 and doom3 and argued which one would look better, and then there was far cry... and suddenly nobody talked about doom3 or hl2 anymore... and for a reason! :D
o rly? :PQuote:
Originally Posted by zerazax
the fastest 1-2 cards are usually beyond reasonable price scaling, or on the limits thereof, but right now even the 5970 has an almost linerily adjusted price compared to prices and perf of other cards:
name - cost - %perf - %price - %/%
HD5970 2gb 600$ 100% 100% 0
GTX295 2gb 525$ 81% 87% +6
HD5870 1gb 400$ 71% 67% -4
HD5850 1gb 325$ 62% 54% -8
GTX285 1gb 375$ 58% 62% +4
GTX260 1gb 225$ 47% 37% -10
HD5830 1gb 250$ 45%*1 42% -3
HD5770 1gb 175$ 43% 29% -14
HD4870 1gb 175$ 44% 29% -15
HD5750 1gb 150$ 37% 25% -12
GTS250 1gb 125$ 36% 21% -15
HD4850 1gb 125$ 35% 21% -14
10 memory chips = 5 64bit channels = 320bit memoryQuote:
Originally Posted by Funky
yepp, sounds like a 470 :)
which is perfectly understandable :DQuote:
Originally Posted by Funky
well thats the power requirements which doesnt necessarily mean they consume the same power... especially in idle... but yes, i think wattage through the roof is exaggerated... but we will see...Quote:
Originally Posted by Amorphous
GTX280 specs:
http://www.nvidia.com/object/product...tx_280_us.html
Maximum Graphics Card Power (W) 236W
Minimum System Power Requirement (W) 550W
GTX285 specs:
http://www.nvidia.com/object/product...tx_285_us.html
http://www.nvidia.com/object/product...tx_280_us.html
Maximum Graphics Card Power (W) 204W
Minimum System Power Requirement (W) 550W
both have the same psu requirements but 15% difference in max tdp
and if you look at actual power consumption...
http://www.xbitlabs.com/articles/vid..._12.html#sect0
GTX280 178W max
http://www.xbitlabs.com/articles/vid...f_7.html#sect0
GTX285 149Wmax
those are a far cry of the actual max power a vga can consume if paired with a 550W psu. you need at least 12A on the 12V rail for the cpu, lets say 20A for the whole system, thats plenty even when overclocking... (240W) so you have 22A left (nvidia wants 42A for the gtx4xx cards) which means there would be 260W of power available for the vga alone.
everybody knows how ridiculous psu ratings are and that they hold no real meaning, so everybody recommend way higher psus than what you really need... so i dont think you can make any statement whatsoever based on the psu requirement...
Compared to the other turd the fantards (now actually worse than Apple's legion justifying the A4 in the iPad) and PR plants are throwing all over the place?Quote:
Originally Posted by trinibwoy
He's not credible by himself in the GF100 scenario- it's the people who try to deride him and then come out with a complete bunch of IQ 50 malarkey that make him credible by default.
Do these fantards all have websites that try to pass off speculation and rumour as fact? I'm not sure what point you're making anyway since there are far more AMD goons than Nvidia goons running amok. Add Charlie to the mix and it's 2:0 in AMD's favor ;)Quote:
Originally Posted by Macadamia
No you said that Doom 3 was just bump mapping everywhere which isn't a subjective opinion - it's a claim about the game engine which is patently false. I don't have an opinion on the game itself since I never played it.Quote:
Originally Posted by saaya
so because i know what bump mapping is, its technical criticism?Quote:
Originally Posted by trinibwoy
i have no idea about programming a game... for all i know jc is a genius or at least pretty damn smart at what he does, that doesnt mean everybody will like their games though...
im pretty sure his goal wasnt to make doom3 the most technological advanced and awesome looking game, not at all... he wanted to play with light and shadows and wanted (or was pushed to) make the game run acceptably on even weak old gpus.
HL2 was easily better than Farcry or Doom. : >
Yes.Quote:
Originally Posted by tajoh111
Slightly shorter from the look of it.Quote:
Originally Posted by zerazax
Huh? Lol you lost me now....maybe you should re-read your post that I replied to.Quote:
Originally Posted by saaya
Yes, we believe him more than nvidia's amorphous PR, thank you.Quote:
Originally Posted by Amorphous
WHAT have you forgotten FUDO, the nvidia pr spinning machine? the guy who in november was still saying fermi was a 2009 launch?Quote:
Originally Posted by trinibwoy
What about Theo? he is not as inherently stupid as Fudo, but he is an nvidia man.
All i know is that if it wasn't of charlie, all the information we would have gotten about fermi by now, is that it was a 2009 launch written up by Fudo LOl, and that is late as hell. At least with charlie we now somethings about the why
I thought I would actually take the time to read the S / A article and this time start idsecting it a bit:
The NDA looks exactly like the one partners had to sign before the GT200-series launch.Quote:
Just before the show started, Nvidia reps ran around and made their partners sign a ham-handed NDA list of what is acceptable and not acceptable to talk about at CeBIT.
I've am in contact with several large competing websites on an almost daily basis. No one linked to it because of the raving contained therein.Quote:
Last week, several sites contacted SemiAccurate saying that Nvidia called them and threatened to cut them off if they linked to the above story.
Haha. Can't say more than that.Quote:
The NDA looks like this, with text below the picture.
Two 6-pin connectors on the 470's pic looks par for the course.Quote:
wattage is through the roof.
Other than overclocked cards, I have yet to see clock speeds written on the boxes of many (if any) manufacturers of previous NVIDIA cards and even current ATI cards.Quote:
Two box samples seen by SemiAccurate had no clocks or specs on them, quite telling for a product that will paper launch later this month.
And the backpedaling begins! Watch this; if a 512SP card is released, Charlie is now covering his tracks after claiming there would never be one.Quote:
Pricing wasn't set according to four partners talked with by SemiAccurate, and availability for the high-end part, at least according to what Nvidia has promised to deliver, is basically a few hundred units per vendor. Worldwide. Ever.
Nobody takes them seriously so they're harmless. Didn't you see how far and wide Charlie's software tessellation spread even though he doesn't know his ass from his elbow on the topic :ROTF:Quote:
Originally Posted by JohnJohn
True, we can thank him for that. Thanks to him we now know exactly what we would've known if we listened to an Nvidia conference call and looked at a calendar. Strip away the vitriol and what else has he told us besides it's late due to yield issues?Quote:
All i know is that if it wasn't of charlie, all the information we would have gotten about fermi by now, is that it was a 2009 launch written up by Fudo LOl, and that is late as hell. At least with charlie we now somethings about the why
Err wha?Quote:
Originally Posted by trinibwoy
Fudo, Theo, THG, Expreview, Focus Group Members.....
BTW at least I know that nobody at DAAMIT ever paid people (marketing agencies who thought this was a good idea) to post on websites without not disclosing themselves as paid. Not rich enough. :rofl:
Journalism of the new age :DQuote:
Originally Posted by SKYMTL
Erm... Nedjo ? :p:Quote:
Originally Posted by Macadamia
64NOMIS ( although I give lots of credit to Simon he seems to be a great guy and he isn't acting like an idiot or just flaming the competition )
Isn't that old news?Quote:
Originally Posted by SKYMTL
Feb 17: http://www.semiaccurate.com/2010/02/...and-unfixable/
Feb 20: http://www.semiaccurate.com/2010/02/...gtx480-scores/Quote:
For those not interested in the 'why' side of things, the short answer is that the top bin as it stands now is about 600MHz for the half hot clock, and 1200MHz for the hot clock, and the initial top part will have 448 shaders.
And people wonder why his credibility is questioned. The hilarity.Quote:
There are two cards, the GTX480 having the full complement of 512 shaders, and the GTX470 with only 448, which is 64 less for the math impaired.
double performance on GTX280 - 236W TDP?Quote:
Originally Posted by Amorphous
http://www.t-shirthumor.com/Merchant...e/iwtb_lg2.gif
We will see next 26/3, I hope :)
Just go on record as a nvidia employee and say the 480 will be more than 5% faster than the 5870 across the board (sure you can have a couple of exceptions, game engines tend to like certain cards over others), and we'll all believe you. :rofl:Quote:
Originally Posted by Amorphous
You quoted someone saying he said that there wasn't going to be a 512sp card, and the you quoted him saying the top card is 512sp.Quote:
Originally Posted by trinibwoy
Yes sir you nailed him :rofl:
Off course he is wrong, there's no problems with fermi, the months of delay and the respins are just because nvidia wants to give amd a chance.
Off course he is wrong, amorphous is right, fermi will be 4x times faster, again nvidia just doesn't want to hurt amd sales, by showing some benchmarks, hence the secrecy.