I like how big that fan is. It's bigger than HD 5850/5870 reference fan, isn't it?
Printable View
Too many people have assured us of too much stuff about Fermi. So I'll take this info with a grain of salt.
That's what I am hoping for as well. Hope these cards will kick some a$$! :D
Yeah, looks it. 4 heatpipes, huh? Reminds me of GTX295 for some reason.
guys at Donanimhaber pointed out that this shot was rumored to be a "GTX 390" some time ago. whatever it is nvidia isn't gonna put FOUR fermi chips on one card that's for sure :D
http://img6.imageshack.us/img6/7158/img1284s.jpg
http://img6.imageshack.us/img6/1328/img1287n.jpg
^ Woah, disappointing. :(
Thanks for mentioning it.
dont be disappointed, from what I've been hearing, GTX 400's will indeed feature reference coolers with heat pipes.
Sorry. I'm screwed from this link :(
http://translate.google.com/translat...-per-il-CeBIT/
I hope the 470 doesn't require an 8-pin connector. My PSU doesn't have any :(
Another set of pics. This is 99% GTX 470 that Nvidia partners received. (No stickers fermi with 2x 6 pin power connectors, hence conclusion that this will be 470)
http://thumbnails25.imagebam.com/701...8270166808.gif http://thumbnails13.imagebam.com/701...5670166809.gif
nah , it doesn't make sense at all for me , fan has to blow most of the air over the heatsink , the way it is made most of the air will flow from the back of the PCB without ...
if that is the intake then the exhaust from the other side "top" , and both ways doesn't make any sense and wont circulate enough over the heatsink ..
Here is a picture of a 280GTX heatsink
http://www.pcper.com/images/reviews/577/heatsink.jpg
I thought the power connectors on fermi were in 90 degree angle.
Its not conventional fan. Its good old centrifugal fan design just like we have seen on other GPUs in modern history. These fans dont direct air downwards. It throws air away from blades around circumference.
Any air that moves down just creates turbulence. So this gap allows air intake from both sides as well as helps in reduction of turbulence.
And about second pic, its just bad cam and shaky hands ;)
i just want to see some prices and performance numbers only because ill be somewhat upset if their sub 500 card is faster than the 5870.
Since when did nvidia start including two sli bridge tabs?
For the sake of competition, why WOULD you be upset? Unless you bought a 5870 in the last 30 days which in that case, that was your decision. I hope the 470 more than matches the 5870 and that the 480 sneaks up on the 5970, for everyones sake. The only thing I'm remotely pessimistic about is the possibility than worse than initial 5870 availability...
If nvidia managed to get the 480 with a similar thermal and noise profile as the 285, I'll be very impressed. I'll hazzard a guess that it will be more 280ish which isn't too bad I suppose but nowhere near ideal (280s are rather loud and hot )
i'm a little bit worried about fermi clockspeeds/yields as we're just seeing pics of gtx 470. i mean, this is cebit. where you show off all you got, don't you? why isn't there a gtx 480 floating around somewhere? let's hope this has nothing to do with problems regarding clock speed and/or SPs of the gtx 480 chip...
I'm kind of disappointed with the stock heatsink (if that is in fact one). For as expensive as this card will be and the high TDP of the card, I wish they had put a bigger fan on the heatsink.
Based on this photograph...
http://img192.imageshack.us/img192/9554/gtx470rear.jpg
Could somebody guess-timate using photoshop how long this card will be? I'm specially interested as I'm modding a case and length is a special issue, and I'm not willing to do a ton of work to find out later that I can't put a high-end card inside :shakes::shakes::shakes:
IMO as that fan is centrifugal it will suck air from the middle (doesnt matter if its top or botton) and throw it outside. If you think about it it's actually good when you have all your cards together in SLI or 3-Way as the cards are better ventilated.
Lets not bash the horse until its here, would ya? :rofl::rofl::rofl:
wild guess: about MSI 5870 Lightning?
9.8 inches? end of the pci-e tab is 5 & 7/8 inches. the 480's pcb is longer i think. looking from the back on that 470 pcb pic, there is 1 mem pad to the left of the top row of memory, and one mem pad on the bottom of the left column on the 480s pcb for 384bit. this thing should be nice at 2x6pin 1280mb 320bit
If my depth perception is worth 2 craps, I'd say the same lenght as GT200 pcb / coolers. Past would dictate the 470/80 will be the same lenght as well.
Is fermi going to have any other tech demo than that rocket sled thing. Its kind of nice, but I prefer the more realistic look, than cartoony look.
Damn, you got me :D I would like some more pics of this bad little fellow though. Four GPUs, wow :eek: That perhaps is one of the best looking PCBs I've seen in my whole life!
That sure is some good pr0n :up:
You don't want to know how dusty cards can get within a few hours... It all depends on where they are used.
€dith: Jesus, would you mind stop quoting pictures?! Just put a dot in there to stop the code from being parsed just like I do...
Even if you take the angle into account the GTX 470 (??) ends up longer than HD 5850. Please take into account three (biiig) Caipirinhas (and one glas of whiskey).
the PCb cut will help the top vga to breathe in close SLI setups.
BUT the flat plastic cover, makes no room for air in TrI-SLI setups, middle card speaking.
Should be angled, like GTS250.
and it looks rather short, I expected a longer vga....
So what's the word on the launch date now, I've been out of the loop and read the last 50 posts or so :)?
march 26
Thanks annihilat0r.... :).
Given the angle the pic is taken at and all that jazz, my guess is roughly 5870 length?
well pcie slot is ~4"..so id say the pcb is 10"
Nice, an actual image of the card, I wish mid-range variants arrive 3 months or so after release of the high end ones.
Well, GTX480 definitely has a 10.5" PCB so 470 being ~10-10.5" would make sense.
Best case is that they might, though realistically it will probably be early to mid Q3.
Also for those that might have missed it or forgot...
GTX480 = $549-579 and GTX470 = $449-$479 is still the best price estimates that have been hinted at.
This looks bigger than 5850 but slightly shorter than 5870.
Here is a pic of 5870 at similar angle.
Any photoshopper having some skills can try to match this with the pic of backside of this card.
I tried ( honestly I am no expert at photoshop) and from what i can tell, its about 1.5 - 2 cm or so shorter than 5870.
So here is reference for those with skills to put them next to each other.
5870
http://thumbnails17.imagebam.com/702...0870243132.gif
GTX470
http://thumbnails13.imagebam.com/701...5670166809.gif
[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.
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.
Except when it comes to Charlie and Nvidia your "fix" isn't relevant at all.
somebody told me but i think he was wrong :D
technical analysis? :confused:
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? :P
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 memory
yepp, sounds like a 470 :)
which is perfectly understandable :D
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...
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?
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 ;)
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.
so because i know what bump mapping is, its technical criticism?
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, we believe him more than nvidia's amorphous PR, thank you.
WHAT have you forgotten FUDO, the nvidia pr spinning machine? the guy who in november was still saying fermi was a 2009 launch?
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:
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
Isn't that old news?
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?
http://www.t-shirthumor.com/Merchant...e/iwtb_lg2.gif
We will see next 26/3, I hope :)
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.
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.
And then he said that there will only be around 3000 512SP cards...
And after a while he'll say that there will be 8000 512SP cards...
And and and...
According to you he's right all the time.
He's right when he's wrong, and he's wrong when he's wrong, so he's right again.
Maybe he's one of the Gods :D
If you want to take everything he says for granted, then good for you :)
http://en.wikipedia.org/wiki/Infinite_monkey_theorem
Basically, if someone writes enough FUD (like a thousand-word rant about NVIDIA, their chips, etc.), some of it may turn out being right but that doesn't equal KNOWLEDGE about the topic.
GTX 480 sale price UK £449.99 :up:
GTX 470 sale price UK £379.99 ;)
dont ask where i got these numbers cause i aint telling, but this is the recommended RRP for these parts come end of march :clap:
I don't take what anyone says about Fermi for granted. Fact is though, a bunch of nV fanbois on this and many other forums were making some bold claims about Fermi (of course they all had veeeeerry reliable source) which all turned out to be nonsense. Add to that nV PR spinster Fuad and Theo and suddenly Charlie looks like a fscking genius.
SO as far as I'm concerned none of you know sjit about Fermi ;)
lolwut, so little difference between 470 and 480?
and 5970-like priced GTX 470? Great! http://www.amazon.co.uk/Sapphire-HD5...7542041&sr=1-1
All this over a graphics card, no wonders there's unrest in the world. Imagine the discussions here if it was over something really substantial, like world peace, Santa Claus... :p:
There's only a Sapphire 5970 under 500gbp and that's so unfortunately... out of stock :p:
Cheapest 5970 in Europe is 535 euros in a few German shops ( didn't check availability ).
If the GTX 480's performance is where I've been informed and said in the past that it might be, meaning close to a 5970, then a UK price of 470gbp GTX 480 doesn't sound bad at all.
But there are 5870's for 300 pounds.
A 470 selling for 380 pounds would be total fail.
Charlie got the Q1 2010 release of fermi right, and everyone who was making fun of him then choose to ignore it now, like, you know, benchzowner et al. It's very unlikely that was a wild guess, so any claims right now should be taken with a grain of salt.. charlie for being an nvidia hater, and most of his detractors for being fanboys.
Rumors are fun though. :)
GTX 285 - £289.99
GTX 470 - £379.99
GTX 295 - £429.99
GTX 480 - £449.99
here is how it sits in the price line up, this is what i am being told by two retail sources (buyers)
and to be honest this looks bang on the button, because you cant get a GTX 295 for love nor money, your only top end option will be the GTX 480.
and from what i am hearing the GTX470 is more than 5% faster than a 5870, more like 10 -15%, coupled with the close relationship with big game houses, this would make sense.
but as benchZowner is saying we still have to wait and see, im not here to get into a debate on what i hear, only to post it ;-)
Yeah just like Q1 is just a monthly timeframe :p:
Let me predict something... Intel will release the 6 core CPUs in 1-4 months from now :D
If there was a chance of Fermi slipping from the initial release plan from nVIDIA ( they intended to have it ready & launched in November { verified from various sources }, unless you believe they planned all this delay for fun :D )
Charlie the chimp can't be ignored because he has a million sites recycling his ironically unbiased and 100% correct "news articles" every single day.
And yes, we everytime we hear about him we laugh out loud :D
He spits 1000 of lines about nVIDIA every day ( no he's not a hater, and he isn't biased at all :D ), he does have some sources [ some good some bad, some biased some unbiased ] but most of the times he pulls more than a couple dirty tricks and "insider info" right out of his as$.
Whenever he's wrong he backpedals in a certain way to make people believe that he was never wrong... right Charlie ? :D
Some new cautious wording from Fudzilla:
http://www.fudzilla.com/content/view/17889/1/
I dunno. This smells bad for me. Everybody pretty much expects that GTX 480 will give HD 5870 a rough time. If it's only "faster in some cases" and quite bigger, then this is big trouble for Fermi as a GPU architecture.Quote:
Nvidia's GTX 480 ends up faster than HD 5870 in some cases and according to Nvidia it is the fastest GPU on the market, or to be precise, it will be faster than the HD 5870 once it finally shows up.
We must note that we can confirm that it does happen in some cases, as Nvidia is still keeping the performance numbers for itself as it wants to keep AMD at its toes. Nvidia reps are pretty sure that AMD is still tapping in dark when it comes to Fermi. The GTX 480 is faster in Unigine Heaven benchmark in some situations where actual GPU power does matter.
All this is pretty good news for Nvidia, but it still has to launch the card, while AMD has been selling its HD 5000 series cards for months now.
He said in 20\02 there was going to be a 512sp, someone said today he was going start to back peddling because in fact there is going to be a 512sp, and you agreed, do i need to make a drawing or it's this enough? And i don't know how why correcting in 20\02 something he said in 17\02 is a bad thing.
The site is called SemiAccurate... get it? So why would that be :rolleyes:
Bravo, your quote coundn't have been more selective and out of context, of course you ignored the rest, it would have only aggravated you nervous breakdown.
So where's the knowledge about the topic? On this thread? :ROTF:
Remindz me when Fudo was more "red" :
Radeon 4850 goes against 8800GT
You know what happened ... :DQuote:
8800GTX/9800GTX will win against Radeon 4850, but they are the other price category that is much closer or level with the price of Radeon 4870.
He was actually correct in that article. That _was_ actually ATI's plan all along. Thing is, they managed to do some lastminute tweaks, which allowed them to use more shaders on the die. Far more than they had initially had hoped for. It even surprised ATI themselves.