GF100 has more ROPs then G200 and less TMUs then G200, but that is not a problem, has new units - raster a geometry units
Printable View
GF100 has more ROPs then G200 and less TMUs then G200, but that is not a problem, has new units - raster a geometry units
One thing is sure. GF100/Fermi is late :p:
I could be wrong and even if they are working on it, prolly nvidia is at full-force right now, but I'm sure of something, we won't see any high-end parts based in 28nm anytime soon after all this delays etc. they have been experiencing trying to adapt a such arch into 40nm, to meet the specs as they were original designed/presented, again I could be wrong but nvidia is known by pulling tricks out of their sleeves, not quite with fermi though, also as we all know nvidia lately is slower making die shrink transitions:yepp:Quote:
Originally Posted by saaya
Nvidia is stretched pretty thin right now with Fermi and Tegra. I don't think they can maintain this pace for much longer, eventually they'll settle down and derivatives will come sooner since they won't be trying to revolutionize the industry with every new architecture.
Yeah, that's the official spin "they're much more efficient" bla bla bla.. it's still BS because the units are THERE, why disable them?Quote:
Originally Posted by Zed_X
Another hint, when Huddy says he expects AMD to have the fastest card for most of 2010, he's not talking about AMD refreshes.
Sounds like 5970 is going to have plenty of time to be king in the spotlight, then. Boring :(Quote:
Originally Posted by neliz
gf100 launch on march, availability on may/june, dual fermi june/july? & fully loaded gf100 on september :p:Quote:
Originally Posted by neliz
Or not so apocalyptic?
while the rest of ati's 5xxx destroys nvidia price/perfQuote:
Originally Posted by Aerwidh
ive seen hardcore nvidia fanboys switch to ati recently... bad sign for nvidia
i thought the 5xxx was worse than any performance comparable 4xxx in price?Quote:
Originally Posted by Newblar
Wow... that is ridiculous.Quote:
Originally Posted by neliz
You give all the info you can that is deemed credible, for both sides.
Just because most of the previous news on GF100/Fermi was bad doesn't mean you are trying to discredit Nvidia, the blame lies soley on Nvidia for that.
If I had (all) the specs for GF100, I'd post them straight up, just like I did with any info before it. Thing is, AMD seems to "not mind" and they actually ask random people to post parts of their specs. NV really seems to be uptight about it's info. That's why I only get the codenames like "Optimus" from their presentations a few days before their embargoed outing.Quote:
Originally Posted by LordEC911
But that's just the difference in companies. A marketing driven closed society versus a business driven open one.
Is post #905 an example of posting stuff straight up?
y-y-yes?:confused:Quote:
Originally Posted by trinibwoy
Ok, then. Just checking :)
So basically it won't be out until summer, then it won't be in stock until fall then it will be time to shrink it already to keep up with ATI/AMD.
Mmmhmm
It dosen't have to mean that GF100 itself is delayed. Nvidia may have delayed that dual GPU version, and from the estimates that we have seen, ATI should retain the performance crown with the HD5970 even after the "GTX380/GTX360" release.Quote:
Originally Posted by ulticool
It does mean GF100 availability has been delayed. I'm saying that MP doesn't start for quite some time, perhaps hinting at an A4 revision.Quote:
Originally Posted by H2O
Now.. Fermi2 otoh.. that might be closer than everyone thinks.
That's going to depend on when the 28nm process is available to nvidia. The design will be ready, but they are still susceptible to delays per the fabs.Quote:
Originally Posted by neliz
I think Fermi design is just crying for a shrink, if what I am hearing is correct.
well ATI seem to have headroom in that... i mean even on their slides they had msot stuff priced 50~100$$$ less plus im sure they could drop it more then that.... so I think they will jsut drop prices to kill the price/perf like last roundQuote:
Originally Posted by Manicdan
as they use tsmc it'll will be 32nm right 28nm is GF and I thought nvidia said they wont use them?Quote:
Originally Posted by xVeinx
is everyone forgetting ATI's supposed launch in september~october of a new complete arch... isn't that kind of bad for DUAL fermi if that seems to be hinting coming out around then :/.... hell if ati get their ass in to gear they may even be able to have it as a tool to take the thunder away from them....
saying that I hope nvidia's fermi suprises us atleast a little bit cos last round they got beat pretty bad(price/perf wise).... and this time it seems worse..... :/
I think it makes sense why ati is ahead right now.
Smaller gpus are easier to design, easier to produce, easier to reach expected speeds since they are less hot. And yields are better, way better.
I really don't understand why Nvidia is still going on with monolithic huge designs, it makes them loose ground to the competition, it's pretty stupid. :shrug:
Even if they get at some point the halo with top of the line products, if you don't have the perf/price ratio right and you cannot invade every segment of the market with products because you cannot shrink the design, then in the long term you will loose more than you gain.
yeah but 5870 is pretty monilithic... i thinks it's just random annoying crap thats keeping them back like solving one thign leads to another thing... :/Quote:
Originally Posted by Florinmocanu
rest i agree with... I think they will end up renaming again.. to stay competitive in the lower segments
TSMC drops 32nm in favor of 28nmQuote:
Originally Posted by Jamesrt2004
well, RV 870 at 330mm2 is a medium sized gpu, i would say over 400mm and more is where the real fun/problems begin, and at over 500 mm, fermi is big, one of the biggest ever made.Quote:
Originally Posted by Jamesrt2004
Still a rumour... plus they showd things at 32nm and saying it's on track so I personally don't believe it but as theres obv' some information out there i'll happily eat my words if wrong :up:Quote:
Originally Posted by Kuroimaho
ok fair point :P i'm just used to small-er chips from the last couple series I guess :up:Quote:
Originally Posted by Florinmocanu
but I agree that going monolithic although having advantages has a pretty heft(huge) downside, I think ati's approach has been much better :/ ok so this round their card was power hungry but as it was only a stop gap until the new arch we cant really complain too much....
I think fermi will show its actual power on the 32/28nm node's (w/e they choose to go with) I just hope it's sooner then later.... im not sure how far tsmc has got.. but we know GF seems to have some pretty complicated wafer's on the go that DON'T look like tests so they are obviously pretty easily on track.
TSMC doesn't seem as vocal.... apart from they showed that 32nm wafer(but I don't know much about it as i havent seen pics of it) but i hope it's pretty advanced/more advanced then GF's 28nm tech as nvidia could be helped quite a bit if they can get the shrink out around the ATI launch (as long as that isn't plagued as much as nvidia's one is/was)
They show 40nm things all the time and look. :D I know it's unconfirmed, I post it as a possibility. Shrunk fermi on 28nm 2011 Q1 anyone ? With free Rocketsled 2 now with split screen multiplayer mode ?Quote:
Originally Posted by Jamesrt2004
I prefer the fastest single gpu card but I am glad Ati was there last year to drag down the prices.Quote:
Originally Posted by Jamesrt2004
Does anyone know if/when we will get any real world Fermi numbers?
I really do hope the Fermi is not a hot (in terms of temperature card).
For those of you who helped me with the DOA GTX 285, thanks :up: I got an RMA for the DOA card yesterday and it POSTS but has other issues (will post a thread on this)
Aside from that I hope that the Fermi is not as flawed as the GT200 series
John
wow that hair demo is sick, once optimized its scary how little power it will take to do it nicelyQuote:
Originally Posted by Andrea deluxe
Thx for sharing buddy :up:Quote:
Originally Posted by Andrea deluxe
wow, that hair demo is really neat! anybody hear the mention of the geometry shader? Thats whats taking the place of some TMUs? Can't wait to see real performanceQuote:
Originally Posted by Andrea deluxe
Nice! Thank you :)Quote:
Originally Posted by Andrea deluxe
Very very nice Sir, thank youQuote:
Originally Posted by Andrea deluxe
Found the hair and the fary cry 2 video great.Quote:
Originally Posted by Andrea deluxe
For comparison my GTX275 does around 40's when OCed in Far cry 2 "2560*1600 with 2-4x AA" thus i think the other pc on the right uses a G200 based gpu but who knows :shrug: Anyways this feels more gtx295 performance-wise..
Also no idea on the settings used?
the vids are private now. a lucky few got to watch them.
scores are about 20% higher than a HD5870. settins were 1920x1200 4xAA 16xAF. the machine on the right scores like a GTX285 at those settings.Quote:
Originally Posted by ajaidev
And seriously.. it's a FC2 benchmark run, you've missed nothing. Other video's were DirectCompute applications, most of which you can download and run on your pc today. (i.e. the water demo)
nuuuuuuu they put it private now :( I watched it at least! very impressive FC2 and Hair demo! the water tessellation was awsome as well!
Damn, I knew I should have saved those videos after watching! Oh well, they were pretty cool demos though :)
http://translate.google.de/translate...n&hl=&ie=UTF-8
according to this:
gt360 overtake 5870 by
Crysis Warhead + 60%
Far Cry 2 + 20%
HAWX + 55%
Fallout 3 + 20%
maybe its fake maybe not.............:shrug:
So, who of you ripped them before they were taking down (private) ey?
If anyone is interested in the FC2 result, a friend saved a SS.
http://www.framebuffer.com.br/sites/...GF100_FC2_.jpg
I've got the results and the FC2 video saved in cache but don't know how to extract if possible?
GF100:
AVG: 84
Max: 125
Min: 65
Card compared with:
Avg: 50,3
Max: 72,8
min: 38,3
I was able to rip 3 off... uploading to youtube now
Also clear high def SS
http://img204.imageshack.us/img204/2563/copyofruko.png
http://img94.imageshack.us/img94/8383/ruko.png
The hair was really impressive, given how hard it is to do realistic hair but don't expect that on any in-game characters soon :)Quote:
Originally Posted by Chumbucket843
According to that same pconline link these numbers could be for a 448 core / 320-bit bus GTX 360 variant. If so, we could see even higher numbers from the flagship.
:clap:Quote:
Originally Posted by ajaidev
High-end models will deploy the full array of 16 Cluster A total of 512 CUDA Core computing unit, to support 384-bit GDDR5 memory body structure, body has a memory capacity of 1.5GB.Fermi 40nm 250W / 300W PCI-E 6-pin + PCI-E 8-pin Although the Fermi core has adopted 40nm manufacturing process, provided that such cards power up to 250W to 300W, and called for PCI-E 6-pin + PCI-E 8-pin power supply. 2 Cluster 1 448 CUDA Core 320-bit GDDR5 1.28GB 225W PCI-E 6-pin A second level model will take away the two Cluster group memory array and a body controller, which fell to 448 CUDA Core and 320-bit GDDR5, memory body with a total capacity of 1.28GB, power consumption is reduced to 225W, and only need two PCI-E 6-pin power supply.
Quote:
Originally Posted by Blacky
Dont be sad, i dont like it when you are sad here see as much as your heart desires :DQuote:
Originally Posted by Sumanji
NEW LINKS TO THE VIDS
http://www.youtube.com/watch?v=jcXmV5Je_gc
http://www.youtube.com/watch?v=5Juxq2tqHyI
http://www.youtube.com/watch?v=4Ie0t_LkImk
If thats true why did they use farcry 2 something that has only 20% advantage, better have used crysis to showcase GF100 instead!!Quote:
Originally Posted by Andrea deluxe
It seems the article talks about GTX360 (448 CC) performance compared with HD5870.Quote:
Originally Posted by Andrea deluxe
Wait & see ...
edit : Thx for those who hadn't seen the vidz ajaidev :up:
Thanks for the vids ajaidev.
Yay, thanks Ajaidev!
If that is GTX360 in those vids, then I'm impressed :D
almost 1000 posts about fermi, and we finally are getting something worth looking at. praise the guys who made the vids
Thanks to fellix
Ray-tracing
Hair
thanks neliz/felix hd yay
BTW i have the HD for the farcry 2 thing uploading it now...
Nice vids, really looking forward to more results!! :up:
The GF100 is about 1.68 times faster than the card on the right in the FC2 demo. I just cant tell if it's a GTX280, GTX285, 5850, or 5870. Anyone know>?
At 84 AVG and 65 MIN, it's a little faster than a 295 according to this:
http://images.hardwarecanucks.com/im...-HD5970-47.jpg
Well, if those GTX360 results are true, just the thought of a GTX 380 on phase gave me a fermi.
wow the hair is very impressive..hopefully this card lives up to the hype and makes up for the delays :)
but this test is in very high not ultra high.......Quote:
Originally Posted by jaredpace
Nice to see a 384-bit and 320-bit bus confirmation. Hopefully that will pull prices down a bit compared to the higher bit buses of last gen.
!!!!! :up: Thank you, so it's much faster than a 295, and that must be a GTX285 on the other PC.Quote:
Originally Posted by Andrea deluxe
so 1.68 x GTX285?! No way!
So compare to this:
http://images.anandtech.com/graphs/5...5500/21348.png
its hard to compare without having identical setups....does anyone know what setup they are using with Fermi..i7 at what clocks etc?
I wonder what kind of shampoo they used on the hair model...
If it's a 448 shader Fermi, then everyone is in for a treat. just noticed the GF100 84fps result says 01/10/2010 and the gtx285 50fps says 01/13/2010.
@Teemax pantene pro v without a doubt.
another great series of benchmarks about mysterious fermi :ROTF:Quote:
Originally Posted by Andrea deluxe
Farcry 2 in HD
http://www.youtube.com/watch?v=W-0iF...layer_embedded
Quote:
Originally Posted by Andrea deluxe
And this info is from...... ???
cheers! :toast:Quote:
Originally Posted by ajaidev
far cry2 does pretty well... if thats only a 360 and it performs as well as a 295... then thats pretty impressive! thats pretty good scaling there... they almost went for a 2.5x increase in specs from gt200 to gt300 and it looks like they actually managed to get close to a 2.5x perf boost as well
did nvidia really say that this was a 360 demo though?
because if they are sneaky, and ahum... nvidia IS, then they might try to get people to think this was only a 360 and the 380 scores even higher, but actually this already was the 380.
even if thats the case, this is good news for nvidia! very good news!
while fermi is most likely very hot, very power hungry, and very expensive, and will arrive very late... the performance seems to be there. and thats the most important thing after all...
now all nvidia needs to do is get the thing out on the market and get the price and power consumptions down. but those arent easy task, especially the latter seems close to impossible without a srhink...
and something nobody mentioned is that since tsmc cancelled 32nm and goes for 28nm, a direct optical shrink of gt300 is NOT possible! :S
that means they cant shrink gt300 as quickly and easily as they did with gt200... and even that took them several months...
and 28nm isnt even close to ready yet... so it seems nvidia will be able to capture the performance crown, but they will probably take longer to get volume up than with gt200, fermi will have a higher price than gt200 from introduction to propper availability and even then, and with the 360 having that many SPs and being as fast as a 295, it doesnt sound like there will be mainstream gt300 parts for quite a while... i wonder if nvidia plans 40nm gt300 mainstream parts at all... they might continue with their recent strategy and just go for a full blown gt300 in 28nm instead and use that as both highend and mainstream part, like they did with gt200... :/
the first SS shows identical results to a gtx285, the second shows the same perf as a 295... so the first was a gtx285 as reference and the second is fermi? gtx360?Quote:
Originally Posted by ajaidev
http://img690.imageshack.us/img690/911/farcry2e.png
:toast:Quote:
Originally Posted by Andrea deluxe
raytracing: "everything is done in raytracing" BS! only the lightning is done in raytracing... thats what really matters, but still, they shouldnt claim its all done in raytracing, im 99% sure its not... and those lightning effects are nice, but its not all that special, there are very similar demos for even dx9 cards... radeon 9700 had several raytracing lightning demos, sure, not nearly as detailled as this one, but im just saying, its nothing really new, just more details...Quote:
Originally Posted by neliz
oh and the guys says they will bundle this app with fermi so people can create their own car and post pics of it on myspace? lol what? :stick:
the hair demo looks sweet! :)
uhmmm did anybody see this?
did the system crash at the end? :lol:
http://www.youtube.com/watch?v=3-1GM...eature=related
the rocketsled demo doesnt look impressive at all to me, the only interesting part is where the house collapses when the sled passes it... that looked very very nice :D
is there an answer that would actually make a difference? ^^Quote:
Originally Posted by SKYMTL
i do wonder what they are planing, they make it sound like you can mod the cars? but i think its just colors or scenery, i dont know if you can change the shapes or anything. but it would be a fun little 3d modeling program for free. too bad it really does nothing, except for facebookersQuote:
Originally Posted by saaya
http://translate.google.de/translate...n&hl=&ie=UTF-8Quote:
Originally Posted by SKYMTL
even if you could model the cars shape... who really cares? :confused:Quote:
Originally Posted by Manicdan
if you could actually race that car you built on your own, THAT would be cool!
but building a car to then... take a screenshots of it.... thats ret4rded imo... :stick:
Sorry if its already been posted but does anyone now when the NDA expires?
This sunday.Quote:
Originally Posted by Bodkin
Great stuff cheersQuote:
Originally Posted by Terrare
17th
Welcome to the world of MySpace and Facebook :welcome: :rofl:Quote:
Originally Posted by saaya
If you were joking, that would have been the best first post, EVER. :rofl:Quote:
Originally Posted by Terrare
Welcome to XS :welcome:
is that at midnight US east coast time? or will i see stuff popping up late saterday from the European guys
Hehe, thanks. Trolling belongs to 4chan, not XS. :)Quote:
Originally Posted by Sumanji
Forgot to say :welcome: cheers for the tip, what can we expect to see after nda? Do reviews have samples?Quote:
Originally Posted by Terrare
Same, I'd kill for a review.
If those numbers belong to the GTX360, nVidia could tag the GTX380 at the price they want to.
Maybe NVIDIA will bundle a free copy of FC2 with every GF100 ;)
Thought tags were not allowed? Failure is not appropret
Sweet... excited now. Glad I didn't bite on a 5850 :D
And Batman ;)Quote:
Originally Posted by Teemax
Don't get what is so cool with this, whatever that card was it's about 15 percent faster than a 5870, and I believe it is the flagship Fermi.
Well for nVs sake I hope this is not the flagship and 15-20% FC2 advantage is not the best it can do. Even a 5000 refresh could come close or match that.Quote:
Originally Posted by annihilat0r
I wanna see some DX11 and tessellation benchmarks, maybe thats where it really shines. Or not.
But if it is the 360, then wouldn't that be cool? At the very least any price reduction from the 5800 series would also be cool (for me and for everyone).Quote:
Originally Posted by annihilat0r
Of course, it would be awesome if it was the 360, but deep down in our hearts we all know it's the 380 :D
they said after players invest a lot of time and money in unlocking and modding the cars.Quote:
Originally Posted by Manicdan
It sounds to me that it could be used as a feature in games to create raytraced images from the cars you unlocked, modded, tuned to show off with your NFS cars.
Hope it doesn't take ages to get a 380 with water block... :cool:
everyone is gonna be so confused when nvidia presents the card with the slides already leaked "guru3d ones somesay its 1000% real"
you make me feel old... :,< :lol:Quote:
Originally Posted by Sumanji
its a LITTLE old dont you think? :DQuote:
Originally Posted by Teemax
plus fc2 would show everybody how awesome in-game physics can be, all without a single line of physix code ;)
if it is, then thats bad news for nvidia...Quote:
Originally Posted by annihilat0r
so what, they didnt say if it was a 360 or a 380?
hmmm isnt that suspicious? why would they hide that its a 360?
if it is a 360, the only reason not to make that clear is to keep ati in doubts over how fast a fully fledged gt300 really is... but whats that going to be good for? ati will squeeze as much out of rv870 as they can either way...
and if its a 360, saying so would make peoples jaws drop, since it means the 380 is even faster... so if they really didnt say whether it was a 360 or 380, id guess it actually was a 380... but maybe im just being pessimisitc again :D
its just that i cant think of a reason to hide the fact that its "only" a 360...
and if it does? all dx11 does is improve performance, which i dont think is a real problem for even the current highend cards... and tesselation.. so then what, you can loop unigine for hours? :DQuote:
Originally Posted by SimBy
well they unlock and mod the cars to then RACE them... not to just look at them... :DQuote:
Originally Posted by Kuroimaho
i think that nvidia guy kinda missed the point ^^
huh use your brain guys who is gonna make presentation with there second fastest card ?
well if my fastest not ready yet and my second fastest is faster then what available now...Quote:
Originally Posted by eric66
why not?
If the fastest is not available anytime soon they will use the available hardware... and since the tesla thing is also 448 shaders this may as well be 448 and not 512 but who knows?Quote:
Originally Posted by eric66
We want to believe...Quote:
Originally Posted by eric66
:P
and how long are you going to wait for that fastest card since they don't even have one example to test on ? god only knows what will hapen by the time when it come to market
Dont really have to wait for the fastest card, i dont thing that AMD can out a single core which can compete against a GTX 380 "If this is the GTX 360"Quote:
Originally Posted by eric66
All we have to wait for is the 5890, if one still does not want to wait you can still get 5870 @ 1ghz factory OCed which is exactly what i am gonna do if the GF100 is not released in end of Jan or start of Feb.
I think clocks will be more decisive than shader count. I don't believe there's gonna be a linear increase from 448 shaders to 512. But then if yields are so bad that you can't manufacture a 512 shader product, chances are that clocks won't be so great either.Quote:
Originally Posted by ajaidev