Even if Nvidia doesn't deliver, AMD's already done that, just need some more 5xxx cards sold and this will be enough to stimulate creating some decently performing clients. :up:Quote:
Originally Posted by ajaidev
Printable View
Even if Nvidia doesn't deliver, AMD's already done that, just need some more 5xxx cards sold and this will be enough to stimulate creating some decently performing clients. :up:Quote:
Originally Posted by ajaidev
Jen-Hsun Huang Trying to figure out why my engineers couldn't fit more TMU's into our GF100 products.
January 14 at 4:26am
humm any thing you guys can predict from this?
That I say (and others laugh at) that the original GF100 design has 128 TMUs. only 64 are active in the final product.Quote:
Originally Posted by Mk
I heard 80 TMU's...
My news letter says that an exciting part is coming in Q3 2010 too :)
And in the land of Quadro it has 6GB of VRAM and even more exciting still...............nearly double the power of the 1.5GB Quadro Part.
Hmm I wonder what this is..
....
Yes that is right, GTX 395* (Name TBC) coming Q3 2010. To add some more detail here, the 1.5GB Quadro part has 448 pipes. (yet the 3GB part has more than double the FX5800 Ultra (assume 512))
Double this.. add some more VRAM and volla.
Voodoo5 6000 external PSU time folks...
John
That's one way to avoid PCI-SIG ;)Quote:
Originally Posted by JohnZS
Q3. .imagine that they planned the product for Q1.
A PCI-SIG employee confirmed to me that neither 5970 nor GTX 295 appear in any of the compliance lists.Quote:
Originally Posted by neliz
And I am the only one who thinks that that facebook account is phoney ?
I think you're the only one who thinks someone doesn't think it's phony.Quote:
Originally Posted by BenchZowner
Jen-Hsun Huang I cannot tell you too much yet, but expect the definite in NVIDIA Graphics in Q3 of this year! Stay tuned!
5 minutes ago "facebook"
Jen-Hsun Huang's facebook is...
http://jenden.us/storage/JD/img/for_teh_lulz.jpeg
This GPU-business is getting more and more like a dirty moneky-business. Not only on this forum, but everywhere else too.
Too many kids are focusing on funny stuff, and few seams care about the technical stuff abut a GPU any longer. Is it possible that some of these are making money based on this kind of monkey business?
http://www.fudzilla.com/content/view/17357/1/Quote:
Retail board to launch in March
Our sources are whispering that Nvidia has promised Fermi chips and Fermi boards in February. The final specification should be closed one of these days if it wasn't closed already, and the partners can start planning the final cards.
The first batch is naturally controlled by Nvidia and partners will only get to buy cards, but at later date someone might even dare to make its own design. This is naturally very unlikely due to big cost and the fact that Nvidia gets too scared that something might go wrong.
The mainstream Femi will end up in partners’ hands and they will manufacture these chips on their own, but this is still at least one quarter after the original GF100 Geforce single chip high-end card.
Nvidia partners barely have anything interesting to sell, and marketing wise they have nothing to advertise due to obvious inferiority compared to ATI’s offer. Hopefully this might change with the introduction of Fermi, but again, we expect shortages of these cards as they should be very desirable.
Sounds like an earlier-in-March launch *shrug*? Beats me :shrug: it's just FUD after all.Quote:
Originally Posted by Mk
^^March retail launch and AIB getting cards in Feb is what I heard as well.
What remains to be seen is how many cards will be out there on the shelves even after retail launch
tags :up: if "chernobyl" was among the first set of tags.. heres
"3 billion transisland accident" (3 mile island accident)
"toxic waste"
"radioactive waste"
"hazardous waste"
"nuclear disaster"
"armageddon"
"pr nightmare"
"fermi paradox" <- :rofl:
pinch in :toast:
http://i1005.photobucket.com/albums/...emp/gtx380.jpg
:banana::banana::banana::banana:
Lol, read my post three pages up or something.. *sigh* some people and reading... :rolleyes:Quote:
Originally Posted by BenchZowner
Quote:
Originally Posted by neliz
http://s003.radikal.ru/i203/1001/13/c72134d71598.gifQuote:
Originally Posted by junkun13
Somehow that pic makes me think of an end of a catacomb that spits out toxic waste, just would need to add a pool of green slime.
*WARNING TROLL POST*
What I am curious about is if those who are criticising the Fermi about having cut DP performance are the same people who said that it does not matter that the Radeon HD 5700 series cards have no DP support?!?
*END TROLL POST*
I am a bit disappointed nVidia have cut DP performance on the consumer cards, but from my understanding this is so they can have higher clocks and more pipes for consumer cards. FYI the Tesla computational cards only have 448 pipes. Quadro has "the best of both worlds", but it will be clocked lower than the consumer part. (yet have more VRAM options)
John
i must have missed that...Quote:
Originally Posted by LordEC911
idk... they cant make money off of a 250$ gpu if they sell the final card for 500$? i find that hard to believe... they are selling 260s for below 200$... and the 260 gpus probably still cost them more than 50$ a piece at this point.Quote:
Originally Posted by LordEC911
i think they can make money selling fermis even with pretty bad yields, its just that they will have a worse margin then selling 285 cards, much worse... almost 0 worse... :D
i dont think bad yields will keep nvidia from launching fermi, i dont think theyd delay it because of bad yields... if yields would be really bad so they cant make much money on the parts theyd do the same they do now with the 285 cards... they drip feed the market and hope for better days :D
so the fact that nvidia hasnt launched yet is not because yields are too bad, i think they simply arent ready yet... they probably couldnt sell a fermi card sooner than what they plan now even if they could sell it for 5000$ a piece...
:lol:Quote:
Originally Posted by Mk
so they removed some physically or just disabled them? any idea why would they disable so many tmus? heat? no notable perf gain? lots of defects in tmus?Quote:
Originally Posted by neliz
oh cool, nice detective work, thx for sharing :toast:Quote:
Originally Posted by annihilat0r
i remember the 9600XT didnt even pass the pci-sig signaling test, right? and it still worked perfectly fine and oems picked it up, even before ati solved the problem... so i dont think the 300W limitation by pci-sig is a big deal... but cooling 300W alone is quite a challange already, so either way you dont really want to go beyond 300W... :D
i dont think so... wouldnt be surprised if that account gets offers of homoerotic services in exchange for hardware :lol:Quote:
Originally Posted by annihilat0r
you mean like people trying to discredit each other at all costs? yeah... pretty annoying... :DQuote:
Originally Posted by Sam_oslo
but tbh idk what trend your talking about, the press, enthusiasts and even normal end users have never been more informed about the actual tech underlying pc hardware as much as today... just 10 years ago most people on enthusiast forums had no clue how chips are manufactured, designed, set up, what gpu units a chip contains and what they do etc...
back then people were arguing on forums based on absolutely nothing but some generic marketing phrases by the big manufacturers, like XX calling their new chip awesome, so it must definately be better than YY companys upcoming card which they instead call a "notable speed improvement" :D
5700 doesnt do dp at all? didnt even know that :p:Quote:
Originally Posted by JohnZS
and havent seen anybody complain...
in the long term cutting down dp might come back and bite nvidia in their 4ss in the future when they try to push cuda some more and devs will ask them how committed they really are... but seeing how fast dp perf evolves and how slow the infra structure evolves... i dont think it will matter...
and is only dp cut down or also sp? cause the fpus are usually the hottest and most power hungry parts of a chip, so maybe they had to cut them down to get power under control?
/worship =p jk lol... nice find!
Quote:
Originally Posted by junkun13
DP is pretty useless for a graphics chip and would just add heat and and die area. since ATi has multiple alu's in each shader they dont have to add much to support DP its just stuck at 5:1 ratio where nv has 2:1. gpgpu is a different story but i dont see how that matters if nobody supports double precision in opencl anyways.:rolleyes:
Don't expect any sorta of availability after launch.Quote:
Originally Posted by Aerwidh
Few thousand cards is what I would expect as a best case.
that's too low, even for a "paper" launchQuote:
Originally Posted by LordEC911
this is gotta be a joke right :confused: i mean its the exact duplicate ATI heat-sinkQuote:
Originally Posted by junkun13
its entirely easy to tell thats photo shopped. especially since ive seen that exact same photo at the exact same angle for an ati card.Quote:
Originally Posted by Mk
You're joking right?Quote:
Originally Posted by Mk
Hehe, you are wrong like always. You meant poor launch availability of Radeons HD5870 and no buyable HD5970 till today, dont ya? For launch, will be enough card for enthusiast!Quote:
Originally Posted by LordEC911
Was thinking the same thing.Quote:
Originally Posted by WeeMaan
Come on dude. It's a DUH moment or you
So you are trying to tell me there will be more than 10k cards at launch?Quote:
Originally Posted by Zed_X
5870 and 5850 had much more than that...
Wrong...
Hmmmmm....Quote:
Originally Posted by Zed_X
Cypress had 20 to 40K available at launch. it took AMD a rough 3 months to produce such an amount. Availability for GF100 will be a lot later. For GF100, think 7800GTX 512 like amounts.Quote:
Originally Posted by Zed_X
I had my 7800GTX 512 on launch date, was a good card, although hideously overpriced *cringe. But, if you have the money, there is no problem really.Quote:
Originally Posted by neliz
Likewise for Cypress.Quote:
Originally Posted by Tim
Duuuh:p:
it was 3:00am and i saw this didn't even zoom just jumped to conclusion :shakes::shrug:
Correct! I had my card on Launch and I didn't even bother to order it in the morning and waited till after work.Quote:
Originally Posted by neliz
smells like green spirit :ROTF:Quote:
Originally Posted by Zed_X
no news in the fermi thread? REPENT!!! THE END IS NEAR!!!! :sofa:
Boring boring boring boring. :(
Is it here yet? Is it here yet? Are we there yet? :p:
I recommend we close this thread, and when someone's got news, he should open a new thread.
^this
+1!Quote:
Originally Posted by annihilat0r
What? There's a thread here? I only came for the tags :rofl::ROTF:
No, It was 300k Cypresses in under 3 months and 500k Junipers in about 2 months. sourceQuote:
Originally Posted by neliz
QFT the tags make me laugh every day!Quote:
Originally Posted by Yvese
They didn't have those numbers for launch though...Quote:
Originally Posted by Eastcoasthandle
He is talking about how long it took them to build the launch numbers.
I think you mistunderstand... when they launch a product it doesnt come out of thin air, right? It takes time to manufacture... i believe he was referring to the time prior to launch that it took to manufacture launch quantities.Quote:
Originally Posted by Eastcoasthandle
Prior to launch then and now is not comparable do to the state that TMSC was in. Once TMSC got it together (more or less) did yield improve (which explains my post). Besides, where is the source offering cards (in those quanities mentioned in his post) ready for sale 3 months prior to the Cypress/Juniper release? I like to read that.
in the beginning of June at COMPUTEX 09 AMD and TSMC demonstrated working Evergreen card and GPU (according to some opinions it was Juniper, and to others it was Cypress). So there you have three months: June-SeptemberQuote:
Originally Posted by Eastcoasthandle
I'm talking about the figures he posted.Quote:
Originally Posted by Nedjo
well if AMD said that they've shipped 800K of chips 'tll the Christmas, I guess there's now way to do it in 25. September - 25 December timeframe!? So they needed to manufacture significant quantities months before the launch. Right?Quote:
Originally Posted by Eastcoasthandle
3 Million DX_11 GPU on 2009 says AMD on the financial conference.Quote:
Originally Posted by Eastcoasthandle
Until March they can reach 10 million or more with the Redwood, Cedar on desktop and the top-bottom mobile GPU's. From January until March is a entire Quarter.
Which would mean that it would have to be more then 20k-40k right? In order to correlate any comparison between Fermi's available to Cypress/Juniper availability we need to know if TMSC was in the same manufacturing capacity. From what I recall TMSC was not. So, I don't see how one can make such a comparison.Quote:
Originally Posted by Nedjo
any link for transcript from the conference?Quote:
Originally Posted by v_rr
For the record playing with tags is against the rules here and we can/do track who adds on and can/do give infractions for messing with the system.
At the end of the day good tags = traffic for XS, so let's leave them be, OK guys?
Roger that.Quote:
Originally Posted by Serra
The old Fermi threads were all started around August/Sept.. so here's to idle specualtion.. and lots and lots of waiting.
Aug09 - when Fermi comes out in the fall its going to DESTROY everything
Sept09 - Ha Ha, Cypress is just 2x 4870. Fermi for the WIN.
Oct09 - Ok ok, a few weeks late.. its TMSC fault, duh.. PS: That's me in the facebook page photo
Nov09 - See!?... see!? Its a real card. They are just putting finishing touches.
Dec09 - DX11 doesn't really matter.
Jan10 - DX11 DOES matter.
Feb10 - ....?
so the guess is late feb to mid march for a fermi launch? i want to buy nvidia but the old ati stuff outruns it for cheaper and the new ati stuff kills it for the same price... i need to replace my 9600GT
god job at pointing out nvidias 180 spin on dx11 deimos :D
they did it in a smart way, they kept saying dx11 doesnt matter, and then started pushing tesselation and how awesome and important tesselation was... well guess what, tesselation is the main feature dx11 is buit around :D
"DX11 doesn't matter" Actually stems from Computex '09.
And NO I don't have any factual numbers on the pre-release amounts besides what I hear from AMD. When I heard 20 to 40.000..
I said.. wafers, right?
no ... cards!
It is rather amusing isn't it, I mentioned that earlier in the thread how ironic that DirectX 11 is not quite so insignificant now ;)Quote:
Originally Posted by saaya
The big n do get rather annoying at times with their PR and Spin, but I can see what they were saying. At the time they made that statement there were no DirectX 11 games available.
John
IT doesn't affect you or me as an end-user. But that statement wasn't meant for you or me. It was directed at System Builders and Developers.Quote:
Originally Posted by JohnZS
Edit: Nedjo, Computex demo's were displayed on Evergreen Broadway : Juniper.
I think that was number of units "sold" as in ordered. Not necessarily produced yet. And pretty sure that report included the Mobility 5000 series products as well, a majority of which have not been delivered yet, and only recently publicly announced. But I'm sure they secured orders for them from the big companies months ago.Quote:
Originally Posted by v_rr
At least, the last report about millions of ATI DX11 GPUs was posted here, someone incorrectly assumed it meant units shipped. But by now, 3 million shipped seems possible. :shrug:
hope too. I have sold in few hours my three GTX 285 and now I have to wait Fermi :yepp:Quote:
Originally Posted by RoadconeTuning
I'm sure Fermi will allow you to play those games you couldn't with your 3 GTX 285's.Quote:
Originally Posted by gifaro
they said 2 million units shipped, i think that makes it pretty clear. Dunno why everyone doubts ati/amds announcement, they delivered on everything they said, all releases came as announced, no delays etc., we have no reasons to doubt itQuote:
Originally Posted by kgk
SLi doesnt scale as well as you'd think especially in mmo type games that are seriously CPU bound; adding to that overhead with SLI does more harm than good.Quote:
Originally Posted by annihilat0r
Im waiting for them as well and infact just picked up a 240GTS for my Physx card; I have full confidence that nVidia can pull this off; abeit very late but meh.
Thats how nv does it and thats why many people don't like it. But I'll be getting like 2 of those fermies when they become avaible anyway... Most likely. Got used to nv iq for the past years too much.Quote:
Originally Posted by saaya
By the way,"it looks like" nv works on the 2nd gen of the fermi cards already:D:
http://www.fudzilla.com/content/view/17387/1/
Fermi 2 is gonna rock since the basic frameworks seems very strong but first they need to send fermi 1 out of the gate
well i dont think the problem with fermi1 was the design... at all... the problem with fermi1 was/is tsmc...Quote:
Originally Posted by ajaidev
unless nvidia realizes that leaning that much on tsmcs shoulders isnt such a great idea and change their overall strategy, i dont think fermi2 will be any more successful than fermi1... :shrug:
Indeed.Quote:
Originally Posted by ajaidev
so now we are waiting for fermi 2
As Jen-Hsun Huang said himself "If you don't have products, you talk about the future"Quote:
Originally Posted by eric66
From the department of Evangelism.
fermi2 tag added :D
idk about you but i cant wait for fermi 2.5...
to hell fermi, let's wait for fermi2 !! :D:rofl:
It's particulary funny to read ya' about fan's brainwashing ... You know really well the problem, isn't it ? :rofl:Quote:
Originally Posted by flippin_waffles
well there are always people ready to wait indefinitely! Our own fellow member Glow9 is awaiting for new platform ever since AMD started platform approach with Spider! He's evaluating Leo, but he'll probably await another year for next platform :)Quote:
Originally Posted by eric66
Nvidia's in trouuuuuuuuuuuuuuuuuuuble! :p:
At least they still have a healthy group of believers to try and spread the message for the time being. Wonder how long that will even last?
ATi stated talking bout 28nm lately, as in this news AMD shows 28nm wafer GlobalFoundries to meet new card? and again a few days later the CEO was out and talking about ATi HD6000 series - "Second Half of 2010 – AMD’s CEO." .
I'm not sure what is going on, but everybody is talking about the end av the year while we are still in Jan. :confused: Is ATi trying to justify jumping over a refresh (or new GPU) to counter the GF100?, and is it possible at nVidia brought it just because ATi was taking bout it?
Anyways what comes in 6 months is not interesting today, either it should come from ATi or nVidia.
So why all the talk about fermi in the last months and the "smart thing to do is wait for fermi"?Quote:
Originally Posted by Sam_oslo
Let's all talk about how great R970 is gonna be and how smart it is to wait for it. :rolleyes:
Back in those days many ppl were expecting a november/december '09 release...Quote:
Originally Posted by LesGrossman
You're being sarcastic but actually, that's not a bad idea.Quote:
Originally Posted by LesGrossman
RV970 discussion topic ahoy!
Because the upcoming GPU is always very interesting. After all Gf100 it is only weeks away, if it doesn't get delayed (in good nVidia tradisjon) again., of course :p:.Quote:
Originally Posted by LesGrossman
The GF100 and a refresh (or new GPU) from ATi to counter it, is the only interesting news right now. Talking about Fermi2, a 28nm ATi or HD6000 is really strange right now, don't you think so?
It's all related, Fermi2 is the new NV35/RV670 and Northern Islands is what was originally planned against Fermi. Saying those things don't matter is showing a complete disregard for History and the future of the current topic at hand.
Why don't we speculate on some "new" AF modes, or a reduction in cost to utilize AF?
Because it WAS units shipped.Quote:
Originally Posted by kgk
You cannot do your numbers like that anymore, thanks to Enron.
I guess all will make sense come March with Fermi launched. Have to say though that simply with Nvidia talking already about fermi 2, G100 would most probably not be as quick as some would hope.
But hey end of the day there should be two great products; 5870/5970 and Fermi. Each will no doubt have their advantages. I am only hoping for a price cut..pretty please!!!
For the love of *** LordEC911 was pretty close with his cost guestimate for GF100 (in reality it's a bit cheaper) We were at an all-time low for graphics card prices. It only goes UP from here.Quote:
Originally Posted by takamishanoku
Is it possible for TSMC to produce 2m 40nm evergreen chips on their phase 4 plant?
They had yield problems and a few other 40nm customers.
What's TSMC's 40nm daily production? I though it was very limited.
Fermi II is probably to be what happened on G80 > G92 , I don't think nvidia is going to release a totally different arch compared to original fermi, since they made fermi arch very flexible... now the potential HD5000 series users that were planning to jump the bandwagon to fermi probably are going to wait for Fermi II :ROTF:
I thought capacity now is around 40k wafer starts per month must've been from a digitimes article.Quote:
Originally Posted by Marios
Nah, I'm jumping ship to the Fermi unless it sucks badly, regardless... I very much dislike my 5870 due to the drivers (from lots of glitches to annoying control panel to fan throttling not working right at defaults in the driver to annoying dual-link DVI artifacts (confirmed with two different 5870's and two monitors one single-link lower rez and one 2560x1600 dual-link one)).Quote:
Originally Posted by Blacky
I love how fermi 2 is mentioned from non other than fudzilla and already people are going into the "lets make fun of nvidia" band wagon.
if this is true, it still gives us nothing more to go on. Shoot this could be another G80/G92 1, 2 knock out punch planned. Nvidia knows refreshes are on the way from ATI. or it could mean major fail. Fermi still stands in the neutral territory with this info. Lots of people here like to see companies fall my god, and for ridiculous reasons.
I'm thinking the thesis would crash and burn in a flaming ball of F-minus. None of what you said makes sense. People dangerously brainwashed? Manipulation of kids? Huh? :wth:Quote:
Originally Posted by flippin_waffles
could you find me one person who would do that? nv doesnt have to do anything at all for them to buy their products. if you are hinting that fermi is going to be a flop you should note that its missing several elements. its just really late. i think your post history could make an interesting case study too.
Well, AMD/ATI is smart. If indeed Fermi2 will be out in Q4 2010. I believe ATI will skip HD6000 series and jump straight to HD7000 or even HD8000 which will be waaayy.. faster and much much more more powerful.Quote:
Originally Posted by ajaidev
eat that NVIDIA!
:D
Nvidia just needs to come to the market ASAP
i dont care which company is better or worse,
u can only have a game if there are 2 teams on the field
i dont know if this was a joke or not, but uh 3 series of GPUs in 1 years, not likely. Lol and skipping a series, thats technically impossible, you cant skip something that never existed. Of course, im guessing this was meant to be sarcasticQuote:
Originally Posted by migunani