nvidia secret event, does nvidia release a new GPU every three months?

evilsofa said:
"there are two possible things that could be happening: 1) Nvidia has finally wizened up and started scrambling the code names of the GPUs to hide their progress from our eyes or 2) the nomenclature is correct and this is something we have never seen before."

I'll go with #1 and a previous rumor I've seen that they're starting Pascal with the biggest chips and working their way down, meaning that could be a flagship Pascal GPU for the Tesla or Quadro lines.
Click to expand...
There is zero chance of this being a Pascal card.
 
misterbobby said:
There is zero chance of this being a Pascal card.
Click to expand...

Maxwell died with the 980ti and the titan cards along with all the GM20x chip code numbers.

So per the quote from the article it basically states we will have a new generation of video cards FINALLY! in 4-5 months. Less heat , less noise, less electricity and more power. They are testing a new chip that costs alot.
anyway Cant wait to dump my 980s and move on from maxwell
The 980ti is basically only good for 4k monitors and not so much for2560x1440 gaming. Owned 980 ti in SLI and they got returned. Id say its great product for anyone upgrading from a 970 and below. Just my personal opinion and many others have expressed the same on this forum that the 980 ti did not meet expectations coming from 980s in SLI.


It is worth noting here that chips spotted at Zauba usually become public in a matter of few months (usually 4-5). Since this coincides with the release date of Pascal – it is a possibility that we are looking at a flagship chip of the 16nm FinFET generation.

Read more: http://wccftech.com/nvidia-graphics-chip-jm-601-gpu-spotted/#ixzz3pdjXKaYC
 
UnrealCpu said:
Maxwell died with the 980ti and the titan cards along with all the GM20x chip code numbers.

So per the quote from the article it basically states we will have a new generation of video cards FINALLY! in 4-5 months. Less heat , less noise, less electricity and more power. They are testing a new chip that costs alot.
anyway Cant wait to dump my 980s and move on from maxwell
The 980ti is basically only good for 4k monitors and not so much for2560x1440 gaming. Owned 980 ti in SLI and they got returned. Id say its great product for anyone upgrading from a 970 and below. Just my personal opinion and many others have expressed the same on this forum that the 980 ti did not meet expectations coming from 980s in SLI.


It is worth noting here that chips spotted at Zauba usually become public in a matter of few months (usually 4-5). Since this coincides with the release date of Pascal – it is a possibility that we are looking at a flagship chip of the 16nm FinFET generation.

Read more: http://wccftech.com/nvidia-graphics-chip-jm-601-gpu-spotted/#ixzz3pdjXKaYC
Click to expand...
Dear lord please not this yet again. STOP equating SLI 980 Ti with a single 980 Ti. You have done that over and over for months now claiming the 980 Ti is only a 4k card.
 
misterbobby said:
WTH are you talking about? The article is about a graphics chip and you had just said it could be the flagship Pascal gpu for Tesla or Quadro.
Click to expand...

A GPU (Graphical Processing Unit) is a graphics chip mounted on a video card.
 
evilsofa said:
A GPU (Graphical Processing Unit) is a graphics chip mounted on a video card.
Click to expand...
What kind of idiotic comment is that? What in the hell do you think the "gpu" will go in? It will go on a pcb and that will be a card will it not? Geforce, Tesla and Quadro are graphics cards last time I looked. :rolleyes:
 
misterbobby said:
What kind of idiotic comment is that? What in the hell do you think the "gpu" will go in? It will go on a pcb and that will be a card will it not? Geforce, Tesla and Quadro are graphics cards last time I looked. :rolleyes:
Click to expand...

It is very strange that you know that a GPU is not a video card but that you cannot understand what I said about the article. Perhaps you didn't read the article?

By the way, Tesla is not a graphics card.
 
It's a lot more plausible that this is a Quadro or Tesla GPU. Although I hope it to be a Dual GPU card!
 
misterbobby said:
What kind of idiotic comment is that? What in the hell do you think the "gpu" will go in? It will go on a pcb and that will be a card will it not? Geforce, Tesla and Quadro are graphics cards last time I looked. :rolleyes:
Click to expand...

What gives you the right to start every sentence in a aggressive manner? You sound like a total douche that is just trolling this forum . Could be because your GTX 980ti is about.to.be obsolete only after 3-4 months? Sounds like you have a chip on your shoulder over the fact this is going.to be one of the best cards we ha e seen in years. Yes the 980 ti to is hot, sucks a 115 volt outlet dry and costs a ridiculous amount already . We believe its Pascal and I for one can't wait to move on from maxwell. Look forward to 4k gaming at 60fps plus in a nice form factor like the amd fury
 
Last edited:
Henri108 said:
It's a lot more plausible that this is a Quadro or Tesla GPU. Although I hope it to be a Dual GPU card!
Click to expand...

Really a dual pascal card already? Rofl , Because pascal artictecture is going to be released in 2016. Or Maybe it is a quadro or tesla pascal..

Nvidia is moving on from the 980 ti .. Its a hot load pos
 
UnrealCpu said:
What gives you the right to start every sentence in a aggressive manner? You sound like a total douche that is just trolling this forum . Could be because your GTX 980ti is about.to.be obsolete only after 3-4 months? Sounds like you have a chip on your shoulder over the fact this is going.to be one of the best cards we ha e seen in years. Yes the 980 ti to is hot, sucks a 115 volt outlet dry and costs a ridiculous amount already . We believe its Pascal and I for one can't wait to move on from maxwell. Look forward to 4k gaming at 60fps plus in a nice form factor like the amd fury
Click to expand...
Maybe you should brush up on your reading comprehension and pay attention to the context. And why in the hell would I be upset if a new GPU came out? I'm looking forward to Pascal as much as any person here.
 
misterbobby said:
Maybe you should brush up on your reading comprehension and pay attention to the context. And why in the hell would I be upset if a new GPU came out? I'm looking forward to Pascal as much as any person here.
Click to expand...


whatever

You like to argue with people on here with your insults and 2 liner sentences with no meaning whatso ever
you sound like a forum troll
 
sigh.... thought there might be something interesting here... nope... just more bullshit arguing.
 
RAutrey said:
sigh.... thought there might be something interesting here... nope... just more bullshit arguing.
Click to expand...

Welcome to the video card forum, the discussion where everything's made up and the posts don't matter.
 
First Pascal series of cards will use GDDR5X memory which is 2 times faster than GDDR5 that the current generation of cards use

The GDDR5X another positive change will be to reduce and the voltage levels required by the chip, and thus their consumption and heat production. This will not only affect the consumption of new graphics cards are beneficial, but it also needs less cooling, which makes the new units will operate more quietly previous-generation counterparts.


http://pcforum.hu/hirek/17589/ketszer-gyorsabb-mukodest-iger-az-uj-gddr5x-memoria


It would seem that only business class servers will get the HBM cards first and then maybe later down the road we will see consumer level HBM cards. They expect to have new cards with GDDR5X by the second half of 2016 , so maybe April 2016.

Combine this with the new Pascal GPU architecture it will put maxwell to shame including the GTX 980ti, titan cards

AMD will have HBM on Arctic Islands based GPUs around the same time Increasing power while decreasing Power consumption. Should be a match until Nvidia gets its hands on HBM or HBM 2. I still think alot of work needs to be done with motherboard manufacturers to somehow incorporate NVlink to obtain a faster PCI-Express interconnect bus. since PCI-E 3.0 is not adequate at 16GB/sec as current cards push 256gb/s or more currently. Plus these GDDR5X cards will be cheaper to produce so another hint we wont see HBM on the first pascal cards.
'Just my opinion
 
Last edited:
UnrealCpu said:
First Pascal series of cards will use GDDR5X memory which is 2 times faster than GDDR5 that the current generation of cards use

The GDDR5X another positive change will be to reduce and the voltage levels required by the chip, and thus their consumption and heat production. This will not only affect the consumption of new graphics cards are beneficial, but it also needs less cooling, which makes the new units will operate more quietly previous-generation counterparts.


http://pcforum.hu/hirek/17589/ketszer-gyorsabb-mukodest-iger-az-uj-gddr5x-memoria


It would seem that only business class servers will get the HBM cards first and then maybe later down the road we will see consumer level HBM cards. They expect to have new cards with GDDR5X by the second half of 2016 , so maybe April 2016.

Combine this with the new Pascal GPU architecture it will put maxwell to shame including the GTX 980ti, titan cards

AMD will have HBM on Arctic Islands based GPUs around the same time Increasing power while decreasing Power consumption. Should be a match until Nvidia gets its hands on HBM or HBM 2. I still think alot of work needs to be done with motherboard manufacturers to somehow incorporate NVlink to obtain a faster PCI-Express interconnect bus. since PCI-E 3.0 is not adequate at 16GB/sec as current cards push 256gb/s or more currently. Plus these GDDR5X cards will be cheaper to produce so another hint we wont see HBM on the first pascal cards.
'Just my opinion
Click to expand...

Yep,

A Hungarian site probably knows more than we do about a GPU designed by an American company and manufactured in Taiwan :p

Just like during the bulldozer launch, that Turkish website had all the "insider" news :p
 
evilsofa said:
"there are two possible things that could be happening: 1) Nvidia has finally wizened up and started scrambling the code names of the GPUs to hide their progress from our eyes or 2) the nomenclature is correct and this is something we have never seen before."

I'll go with #1 and a previous rumor I've seen that they're starting Pascal with the biggest chips and working their way down, meaning that could be a flagship Pascal GPU for the Tesla or Quadro lines.
Click to expand...

I Agree HBM on telsa/quadro lines in the next 4-6 months while GDDR5x on pascal for consumers
 
Zarathustra[H];1041936150 said:
Yep,

A Hungarian site probably knows more than we do about a GPU designed by an American company and manufactured in Taiwan :p

Just like during the bulldozer launch, that Turkish website had all the "insider" news :p
Click to expand...
All the chips and boards are designed overseas so yes they know.more. it only makes sense
http://techarx.com/amd-rumoured-to-be-using-gddr5x-on-its-next-gen-gpus/

And then there was the old rumor claiming AMD wouldn't be able to source HBM2 at all.
http://www.tweaktown.com/news/47773/amd-trouble-sourcing-hbm2-next-gen-video-cards/index.html
It will come before nvidiA and I bet before summer 2016
Hopefully GDDR5X is cheaper than HBM.[/QUOTE]

It most certainly will be
 
UnrealCpu said:
All the chips and boards are designed overseas so yes they know.more. it only makes sense
http://techarx.com/amd-rumoured-to-be-using-gddr5x-on-its-next-gen-gpus/

And then there was the old rumor claiming AMD wouldn't be able to source HBM2 at all.
http://www.tweaktown.com/news/47773/amd-trouble-sourcing-hbm2-next-gen-video-cards/index.html
It will come before nvidiA and I bet before summer 2016
Hopefully GDDR5X is cheaper than HBM.
Click to expand...

there's no way AMD will beat NVIDIA to market in 2016, arctic islands is only just rumored to tape out now while pascal taped out 6 months ago.

AMD is fucked in 2016, they'll be lucky to be out in a year, let alone next summer.
 
cocdod said:
there's no way AMD will beat NVIDIA to market in 2016, arctic islands is only just rumored to tape out now while pascal taped out 6 months ago.

AMD is fucked in 2016, they'll be lucky to be out in a year, let alone next summer.
Click to expand...

They have first right to HBM so 2016 AMD will be fine, also Samsung is working with AMD
Also you read the news lately?

Just dont come in here without quoting something to backup your claim.

I can say AMD will be doing just fine simply saying

GlobalFoundries and AMD Announce First 14nm FinFET Sample Production Success
by Ryan Smith on November 5, 2015 9:00 PM EST

now i will post a link

http://www.anandtech.com/show/9764/...e-first-14nm-finfet-sample-production-success

Overall the message from the two companies is clear that they are making progress on bringing up 14nm manufacturing at GlobalFoundries ahead of mass production in 2016.

GlobalFoundries, AMD’s former chip manufacturing arm, is a fab that has seen some hard times. After being spun-off from AMD in 2009, the company has encountered repeated trouble releasing new manufacturing nodes in a timely process, culminating in the company canceling their internally developed 14XM FinFET process. Charting a new course, the in 2014 the company opted to license Samsung’s 14nm FinFET process, and in some much-needed good news for the company, today they and AMD are announcing that they have successfully fabbed AMD’s first 14nm sample chip.
 
Liger88 said:
I'm sold that this secret meeting never happened and this thread was along with the story was a complete fabricated hoax at this point.
Click to expand...
It was almost certainly just the new 'mobile' GTX 980.
All the other speculation was fluff.
 
UnrealCpu said:
They have first right to HBM so 2016 AMD will be fine, also Samsung is working with AMD
Also you read the news lately?

Just dont come in here without quoting something to backup your claim.

I can say AMD will be doing just fine simply saying

GlobalFoundries and AMD Announce First 14nm FinFET Sample Production Success
by Ryan Smith on November 5, 2015 9:00 PM EST

now i will post a link

http://www.anandtech.com/show/9764/...e-first-14nm-finfet-sample-production-success

Overall the message from the two companies is clear that they are making progress on bringing up 14nm manufacturing at GlobalFoundries ahead of mass production in 2016.

GlobalFoundries, AMD’s former chip manufacturing arm, is a fab that has seen some hard times. After being spun-off from AMD in 2009, the company has encountered repeated trouble releasing new manufacturing nodes in a timely process, culminating in the company canceling their internally developed 14XM FinFET process. Charting a new course, the in 2014 the company opted to license Samsung’s 14nm FinFET process, and in some much-needed good news for the company, today they and AMD are announcing that they have successfully fabbed AMD’s first 14nm sample chip.
Click to expand...

Having priority access to HBM is meaningless in the context (AMD doesn't have priority access btw, nvidia has complete shut them out of hbm2 supply and will be the only company pushing our hbm2 gpus in 2016). The problem here is that AMD's chip design has finished months after nvidia's has, they will need a year to iron out all the issues at the least. This is demonstrably proven by years and years of evidence.
 
You must log in or register to reply here.
Back
Top