** The Official Nvidia GeForce 'Pascal' Thread - for general gossip and discussions **

[Kaapstad]

They'll want a reason to get more money out of their customers, the only differentiator between Titan and x80ti is memory capacity. It's very likely Titan will have more memory than the x80ti and there isn't an easy way to do it other than using double capacity stacks.

To TitanXs do have more active cores as well.

I also prefer using the TXs are they don't need to be clocked so high to get the same result.

 

[drunkenmaster]

Total rubbish, his stories is based of the same info as everyone else's, he has just interpreted it differently.

He may be right, but he may just as easily be wrong.

All of the rumours and synopsis from all the various sites and forum dwellers like all of us here, are guessing, we are going by the same piece of shipping info and interpreting it differently.

Myself, I would love to see a Pascal titan shown and released (paper release ?) at GTC, is that possible, well of course it is, is it likely, who knows for sure. Not long to wait and see.

No, the information is there and he didn't interpret it just differently, he interpreted it correctly. Nvidia isn't shipping $8000 gpus, it's that simple, also the parts have names/identifying information which puts them as bracket/retainer/tec, they are testing materials. They are categorically NOT confirmation of Pascal chips being shipped. Regardless of if Pascal chips have shipped elsewhere at other times THAT manifest absolutely does not have Pascal chips being shipped and that manifest is the basis of all these other sites saying Pascal will ship in April, it's that simple.

There are parts of the article that are speculation and parts that aren't. The rumours of Pascal in April are completely made up based off a manifest that very clearly doesn't say what people thought it did, that is fact. If and when Pascal silicon has come out of the fab is speculation, though there is no reason at all for Nvidia to have shown Maxwell and lied saying it was Pascal if they had silicon back by that point(well, lets say anytime up to about a week before CES).

 

[Simon Belmont]

No, there is no chance of a April release unless it's a paper launch and not available for months.

Far too many times, well, read the semiaccurate article. One guy says something on a forum, one of the usual suspects puts it up as a 'story' and every other questionable tech news site just repeats the same 'story' as if it's fact.

There is nothing at all, no news whatsoever that points to an April release. The only sites pointing to April release are using completely incorrect information.

I'm going to save this post for if pascal releases in April I can come back here to point it out to others to see.

 

[Orangey]

Still saying April at this point is desperation.

Read the whole article, it's a masterpiece and if the tech blogs had any shame they'd publish apologies. But instead they're far more likely to be butthurt and find some way to strike back.

 

[Kaapstad]

I'm going to save this post for if pascal releases in April I can come back here to point it out to others to see.

Even if NVidia could launch Big Pascal in April they would gain nothing and could lose a lot.

Their existing cards like the 980 Ti are still selling well.

There is no serious competition from AMD at the moment.

NVidia will get better yields on Small/Medium Pascal.

NVidia only need a card that is marginally faster than the 980 Ti to get people to open their wallets and upgrade.

 

[Gregster]

I am still hopeful for April, not that I will be buying but GTC 2016 would be a good time for a release or at least some news.

 

[Orangey]

So back to Nvidia and Pascal, assuming these manifests are for Pascal related tools. What does it mean? Does it mean Nvidia unquestionably has Pascal silicon? Nope, it means Nvidia unquestionably does not have Pascal silicon and probably wouldn’t for at best several days after they got these tools in-house. In short these tools shipping pretty much prove that Nvidia was anticipating Pascal early silicon, and not an enthusiast part, as the last few days of 2015 passed. If you assume a day to ship these parts from Taiwan to Bangalore, a few days to set it up, and a few more to test and prep, and a few more just in case something goes wrong, there is no way Nvidia had Pascal silicon when Jen-Hsun claimed they did, and he had to have known it.

It also proves the ‘reports’ of Pascal taping out in June were unquestionably wrong, if it taped out in June there is no way silicon would be getting back to Nvidia in early January, best case. ~7 months is enough for a full production wafer to go through TSMC, a month of debug, and another full production run after that. If you hot lot the silicon as they most assuredly did, there would be room for 3-4 full cycles, more than any GPU in recent memory bar Fermi needed. They would have shown it off months before AMD showed running Polaris’ in early December.

So dozens of sites all echoed the June tapeout date, and since it was on so many sites it was definitely true, only it really wasn’t. Pascal likely taped out in early November and wasn’t back from the fabs when Jen-Hsun gave his CES speech, at least that is what the data shows. The only thing the tapeout story really proved is none of the dozens of tech sites has any real sources and will all copy from anyone without bothering to so much as verify a fact. It isn’t news or journalism, it is plagiarism in a socially accepted form. And it is wrong.

The same happened with the Zauba manifests, again assuming they are for Pascal. These manifests unquestionably prove that there is no Pascal silicon in hand, not that Nvidia would be insane enough to ship such valuable parts via commercial cargo anyway, they prove that it didn’t exist at the time of shipping. Once again if the dozens of sites that reported Pascal silicon arrival had any technical knowledge, any sources, or even the most basic urge to do their jobs and verify the data, they would have known it was false. Literally no site other than SemiAccurate did this, they just copied from the nearest source and posted it as fact. And they were once again all wrong.

This is the sad state of technical ‘journalism’ and ‘news’, no sources, no checking, just copying. A random post on a forum with questionable validity turns into dozens or hundreds of articles, all of which were wrong. If any of the people involved had any sort of technical knowledge, or even read the data they were writing ‘facts’ about, they would know it was wrong. Now do you see why my head hurt?

 

[pmc25]

I'm going to save this post for if pascal releases in April I can come back here to point it out to others to see.

There is absolutely zero chance of this. They may still not even have ANY (let alone big chip) working silicon for Pascal chips. They certainly didn't at or immediately prior to CES.

Being optimistic, early Q4 for small chips, and maybe just maybe first delivery of the big Tesla chips to their supercomputer partners. Being realistic, mid Q4 for small and mid range chips and automotive, Q1 '17 big Tesla NVLink chips, mid Q2 '17 Titan & x80Ti chips. Pessimistic - their Titan and x80Ti chips release about the same time as AMD debut their replacement for Polaris around E3 '17 or are canned entirely and respun versions of the designs finally launch in autumn / winter of '17.

 

[Shockwave]

I don't see any point in Nvidia releasing the card in April. I also would take anything Charlie posts with a pinch of salt. Semi-accurate is such an anti nvidia/intel site.

 

[pmc25]

I don't see any point in Nvidia releasing the card in April. I also would take anything Charlie posts with a pinch of salt. Semi-accurate is such an anti nvidia/intel site.

The point is that the point of it is moot. They absolutely cannot launch Q2.

Doesn't matter if the information is correct, which it is. If he'd wanted to stick the knife in, he could have talked about a pessimistic Fermi-esque scenario or huge delays per my pessimistic timeline. He didn't, though he has already stated that they are probably going to be at least 6 months behind AMD to market.

 

[Gregster]

CD is so anti nVidia, it just isn't a good read when he writes about them.

 

[AlamoX]

Nvidia could be just hiding well their cards up their sleeves, or could be 1st chip encountered an issue at debugging and sent back for another sample, there is a lot of new stuff for their nvidia's next architecture, changing scheduling to async, HBM, die shrink.
AMD for once caught a break, when samsung started mass production on january instead of Q2 like they were expecting, now have to see if AMD is part of the on going production or still in queue and start later.
but anyway can't wait for some news, because it's been kinda quiet.

 

[Rroff]

There is absolutely zero chance of this. They may still not even have ANY (let alone big chip) working silicon for Pascal chips. They certainly didn't at or immediately prior to CES.

Being optimistic, early Q4 for small chips, and maybe just maybe first delivery of the big Tesla chips to their supercomputer partners. Being realistic, mid Q4 for small and mid range chips and automotive, Q1 '17 big Tesla NVLink chips, mid Q2 '17 Titan & x80Ti chips. Pessimistic - their Titan and x80Ti chips release about the same time as AMD debut their replacement for Polaris around E3 '17 or are canned entirely and respun versions of the designs finally launch in autumn / winter of '17.

Pascal isn't going into (big) supercomputers. They have some middle level industry contracts for Pascal (i.e. automotive), the big stuff is going to be Volta which will be a focus for them through 2017 - consumers however won't get their hands on Volta for quite awhile.

TSMC ramped to GPU size volume production nearly a couple of months back its perfectly possible (albeit I don't think nVidia is quite there yet) for an April launch.

EDIT: PS Don't think people will be talking in context of Titan and x80ti come this release.

though there is no reason at all for Nvidia to have shown Maxwell and lied saying it was Pascal if they had silicon back by that point(well, lets say anytime up to about a week before CES).

Bare in mind though that nVidia historically has not shown off actual silicon a few times only a few weeks or months before an actual release - while there are some things that can't be tested or verified until you have actual silicon infront of you nVidia have massive hardware emulation systems these days which hugely reduce the amount of debugging they need to do with actual silicon. (Which makes me laugh sometimes when people talk about physics and the CPU being adequate for any possible physics effect as CPU equivalent of the processing power a GPU can bring to some physics calculations literally would occupy the space of a 40 foot lorry trailer).

 

[Simon Belmont]

http://www.pcgameshardware.de/Nvidi...ascal-Verzoegerung-GP100-GP104-GP106-1185107/

 

[KillBoY_UK]

http://www.pcgameshardware.de/Nvidi...ascal-Verzoegerung-GP100-GP104-GP106-1185107/

I spot reading at "The website semiaccurate.com....."

 

[Simon Belmont]

I spot reading at "The website semiaccurate.com....."

Just in case there was anything new

 

[LambChop]

CD is so anti nVidia, it just isn't a good read when he writes about them.

CD and SA is the pied piper of hamelin, all of his followers are spread about forums including this one, typing similar nonsense.

 

[Gregster]

CD and SA is the pied piper of hamelin, all of his followers are spread about forums including this one, typing similar nonsense.

He really comes across as a spoiled brat with his toys out the pram because nVidia wouldn't bow down to him - hence the hate campaign he has. What makes me chuckle is the amount of $1000 per year he charges for people to read his rants and I guess people pay it

 

[pmc25]

Nvidia could be just hiding well their cards up their sleeves, or could be 1st chip encountered an issue at debugging and sent back for another sample, there is a lot of new stuff for their nvidia's next architecture, changing scheduling to async, HBM, die shrink.
AMD for once caught a break, when samsung started mass production on january instead of Q2 like they were expecting, now have to see if AMD is part of the on going production or still in queue and start later.
but anyway can't wait for some news, because it's been kinda quiet.

That is the point a few of us have been making for a long time. They've left a huge amount for themselves to do with this transition, and historiically they've dealt very poorly with new nodes and memory controllers.

Nonsense. LPP was always scheduled for mass production mid Q4 '15, because that's when Samsung needed it for the new Exynos chips. If Q2 '16 was original schedule for mass production, there's no chance in hell AMD wouldn't have gone for TSMC. I expect mass production of the small / laptop Polaris to start this month or very early next. It needs to for Apple.

 

[Ayahuasca]

I'm really not bothered how long Pascal takes, if AMD arrive earlier with Polaris that offers an exceptional performance leap then most of us wouldn't have a problem switching, it would also help competition massively.

In fact I'd prefer it to pan out like the above, AMD simply can't afford a massive Pascal win for nVidia this time around.