• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

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

D.P. said:
Under NDA so can't say much but there are some big players who are getting plenty of press that are working with nvidia. Spoke to one engineer from an unamed company working of ADAS and he was extremely excited about drive PX2 and Nvidia deep learning software.

Auto OEMS like to have multiple providers, they rarely go exclusive. For map data one model year might have one provder, the next model year has a different, the next year yet another, heck different models within the sqme yesr couks use duffetebt suppliets. When looking at CPU solutions the OEMS are mostly looking at costs, if Qualcomm is $5 cheaper and meets minimum spec then they will get the deal. Althoigh it shpuld ne noted NV ate doing much more thsn hardware, their software solutions is part of their value add which others ignore.

I can say is last yearn nVida were no where to be seen at CES automotive, now they have a giant booth slap bang next to Ford. Their booth was packed with senior Toyota honda, chevy GM, autoliv, samsung, etc. execs. They are providing several OEMS with hardware I can't discuss.
Click to expand...

Its true NV are proving hardware to several OEMS but a lot of that hardware has a competitor chip on doing the ADAS (instead of Tegra). Audi being an example they used a NVidia solution which was really powered by a competitor chip not Tegra. NVidia dropped the competitor chip and Audi appear to have dropped NVidia going forward. From what I can see most of NVidia deals are from using the competitor chip. There own solutions without the competitor chip isn't doing very well.

NV still have a few things going forward from the old pipeline but they have been losing key deals over the past year. I know 1 competitor that won 22 of the top 25 OEMs across 252 models and is expected to have 75%-80% of the value of the global ADAS market for the next 5-7 years. Others competitor have won deals as well not leaving much market left for NVidia. Just look at the car hardware NVidia have a 250watt liquid cooled solution against the competitor’s using 3 watt that doesn’t need liquid cooling.
 
D.P. said:
Under NDA so can't say much but there are some big players who are getting plenty of press that are working with nvidia. Spoke to one engineer from an unamed company working of ADAS and he was extremely excited about drive PX2 and Nvidia deep learning software.

Auto OEMS like to have multiple providers, they rarely go exclusive. For map data one model year might have one provder, the next model year has a different, the next year yet another, heck different models within the sqme yesr couks use duffetebt suppliets. When looking at CPU solutions the OEMS are mostly looking at costs, if Qualcomm is $5 cheaper and meets minimum spec then they will get the deal. Althoigh it shpuld ne noted NV ate doing much more thsn hardware, their software solutions is part of their value add which others ignore.

I can say is last yearn nVida were no where to be seen at CES automotive, now they have a giant booth slap bang next to Ford. Their booth was packed with senior Toyota honda, chevy GM, autoliv, samsung, etc. execs. They are providing several OEMS with hardware I can't discuss.
Click to expand...

That's lovely, would make a good story for those on some car technology forum.

Lets keep this thread for Pascal news, shall we :)
 
Orangey said:
When did DP's dad start working for Nvidia?

Mine works for Nintendo so any stuff I make up can't be proved wrong either.
Click to expand...


Finally the inspiration starts to make sense. :D

thSoX3b.png
 
Dave2150 said:
That's lovely, would make a good story for those on some car technology forum.

Lets keep this thread for Pascal news, shall we :)
Click to expand...

Drive PX2 uses Pascal technology, nvidia cams their demo board has real pascal chips and is 6x faster than a TitanX for image recognition. That is plenty of GPU news.
 
I'm wondering, whats the format this year for pascal, low/mid to high end or the other way around?

As i'm looking to go into a notebook, figure its a bit naff doing it when a big jump in grunt is upcoming.
 
StriderX said:
I'm wondering, whats the format this year for pascal, low/mid to high end or the other way around?

As i'm looking to go into a notebook, figure its a bit naff doing it when a big jump in grunt is upcoming.
Click to expand...

Big Pascal taped out first and will be sent to IBM soon, what that means for gamers no one knows though. Smaller Pascal based chips exist in small quantities for PX2 shipping to partners around April, mass production is end of the year for PX2 but no one need PX2 before then anyway so it simply won't have the priority at TSMC.

If AMD is really pushing their 750ti and 950Ti competitors then Nvidia will likely respond in a timely manner.
 
NVIDIA GP104 and first Polaris GPUs supposedly spotted on Zauba

NVIDIA GP104, smaller die, more pins

Let’s start with GP104, the GM204 successor, based on Pascal architecture. We are probably looking at first high-end GPU from upcoming 1000 series (the name has not been confirmed). Same as with Kepler and Maxwell, we don’t expect the ‘Big Pascal’ to be released anytime soon. Instead much smaller GPU, that should theoretically be faster than both GTX 980 Ti and TITAN X is expected in the coming months.
Click to expand...

http://videocardz.com/58101/nvidia-gp104-and-first-polaris-gpu-supposedly-spotted-on-zauba
 
It wouldn't surprise me to see Pascal with GDDR5X instead of HBM in the first releases for mainstream discreet GPUs and I was kind of expecting it.
 
Gregster said:
It wouldn't surprise me to see Pascal with GDDR5X instead of HBM in the first releases for mainstream discreet GPUs and I was kind of expecting it.
Click to expand...

Judging by the very lack lustre performance of HBM1 @1080p compared to GDDR5 I would rather have GDDR5X instead of HBM2.

HBM is ok at high resolutions but very poor at low resolutions in the Fiji cards.

IIRC 8 Pack overclocked a Fury (P) on LN2 with a massive overclock on the HBM which yielded good gains in performance. This only goes to prove that under normal conditions the clockspeed of HBM is way too slow.
 
You must log in or register to reply here.
Back
Top Bottom