Nvidia Newest Samsung Foundry Customer For 14nm Process, After Apple And Qualcomm

  • Thread starter Thread starter Guest
  • Start date Start date
Toggle sidebar Toggle sidebar
Status
Not open for further replies.
Too bad it's probably for the mobile platform, I don't see 14nm desktop GPUs coming anytime soon. Hopefully they surprise us though. I'm actually surprised Samsung would work with Nvidia after the rocky last 2 years. I don't even think Nvidia put Samsung memory on any recent GPUs. I could be wrong though.
 


We will be seeing 14nm GPUs this year, this is why we didn't see the GTX 970 and 980 launch at 20nm because TSMC was going to skip the drop from 28nm to 20nm and go straight to 14nm to keep up with other foundaries.
 
Nvidia's SOCs are on a yearly release. The X1 was just announced and the next announcements will be Jan 2016 at CES.

If Nvidia is really going to use Samsung's 14nm FinFET process then it will be for next gen GPUs and a possible Denver SOC later in the year.
 
...That is, unless Nvidia has completely given up on Denver, in which case the company will probably use Cortex A57 and Cortex A53
Click to expand...

I don't really give a poop one way or the other, but you should likely avoid such supposition without basis (new road map, report, press release, etc.).



 
Businesses like these always surprise me. They announced they will be working together all the while they sue the nuts off eachother. Or has the whole velocity micro incident been resolved?
 
Skip from 28nm to 14nm? I don't see it. I can see Nvidia milking 20nm and 16nm finFET, but going to 14 is a big leap. I guess I'm just used to the minor jumps now, used to jump 20 and more at a time.
 
Why don't the quote button ever work?
Click to expand...
What do you mean? It's not a flaw but a feature that you have to click "add comment" in the top, then click on the comment window in the bottom followed by pressing the quote button!

And is the broken quote system really the biggest gripe for tom's? How about the completely messed up "previous page" feature? Or the fact that comments some times simply doesn't work? Or that you're expected to go into the forum to continue the discussion on an article? Or maybe the completely broken reverse link from a reply in the forums to the articles on the UK site? Or the..................................... 😛
 
tslot05qsljgo9ed wrote: "Nvidia's SOCs are on a yearly release. The X1 was just announced and the next announcements will be Jan 2016 at CES.

If Nvidia is really going to use Samsung's 14nm FinFET process then it will be for next gen GPUs and a possible Denver SOC later in the year. "

The Tegra K1 with the Denver core came out months after the Tegra K1 with the ARM cores. In that case it was on the same process and announced during CES 2014. But, NVIDIA has recently said Denver is not dead and it will be released on a finfet process. There's nothing to say that NVIDIA can't release a version of the X1 with the same Maxwell GPU and a Denver core on 14nm finfet later this year.
 

Even though the gate pitch wouldn't change the gains for lower power consumption will be there, which is what the 20nm SD810 is suffering from. So, yeah, Samsung's 14nm technology will be that much better, even if they'd build their chip using the same ARM blueprint as Qualcomm.
http://semiaccurate.com/2013/05/28/samsung-talks-about-their-14nm-finfet-process/

 


Yes, I don't know why anyone would assume they gave up when time to market forced X1 to use A57 ARM IP instead of Denver. They said directly they'd be back in future stuff (IE, 14nm, I don't think they'll do a 20nm with X1 out soon). Denver will be in the chips from samsung's 14nm process. Either the same version, or tweaked since they should have had time to do some work on optimizing it another 6-9 months. You don't hire a crack cpu team to give up in one chip when it really wasn't designed for the 28nm it ended up on. The fruits of Denver will be seen on samsung's process along with maxwell. Both should easily enable phones with a modem from qcom or samsung.

I could easily see samsung going NV as they'd rather help the smaller player than qcom here (who is in modem competition with them too, where NV really isn't a threat there), for the chips that are usually qcom I mean. That half of the galaxy's that use qcom instead of samsung chips for half the world. I'm fairly certain they'd rather fab & use an NV soc with samsung modem, vs. fabbing for qcom and using those for some products. Qcom makes 6B vs. NV 1/10 that. You don't want to help qcom any more than you want to help Apple if you're samsung. The enemy of your enemy...
 
But Samsung is making chips for Apple, they just had a contract for the new ARM processors Apple wants. Yeah they sue each other every other week, and Apple usually embarrasses Samsung, but money is money and Apple pays the big bucks.
 
As for Apple paying the big bucks the case has been that they paid very little =P

As is though volume helps.

But shouldn't it be more of a decision over at Apple rather than Samsung?
And if Apple pick Samsung even though they are the competition and they have disputes and Apple likely wouldn't want to support them unless they had too can't one assume Samsung actually provides something others may not?
 


Well, I was being overly sarcastic, which should have been obvious as I listed a few of the most annoying bugs on this site that they've never bothered to fix.... 😀
 
Status
Not open for further replies.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom