Array
(
    [content] => 
    [params] => Array
        (
            [0] => /forum/index.php?threads/gelsinger-will-compete-with-amd-nvidia-in-2-years.14865/
        )

    [addOns] => Array
        (
            [DL6/MLTP] => 13
            [Hampel/TimeZoneDebug] => 1000070
            [SV/ChangePostDate] => 2010200
            [SemiWiki/Newsletter] => 1000010
            [SemiWiki/WPMenu] => 1000010
            [SemiWiki/XPressExtend] => 1000010
            [ThemeHouse/XLink] => 1000970
            [ThemeHouse/XPress] => 1010570
            [XF] => 2021071
            [XFI] => 1050270
        )

    [wordpress] => /var/www/html
)

Gelsinger will compete with AMD & NVIDIA in 2 years

Arthur Hanson

Well-known member
Just got done listening to Gelsinger and he says Intel will advance 4 nodes in five years. Also, he stated they will come out with a technology called ribbon technology for transistors that will be superior to what's out there. He also says they set to dominate the software end with fifteen thousand developers. He also stated they will compete with Nvidia and AMD in two years. Any thoughts or opinions on this are appreciated as I'm an outsider to the industry.
 
  • Like
Reactions: VCT

count

Well-known member
There was a pretty good summary here:


My view:
1. I don't think they can successfully advance 4 nodes in 5 years. That's fantasy.

2. Ribbon is Intel jargon for Gate All Around (GAA). Kind of like how Tri-Gate was the Intel term for FinFET and Tile is the Intel term for chiplets. Samsung will be the first to move to GAA at 3nm, and I think TSMC will make the move at 2nm. I doubt there is anything that special about Intel's Ribbon FET vs GAA FETs.

3. Intel has failed over and over at software. Out of the 15,000 software developers at Intel I bet the majority at coding all the bloatware that gets preinstalled on your PC that everyone just uninstalls later. That said, Pat has some credibility running a software company, and it's possible that software can become a strength for the company.

4. Pretty unlikely they will catch up to NVidia for GPU at this point. If they execute well they could take the lead back from AMD, but AMD has become formidable under Lisa Su. I don't think Pat can hold a candle to Lisa Su as far as leadership goes.
 

Daniel Nenni

Admin
Staff member
4. Pretty unlikely they will catch up to NVidia for GPU at this point. If they execute well they could take the lead back from AMD, but AMD has become formidable under Lisa Su. I don't think Pat can hold a candle to Lisa Su as far as leadership goes.

And I don't think Pat is in the same class as Nvidia CEO Jensen Huang, not even close.
 

slin

Member
Not saying it's impossible but Pat and Intel is challenging multiple companies, each of which has the entirely company solely focused in that one field. AMD with CPUs, NVidia with GPUs, and TSMC with pure-play foundry business. It's an incredibly lofty goal for Intel to regain dominance in those 3 areas in the next 5 years. Given the track record in the past decade or two I seriously doubt it.
 

count

Well-known member
Not saying it's impossible but Pat and Intel is challenging multiple companies, each of which has the entirely company solely focused in that one field. AMD with CPUs, NVidia with GPUs, and TSMC with pure-play foundry business. It's an incredibly lofty goal for Intel to regain dominance in those 3 areas in the next 5 years. Given the track record in the past decade or two I seriously doubt it.
Intel still sees itself as the 800lb gorilla of the semiconductor industry, even though these days it's really just a chimp.
 
P

Portland

Guest
Yes, Jensen Huang is a true visionary. He has done an absolutely incredible job.

Nvidia made the right vertical integrations and partnerships with tsmc. Look at the rivalry between ati and nvidia it looks like a lot of guessing and a lot of luck.
 

count

Well-known member
I wonder, is Lisa Su in the same class as Jensen?
I personally believe she is even better.

Jensen benefited from the tailwind of AI, where people started to use GPUs for training their deep learning models. NVidia, with it's strength in GPUs was perfectly positioned for this trend by chance. Jensen was successful because he immediately recognized the importance of AI and pivoted his company over with a brand new vision of becoming the leader in AI/ML which they have succeeded at wildly. Their previous strategy for getting into HPC was "Project Denver" which was an ARM based design.

Lisa Su on the other hand, brought AMD back from the dead. She focused the company on it's strengths and oversaw the greatest turnaround in tech history since Apple. This is in spite of vicious competition from Intel and NVidia. AMD did not benefit from the same tailwinds as NVidia, and in fact was disadvantaged as everyone had already adopted CUDA for AI. However under Lisa Su, AMD has managed to claw it's way back from a disadvantage in both CPU and GPU and is competing successfully against both Intel and NVidia.
 
Top