TSMC 7 nm Second-Generation EUV Chips Taped Out, 5 nm Production in April 2019

>5 nm Production in April 2019
AAAAAAAHHHHHHHHHHHHHHHHHHHHHHHHHHHHH

Attached: 1539079592707.jpg (1035x1000, 189K)

Attached: 1502417195657.jpg (2999x2991, 826K)

>pins
>transistor
Wot

Node names have been based on a lot of things in the past, now everyone pretty much just follows what ASML says. Technically a 7nm class node should be able to etch a 7nm line with one pass, thats what the tooling and light source should be capable of.

it's literally marketing.
like how tsmc 12nm is just a tweaked 16nm.

lmao, that CPU has 200MHz less boost clocks than the 8130U (14nm variant) and a non-functional GPU.

Even Intel admits their 10nm is a generation behind their 14nm++++, look at pic related.

Attached: 9110881-15271462873472602_origin.png (1467x722, 201K)

>pins
im talking about the metal things sticking out of this

Attached: transistor.jpg (1000x1000, 156K)

My guy, the transistors in a processor aren't the same thing.

TSMC's 5nm is behind Intel's 10nm if you know anything about microprocessors.

7nm is already ahead of Intels 10nm
but then again Intels 14+++++ is ahead of intels 10nm

No, it isn't. See What you're gonna see when Intel releases 10nm processors in late 2019 is far less overclockability, low yields and high prices. Why? Because their 10nm is garbage and the 8121U has a non-functional GPU, lower clocks and still consumes the same power.

Do you remember what a disaster 14nm was at launch? It had lower yields, was more expensive, had worse overclocking and people kept buying last-gen. The same thing is gonna happen with 10nm except this time it's worse.

what do they look like inside a CPU then?

i've never imagined that i'd see a reference from that faggot.
he is completely illiterate when it comes to tech, and ofc cpus.

Nice try kike

Attached: 1539107572471.png (698x396, 34K)

Not only illiterate - he is completely nuts.

Attached: gfins.jpg (242x184, 10K)

fugg how do they even produce that

Elves, tiny ones.

fucking boomers somebody should fucking shoot this fuck

For stock performance you have to account for the enabling of avx-512 in those cpus. But in terms of peak clock speeds it will be a regression even without the use of avx.

>According to an EE Times article, the Intel engineer who presented the details of the company's 10-nanometer SRAM bit cells at the ISSCC admitted that Intel's technology was "within 15 percent of the smallest reported 7-nm cell."

google.com/amp/s/www.fool.com/amp/investing/2018/02/19/intel-admits-that-its-10-nano-technology-is-behind.aspx

The picture of a transistor you posted, the metal pins are just connections to the actual transistor. A transistor looks nothing like that. It's mostly silicon in a special geometry layout with some other atoms thrown in at at specific places, and the feature size is equivalent to half of what the smallest part of the transistor can be in width or length. So say 5 nm, that means the gate metal of a transistor could be 10 nm across, but the whole transistor is much larger than that.

so where is 7nm cpus and gpus????

Same way they've been making them for a few decades, pal.
Advance lithography, atomic scale deposition, dozens of lith masks, high tech fancy photo resist compounds, a whole plethora of metals, lab created isotopes of materials engineered for one specific use.
Its a multi billion dollar industry, rightfully so.

>For stock performance you have to account for the enabling of avx-512 in those cpus. But in terms of peak clock speeds it will be a regression even without the use of avx.
Intel's consumer CPUs do not have AVX-512. Their server CPUs will, but unless you need AVX-512 then AMD EPYC is the better choice atm.