[G3D] TSMC Unveils 6-nanometer Process - Page 4 - Overclock.net - An Overclocking Community

Forum Jump: 

[G3D] TSMC Unveils 6-nanometer Process

Reply
 
Thread Tools
post #31 of 36 (permalink) Old 04-17-2019, 09:31 AM
Otherworlder
 
epic1337's Avatar
 
Join Date: Feb 2011
Posts: 7,186
Rep: 214 (Unique: 122)
Quote: Originally Posted by DNMock View Post
My gut is telling me I'm ignorant on something here, so help me out please:

I had always assumed reducing the gap size was increasing the density. As in the the density of transistors over a given area increases.
that does increase density, but decreasing gap size increases parasitic capacitance, you can check the formula on "capacitor wiki".

on a side note, this does mean that there'd be a limitation to how small they can shrink things, either sacrifice clock speed for density, or sacrifice density for clock speed.

trolling an adult is very dangerous, don't try it at home nor at work. you don't want to play tag with a rabid man.

Last edited by epic1337; 04-17-2019 at 09:38 AM.
epic1337 is offline  
Sponsored Links
Advertisement
 
post #32 of 36 (permalink) Old 04-18-2019, 12:45 AM - Thread Starter
Performance is the bible
 
Join Date: Apr 2009
Posts: 6,610
Rep: 433 (Unique: 299)
Quote: Originally Posted by epic1337 View Post
that does increase density, but decreasing gap size increases parasitic capacitance, you can check the formula on "capacitor wiki".

on a side note, this does mean that there'd be a limitation to how small they can shrink things, either sacrifice clock speed for density, or sacrifice density for clock speed.
It is just about finding a balance.
You can make things a bit more dense, not 18% dense but enough that you will also need less power, and also reduce the gap a bit.
You don't have to sacrifice this or this or this. You can move a step forward on all fronts instead of 5 at the expense of the rest.


Defoler is online now  
post #33 of 36 (permalink) Old 04-18-2019, 02:46 AM
Otherworlder
 
epic1337's Avatar
 
Join Date: Feb 2011
Posts: 7,186
Rep: 214 (Unique: 122)
Quote: Originally Posted by Defoler View Post
It is just about finding a balance.
You can make things a bit more dense, not 18% dense but enough that you will also need less power, and also reduce the gap a bit.
You don't have to sacrifice this or this or this. You can move a step forward on all fronts instead of 5 at the expense of the rest.
yeah but i meant the limits of density increase.
if they further pursue density increase then they'd inevitably hit a brickwall, of course this includes the fact that a single-atom thick gap is also an absolute limit.

trolling an adult is very dangerous, don't try it at home nor at work. you don't want to play tag with a rabid man.
epic1337 is offline  
Sponsored Links
Advertisement
 
post #34 of 36 (permalink) Old 04-19-2019, 09:32 PM
New to Overclock.net
 
os2wiz's Avatar
 
Join Date: Sep 2012
Location: Brooklyn, New York
Posts: 2,325
Rep: 61 (Unique: 40)
Quote: Originally Posted by Defoler View Post
Source


Feels like they are going to be the top chip manufacturer very soon, considering intel are still only on their way to 10nm.

Both AMD and Nvidia could bring much stronger GPUs than 7nm could bring. And this could be great for nvidia considering they have been waiting on 7nm, so they might bring cards in 6nm instead next year, since AMD are planned to bring navi this year, so 7nm.

And if AMD could use that higher density for lower power, and if manufacturing allows, higher speeds, they could finally and completely de-throne intel from the top CPU market with 6nm CPUs and high core speeds.


Feels like this year is only going to be a preview of all the great things coming in 2020.
I hate to douse your speculation with water. But it needs it. Nvidia has gone all in on 7 nm with Samsung not TSMC. It is doubtful after accomodating AMD graphic and cpu needs that TSMC will have adequate capacity for Nvidia as well. TSMC processes are definitely higher performing than Samsungs anemic 7nm. If anybody is in a bit of a corner here it is Nvidia.

Technology is NOT a liberator . Whom technology serves is a class question.
Technology wielded by the capitalist class benefits only the exploiters.
Technology wielded by the working class will serve ALL Humanity.
Ryzen 2700X at 4.2 GHZ, MSI X470 Gaming M7 AC motherboard, AlphaCool Eisbaer 360 cpu AIO cooler
16 GB G.Skill TridentZ DDR4 4266mhz CL-19-19-19-39 1.40v Samsung 960 EVO m.2 NVMe 512 GB
MSI Gaming X GEForce GTX 1080 Ti
os2wiz is offline  
post #35 of 36 (permalink) Old 04-20-2019, 11:33 AM - Thread Starter
Performance is the bible
 
Join Date: Apr 2009
Posts: 6,610
Rep: 433 (Unique: 299)
Quote: Originally Posted by os2wiz View Post
I hate to douse your speculation with water. But it needs it. Nvidia has gone all in on 7 nm with Samsung not TSMC. It is doubtful after accomodating AMD graphic and cpu needs that TSMC will have adequate capacity for Nvidia as well. TSMC processes are definitely higher performing than Samsungs anemic 7nm. If anybody is in a bit of a corner here it is Nvidia.
Do you have a source for this beside speculations?
Nvidia are rumored to consider using samsung for 2020 because TSMC already full with orders (including nvidia ones).
The speculations are that nvidia will use both to gain some extra production.


Defoler is online now  
post #36 of 36 (permalink) Old 04-22-2019, 07:15 AM
New to Overclock.net
 
EniGma1987's Avatar
 
Join Date: Sep 2011
Posts: 6,147
Rep: 330 (Unique: 241)
Quote: Originally Posted by os2wiz View Post
I hate to douse your speculation with water. But it needs it. Nvidia has gone all in on 7 nm with Samsung not TSMC. It is doubtful after accomodating AMD graphic and cpu needs that TSMC will have adequate capacity for Nvidia as well. TSMC processes are definitely higher performing than Samsungs anemic 7nm. If anybody is in a bit of a corner here it is Nvidia.

Actually Samsung's 7nm process is built from the start for EUV. The 7nm process AMD is using at TSMC is not. While we have not seen the capability of either, the Samsung process will have much sharper transistor features and should lead to better performance due to more transistors being "perfect". AMD will convert over to 7nm+ next year, which will be the same 7nm process node just with EUV as well. Then we will actually be able to see which process is superior. I would suspect still Samsung, as it was built from the beginning with the new light source in mind. This new 6-nm process from TSMC is really closer to what the new should be set up as if it had been designed from the beginning with EUV just like Samsung did.

EniGma1987 is offline  
Reply

Quick Reply
Message:
Options

Register Now

In order to be able to post messages on the Overclock.net - An Overclocking Community forums, you must first register.
Please enter your desired user name, your email address and other required details in the form below.
User Name:
If you do not want to register, fill this field only and the name will be used as user name for your post.
Password
Please enter a password for your user account. Note that passwords are case-sensitive.
Password:
Confirm Password:
Email Address
Please enter a valid email address for yourself.
Email Address:

Log-in



Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may post new threads
You may post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off