TSMC Starts 10nm production in 2016

Published by

Click here to post a comment for TSMC Starts 10nm production in 2016 on our message forum
https://forums.guru3d.com/data/avatars/m/227/227853.jpg
So, news announcements from the big companies have now turned into click-bait bull**** too. In reality, MAYBE we will all see 14nm used for actual processors by mid 2016. Also, intels 22nm is the same size as other`s 32nm or so. They basically just lie. Yes, thats right. FCUK IT! Lets just LIE.
Source? Also Broadwell is 14nm.
https://forums.guru3d.com/data/avatars/m/80/80129.jpg
what sort of things are you punching into google? I tried to find these sources you talk about but its just coming back with CPU / GPU spec-sheets referring to their die wafer size. I'm having a tough time making sense of your post.
That's because his post makes no sense. There is more to "14nm" then just the number. There is gate pitch, m1 pitch, fin pitch, cell size, etc. While some of Intel's nodes are behind competitors, some of them are also in front of others. With 14nm for example, it beats Samsung's process in every single category. http://www.extremetech.com/wp-content/uploads/2015/02/Intel-Features.jpg
data/avatar/default/avatar32.webp
In reality TSMC will announce later that 10nm production is delayed by 9 to 12 month, basically what Intel said about 10nm, put it back because yields are very poor. Remember TSMC announced 20nm then was delayed....hence why NV/AMD stayed with 28nm an skipped 20nm for 16nm products. Key point TSMC hasn't even got 16nm out the door, Intel already makes 14nm products now. Intel saying 10nm will be delayed now...where as TSMC saying 10nm is on track. I know who I believe an sure is ain't TSMC
https://forums.guru3d.com/data/avatars/m/227/227853.jpg
I cant be asked to go find you a specific source. This has been heard many times from those who pay close attention how intel calculates their nm. They ALL calculate the nm, and then advertise the calculated results. On intels 22nm, the smallest measurable distance is somewhere about 50nm for example. Similar on AMD`s 32nm. AMD calculates the nm they are going to advertise the same way most do. Intel goes an extra step and lies a bit more. This IS common knowledge for those who have been paying attention to this tech world for a while. If you really care about it, invest your own time a bit, im sure you`ll find many sources (usually with microscope pics of transistors and stuff).
Do you have a source or not? What you said doesn't make any sense. 'Smallest measurable distance' of 50nm, I have no idea what that means because the collocation itself makes no sense when talking about fab processes. The smallest measurable distance in physics is 1 planck length, I have no idea how you correlated that with fab nodes. The process' 'number' measures the average half-pitch, do note AVERAGE. It is known that in Intel's 22nm process there can have half-pitches of up to 26nm, but also smaller than 22nm half-pitches, virtually making the average around 22nm. The max half-pitch doesn't matter, it just proves that the process is not perfect. A perfect process would mean 100% yields which would also mean we've mastered the current node at such a high degree that we would switch to the next node immediately. And lastly the half-pitch isn't everything, '14nm' is just a simplification. Read Denial's post. I have seen no evidence that Intel is lying about their fab nodes and until you provide hard proof your argument will remain baseless.
https://forums.guru3d.com/data/avatars/m/164/164033.jpg
I cant be asked to go find you a specific source. This has been heard many times from those who pay close attention how intel calculates their nm. They ALL calculate the nm, and then advertise the calculated results. On intels 22nm, the smallest measurable distance is somewhere about 50nm for example. Similar on AMD`s 32nm. AMD calculates the nm they are going to advertise the same way most do. Intel goes an extra step and lies a bit more. This IS common knowledge for those who have been paying attention to this tech world for a while. If you really care about it, invest your own time a bit, im sure you`ll find many sources (usually with microscope pics of transistors and stuff).
If you claim something you should provide some frikkin sources to support your claims... Else all this is just pointless jargon from you. And yeah right tsmc lets see in 2017 when you get possibly some 10nm out!