Jespi:

I think we will reach the limits of a sillicon, rather than limits of physics. I think, that around 1nm - 0.8nm we will have to switch from sillicon to other materials, i´ve already read about something about graphene and other complex substances, which are too expensive. I think manufacturers are trying to squeeze maximum from the sillicon just because the prices are "low" ... nobody would pay 5000€ for mid-range desktop processor based on let´s say graphene. People are mentioning quantum computing, but i think that´s rather 15-20 years into the future (when we will be able to build quantum computer, that can do general computing not pre-defined specific things)

More than likely we will reach the cost limits for R&D well before materials or physics hit a road block. When manufactures moved on from 2D transistors and started using FINFet the process nm names didn't mean that much anymore. Its kind of like asking how big a person is and providing only the shoulder width of 44 inches which in itself is almost pointless(FinFET transistors have more than one dimension). The fact is even TSMC's enhanced 7nm node has many features larger than 7nm so even calling it 7nm is a bit misleading. We have a long ways to go before we are talking about having transistors that are actually 2nm. However if someone starts doing some math on whats expected densities look like I suspect somewhere around 2-3nm on TSMC's processes(not real 2-3nm) we will be able to have enough transistors in a chip for any normal every day activities like gaming, encoding etc. Once you reach that point where you have more compute or graphics you need in a single CPU/GPU the APU's are going to take over and tech will start to become a commodity. We already see this with core counts and it wont be long with GPU's either. The big transistor change is moving from FINFet to some version of gate all around if you are interested. https://www.anandtech.com/show/14333/samsung-announces-3nm-gaa-mbcfet-pdk-version-01

ToxicTaZ:

Funny how you forget about Core 2 days already lmao... Was double P4... That's 100% by the way... Unfortunately for you it's not optimism! It's a what's actually is coming! Now that Jim Keller the guy that invented your AMD Ryzen is now working for Intel building Ryzen Killer! Ocean Cove is over 80% IPC Gain over Intel's 10th generation "SkyLake" Architecture! Ocean Cove is in Meteor Lake Intel first 7nm+ PCIe 5.0 platform (Q4 2022) The first of 3 cores updates is PCIe 4.0 Rocket Lake with "Willow Cove" (Q1 2021)...... Willow Cove is 25% IPC Gain over Intel's 10th generation. Yes it's just a little more than 7 months away. Then... Brand new architecture based upon 16 cores big.Little The second is PCIe 4.0 Alder Lake with "Golden Cove" cores design that has 50% IPC Gain over Intel's 10th Gen SkyLake architecture...(Q4 2021) Intel 7nm+ is coming prepare yourself for extreme power efficiency and extreme IPC performance! Intel Fab42 7nm is almost online. Q Boom.

AMD Zen 4 is coming with a 200% IPC increase on IBM IP 3D technology in-conjunction with TSMC's 5nm graphene transistors. Brand new architecture, 5 way SMT and uses XDRAM as a cache accelerator for games. I'm honestly not sure how Intel is ever going to compete - Jim Keller is already talking about leaving to go back to Apple.

ToxicTaZ:

Funny how you forget about Core 2 days already lmao... Was double P4... That's 100% by the way...

Are you trying to imply that the current core series is a bad architecture like the P4? Anyone can make a bad architecture design and therefore show large percentages because they fixed their mistake. (Example: AMD with their FX bulldozer compared to AMD Ryzen) So is Intels current architecture a mistake? Is that what you're trying to imply? Honest question dude: Are you employed by Intel to market make-believe products to try and drive hype away from real, current products? What is your goal here? Do you even have a goal? Is it to simply troll? Why are you here?

He's out.

☐ not rekt ☑ rekt