Not sure if this is what Pat originally meant by “5 nodes in 4 years”
So finally leaving 10nm behind for dies that could greatly benefit from more advanced nodes?
I guess it won’t take too many years from now on to get good battery life from an Intel laptop. Most AMDs still get 1.5-2 times the battery life, all else equal. This can only be good.
Meteor Lake is promising 50% more efficiency for the same level of performance as 13th gen. That should more than bring the laptop efficiency fight back to AMD. I’m not sure about the 2 times, they would usually have a few hours more, unless you mean the H series under load or something.
Most AMDs still get 1.5-2 times the battery life
So Intel right now is just another AMD? Outsourcing their cutting edge chips to TSMC is basically admitting that their own leading process is not good enough. What happened to their ambitious plan to come back as the leader of foundry business?
This is part of that plan. This doesn’t make Intel just an AMD.
It totally depends on what chips are outsourced. Intel has been a TSMC customer for many years.
Intel 4 and Intel 20A are not library complete nodes - they’re optimized specifically for x86 compute tiles. Intel 3 and 18A are the refined, library complete versions of these nodes.
Lunar Lake combines NPU, iGPU, and Compute on a single tile, so Intel 4 and 20A are not viable nodes. Arrow Lake has Compute on its own tile, so it’s using 20A. Lunar Lake either be delayed 6 - 9 months and wait for 18A, or it can launch on TSMC N3. N3 was likely a better choice than Intel 3 - either due to capacity constraints (Granite Rapids and Sierra Forest will be launching on Intel 3 in the same year), or it could be due to performance (N3 could just be better suited for GPU - or it would be too costly to try to and port the Arc iGPU to Intel 3 just for Lunar Lake).
Intel’s business structure has changed. Their nodes and design teams aren’t working tightly together anymore like in the past, where Intel nodes were highly optimized to work with their own designs, and their designs were not portable to other foundries: Intel Fab designs standardized nodes now that compete for customers, and Intel design has more flexibility in which fabs they choose for their now-portable designs. (One recent change is that Intel design teams has to pay for foundry steppings from their own budget, rather than Foundry eating the cost)
It’s a good tactic for Intel - stuff TSMC foundries with Intel wafers so other companies have to use intels foundries
uhh Apple straight up buys entire runs of TSMC nodes. AMD, Nvidia and Intel combined wouldn’t have enough money for that strategy to work.
Nvidia does. AMD doesn’t.
Intel probably could, but their margins are too slim.
Intel’s strategy is probably shifting towards higher margin products, I’d guess a lot of older intel nodes get bought for mil purposes.
Isn’t that the Norwegian strategy?
“Intel has outsourced part of its production to external foundries for decades, and the benefits outweigh the negatives. Additionally, the company is in a better technological position than when it struggled to get 10nm chips out the door. If Intel had allowed itself to produce its CPUs at TSMC back then, then perhaps it wouldn’t have lost as much market share to AMD as it has in recent years.”
What happened to the investments with chips act?
These contracts and roadmaps for chips are done years in advance.
Intel will still produce domestic CPUs. There is too large a market not to, in the corporate and government sectors.
They’re just also doing orders externally. These will largely all be consumer-only segments.
Too lazy to read. Are they manufacturing in Taiwan now?
Eh I’m too lazy to answer
Besides they need an edge for their gpu
its already really good and the most interesting thing happening right now… but they need to implement frame generation.
get driver performance up to par then worry about that. in games where they’ve fixed drivers the performance is fantastic for the price. in other games they’re slow or unplayable.
though I suppose the next design may need less time on drivers since apparently the current design has memory handling issues resolved by software.
Yeah that have it is called xess.
has it expanded into frame generation yet, or is it like early DLSS and FSR where it’s primarily upscaling an image. Frame generation is a bit different than that.
I guess then no there is no interpolated frames on Intel’s implementation yet.
No frame generation yet. Only AI upscaling.
Frame generation is ass even on Nvidia GPUs
amd made the smart move to ditch semiconductor production and focus on designing and outsourcing, cheaper and more profitable, intel probably realize it by now
healthy relationship, good for the market, and smart play from Intel
It’s good for Intel, not for the market.
I remember Nvidia spent like $11 Billion in 2021 in preparation of current gen chips so $14 billion on chips including thos on N3B checks out
Maybe my next platform will be Intel again, if the Intel finally makes a CPU that’s faster than the competition and at the same time doesn’t consume as much power as a high end GPU. Basically I want the Haswell times back (AMD was slower ,inefficient, hot and a power hog compared to Intel even if Intel only had half the cores)… until then I’ll stay with Ryzen.
🤡
Big question is when would Intel start getting chips from N3B. If its H2 of 2024 then capacity wont be an issue as TSMC should have both N3B/N3E in production and can support anyone who is ready to pay. Obviously Intel is using TSMC for all GFX tiles across all chips and also all non CPU products like Gaudi though Gaudi 3 is just on N5(which is weird in 2024).
If you can’t beat them, join them.
so when are these new ‘next gen’ intel 5nm?? hitting the shelfs exactly? I am sitting on 9900k waiting to be finally put to rest, but it was a great cpu even now playing everything at 4k60+ with 4090
The 5nm are for data centers only I believe. Arrow Lake for customers is 2nm and that’s supposed to be late 2024/ early 2025
I don’t know of any Intel nodes called “5nm”, but Intel 4 based client chips are launching in laptops on Dec. 14th, and 20A based desktop chips are launching sometime next year, likely in the typical October - November timeframe Intel usually launches desktop chips.