The Hacker News discussion revolves around Intel's potential development of ARM-based chips and the broader implications for the semiconductor industry. Here's a breakdown of the key themes:
Intel's Potential ARM Chip Strategy
A central theme is the speculative idea of Intel releasing an ARM chip that incorporates native x86 translation capabilities. This would allow it to run existing x86 software efficiently.
- Dlojudice initiated this line of thought: "Very unlikely to happen but Intel could release an Arm chip with native x86 translation. Arm and AMD IP would be needed but this would be the best chip for Windows".
Clarification on "Native x86 Translation"
There's a discussion about what "native x86 translation" actually means in this context, with comparisons made to Apple's approach.
- Mort96 questioned the concept: "I don't understand what the difference is between 'an ARM chip with native x86 translation' and a dual-ISA x86 and ARM chip." He further elaborated on why a dual-ISA chip might be complex and less advantageous.
- LoganDark offered an interpretation, referencing Apple's M-series chips: "Look at Apple's Rosetta 2 for an example. M-series Apple Silicon has special undocumented modes that mirror x86 architectural quirks that don't usually exist in ARM, in order to support AOT-translated machine code. The chip doesn't support x86 instructions, but it has the amenities to support x86 code. That could be what 'native x86 translation' meant?"
- Mort96 agreed with this interpretation as a possibility, stating: "That's what I suggested in my comment's last paragraph. I don't think that counts as 'an ARM chip with native x86 translation', but really the only person who can say whether that's what dlojudice meant is dlojudice."
Hardware-Level Memory Ordering and Compatibility
A more technical point raised is the implementation of memory ordering guarantees, which are crucial for software compatibility and performance, particularly when emulating or translating instruction sets.
- Mort96 suggested a modification to ARM to improve compatibility: "What they could do is to release an ARM chip with a slightly extended ISA to add the select features which are difficult to emulate in software, such as loads and stores with the memory ordering guarantees x86 provides but ARM doesn't. Apple does this AFAIK, and it's one part of why Rosetta 2 is so good. But any ARM CPU maker could do this."
- Astrange pointed out that other companies are already implementing strong memory ordering: "Fujitsu and Nvidia also implement (at least) TSO."
- Murderfs provided context for Nvidia Denver's design choices: "Denver does it because it was supposed to be an x86 CPU, but they couldn't get an agreement with Intel for patent licensing, so they pivoted into being the first available aarch64 CPU since decode was happening entirely in software."
Intel's Fab Strategy and External Customers
A significant portion of the discussion shifts to Intel's manufacturing (fab) capabilities and the need for external customers to validate and improve their advanced process nodes.
- Threatripper questioned Intel's ability to finance future nodes and suggested a split-up of its fab and processor businesses: "If we assume that intel gets successful with 18A with their x86 processors, would they even have the money to finance the node after that? And the node after that which gets exponentially more expensive? ... For the processor business it may be better to use TSMC for production. For the fab it may be necessary to manufacture for many customers and take a premium for being based in a country in need. So, a split-up may be inevitable and this fabbing a competitive ARM chip surely helps in attracting more customers."
- Blackoil agreed that external funding from companies like Apple, Nvidia, and the US government could be crucial for Intel's fab, especially to break TSMC's monopoly: "Apple, Nvidia and US govt can provide the required funds if they have confidence in its ability to deliver. These companies will benefit from breaking current monopoly of TSMC."
- Cromka added Amazon and Google as potential funders: "Amazon and Google probably as well?"
- Zimpenfish expressed skepticism about Apple's willingness to fund Intel, given Apple's past experiences: "Given Apple's history with Intel's ability to deliver, I'm guessing the confidence there isn't high."
- Walterbell sought clarification, asking: "Are you referring to 5G radio modems or another chip?"
- Mallets contrasted Intel's position with Samsung's, highlighting Samsung's experience with external customers: "Samsung is already in a much better position for this. They have external customers and experience facilitating them. Unlike Intel's track record which doesn't inspire confidence at all."
- Testdelacc1 detailed the "chicken and egg" problem Intel faces with its fab, needing a customer to iterate and prove its capabilities but struggling to attract one without guaranteed yields: "Assuming theyâre telling the truth, theyâve successfully built one chip from that fab. Thatâs good, but it doesnât mean the fab is capable of manufacturing at scale while turning a profit. They need an external customer for the fab so they can iterate and work out the issues. Itâs anyoneâs guess if someone trusts intel to manufacture on their behalf instead of sticking with an established player. Theyâre stuck in a chicken and egg situation - canât reach high yields without a customer, but a customer only wants to sign up if the yields and future deliveries are guaranteed. Intels only hope might be that someone, not naming names, coerces an established company to sign up."
- Baq offered a more optimistic view, suggesting that customers are motivated to solve the unproven process problem to avoid monopoly issues: "That's too pessimistic. In general, customers don't want to be dealing with a monopolist and foundry customers are no different. It's in everyone's interest to solve the unproven process problem, so if Intel has evidence that the process isn't bust, customers will find a product which can be used as a pipe cleaner for mutual benefit."
- YetAnotherNick noted that companies with high profit margins, like Nvidia, might see the benefit of diversifying foundry partners as outweighing the risks associated with Intel's unproven process: "Specially companies like Nvidia for which the gross profit margin is so high their risk of losing TSMC is higher than risk of losing money."
- ExoticPearTree clarified the statement about iteration, implying Intel would bear the initial cost: "I guess you mean Intel to iterate using its own money to get the customer's chip right, no?"
Intel's Arm Licensing and the RISC-V Alternative
The discussion also touched upon Intel's historical relationship with ARM architecture licensing and a suggestion to consider RISC-V.
- Nxobject asked a direct question about the origin of the ARM core design: "Random question: where did the ARM core design come from?"
- Unwind provided information on Intel's licensing status: "Intel are believed to hold an Arm architectural license [1] as far as I know, they have made Arm-based things in the past."
- Sylware proposed RISC-V as a better alternative for Intel: "It should be RISC-V... who is in charge at Intel??"
- Sylware also speculated about potential financial ties: "Is this related to the rumors of softbank (ARM) money injection in Intel?"
- FirmwareBurner responded to the RISC-V suggestion by asking for the reasoning: ">It should be RISC-V... who is in charge at Intel?? Why?"