Weekend Depth: Position, Boundary, and Endgame of Independent Laser Players from the Perspective of CPO+ELS Light Source Trends The bottleneck of AI computing power is shifting from computation to bandwidth. With the expansion of GPU scale, communication between nodes approaches N ² growth, electrical interconnection reaches its peak in power consumption and distance, and optical interconnection changes from "optional" to "essential". During this process, CPO (Co Packaged Optics) and ELS (External Laser Source) began to reconstruct the industry chain: lasers were stripped from the module and became system level resources. Independent laser player SIVEF is at a critical juncture of this change. 1、 What does SIVEF do The core of the company is a WDM DFB laser array based on the InP platform. Simply put: DFB: Stable single wavelength laser WDM: Multi wavelength Multiplexing Array: Multi laser integration The essence is not to sell "lasers", but to provide multi-channel optical bandwidth capability. Under the CPO+ELS architecture: Traditional: One laser per module New architecture: One light source for multiple channels The laser has transformed from a "distributed component" to a "centralized resource", which is the starting point of value redistribution. 2、 Why WDM DFB array The constraints of AI data centers are clear: single channel rates are approaching their limits, electrical interconnect power consumption is not scalable, and bandwidth must rely on "parallelization" The only expandable path is: Multi wavelength division multiplexing (WDM) The premise of WDM is a stable and controllable single wavelength light source (DFB) Therefore, WDM DFB array is currently the optimal solution in engineering. Although not the most advanced theoretical solution, it is the only one that can be scaled up and implemented. 3、 The essential advantages of SIVEF The advantage of SIVEF lies not in "technological monopoly", but in three points: 1) No historical baggage Without modular business, products can be designed entirely around CPO+ELS. 2) System level adaptation The product was designed from the beginning for SiPho/CPO, not for general-purpose lasers. 3) Entering the ecosystem first I have entered the Ayar Labs system and belong to the 'selected players'. This means that current advantage=first mover+architecture matching, not barriers 4、 Competitive landscape First tier: Traditional laser giants Lumentum Holdings Coherent Corp. Advantages: Capacity, Customers, Full Stack Capability Disadvantage: Path dependency Second tier: System company Broadcom Inc. Ayar Labs Advantage: Defining Architecture Risk: Upward integration of light sources Third tier: Light Source Focused on Players SIVEF Features: Flexible and adaptable to new architectures Problem: No scale, no capacity control 5、 The essence of power consumption advantage The advantage of SIVEF is not that a single laser has higher efficiency, but rather: System level efficiency improvement brought about by architecture changes Core changes: reduced number of lasers, shortened optical path, optimized thermal environment The result is a several fold decrease in system power consumption (rather than single point optimization) 6、 SiPho complexity and tuning barriers The difficulty of SiPho system lies not in individual devices, but in multi-layer coupling: wavelength matching, optical coupling, thermal management Calibration is a continuous process, not a one-time design. This brings engineering experience and data accumulation, with a long verification cycle (12-24 months). Therefore, it will form an engineering lock+time lock. But it does not form a technological monopoly Possible flywheels it may form: Design in → Data tuning → Performance improvement → More orders → Further optimization But this is a 'conditional flywheel' that depends on: 1) ELS has become the mainstream architecture 2) Customer generated switching costs 3) The company has the ability to expand production not a single one can be omitted. The real barrier to this track lies in system validation and customer import, rather than the components themselves. 7、 Technological evolution WDM DFB light sources will ultimately be subject to three types of physical constraints: linewidth and noise Spectral density, energy efficiency limit At present, there are still: power consumption: 3-10 times optimization space; Wavelength density: 2-4 times increase space But the limit is system level, not device level. System level players AVGO and ALAB are more likely to become industry chain leaders In the long run, WDM DFB will face the threat of frequency comb Frequency comb is essentially a laser that generates all wavelengths and can theoretically replace DFB arrays. But currently it is still in the laboratory stage and difficult to engineer. It may take 5-10 years to have a marginal impact. Due to space limitations, this article will not be elaborated. 8、 Conclusion SIVEF is in a typical "architecture transition dividend period": the current advantage comes from first mover and adaptation, the mid-term depends on whether the design in is converted into an order, and the long-term is constrained by scale, production capacity, and system integration This is a dynamic competitive track driven by time difference and learning curve. The key is to compete for the customer orders needed to transition from technical validation to large-scale production. Disclaimer: I hold the subject matter mentioned in the article, and my views are inevitably biased and not investment advice. The investment risk is enormous, and I need to be extremely cautious when entering the market