

MtRainier
627 posts







THE OPTICAL PHOTONICS BOTTLENECK As AI clusters scale past copper’s physical limits, the bottleneck shifts to optical & these are the companies building that layer across the stack: 1. $AAOI building the transceiver layer of the AI network through vertically integrated U.S.-based InP laser manufacturing. It has already secured over $200M in its first volume 1.6T order from one hyperscale customer followed by another $124M in 800G orders from a second. 2. $AEHR building the reliability layer for the optical & AI hardware stack through burn-in & test systems. It just received a record $41M follow-on order from its lead hyperscale customer reinforcing the idea that Sonoma is becoming a key production burn-in platform for high-power AI ASICs. 3. $CRDO building the connectivity layer that helps AI clusters move data faster through active electrical cables, retimers & high-speed interconnect silicon. The DustPhotonics acquisition also extends that platform into silicon photonics before copper becomes a real constraint. 4. $LITE building the laser layer of the AI optical stack through EMLs, optical components & optical switching exposure. The setup is backed by a $2B $NVDA strategic investment & optical circuit switch backlog above $400M with orders reportedly extending through 2028. 5. $VIAV building the testing & validation layer of the optical stack through network instrumentation & photonics measurement tools. It is the picks-and-shovels layer of the transition because every high-speed optical buildout still needs to be tested regardless of which transceiver vendor wins. 6. $COHR building one of the core photonics bottlenecks through indium phosphide lasers, optical engines & communications components tied to next-gen AI networking. It also has a $2B $NVDA strategic investment behind it & is doubling InP device capacity into the 1.6T ramp. 7. $MRVL building the DSP & optical infrastructure layer through electro-optics, PAM DSPs, interconnect silicon & custom networking chips. The Celestial AI deal & NVLink Fusion exposure both strengthen its position as photonics becomes more central to AI cluster design.





GPT2 x 建筑图鉴 x 提示词 知识性拉满,而且它会联网喝茶信息准确定 然后仔细作图,真用心呀 顺着图片看, 我学到了不少信息 肝了3个小时, 还是没有达到我要求的手绘感觉 但是刊载知识性强烈的面上, 果断发布 你可以最后自定义尺寸和建筑名,让它提供图解。 💬Prompt 见评论区一楼









我最近也是类似的想法 VR300配备220TB LPDDR5,一年销量大概是5.5EB(甚至没算HBM) 手机市场LPDDR平均内存大概6~7GB,到2027年手机市场按30%+衰退计算(1.2->0.8B销量),有概率还不到5.5EB Nvidia一个公司单个产品的LPDDR用量,超过了全人类手机消费电子的用量,而且以后GPU的HBM消耗是指数型上涨 人类内存消费增长不大,十年DDR单机装机容量才三倍(7.xGB->23GB),而ai内存消费每一代架构都在翻倍 也就是说ai对内存的消耗是人类的倍数越来越大,指数型拉开差距 唯一限制这个差距拉大的,只有产能扩充的速度

卧槽!知名记者指出苹果CEO库克的管理盲区,过度依赖表格数据将导致无法应对地缘风险。 在最近的一段访谈中,记者Patrick McGee透露了库克被称为表格先生的原因。他分享了一个轶事,库克第一次接手原本只需两小时的每周数据复盘会时,竟然开了13个小时。他对细节有着非常执着的要求,并且把这种工作方式要求到了所有下属身上。 他提到,1998年库克刚加入苹果时,团队里戴眼镜的人并不多。但没过几年,大家几乎都戴上了眼镜。因为他们每天都要盯着特大号的纸张,核对无数个Excel表格里的供需数据。一部手机里有上千个零部件,他们必须检查所有细节来掌控全球供应链。 随后McGee提出了他的担忧。他认为库克确实完美践行了能被测量的才能被管理这句名言,这也是苹果能拥有超高利润率的核心所在。 但问题是,那些无法被写进Excel表格里的隐患,比如给竞争对手国家提供供应链支持所带来的地缘风险,到底该填在表格的哪一列。他总结说,这种对于量化数据的绝对依赖,恰恰是库克现在最大的盲点。


@aleabitoreddit Did you look into superconductor companies like $asmc yet?

AXT昨天定价了一笔5.5亿美元的增发,今天完成交割。募资用途很明确,扩InP衬底产能。 一家年收入8800万美元、还在亏钱的公司,一次增发募了超过年收入6倍的钱。市场在为一件事付溢价。它赌AI光互联的上游材料会长期紧缺。 磷化铟衬底是光模块产业链里存在感最低的环节,也是供给弹性最小的环节。光模块的故事讲了两年,800G放量、1.6T送样、头部公司业绩翻倍,注意力一直在下游。产业链里谁的产能最难扩,谁就卡着整条链的节奏。 InP是做光芯片的基础材料。激光器、探测器、调制器,底层都长在InP衬底上面。1310nm和1550nm波段的高速光发射,硅做不了,只有InP行。你走EML方案也好,走硅光加外部光源也好,物理起点都是一片InP晶圆。 很多人以为硅光是替代InP的。反过来。硅光方案里硅负责波导和调制,发光还是得交给InP。每个硅光引擎配一颗InP的CW激光器。英伟达Quantum-X一台交换机装18个硅光引擎,18颗InP光源。硅光越火,InP用量越大。 四家公司控制全球95%以上的InP衬底产能。住友电工大约六成,AXT通过北京通美大约三成半,加上JX金属和少量其他厂商。半导体材料里很难找到比这更寡头的格局。 缺口的数字很直白。2025年全球InP器件需求大约200万片,产能60万片左右,差了七成。头部订单排到2026年。住友说2027年前扩产40%,AXT说翻倍,JX说扩20%。听起来力度不小。但需求那边是指数型增长。一个800G模块用4到8颗InP激光器芯片,到1.6T每个光引擎对衬底面积需求是800G的四倍。供应商涨两三成,需求可能已经翻了一倍。 扩产又快不起来。InP单晶生长靠VGF或VB法,设备定制化程度高,良率爬坡按年算。跟硅晶圆那套成熟体系完全不是一个速度。2英寸衬底价格从8000元涨到1.67万,涨幅超100%,照样供不应求。 所以AXT这笔5.5亿美元增发的逻辑就很清楚了。去年12月刚融了1亿,四个月后又融5.5亿,募资规模翻了5倍还多。它在抢时间窗口。趁股价在高位把钱拿到手,砸进北京通美的产线。这个窗口关不关得上,取决于扩产速度能不能追上需求曲线。目前看,追不上。

The Head of Claude Code at Anthropic said he hasn’t written code by hand in months. In 2 days he shipped 49 full features. All written 100% by AI. He just dropped a 30 min talk on exactly how he does it. Worth more than any $500 vibe coding course. Bookmark it: