自制海上步行机穿越爱尔兰海 nike创新足尖舞蹈鞋 hiriko环保概念小汽车 能感应情绪变化的大耳朵 惊悚的大脑蜡烛 迷你“睡箱” 透明遥控器 钉子开瓶器 球状手机音箱 巨大的烧瓶艺术 会写毛笔字的机器人 可以自动跟随的行李箱 概念手环触摸MP3 一屏幕显示多内容技术 超酷0℃概念直升机 排插理线器 各款iphone外置播放器欣赏 指尖上的名人 节省空间与材料的新型建筑结构 骑行运动浴缸 微软新出手势感应设备 超级方便实用的微波炉和电磁炉 玻璃触摸键盘与鼠标 三维折纸秀 微型圆盘办公套装 海里的餐厅 MC 一片木锤 遥控蝙蝠 全球最薄的磁性传感器 高速书本扫描仪 创意LED发光键盘 彩虹镜子 苔藓地毯 通过行走产生能量的鞋 数字绘画笔 创意食品服装 通信手套 纸质布质太阳能电池 雪铁龙C2汽车人 折叠环保袋 照片讲述机 极具创意的楼梯设计 创意纺车时钟 多用折叠垫 创意发光陶瓷杯 智能水龙头 电动降噪耳塞 魔术电脑屏幕贴膜 无线太阳能充电桌 创意太阳能加热饭盒
Amazon hires Disney SVP Kyle Laughlin as director of Alexa Gadgets
Apple’s battery cases return for the iPhone XS and XR
The Future Of TV Is HTML
This Week On Bullish: Can The Bold Italic Come Back?
Microsoft Drops New Windows 10 Mobile Build, Promises Faster Release Cadence
Cisco Pops 2.5% After Reporting Better-Than-Expected FQ4 Revenue Of $12.84B
Yahoo Sags 4% After Alibaba's Q2 Earnings Disappointment
Microsoft Postpones Ship Date For Its 'Surface Hub' Wall Computer To January
You Can Now Run Windows 10 On Your Mac
Microsoft Drops Another Windows 10 Update
Zulily's Buyout Spike Leaves It Less Valuable Than When It Went Public
Microsoft Launches New Windows 10 Build, Vows To Keep Its Early Access Program Alive


漂流瓶终于彻底拜拜 微信7.0.4新版体验
微信漂流瓶被玩坏了 聊聊漂流瓶里那些事
微信关闭漂流瓶 它曾经满足了我们对世界的好奇
微信暂停漂流瓶功能:对色情内容零容忍
[视频]惠普Chromebook x360 14 G1评测:搭载Chrome OS的商务变形本
特斯拉:北京客户可三年免息融资购车并免费租赁车牌
借贷宝:停止催收百名裸条女大学生 未满23岁将不得借贷
京东白条多地频现盗刷 消费者遭催收公司“逼债”
借款野蛮催收行为将被规范 真是几家欢喜几家愁
为规范网贷催收 上海互金协会发行业倡议书
腾讯解释为什么微信没有夜间模式 真相你相信吗?
一张发行8年的微信唱片:只收录了4首歌曲


GlobalFoundries Gives 7 nm Capacity Update, Mulls Skipping 5 nm

当前位置: 艾金森 > 门户 > 海外资讯

点击量 25
编辑: 1   作者: 3Ddayin.net   时间: 2018/12/26 8:34:29  

High-ranking executives of GlobalFoundries this month gave several updates concerning future plans of the contract maker of semiconductors. As it appears, in a bid to provide more tangible advantages to its customers and not to invest in short-lasting nodes, the company is mulling skipping 5 nm manufacturing technology like it did with the 10 nm fabrication process. In addition, the company admits that its only leading-edge fab may not have enough capacity for all of its customers at 7 nm, and some of its clients may need to work with other foundries to meet their demand.

7LP On Track for HVM, But Capacities Are Concern

Gary Patton, CTO of GlobalFoundries, said in an interview with EETimes that the company’s first-generation 7LP process technology (7 nm leading performance), which relies exclusively on DUV step and scan systems, is on-track for high-volume manufacturing in the second half of this year. One of the first customers to use this tech will be AMD, which briefly announced its 7LP Vega GPU for a couple of times already. Not everything is rosy with 7LP though.

According to Mr. Patton, since the Fab 8 produces leading edge chips for all of the company’s clients, its 7LP capacity may be limited, which is why AMD may not get enough capacity. The good news is that GF’s 1st Gen 7LP is similar to TSMC’s CLN7FF (which also does not use EUV) and therefore clients with significant demand will be able to port their designs from one foundry to another with some ease. Under the wafer supply agreement signed in 2016, AMD has rights to use foundries other than GlobalFoundries in certain situations. Though whether this capacity-constrained situation actually comes to pass, and whether or not AMD and other partners of GlobalFoundries will have enough time to develop all the designs they need for both foundries should it occur, remains to be seen.

Meanwhile one thing that should be kept in mind is that GlobalFoundries is planning for three generations of 7LP processes, with two of them using EUV lithography and ASML’s Twinscan NXE steppers. As of early 2018, Fab 8 had only one EUV stepper installed with another being constructed. Meanwhile, there is a space for two more, so eventually GlobalFoundries will have four Twinscan NXE machines in Fab 8.

Capacity Expansion May Be Needed for Next Major Node

Earlier this year GlobalFoundries said that to provide its customers a tangible performance, transistor density, and power reduction boost from 7LP, it might need to move to new transistor structure. The company did not disclose any names for its next-gen node and for a reason. Apparently, GlobalFoundries is mulling whether to skip what the industry calls the 5 nm node and jump right to 3 nm. No final decision has been made and there are other concerns.

As leading edge process technologies get more complex and expensive to develop amid rising demand for chips, foundries need to boost their production capacities to remain profitable. Companies like TSMC and Samsung Foundry can land orders from large makers of consumer electronics (Apple, Samsung Electronics, etc.) that can offset development costs of nodes and equipment by ordering huge amounts of chips at premium prices. As a result, to be successful in the long term, GlobalFoundries will not only need to develop competitive process technologies, but also expand its leading edge capacities. Alternatively, GlobalFoundries is open to another process-sharing deal, like the one it signed with Samsung for their 14LPP technology.

Several years ago, GlobalFoundries already considered increasing capacity of the Fab 8 by building an additional module, but the plan was shelved back then. Instead, the company switched some of the equipment in Fab 8 to improve 14LPP supply in 2017. GlobalFoundries is still mulling whether to construct what would be Fab 8 Module 2, but a project like this might require support from the federal government, which is something that the company may not get in the USA. Meanwhile, since GlobalFoundries runs fabs not only in the U.S., but also in Germany and Singapore, another way of adding leading edge capacity would be to upgrade one or two of the non-U.S. facilities assuming that the respective governments provide incentives to the chip maker.

At present it is too early to make any predictions regarding GlobalFoundries’ plans for 5 nm or 3 nm fabrication technologies because they are years away. Meanwhile, if GlobalFoundries is seriously thinking about expanding its leading edge capacities for its next major node, it may have to make its decision rather sooner than later.