Wednesday, September 6th 2023

ASML's First Pilot Tool for Next-gen Products to be Delivered in 2023

ASML's CEO, Peter Wennink, has announced that his team will be shipping out the first pilot tool (a high-NA EUV system) in its next product line before the end of this year. Reuters reports that supply chain problems have caused hold-ups along the way, but the Dutch multinational corporation is confident in delivering its next-gen opening salvo—these high numerical aperture EUV machines are large enough to warrant transportation via truck, and their per unit cost is over €300 million (~$322 million). The most demanding of chipmakers will be snapping up ASML's behemoth apparatuses in order to produce improved (i.e. smaller) chips over the next ten years.

Wennink spoke to Reuters at an industry event (that took place in Eindhoven): "A few suppliers had some difficulties in actually ramping up and also giving us the right level of technological quality, so that led to some delay. But in fact the first shipment is still this year." The CEO expects to see a growth in revenue thanks to burgeoning interest in AI-oriented silicon—new manufacturing facilities in Arizona and Taiwan are primed to adopt high-NA EUV machines in 2024. Key clients will be experimenting with these new machines (EXE:5200), before a full push into commercial production—logic chip makers have demanded that they get priority access over memory manufacturers. Intel has made declarations, in the recent past, that its foundries are first in line to receive ASML's latest and greatest tools.
Sources: Reuters, Hardware Info NL, Computer Base DE
Add your own comment

3 Comments on ASML's First Pilot Tool for Next-gen Products to be Delivered in 2023

#1
Nanochip
So what intel process will utilize these delayed High-NA machines ? Doesn’t seem like intel 18A is based on the tech. So will it be intel 16A (or whatever the successor to 18A will be called)?
Posted on Reply
#2
Wirko
It looks like all the world's data centers will have to stay on existing Xeons and Epycs forever because the available chip manufacturing capacity will be entirely occupied by "AI".
Posted on Reply
#3
TumbleGeorge
WirkoIt looks like all the world's data centers will have to stay on existing Xeons and Epycs forever because the available chip manufacturing capacity will be entirely occupied by "AI".
"AI" is also composed of data, from end to end.
Posted on Reply
May 3rd, 2024 19:09 EDT change timezone

New Forum Posts

Popular Reviews

Controversial News Posts