Array
(
[content] =>
[params] => Array
(
[0] => /forum/whats-new/latest-activity?before_id=114573
)
[addOns] => Array
(
[DL6/MLTP] => 13
[Hampel/TimeZoneDebug] => 1000070
[SV/ChangePostDate] => 2010200
[SemiWiki/EmailDomainReplace] => 1000010
[SemiWiki/Newsletter] => 1000010
[SemiWiki/WPMenu] => 1000010
[SemiWiki/XPressExtend] => 1000010
[ThemeHouse/XLink] => 1000970
[ThemeHouse/XPress] => 1010570
[XF] => 2031070
[XFI] => 1060170
)
[wordpress] => /var/www/html
)
You are currently viewing SemiWiki as a guest which gives you limited access to the site. To view blog comments and experience other SemiWiki features you must be a registered member. Registration is fast, simple, and absolutely free so please,
join our community today !
JavaScript is disabled. For a better experience, please enable JavaScript in your browser before proceeding.
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an
alternative browser .
K
I worked with some folks in the 90s that came out of IBM. Some of the sharpest brains I ever had a chance to work with. And easy to work...
Tuesday at 8:15 AM
K
China can build a self-sufficient equipment ecosystem for 28nm and above, and gradually into selective 14/7/5nm production with...
Tuesday at 8:13 AM
O
The link you provides has good current information. The three volume DRAM players plus CXMT. Nanya also is mentioned.
One thing that...
Tuesday at 8:12 AM
O
I’m on the same page as you, benb. We’ve had an upward demand shock at possibly the worst time imaginable: prior years was low CAPEX and...
Tuesday at 8:12 AM
B
In my opinion I think TSMC feels there is an AI bubble so they are not building as many fabs since they will be hard to fill after the...
Tuesday at 7:59 AM
O
AFAIK, yes. They are planning a shift to 3DRAM in this time window. Though, IDK if that explains the less aggressive Samsung build out.
Tuesday at 7:49 AM
B
The report projects Samsung to have the lowest number of new wafers, could they be planning on switching to 3D DRAM?
Tuesday at 7:45 AM
B
CXMT is probably burning a lot of wafers to build HBM, which they have a lot to catch up.
YMTC maybe of help because their tech. is...
Tuesday at 7:28 AM
S
That would land M0 at 22.5nm and if it's 6 Track the cell height would be 135nm and the CGP would likely be 45nm pure guess work on...
Tuesday at 7:27 AM
S
In my opinion I think TSMC feels there is an AI bubble so they are not building as many fabs since they will be hard to fill after the...
Tuesday at 7:26 AM
S
You could, but it would be too slow without chip-level hardware acceleration. You need specialized hardware acceleration to make memory...
Tuesday at 7:26 AM
T
They can. Because China to this day still imports large amount of memory，which is a demand force drive up the memory market price. If...
Tuesday at 7:06 AM
B
The link you provides has good current information. The three volume DRAM players plus CXMT. Nanya also is mentioned.
One thing that...
Tuesday at 7:05 AM
My brief delve into technical details of "AIs" showed just how amazingly computationally inefficient they are.
No signs of anyone with...
Tuesday at 7:02 AM
People who write physics simulation software have long solved most of such memory, cache, dynamic range, useless computation elimination...
Tuesday at 7:02 AM
Like.
I worked with some folks in the 90s that came out of IBM. Some of the sharpest brains I ever had a chance to work with. And easy to work...
Daniel Nenni reacted to Paul2's post in the thread Intel and AMD are teaming up to make x86 CPUs better at AI, without needing a GPU with