Array
(
[content] =>
[params] => Array
(
[0] => /forum/whats-new/latest-activity?before_id=107405
)
[addOns] => Array
(
[DL6/MLTP] => 13
[Hampel/TimeZoneDebug] => 1000070
[SV/ChangePostDate] => 2010200
[SemiWiki/Newsletter] => 1000010
[SemiWiki/WPMenu] => 1000010
[SemiWiki/XPressExtend] => 1000010
[ThemeHouse/XLink] => 1000970
[ThemeHouse/XPress] => 1010570
[XF] => 2030970
[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 .
I
HBM was the most sensitive, the error rate is in the attached snippet. Good enough for AI inferencing. This was based on 10mm aluminum...
Friday at 3:22 AM
I
There's no doubt that for specific (inference?) tasks which map well onto the Cerebras architecture (*enormous* local memory bandwidth...
Friday at 3:19 AM
I
True, but also if you're using core chiplets like AMD there's less advantage to going to the next node because density (gates/mm2)...
Friday at 3:07 AM
P
The thing is, tariffs will not influence them in any way. Only DIY sector, and the last of commercial PC assemblers do import DRAM as...
Friday at 2:05 AM
Yes, I have understood that distinction for many years. If you have time, please read my previous three posts this thread.
I wrote that...
Friday at 1:04 AM
B
This all seems like another "once in a lifetime" opportunity to jack up the prices.
Not so much benefit for those further down the totem...
Feb 12, 2026
F
Seems early, but probably so, if the companies allow.
Feb 12, 2026
F
A zoom-in on the slide confirms this picture as well...
Feb 12, 2026
It is clear you and I will never agree on this point, though personal insults are not necessary. I'm not particularly interested in an...
Feb 12, 2026
Probably controversial for this forum, but I think Pat tried to break that cycle by building as much leading edge fab capacity as...
Feb 12, 2026
Fred - does this address (partially or fully) the energy per bandwidth gap between current/traditional DRAM solutions and HBM?
Feb 12, 2026
So DRAM and NAND are different…
I think Nanya is mainly DRAM.
Feb 12, 2026
So let me understand this edited version. You are thinking that Intel was showing a lack of foresight about AI software using NAND...
Feb 12, 2026
a nice article by SEMIVISION, link = https://tspasemiconductor.substack.com/p/can-zam-replace-hbm-intel-and-softbank
Feb 12, 2026
Nonsense. Intel exited the DRAM business 40 years ago. Until four months ago, Micron stock was not an especially great investment over...
Feb 12, 2026
Like.
HBM was the most sensitive, the error rate is in the attached snippet. Good enough for AI inferencing. This was based on 10mm aluminum...
hist78 replied to the thread Softbank Corp, Intel announce memory chips collaboration.Yes, I have understood that distinction for many years. If you have time, please read my previous three posts this thread. I wrote that...
tomatoma reacted to Artificer60's post in the thread Intel Foundry is way behind TSMC, but the goal is #2 by 2030 with
Haha.
Nonsense. Intel exited the DRAM business 40 years ago. Until four months ago, Micron stock was not an especially great investment over...