this post was submitted on 31 Jul 2024
13 points (84.2% liked)

Hardware

464 readers
90 users here now

All things related to technology hardware, with a focus on computing hardware.

Rules:

  1. Follow the Lemmy.world Rules - https://mastodon.world/about

  2. Be kind. No bullying, harassment, racism, sexism etc. against other users.

  3. No Spam, illegal content, or NSFW content.

  4. Please stay on topic, adjacent topics (e.g. software) are fine if they are strongly relevant to technology hardware. Another example would be business news for hardware-focused companies.

  5. Please try and post original sources when possible (as opposed to summaries).

  6. If posting an archived version of the article, please include a URL link to the original article in the body of the post.

Some other hardware communities across Lemmy:

Icon by "icon lauk" under CC BY 3.0

founded 1 year ago
MODERATORS
top 4 comments
sorted by: hot top controversial new old
[–] linearchaos@lemmy.world 5 points 1 month ago (2 children)

I find it extremely hard to believe with their level of espionage that they can't come up with a clone of a 5 year old GPU and slam 32GB of vram on it.

[–] Alphane_Moon@lemmy.world 5 points 1 month ago (1 children)

The issue is access to leading edge semiconductor fab nodes and high end lithography machines (if they want to build their own leading edge node at SMIC).

[–] tjsauce@lemmy.world 2 points 1 month ago

Asianometry on YouTube has great content on this topic, he's great at digesting complex industrial history

[–] Dudewitbow@lemmy.zip 3 points 1 month ago

the hard part for them is actually that their fabs have a hard time making larger dies.

although china has made significant fast process on the capability of their fabs, and being able to already make 5nm yields, they have a problem with doing it with enough volume (possibly due to having high defect rate). its likely part of the reason for example, major companies like huawei havent switched fully to in country fabs for their devices.

GPU dies are absolutely huge and when you have a high defect rate on cuts of silicon, getting working dies are exponentially hard. China also does not have as many driver devs with enough experience to fully switch off it just yet. GPU drivers arent easy, and intels forray into doing dGPUs give you an example that even for multi billion dollar companies, getting a competent driver team isnt easy.