Rekall is a company that provides memory implants of vacations, where a client can take a memory trip to a certain planet and be whoever they desire.
State-backed chipmakers have been trying to push the limit of ASML’s older deep ultraviolet lithography machines, the Dutch supplier’s second-best lineup (after EUVs), with the so-called quadruple patterning technique.
That requires lithographic machines to perform up to four exposures on a silicon wafer, with a total margin of error of no wider than hundredths the diameter of a human hair. Compared to EUV lithography, the multi-patterning technique with DUVs is not only resource-intensive but also prone to alignment errors and yield losses, according to Ying-Wu Liu, an analyst at research firm Yole Group.
That effort isn’t going well because of subpar local gear used in conjunction with ASML’s DUV systems, according to one of the people. On at least one trial production line, engineers have been forced to replace Chinese gear with foreign equipment to ensure reasonable output, the person said.
“Multi-patterning inherently introduces more process steps, increasing the risk of defects and variability,” Liu said. “Additionally, the higher complexity and cost of multi-patterning make it less economically viable for high-volume production of advanced nodes like 5nm.”
I was always skeptical about China’s 7nm fab process. It did not seem economically viable.
Apologies for calling you a sir, miss!
They can’t even get the Android tablet experience figured out after nearly 20 years and now they are going to have deskop/laptop UI as well?
I had an ASUS Transformer Pad 10 years ago. Cool device, but the OS experience with Android was atrocious. I am sure they’ve improved a lot in the last ~10 years, but I have my doubts whether they will provide long term support for laptop version of Android (not to mention getting 3rd party apps working with optimized UI).
That’s one solid heatsink!
Well played, sir!
Good to hear at least some upside for Mastadon from the recent rise in Bluesky users.
Hopefully this trend will continue and and Mastodon can retain at least a solid percentage of those why try it.
By the time these systems become fully operational, TSMC is expected to have advanced to its A10 node, representing several technological generations beyond its current capabilities. This timeline aligns with TSMC’s broader roadmap for advancing chip manufacturing processes.
I can only imagine what the costs would be for A10 fabrication. I remember doing some back of the napkin calculation based on public info and the latest TSMC node represented about ~20% of the cost of a Zen 5 CPU. I wouldn’t be surprised if this increases to 30-40% even with a significant in base prices of leading CPUs.
120kW per rack
I knew GPU compute took a lot of energy, but I didn’t realize it was 120 kW per rack. That is a stupid amount.
I am not familiar with Zack Freedman’s work, but a quick review of via image search honestly does not inspire confidence.
That being said, my tastes are limited to device design where you can’t tell the difference from a normal pair of medical glasses.
I am pretty sure that’s a mockup (likely AI generated).
But nevertheless, isn’t this true of all AR glasses?
The warranty issue was a big fail on their part.
They would be leaving money on the table by not addressing it.
I did not know ASUS “ProArt” series include CPU coolers.
The design has a rather pleasing miniamlist style (especially compared to most “gamer” components with RGB and an overbearing aesthetic). Nothing wrong with that, but it’s good to have choice.
Sounds like they feel they have enough leverage over US companies where they can keep the leading edge node “exclusive” to in-country manufacturing.
And they do have very strong leverage.
It’s solid by iGPU standards, but you’re not going to be able to run modern AAA games at native resolution and above minimal graphics settings.
I can’t comment on the political dimensions, but I can say that corporations want to continue selling into China (huge market) and oppose tariffs or restrictions on sales of bleeding edge tech.
I believe many of major the drivers issues were sorted out after releases. Although I doubt support is anywhere close to being as good as AMD, let alone Nvidia.
1 month is nothing and the data from Statcounter is likely to be more directional (since from my understanding it’s not based on shipments or POS transaction aggregation). If they saw multiple quarters of significant gains in market share, that would be a different story.
The script for Part 3 is still a work in progress. There might be significant revision to the initial draft (could make for a more exciting part 3).
I was really hoping to see more competition in the dGPU space. But considering Intel’s overall troubles and the challenges with gaming dGPUs (even AMD can’t come anywhere close to Nvidia in the gaming dGPU space) this is to be expected.
That’s not what the information implies.
The quadruple patterning approach does not seem viable even if they can deliver products with 7nm.