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!
Will Micron/TSM form an alliance? They did form an alliance to serve their chiplet customers long times ago. TSMC chair’s comments: "I think everyone should not forget who is the technology leader in memory? It is Micron, and Micron has already surpassed Samsung.”
What he didn’t say is far more...
According to some local news, phase one (presumably Fab01 of the site plan) will produce N5P and N4 processes in the 5nm family. The upcoming phase two (Fab02) will adopt a N5 and N3 mixed production mode. This mode is currently in operation at Tainan Fab 18B.
https://semiwiki.com/forum/index.php?threads/tsmc-phoenix-arizona-fab-site-plan.16066/
According to the fab site plan, Fab01 and Fab02 are under construction.
For the development of a new process, it requires close cooperation between process and design teams. Process engineers can only do that much. The last mile in tuning the yield must come from the inputs of a very experienced design team. It is not entirely accurate to say Samsung vs TSMC in...
My understanding is that Meteor Lake is Intel’s first attempt to put many different tiles (CPUs, GPUs, memory,..) together, and to challenge other HPC competitors in a meaningful way. It is a new ball game even for Intel. Many things including process, packaging, testing, software driver, and...
The recent story of TSMC 28nm (or legacy nodes) is actually quite entertaining. Prior to covid, TSMC charges about 10-20% higher than its peers. For a long time, 28nm’s utilization rate was not impressive. As I recall, the TSMC chair once said the world had enough 28 nm Fabs. Then chip shortage...
Glad I can help. Indeed, it is very significant to learn the defect density of each company’s processes. If you find something conclusive, please share it with us.
Yield is a combo of design and process. In the design end, die size, corner simulation, and testing criterium will all change the yield dramatically. This is why the foundry cannot promise the yield for its customers. In some recent reports, Snapdragon 4 Gen 8 yields at 35%, and, at the same...
As I see it, the US is hiring TSMC to build a military-grade 5nm process line in AZ so that the US has an unquestionable lead over Russia/China in digital aerospace/military ICs for the next 20 years.
The military-grade process is much harder than the auto-grade process at the same node. TSMC...
From that interesting news, “Four Taiwanese trade groups and three nonprofit organizations representing the whole of Taiwan’s high-tech industry signed the deal.”
No government? No TSMC? No UMC? No MKT? And, they represent the whole of Taiwan’s high-tech industry? Must be joking. I am curious...
Another important announcement is to build a 22-28nm fab in Japan. It is interesting to know that Japan is willing to take a “half-node” approach (the best Japan’s logic fab is 40 nm) and to pay 50% upfront (~3.5 Billion USD). The caveat is that the chips should supply Japan’s internal use...
The whole report is pretty much summarized again in that graph. If we look at the issue from the CPU design team’s point of view, they are handicapped by the incompetence of the process division. We know the problem is “staying away from EUV for too long”. Now, with 3nm at their disposal, they...
That UDN report said nothing about “box out ….”. We all know that is not how it works in the foundry business.
There is nothing new about Intel’s 3nm order however that report did fill in more detail about the order. It said Intel has completed 3 CPUs and 1 GPU engineering runs in TSMC Fab 12...
Nvidia is the most powerful company in AI/ML. What is what Nvidia has done but not by Intel? Nvidia provides total solutions. From all I can see, AI training was done on Nvidia’s workstations.
In Nvidia’s picture, GPU-CPU is just like big.LITTLE core: GPU handles all heavy computation, and CPU...
