Subscribe
Sign in
Home
Meetings
Accelerator Industry Model
Datacenter Industry Model
Compliance Policies
Archive
About
New
Top
Discussion
Nvidia’s Optical Boogeyman – NVL72, Infiniband Scale Out, 800G & 1.6T Ramp
Transceiver to GPU Ratio, DSP Growth, Revealing The Real Boogeyman
Mar 25
•
Dylan Patel
and
Daniel Nishball
74
Share this post
Nvidia’s Optical Boogeyman – NVL72, Infiniband Scale Out, 800G & 1.6T Ramp
www.semianalysis.com
Copy link
Facebook
Email
Note
Other
5
Nvidia B100, B200, GB200 - COGS, Pricing, Margins, Ramp - Oberon, Umbriel, Miranda
The B Stands For Jensen's Benevolence
Mar 18
•
Dylan Patel
,
Myron Xie
, and
Chaolien Tseng
82
Share this post
Nvidia B100, B200, GB200 - COGS, Pricing, Margins, Ramp - Oberon, Umbriel, Miranda
www.semianalysis.com
Copy link
Facebook
Email
Note
Other
15
Astera Labs IPO - The Next Connectivity Superhero or Steamrolled By Competition?
Bottoms Up Model, Units, ASP, Revenue, By Hyperscaler Analysis, EPS & Cashflow, Competitive Analysis
Mar 17
•
Dylan Patel
and
Jeremie Eliahou Ontiveros
66
Share this post
Astera Labs IPO - The Next Connectivity Superhero or Steamrolled By Competition?
www.semianalysis.com
Copy link
Facebook
Email
Note
Other
4
CXL Is Dead In The AI Era
AI Accelerator Beachfront Considerations, Memory Pooling Negatives, Custom Silicon Adoption
Mar 16
•
Dylan Patel
and
Jeremie Eliahou Ontiveros
82
Share this post
CXL Is Dead In The AI Era
www.semianalysis.com
Copy link
Facebook
Email
Note
Other
4
AI Datacenter Energy Dilemma - Race for AI Datacenter Space
Gigawatt Dreams and Matroyshka Brains Limited By Datacenters Not Chips
Mar 13
•
Dylan Patel
,
Daniel Nishball
, and
Jeremie Eliahou Ontiveros
99
Share this post
AI Datacenter Energy Dilemma - Race for AI Datacenter Space
www.semianalysis.com
Copy link
Facebook
Email
Note
Other
30
February 2024
Groq Inference Tokenomics: Speed, But At What Cost?
Faster than Nvidia? Dissecting the economics
Feb 21
•
Dylan Patel
and
Daniel Nishball
108
Share this post
Groq Inference Tokenomics: Speed, But At What Cost?
www.semianalysis.com
Copy link
Facebook
Email
Note
Other
5
Hybrid Bonding Process Flow - Advanced Packaging Part 5
BESI, EV Group, AMAT, TEL, ASMPT, SET, Shibaura, SUSS Microtec
Feb 9
•
Dylan Patel
,
Myron Xie
, and
Jeff Koch
81
Share this post
Hybrid Bonding Process Flow - Advanced Packaging Part 5
www.semianalysis.com
Copy link
Facebook
Email
Note
Other
6
January 2024
Neural Network Quantization & Number Formats From First Principles
Inference & Training - Next Gen Hardware for Nvidia, AMD, Intel, Google, Microsoft, Meta, Arm, Qualcomm, MatX and Lemurian Labs
Jan 11
•
Dylan Patel
145
Share this post
Neural Network Quantization & Number Formats From First Principles
www.semianalysis.com
Copy link
Facebook
Email
Note
Other
10
Intel GenAI For Yield, TSMC CFET & 3D Stacking, AMD 3D Device Modeling, Applied Materials Material Innovation, SK Hynix HBM4, Micron 3D DRAM…
China’s CXMT Sanctions Violation, Samsung 1000-layer Vertical NAND (VNAND), Kioxia highest density CMOS bonded to array (CBA) NAND, Micron non-volatile…
Jan 3
•
Dylan Patel
,
Jeff Koch
,
Myron Xie
,
Daniel Nishball
, and
Anand Chamarthy
93
Share this post
Intel GenAI For Yield, TSMC CFET & 3D Stacking, AMD 3D Device Modeling, Applied Materials Material Innovation, SK Hynix HBM4, Micron 3D DRAM & FeRAM, Hybrid Bonding vs TCB - IEDM 2023
www.semianalysis.com
Copy link
Facebook
Email
Note
Other
4
December 2023
Inference Race To The Bottom - Make It Up On Volume?
Mixtral Inference Costs on H100, MI300X, H200, A100, Speculative Decoding
Dec 18, 2023
•
Dylan Patel
and
Daniel Nishball
103
Share this post
Inference Race To The Bottom - Make It Up On Volume?
www.semianalysis.com
Copy link
Facebook
Email
Note
Other
10
ASML Dilemma: High-NA EUV is Worse vs Low-NA EUV Multi-Patterning
Cost model for low- & high-NA EUV, Technical Challenges
Dec 11, 2023
•
Dylan Patel
,
Jeff Koch
, and
Lithos Graphein
62
Share this post
ASML Dilemma: High-NA EUV is Worse vs Low-NA EUV Multi-Patterning
www.semianalysis.com
Copy link
Facebook
Email
Note
Other
15
AMD MI300 Performance - Faster Than H100, But How Much?
MI400 Broadcom + AMD Anti-Nvidia Alliance Coming With UEC and Open XGMI
Dec 6, 2023
•
Dylan Patel
and
Daniel Nishball
79
Share this post
AMD MI300 Performance - Faster Than H100, But How Much?
www.semianalysis.com
Copy link
Facebook
Email
Note
Other
16
This site requires JavaScript to run correctly. Please
turn on JavaScript
or unblock scripts