Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
SemiTom
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
21 ms
·
151.
▲
RISC-V Markets, Security and Growth Prospects
(semiengineering.com)
1 points
by
SemiTom
7y ago
|
0 comments
152.
▲
Die-to-Die Connectivity
(semiengineering.com)
3 points
by
SemiTom
7y ago
|
0 comments
153.
▲
Why Standard Memory Choices Are So Confusing
(semiengineering.com)
1 points
by
SemiTom
7y ago
|
0 comments
154.
▲
GDDR6 Drilldown: Applications, Tradeoffs and Specs
(semiengineering.com)
1 points
by
SemiTom
7y ago
|
0 comments
155.
▲
Power Complexity on the Rise
(semiengineering.com)
1 points
by
SemiTom
7y ago
|
0 comments
156.
▲
October’19 Startup Funding: Mega Harvest
(semiengineering.com)
2 points
by
SemiTom
7y ago
|
0 comments
157.
▲
Revving Up For Edge Computing
(semiengineering.com)
47 points
by
SemiTom
7y ago
|
18 comments
158.
▲
How Secure Is Your Face? Racing Against Hackers in Biometrics
(semiengineering.com)
2 points
by
SemiTom
7y ago
|
0 comments
159.
▲
by
SemiTom
7y ago
Interesting to read Thomas Dolby's thoughts on music/technology interfaces--particularly with VR https://semiengineering.com/thomas-dolbys-very-different-vie...
160.
▲
IP Management and Development at 5/3nm
(semiengineering.com)
1 points
by
SemiTom
7y ago
|
0 comments
161.
▲
by
SemiTom
7y ago
https://semiengineering.com/5nm-vs-3nm/ TSMC’s 5nm technology is 15% faster with 30% lower power than 7nm. A second version of 5nm, due out next year, is 7% faster. Both versions also will use EUV.
162.
▲
Utilizing Computational Memory: Processing near memory is a game changer
(semiengineering.com)
1 points
by
SemiTom
7y ago
|
0 comments
163.
▲
RISC-V Challenges and Opportunities
(semiengineering.com)
2 points
by
SemiTom
7y ago
|
0 comments
164.
▲
What’s the Best Advanced Packaging Option?
(semiengineering.com)
2 points
by
SemiTom
7y ago
|
0 comments
165.
▲
Building an MRAM Array
(semiengineering.com)
1 points
by
SemiTom
7y ago
|
0 comments
166.
▲
by
SemiTom
7y ago
FPGAs are getting larger, more complex, and significantly harder to verify and debug https://semiengineering.com/fpga-design-tradeoffs-getting-mo...
167.
▲
Thomas Dolby’s Different View of Progress:Bridging Gap Between Music and Tech
(semiengineering.com)
1 points
by
SemiTom
7y ago
|
0 comments
168.
▲
Pushing Memory Harder
(semiengineering.com)
1 points
by
SemiTom
7y ago
|
0 comments
169.
▲
Using Multiple Inferencing Chips in Neural Networks
(semiengineering.com)
3 points
by
SemiTom
7y ago
|
0 comments
170.
▲
New Security Risks Create Need for Stealthy Chips
(semiengineering.com)
1 points
by
SemiTom
7y ago
|
0 comments
171.
▲
Over $7B Raised in Mega-Rounds by 27 Startups
(semiengineering.com)
2 points
by
SemiTom
7y ago
|
0 comments
172.
▲
Solving the Memory Bottleneck
(semiengineering.com)
2 points
by
SemiTom
7y ago
|
0 comments
173.
▲
by
SemiTom
7y ago
Scaling will continue in conjunction with packaging, but whether that’s Moore’s Law is debatable. The economics of traditional shrinking are gone. It now costs more per transistor at each new node. And the power/performance improvement
174.
▲
The Great Data Flood Ahead
(semiengineering.com)
22 points
by
SemiTom
7y ago
|
8 comments
175.
▲
SiC Market Moves into Overdrive
(semiengineering.com)
1 points
by
SemiTom
7y ago
|
0 comments
176.
▲
MRAM: Magnetic Memories Reach for Center Stage
(semiengineering.com)
1 points
by
SemiTom
7y ago
|
0 comments
177.
▲
Why DRAM Won’t Go Away
(semiengineering.com)
1 points
by
SemiTom
7y ago
|
0 comments
178.
▲
The New CXL Standard: Critical for Moving Large Amounts of AI/ML Data
(semiengineering.com)
1 points
by
SemiTom
7y ago
|
0 comments
179.
▲
FPGA Design Tradeoffs Getting Tougher
(semiengineering.com)
1 points
by
SemiTom
7y ago
|
0 comments
180.
▲
Making Better Use of Memory in AI
(semiengineering.com)
2 points
by
SemiTom
7y ago
|
0 comments
More ›