📃 Next-Gen Chips Will Be Powered From Below

Buried interconnects will help save Moore's Law

Delivering the current to billions of transistors is quickly becoming one of the major bottlenecks in high-performance SoC design. As transistors continue to be made tinier, the interconnects that supply them with current must be packed ever closer and be made ever finer, which increases resistance and saps power. This can't go on: Without a big change in the way electrons get to and from devices on a chip, it won't matter how much smaller we can make transistors.

Power consumption is crucial whether you're worrying about the cooling bill for a data center or the number of times you have to charge your smartphone each day.

In today's processors both signals and power reach the silicon from above. New technology would separate those functions, saving power and making more room for signal routes.

More information...

#articles #electronics

📟 Magnetic Field Visualizer

This project is about developing a simple magnetic field visualizer. Here, the XMC1100 is the system MCU with fetches magnetic field data of a nearby magnet from the TLV49 3D magnetic sensor.

The MCU communicates with the sensor over I2C protocol. For visualizing the data, an E-Ink display is added to the system. The display is 3.3 V like rest of the system and communicates with the host MCU over SPI.

This visualizer can be used for :
▫️ Identifying North/South poles of a magnet
▫️ Magnet's orientation/heading on XY plane
▫️ Visualize virtual Field lines
▫️ Detect effective field area of a magnet
▫️ Compare magnets strength qualitatively

More information...

#projects #sensors

📃 The Tesla Dojo Chip Is Impressive, But There Are Some Major Technical Issues

An interesting analytical article from SemiAnalysis about recently presented chip from Tesla

One of Tesla’s biggest draws is that the chip-to-chip interconnect is incredibly powerful. It’s more an order magnitude higher bandwidth than the competing ASICs and Nvidia. Tesla achieved this by packing every edge of the die with 112G SerDes. This IP is entirely licensed from Tesla’s partnership with a 3rd party provider. With a total of 576 of these SerDes, they reach a mindboggling 8TB/s IO.

In order to get this massive IO out of the chip, Tesla has to rely on very exotic packaging. Their chip packaging is ludicrously expensive, but it is the only way to get 8TB/s of IO out of such a small package. Normal LGA style packaging that a desktop or server CPU would utilize would not have anywhere near enough pins.

More information...

#articles #electronics

🎬 ESP-NOW - a new protocol for Wi-Fi data transfer between ESP32 or ESP8266

ESP-NOW is a protocol ideally suited to sending out small amounts of data (up to 250 characters). In this video, author uses ESP32 as receiver and an ESP8266 as transmitter and demonstrates the work of the protocol.

https://youtu.be/lkHdk_mKGUw

#video #esp32

📃 Tradeoffs Between Edge Vs. Cloud

Analytical article on the latest tendencies in IoT field

Edge computing is on the upswing because of the process technology, innovations in the AI hardware design, and because we are doing processing in fixed point, like 8-bit, you can put a lot of processing power on devices like mobile platforms.

But computing at the edge also has limitations. When you cannot do something on the edge, because devices are not powerful enough, you have to do them in the cloud. The real value of the cloud, in this case, is a broader analysis of processed data that is less time-sensitive. An edge system may need to do its own analytics for its primary task.

An example is a drone that is examining an electrical tower. It may detect something and immediately has to act on that to figure out what the issue is. When it returns to its home base, it may transfer a lot more data that you want to analyze in the cloud. There are different layers of processing that could be done.

#articles #IoT