top of page

POSTS

tld image.jpg

AUTONOMOUS DRIVING

Autonomous vehicles are the future, already started harnessing compute intensive deep learning platforms with lots of high definition Cameras, Sensors, Sonars, Lidars, etc.  and supercomputing level performance processing towards fully computer operated vehicles in a decade from now. Time lines for levels is: L1 in 2000, L2 in 2013, L3 in 2018, L4 in 2024, L5 in 2027-30

tld image.jpg

DRONE DELIVERY

Flipdish restaurants in Dublin will have the option to provide drone delivery to customers using their online ordering system. This service promises a three-minute delivery time to orders within a 2 km radius of the restaurant, keeping the food fresh and hot. Using technology to navigate the skies, the drones will be able to draw a straight line between the restaurants’ locations and the destination, cutting out complicated, tiresome routes for drivers and avoiding traffic lights and traffic jam

tld image.jpg

AUTOMATED GUIDED VEHICLE​

Automated Guided Vehicle (AVG), is a smart vehicle designed and built with features that reduce costs, increase productivities and efficiencies in large facilities. Typically, embedded machine learning yields autonomous route and re-route planning, automatic obstacles avoidance, free navigation without auxiliary facilities such as floor magnet, navigational beacons, color tapes. Carry 1000 to 500 Kg and move at walking speed

tld image.jpg

CPU & GPU​

GPUs are more of ALU's than processors, that are superior in doing arithmetic operations typically used in graphics or games. CPU & GPU are like Brain & Brawn, cooperating on code; the CPU executes instructions and the GPU executes the compute-intensive arithmetic on large data sets. Today, there is a trend now of enhancing the GPUs capabilities to execute general instructions, General Processing on GPU (GPGPU). GPU cores are simple 1 to 2 Million transistors each deep pipelines (100's of stages), while CPU cores are complex 100 to 200 Million transistors each with shallow pipelines (10-30 stages). CPU cores have larger caches, while GPU have higher memory bandwidth (faster and wider data channels, may reach 4069 bits). GPU cores run at lower frequency and lower voltage, and yet may consume up to 1200W of power

tld image.jpg

IT MARKET SHARE​

Information Technology industry has been leading all others including Energy, with 46% increase in the 2019 and 305% over the last decade. The current market share of Information Technology is 26%. Market values in Billions of US dollars: Apple 962, Microsoft 947, Amazon 916, Alphabet 863, Berkshire 516, Facebook 512, AliBaba 481, ..., Samsung 272, Proctor & Gamble 265, Royal Dutch Shell 265, Intel 263, Cisco 248, ...

tld image.jpg

TESLA GV100

The world’s first 21 Billion transistors 815 square mm using in 12 nm process is here. The 5120 CUDA Cores, utilizing 16 GB HBM2 operating at 900 GB/s, to achieve 15 TFLOPS. The cost of development was 3 billion dollars

tld image.jpg

CAN BUS

A modern automobile may have as many as 70 electronic control units (ECU) for various subsystems, to control the engine, transmission, airbags, braking, cruise control, steering, audio systems, power windows, doors, mirror adjustment, climate, parking, and others. Communication among those is essential, and the CANBUS is the standard due to its wide range of safety, economy and convenience features to be implemented using software alone.

tld image.jpg

BOEING 737 MAX

737 Max is a modern aircraft that builds on the 737 model, a successful and popular model launched in 1968, just to go faster and carry more. And for accommodation at airports with less complex facilities, they made it a bit closer to the ground, and this required changing the wings and the engines positions, and hence some sensors based computations. Boeing is blamed for faulty design, and the model got grounded and marked not safe to fly, because two flights went down killing 350 people due to incorrect sensor computations that controls the nose. The fix was only a software update

tld image.jpg

CODE & LANGUAGE

Python, C#, C++, Java, etc., are they truly Computer Languages?. This is misleading, because those languages are in fact for us as humans to communicate and exchange thoughts about how to instruct computers to solve problems, we use special software to translate those ideas to something the machine understand. This is because processors, the brains of computers, understand only sequences of binary codes in the form of streams of 0's and 1's. Intel processors still use the same vocabulary (the x86 instruction set) and yet a new language pops from time to time, simply because it is meant for us and not for the processor. We develop new language for better communication among programmers, and better ways to express our ideas about solving problems, and then invent a compiler that translates it to the fixed language the processor understand

tld image.jpg

TESLA SELF DRIVING

Tesla announced its chip for autonomous driving, “the best chip in the world” is 260 square millimeter piece of silicon, with 6 billion transistors, capable of 6 Trillions operations per second with 72W of power. The autonomous driving unit has two chips

© 2002 − 2025, Dr. Taisir Eldos, Techno

bottom of page