Understanding Anamorphic Lenses' Distortion

"FEM has been actively developed in the v1.1 development cycle. Such progress wouldn’t be possible without the invaluable contributions of marioalexis84 and ickby, as well as other devs occasionally contributing to FEM. There’s still a lot to improve, such as: more CalculiX and Elmer features; mixed and multiple meshes; multistep analyses; further postprocessing usability improvements (including unit handling and color legend issues); a framework for additional specialized solvers." https://lnkd.in/ebZX978r

While exploring more about system design I came across another classic rate limiting algorithm: The Fixed Window Counter Algorithm. Here’s how it works. The timeline is divided into fixed-sized time windows, and each window has a counter. Every request increments the counter by one. Once the counter hits the threshold, new requests are dropped until the next window begins. For example (refer image 1) , imagine a system that allows a maximum of 3 requests per second. If more than 3 requests arrive in that second, the extra ones are dropped. Sounds simple, right? And it is. The algorithm is easy to understand, memory efficient, and resetting quotas at the end of a window works well in certain use cases. But there’s a catch (refer image 2) . A burst of requests at the edges of two windows can let more traffic through than intended. Say a system allows 5 requests per minute. If 5 requests arrive at the end of one minute and another 5 come in immediately at the start of the next, that’s 10 requests within a 60-second span. Twice the allowed limit. This is the major drawback of the Fixed Window Counter Algorithm. This is a reminder that even simple algorithms come with trade-offs. To address this limitation, the Sliding Window Log Algorithm offers a more adaptive solution, which I’ll explore in my next post. #SystemDesign #Algorithm

The following #Matlab and #Simulink webinar on system identification (SysID) and model predictive control (MPC) design using low-cost bi-copter hardware is relevant for Control Systems and GN&C engineers, including students and enthusiasts. Input-output data is collected through design of experiments, then used to identify a model of the bicopter. This model is used as the prediction model in the design of MPC, which is deployed to bi-copter for angular position control. If you work in #Python but are not familiar with the domain of SysID and dynamic systems modeling, there’s the open-source SysIdentPy library. It supports both linear and nonlinear models for MIMO (multi-input, multi-output), MISO (multi-input, single-output), SISO (single-input, single-output), MO (multi-output, i.e, multivariate time series), and SO (single-output, i.e., univariate time series) systems. I use SysIdentPy for MISO forecasting with Nonlinear-ARX, and the results are competitive with (or, on some datasets, even better than) those of Transformer-based or time series foundation models. Moreover, the execution speed of SysIdentPy models, both training and testing, is very fast. For example, some tasks with time series of length 1000 can take less than 10 seconds.

These tech CEO’s are unfortunately right…it’s time to double down on properly calibrating that internal physics engine. It won’t be about coding. It never really was. But it’ll explicitly be about how well you’re able to reasonably accommodate and approximation of a scale model universe in your application architecture planning. Then…work on your communication skills so you don’t keep accidentally building ti-84 calculator game generators. Finally…accept it’s time to become much more declarative about your work. *deep sigh* E no easy o…

When people hear "partition," they often think of land disputes. In chip design, however, partition-based synthesis (PBS) focuses on improving design efficiency. PBS MiM allows for optimizing a logic block once and applying it across all instances, leading to faster runtimes and less redundancy. This method supports efficient and scalable SoC designs. Read this blog to learn morehttps://https://ow.ly/8c3Z50WuUSq

Not all “new software” is new… some are just sequels with better graphics Not every “new” software release is truly new. Sometimes it’s less like a breakthrough… and more like a movie remake. Shiny packaging, same old story. Here’s how to tell the difference: Does it actually solve new problems for you? Is it built on a different foundation (not just a facelift)? Would your day-to-day work change because of it? If the answer is “no” to most of these, you’re looking at a remake, not a revolution. Before upgrading, ask: Is this really new, or just rebranded? #Innovation #TechTrends #Software #DigitalTransformation #Productivity

While doing R&D for an upcoming water FX automotive project, I realized I could finally explain how Surface Tension works in FLIP clearly, and even how to implement simple VEX fill-hole techniques to boost art-direction. Some Houdini users may not fully understand the concept, or might have misunderstood the explanation why the Surface Tension in FLIP solver be like that, only one scale parameter, so here’s my take: Houdini FLIP Surface Tension Demystified. Link to article full explain with file download link: https://lnkd.in/gGRKm3RP One-day dev for this article and sample files, fueled by years of grind and the smart Houdini gang before me. One shot, one hit, low computing cost is my goal. You’ll get a free HIP setup with built-in FLIP tools. If that’s not enough, buy me a beer for the VEX fill-hole version, it’s a simple setup based on position distance, but I’ve already hinted in the article how you can mask mean curvature and push it into advanced setups. For larger scales or more complex motion, you’ll need other methods, but the mindset stays the same. Let me know how your own explorations go!

Layering vs. Composition in Flox Both let you combine environments. But they serve different purposes: ✅ Layering is an imperative, runtime pattern. Perfect for ad hoc debugging, quick experiments, or extending your workflow without rebuilding everything. ✅ Composition is declarative, resolving dependencies as soon as you save your edited manifest. Ideal for team projects, CI/CD, or anywhere predictability matters. Think of it like this: - Layering is flexible → grab what you need, when you need it, on-demand. - Composition is more deterministic → define once, reuse everywhere, get predictable behavior. And the best part? You don’t have to choose. Flox lets you mix and match: use composition for solid, portable stacks, then layer in extras when your workflow demands it. Full guide linked in the comments!

Ever notice how your model suddenly lags like it’s wading through treacle when all you want to do is rotate, pan, or just look at your beautiful design? Here’s a little trick I use to make models faster: hide annotations. Sounds too simple, right? But think about it - every dimension, text note, tag, or marker the software has to render adds to the brain strain of your computer. Hide them, and suddenly your model feels like it’s on roller skates instead of crawling uphill. But I’m curious… what’s your go-to “speed up the model” hack? Do you: Turn off layers you don’t need? Use lightweight representations? Purge old stuff constantly? Or just pray really hard that the computer doesn’t freeze? I want to hear the weird, the genius, the “I can’t believe this actually works” tricks. Let’s start a little discussion - maybe we can all save a few precious seconds (or minutes!) of our lives that would otherwise be spent staring at the spinning wheel of doom. Because honestly, isn’t speeding up the model just the adult version of “please make it go faster”?

🚀 Introducing the PyAEDT Via Design extension Say goodbye to hours of manual via modeling. With the Via Design extension, you can generate a fully parameterized PCB or package via design in just seconds, using a simple configuration file. ✨ Key Highlights: Define stackup, pitch size, and parameters through an easy-to-manage configuration file. Create via variants instantly: ▸ RF via ▸ Differential via pair ▸ PCB standalone, package standalone, or PCB–package transitions Build consistent, reusable, and customizable models without repetitive manual work. 🛠️ Benefits: Cut design time from hours to seconds. Ensure accuracy and repeatability with parameterized definitions. Simplify management with configuration-file-based modeling. 🎥 See the Via Design in action: https://lnkd.in/dTFFKxqf #PyAEDT #Synopsys #Ansys #Python #HFSS #Via #PCB #EDA