1Better Code: Contracts Dave Abrahams & Sean Parent What's holding our software together? Can we do better than duct tape and good intentions? 2© 2023 Adobe. All Rights Reserved. Adobe’s Software review© 2023 Adobe. All Rights Reserved. C++ at Adobe! Careers Events Training Videos (STLab Better Code Series!) Blog Posts … 132 developer.adobe.com/cppQ & A 133© 2023 Adobe. All Rights Reserved

Better Code Exploring Validity David Sankel | Principal Scientist CppCon 2023© 2023 Adobe. All Rights Reserved. Adobe’s Software Technology Lab Sean Parent Senior Principal Scientist Manager, Software ”© 2023 Adobe. All Rights Reserved. C++ at Adobe!  Careers  Events  Training Videos (STLab Better Code Series!)  Blog Posts  … developer.adobe.com/cpp© 2023 Adobe. All Rights Reserved. Comments/Questions©

w w w . a t o m o s s p a c e . c o m Newer Isn’t Always Better Investigating Legacy Design Trends and Their Modern Replacements Katherine Rochaatomosspace.com | 3 • Software Engineer at Atomos Space Programming vs Data-Oriented Design Other Potential Evaluationsatomosspace.com | 59 • Newer Isn’t Always Better • Consistently Reevaluate Alternatives • Use Case Determines Usability ConclusionGET TO YOUR PLACE

Better x86 Assembly Generation with Go Michael McLoughlin Gophercon 2019 Uber Advanced Technologies Group Assembly Language Go provides the ability to write functions in assembly language. Assembly

P. All rights reserved. Exceptionally Bad: The story on the misuse of exceptions and how to do better Core C++ June 7, 2023 Pete Muldoon Senior Engineering Lead© 2023 Bloomberg Finance L.P. All rights The story on the misuse of exceptions and how to do better Core C++ June 7, 2023 Pete Muldoon Senior Engineering Lead Exceptions in C++ Better Design Through Analysis of Real World UsageWho Am I Look at Exceptions in both their use and design as seen in real code • Better thinking on the handling of error use cases • Better thinking on the design of an Exception class Warning May be controversialThings

GPUs. It includes support for 8-bit floating point (FP8) precision on Hopper GPUs which provides better training and inference performance with lower memory utilization. Transformer Engine also includes relative positional encoding. The enhancements that were introduced in Transformer-XL help capture better long-term dependencies by attending to tokens from multiple previous segments. Our implementation implementation uses modified model architecture hyperparameters, our modifications were made to achieve better hardware usage and to take advantage of Tensor Cores. This model script is available on GitHub.

Things happening in SG14 Making C++ better for game development & other low-latency domains Patrice Roy, CeFTI, Université de Sherbrooke, Collège Lionel-Groulx Patrice.Roy@USherbrooke.ca, Patrice.Roy@clg systems). • Since pandemic times, SG14 has been brewing a number of proposals meant to make C++ «better for game developers». 6Things happening in SG14 • The abstract of this talk is as follows: • « systems). • Since pandemic times, SG14 has been brewing a number of proposals meant to make C++ «better for game developers». • We will look at the principles behind this effort, the output of this work

(hopefully) know that O(n) is better than O(n2) 4Algorithmic Complexity @ CppCon 2021 Performance is the name of the game You all (hopefully) know that O(n) is better than O(n2) But there is still 2021 Why this talk? Performance is the name of the game You all (hopefully) know that O(n) is better than O(n2) But there is still important stuff that might be overlooked And… 6Algorithmic Complexity 2021 Why this talk? Performance is the name of the game You all (hopefully) know that O(n) is better than O(n2) But there is still important stuff that might be overlooked And… the academic answer

is more predictable according to some artists, compositing-wise. The PS way leads to a cleaner and better ordered layer stack visually wise. Multi Layer Transform or Move You can select multiple layers prepare your work for print, or to do tricks with the LUT docker, so you can diagnose your image better. For example, using the LUT docker to turn the colors grayscale in a separate view, so you can see stylus used to draw directly on the screen. These are better to use than a mouse because it’s more natural to draw with a stylus and generally better for your wrists. With a properly installed tablet stylus

CPU increase translates into 110+ times more transactions per minute (tpmP), making the case for better server utilization when using NVMe SSDs. Mean CPU % utilization (User+Sys) on the Quad Socked Server and Percona Server for SSDs We observed that the NVMe SSDs performed either 200% or 700% better than SATA SSDs, depending on the selected configuration. With that in mind, we then proceeded to Observe that the SATA SSD result is 32+ times better than that of the HDD, and that NVMe SSD throughput is two to five times better than SATA, and up to 51% better than SAS SSD. Throughput in tpmP – Quad