performance metrics Archives

Winsage

December 27, 2024

Mythbusters: Windows IoT Enterprise LTSC for Arm

Advantech's infographic addresses myths about Microsoft Windows IoT Enterprise LTSC for Arm, distinguishing it from Windows Home or Professional editions. Myths about the operating system include: 1. Windows on Arm consumes significant power and resources, leading to poor performance. 2. The image size of Windows on Arm is greater than or equal to that of Windows on x86. 3. Desktop mode is unavailable on Windows on Arm. 4. Windows on Arm is a stripped-down version of Windows. 5. Arm architecture may face limitations with complex tasks. 6. All Windows devices are power-hungry, and switching to Arm offers no improvement. Key facts include: - Windows IoT Enterprise LTSC typically occupies about 6GB of storage, with a minimum requirement of 16GB. - Desktop mode is available on Windows on Arm. - Windows on Arm retains core functionalities, though some hardware drivers may have compatibility issues. - Arm processors are generally more efficient than x86 counterparts. Myths regarding applications and workload include: 7. Windows on Arm exclusively supports Arm64 applications. 8. Multimedia applications perform poorly on Windows on Arm devices. 9. Windows Copilot can be extended to all versions of Windows running on Arm. Key facts include: - An emulation system allows x86 applications to run on Arm hardware. - Multimedia performance depends on hardware acceleration, which is supported by major SoC manufacturers. - Cloud-based services like Windows Copilot operate effectively on Windows IoT Enterprise LTSC for Arm, but on-device Copilot+ requires specific hardware capabilities. Myths about licensing and development include: 10. The Windows license fee for x86 is expensive, and the fee for Windows on Arm is similarly high. 11. Windows on Arm has limited support for certain Windows frameworks. 12. Development and debugging on Windows for Arm architecture present challenges requiring extended transition periods. Key facts include: - Windows IoT Enterprise LTSC is priced between 0 and 0, offering a decade of security updates. - Transitioning from x86 to Arm for application development is streamlined, with common programming languages and frameworks remaining compatible. Tools like Visual Studio facilitate the development process.

Winsage

December 21, 2024

Core Ultra 9 285K is faster in Linux than in Windows

Intel's Core Ultra 9 285K shows an average performance increase of 6% on Ubuntu 24.10 compared to Windows 11 with the 24H2 update. Upgrading to Linux kernel version 6.13 provides an additional 2% boost, totaling an 8% improvement. The processor struggles with tasks like rendering, ray tracing, and encoding on Windows, performing nearly double the speed on Ubuntu in encoding tests. The 14900K slightly outperformed the 285K in benchmarks, but no gaming tests were conducted due to limited Linux game ports. The P-State performance governor did not enhance speed as expected, and Intel's Battlemage B580 GPU also performs better on Linux but still lags in gaming compared to Windows.

Winsage

December 18, 2024

Elevate your PC gaming experience with Windows 11 this holiday season

Windows 11 offers new gaming experiences for PC Game Pass subscribers, including titles like Microsoft Flight Simulator 2024 and Indiana Jones and the Great Circle™. Key features include: - Auto HDR: Enhances SDR games to HDR for improved colors and contrasts. - DirectStorage: Reduces load times using NVMe SSDs, optimizing data loading for better performance. - Game Bar Compact Mode: Simplifies navigation on smaller screens, providing quick access to games and performance metrics. - Dynamic Lighting: Allows synchronization and customization of RGB lighting across devices. - Optimizations for Windowed Games: Introduces Variable Rate Refresh (VRR) to reduce tearing and stuttering in windowed modes. - Advanced Hybrid (CASO)/UMA: Improves performance on hybrid laptops with integrated and discrete GPUs. - Wi-Fi 6E and Wi-Fi 7 support: Enhances online gaming with faster speeds and lower latency. - Color filters for colorblind players: Helps colorblind players differentiate colors while improving frame rates. - Automatic Super Resolution for Arm Copilot+ PCs: Upscales lower-resolution content in real-time. - Xbox app on Windows PC: Manages PC gaming content and facilitates cloud gaming and cross-platform connections.

AppWizard

December 18, 2024

Five quick methods to measure the frames/s in PC games

To measure frames per second (FPS) in PC gaming, several methods can be utilized: 1. Steam's Built-in FPS Counter: Activate the FPS counter in Steam's settings under the In-game tab, allowing customization of its position on the screen. 2. Nvidia's GeForce Experience: For Nvidia graphics card users, enable the In-game overlay in the General settings and customize the FPS counter's position in the HUD Layout settings. 3. AMD Radeon Software: AMD users can enable the Overlay feature in the Performance tab to display FPS, frame time, and CPU/GPU utilization, with the option to toggle the overlay during gameplay. 4. In-Game FPS Counters: Many modern games have built-in FPS counters that can be activated through graphics options or specific key combinations. 5. MSI Afterburner: This software can monitor FPS across various graphics card brands. Enable the Rivatuner Statistics Server during installation and activate the Frame rate option in the Monitoring tab. Using the integrated Game Bar in Windows is another option, but it may negatively impact performance.

Winsage

December 16, 2024

“Reinventing the Wheel: New Computing Concepts Using Windows NT Architecture and PowerShell”

The Windows NT architecture continues to support a significant portion of global IT infrastructure, with millions of installations across Windows Server, Windows 10, and Windows 11. It can be leveraged alongside modern PowerShell techniques to create next-generation computing solutions. A secure and distributed file system can be implemented using Windows NT's Distributed File System (DFS) with encryption capabilities through PowerShell scripts. This allows organizations to create a secure, fault-tolerant file-sharing mechanism. PowerShell scripts can also be used to establish a real-time health monitoring dashboard that aggregates data from event logs, system performance counters, and custom triggers, enabling system administrators to swiftly identify failures and monitor system health. Automating patch management can be achieved through PowerShell by utilizing Windows Update Services (WSUS) to streamline the detection of missing updates, apply patches, and audit systems for compliance. PowerShell can enhance identity and access management (IAM) processes by automating compliance and monitoring permissions, ensuring continuous auditing of user access rights and adherence to corporate policies.

Winsage

December 14, 2024

Rising to the TOPS: How will NPUs and Windows AI grow in 2025?

In 2024, Microsoft introduced the "Copilot+ PC" branding for AI-capable laptops, while Apple launched Apple Intelligence. These developments have led to mixed outcomes, with features like real-time translations and on-device speech-to-text being beneficial, but others, such as Windows Recall, still proving their value. By 2025, mainstream developers are expected to integrate on-device AI into Windows applications, influencing consumer purchasing decisions. The term "TOPS" (Trillions of Operations Per Second) is becoming important for evaluating the AI performance of Windows laptops, with a minimum of 40 TOPS required for Microsoft's "Copilot PC+" designation. Qualcomm's Copilot+ PCs reported around 45 TOPS, significantly higher than Intel's 11 TOPS. By the end of 2024, premium Windows laptops are expected to see a three- to four-fold increase in NPU performance compared to 2023 models. Analysts speculate further performance improvements may occur towards the end of 2025. Despite the potential for a two- to three-fold enhancement in on-device AI performance, experts caution against overemphasizing TOPS figures, which may not accurately reflect real-world performance. The lack of a unified API for leveraging NPU capabilities in Windows complicates matters for users of Copilot+ laptops without Qualcomm chips. Although AMD and Intel have released competitive chips, Qualcomm currently holds an advantage with exclusive support for certain applications. Microsoft is promoting its low-level machine learning API (DirectML) and the Windows Copilot Runtime, which may enhance the Copilot+ PC ecosystem. While cloud-based AI solutions remain an option, the cost of these services is expected to rise, making on-device AI more appealing. The introduction of ChatGPT Pro highlights the financial implications of cloud access compared to on-device NPU usage, which incurs no additional costs. The pace of on-device AI adoption in Windows' software ecosystem is anticipated to accelerate in 2025.

Winsage

December 13, 2024

Windows PCs are full of ads. These 9 settings turn off the worst ones

Jon Martindale is a tech enthusiast focused on computer hardware and software, particularly graphics cards, processors, and displays. He analyzes the latest models of graphics cards, evaluates cutting-edge processors for gaming and professional use, and assesses new display technologies for quality and features. Jon also experiments with various AI applications to understand their functionalities and implications. Additionally, he advocates for ergonomic solutions in the workplace, reviewing standing desks that promote better posture and health.

Tech Optimizer

December 13, 2024

Benchmark Amazon RDS for PostgreSQL with Dedicated Log Volumes

Organizations face challenges in maintaining the performance of mission-critical databases under demanding workloads. Amazon Relational Database Service (Amazon RDS) offers a managed, cost-effective, and scalable solution. In PostgreSQL, transactions are logged in a Write-Ahead Log (WAL) buffer before being stored in a WAL segment, which is crucial for data recovery and replication. The performance of these transactions is linked to the storage system for WAL segments. By default, PostgreSQL stores all files, including WAL, in one location, but separating WAL storage can enhance performance. Amazon introduced the Dedicated Log Volume (DLV) feature, allowing users to store WAL segments on a separate volume of 1,024 GiB with 3,000 Provisioned IOPS. DLV features write coalescing, reducing IOPS demand by consolidating log writes into larger operations, improving throughput and efficiency while potentially lowering costs. A benchmarking solution involves executing a script on an Amazon EC2 instance to evaluate performance across two RDS for PostgreSQL instances: one with DLV enabled and one without. To benchmark, prerequisites include launching an Amazon EC2 Linux instance, installing pgbench, and creating two RDS instances—one with DLV and one without. The benchmarking script resets WAL statistics and simulates a workload with 64 concurrent clients processing half a million transactions each. The infrastructure for benchmarking includes an EC2 instance with 16 vCPUs and 128 GiB memory, and two RDS instances with 64 vCPUs and 512 GiB memory each, using IO2 storage with 10,000 PIOPS. Key database configuration parameters impact performance, particularly those related to commit, checkpoint, and WAL settings. The benchmarking script uses pgbench to create a TPC-B-like workload and measures performance metrics such as transactions per second (TPS) and latency. Results indicate that enabling DLV significantly reduces runtime from 77 minutes to 40 minutes (a 92.5% improvement), increases TPS by 91.83%, and reduces latency by 95.56%. Key WAL statistics show improvements in write and sync times, confirming DLV's effectiveness in enhancing database performance. Monitoring techniques for DLV include using Amazon CloudWatch for metrics like IOPS and latency, and PostgreSQL's pg_stat_wal for detailed WAL activity statistics. The DLV feature is compatible with PIOPS storage types and supported in specific PostgreSQL versions across all regions.

AppWizard

December 11, 2024

All-in-one SDK is exactly what your app needs for effective monetization

Users globally spent 1 billion on mobile app purchases in 2023. Over 95% of mobile apps in the global AdTech market utilize SDKs. Custom SDKs enable advanced features like targeted advertising, real-time bidding, and diverse ad formats. SDKs facilitate comprehensive tracking of user engagement and ad performance metrics. Partnering with a programmatic advertising company can simplify ad management processes for publishers. The integration of SDKs is expected to remain in high demand through 2025 and beyond, aligning with regulatory frameworks like GDPR and CCPA.

Winsage

December 6, 2024

Phi Silica, small but mighty on-device SLM

The Applied Sciences team has developed the small language model (SLM) Phi Silica, which enhances power efficiency, inference speed, and memory efficiency for Windows 11 Copilot+ PCs using Snapdragon X Series NPUs. Phi Silica is designed for on-device use and supports multiple languages, featuring a 4k context length. Microsoft announced that developers will have access to the Phi Silica API starting January 2025. The Copilot+ PCs can perform over 40 trillion operations per second, achieving significant performance improvements when connected to the cloud. Phi Silica utilizes a Cyber-EO compliant derivative of Phi-3.5-mini, and its architecture includes components such as a tokenizer, detokenizer, embedding model, transformer block, and language model head. The model's context processing consumes only 4.8mWh of energy on the NPU, with a 56% improvement in power consumption compared to CPU operation. Phi Silica features 4-bit weight quantization for efficiency, rapid time to first token, and high accuracy across languages. The model was developed using QuaRot for low-precision inference, achieving 4-bit quantization with minimal accuracy loss. Techniques like weight sharing and memory-mapped embeddings were employed to optimize memory usage, resulting in a ~60% reduction in memory consumption. Innovations such as a sliding window for context processing and a dynamic KV cache were introduced to expand context length. The model has undergone safety alignment and is subject to Responsible AI assessments and content moderation measures.