Can Your NPU Run DOOM? Chimera Can.
Is your NPU DOOMed? Quadric's Chimera GPNPU runs every AI model — and a complete DOOM engine. Find out why Quadric is different.
Vendor:
Chips&Media, Inc.
Category:
NPU
Specialized Video Processing NPU IP for SR, NR, Demosaic, AI ISP, Object Detection, Semantic Segmentation
WAVE-N is a high-performance, video-specialized NPU IP designed to deliver real-time, deep learning-based image enhancement for e…
Overview
WAVE-N is a high-performance, video-specialized NPU IP designed to deliver real-time, deep learning-based image enhancement for edge devices. By utilizing a proprietary 'Line-by-Line' processing architecture, it significantly reduces DRAM bandwidth and achieves 4x to 10x faster processing speeds compared to conventional NPUs. Optimized for Super Resolution, AI-ISP, and Noise Reduction, it offers a small silicon footprint and high energy efficiency to provide superior video quality in power-constrained environments.
Key features
- Proprietary Line-by-Line Processing: Minimizes DRAM access, delivering 4x to 10x faster processing speeds compared to conventional NPUs.
- Compact & Efficient Architecture: High MAC utilization with only 4M logic gates for 1024 FP16 MACs.
- Fully Programmable Core: Features a proprietary ISA for instruction-level coding and flexible algorithm implementation.
- Native 16-bit Floating Point Unit: Simplifies network development by eliminating the need for complex fixed-point quantization.
- Versatile AI Video Processing: Optimized for Super Resolution (up to 4K60fps), Noise Reduction, Demosaic and AI-ISP, Object detection, Semantic Segmentation.
Block Diagram
WAVE-N block diagram
Benefits
- Drastic Bandwidth Savings: Reduces system-level power consumption and bus contention through optimized data flow.
- Reduced Time-to-Market: Floating-point support and a SW simulation package (Compiler, Simulator) ensure rapid deployment without precision loss.
- Extreme Area Efficiency: Achieves high performance in a significantly smaller silicon area, ideal for cost-sensitive edge devices.
- Future-Proof Flexibility: A fully programmable engine allows easy updates to the latest CNN models even after silicon production.
Applications
- Super Resolution
- Noise Reduction
- Full AI-ISP
- Object Detection
- Semantic Segmentation
What’s Included?
- Synthesizable RTL
- Pre-trained AI Models for SR, NR, Demosaic
- Software Simulation Package (Compiler/Simulator)
- Evaluation Kit
- Technical Documentation
Files
Note: some files may require an NDA depending on provider policy.
Specifications
Identity
Part Number
WAVE-N
Vendor
Chips&Media, Inc.
Provider
Chips&Media, Inc.
HQ:
Rep of Korea
Chips&Media is a GLOBAL LEADING VIDEO IP (Intellecual Property) Company that develops and designs the professional video IP by implementing video compression algorithm into hardware.
Chips&Media’s video processing engines have been developed for a variety of multimedia application such as Mobile phones, Home Consumer Devices, Wired and Wireless Communications, PMP and Tablet PCs. Lately new IT Products have constantly been launched like Smart Phones, Tablet PC and so on, this makes the digital devices to have a variety of multimedia functions. Therefore, Chips&Media has focused on developing video processing engine, which is adapted to its own hardware structures for high speed, high performance and ultra-low power.
Learn more about NPU IP core
Heterogeneous NPU Data Movement Tax: Intel's Own Slides Tell the Story
At Quadric, we have long argued that heterogeneous NPU designs — those that stitch together multiple specialized fixed-function engines — carry an unavoidable hidden cost: data has to move. A lot. And data movement burns power, adds latency, and creates silicon-area overhead that scales with every new generation of AI models. Now, Intel has made that case for us.
The Upcoming NPU Shakeout
The IP industry is no stranger to boom and bust cycles, and it looks to be at the crest of another wave.
One Instruction Stream, Infinite Possibilities: The Cervell™ Approach to Reinventing the NPU
AI is evolving faster than the chips designed to run it. Models like large language transformers and generative networks are shifting rapidly–while silicon development cycles remain long and rigid. Traditional NPUs, built around proprietary instruction sets and opaque compilers, simply can’t keep up.
Legacy IP Providers Struggle to Solve the NPU Dilemna
Just bolting a matrix accelerator onto existing processor IP leads to long-term challenges.
Can You Rely Upon your NPU Vendor to be Your Customers' Data Science Team?
Frequently asked questions about NPU IP cores
What is Specialized Video Processing NPU IP for SR, NR, Demosaic, AI ISP, Object Detection, Semantic Segmentation?
Specialized Video Processing NPU IP for SR, NR, Demosaic, AI ISP, Object Detection, Semantic Segmentation is a NPU IP core from Chips&Media, Inc. listed on Semi IP Hub.
How should engineers evaluate this NPU?
Engineers should review the overview, key features, supported foundries and nodes, maturity, deliverables, and provider information before shortlisting this NPU IP.
Can this semiconductor IP be compared with similar products?
Yes. Buyers can compare this product with similar semiconductor IP cores or IP families based on category, provider, process options, and structured technical specifications.