OUR CUSTOMERS
Live Streaming Platforms
The insatiable appetite of streamers
They want nonstop, never-ending, high-resolution, non-buffering content everywhere. And they want it NOW!
Fueled by the pandemic, viewers have developed an addiction to continuous streaming and binge viewing. The average American spends more than 13 hours a day using some type of digital media. That’s more than half the day spent streaming, listening, and gaming. To support this accelerating rate of consumption, streaming platform providers must increase transcoder hardware investments, build more data centers, and boost energy consumption.
Global live streaming market will 2x in 4 yrs: $1.23 bil in 2022 - $2.61 bil by 2026.
Market Growth Factors:
5G internet’s low latency and rapid adoption of live streaming via social media are driving factors with OTT streaming in developing countries expected to grow highest
Market Restraining Factors:
Rising concern related to content piracy and protection hinder expansion.
SOURCE: www.researchandmarkets.com
Global live streaming market will 2x in 4 yrs: $1.23 bil in 2022 - $2.61 bil by 2026.
Market Growth Factors:
5G internet’s low latency and rapid adoption of live streaming via social media are driving factors with OTT streaming in developing countries expected to grow highest
Market Restraining Factors:
Rising concern related to content piracy and protection hinder expansion.
SOURCE: www.researchandmarkets.com
Smart VPU
ASIC Video Processing Unit, powered by AI
We created a new category of Smart VPUs to disrupt the previous limitations of video encoding for surveillance platforms. It is unique because it’s dense, cost effective, flexible, and AI-powered making it the smart technology for platforms to cut streaming costs 80% and hyper-scale profitably.
Smart VPU Trifecta
Highest Density
You can install 20 or more VPUs into a single server, minimizing your data center footprint.
Lowest Cost
VPUs deliver more streams per device than CPU or GPU, which translates to the lowest available cost per stream (CAPEX).
Lowest Power
Smart VPUs consume a fraction of the energy of other technologies, minimizing OPEX and carbon emissions.
NEW FEATURE
All upside, no downside.
Dynamic bitrate reduction during live broadcast preserves bandwidth, cost and quality.
Bandwidth is a significant cost for live-streaming productions. Capped CRF is a bitrate control technique that reduces the bitrate of easy-to-encode footage while “capping” the bitrate of hard-to-encode footage so it doesn’t exceed the designated cap. You save bandwidth on the easy sequences and transcode harder sequences at existing bandwidth and quality levels. Smart VPUs watch your bottom line, so you don’t have to.
Mosaic display for conferencing
Smart VPUs efficiently scale in hardware using the FFmpeg Xstack filter, enabling customizable and responsive multi-window mosaic displays with minimum CPU utilization.
Real time feels real
Logan and Quadra products deliver positive user experiences with their low latency capabilities.
Smart VPUs for Live Streaming
ASIC VPUs are engineered for live, high-volume streaming, interactive applications and the metaverse. They’re inexpensive, capable, flexible, exceptionally efficient and have built-in AI. Best of all, VPUs offer a simple upgrade path with drop-in replacement into enterprise-class NVMe servers.
Here’s a glance at all NETINT ASIC VPU products:
 |  |  |  |  |  |  |  | |||
|---|---|---|---|---|---|---|---|---|---|---|
1st Generation – VPU | 2nd Generation – Smart VPUs | |||||||||
Modules | Server | Server | Modules | |||||||
ASIC VPU
T408 | ASIC VPU
T432 | G4 VPU
Logan | G5 Smart VPU
Quadra | Ampere Edition
Quadra | Smart VPU
T1S | Smart VPU
T1U | Smart VPU
T1A | Smart VPU
T2A | ||
Architecture | x86 | x86 | ARM | |||||||
CPU | AMD EPYC 7232P, 8/32/64-core options | AMD EPYC 7232P, 8/32/64 core options | Ampere Altra Max
M96-28, 96-core | |||||||
Server – 1RU | Supermicro 1114S-WN10RT | Supermicro 1114S-WN10RT | Supermicro Mega DC ARS-110M-NR | |||||||
ASIC Codensity chip | G4 | G4 (4x) | G4, T408s (10x) | G5, T1Us (10x) | G5 | G5 | G5 | G5 (2x) | ||
Price | $300 | $1,200 | starting at $7,000 | starting at $19,000 | quoted | $1,500 | $1,500 | $2,750 | ||
Form Factor | U.2 | AIC, HHHL | 1RU Server | 1RU Server | SODIMM-260 | U.2 | AIC, HHHL | AIC, HHHL | ||
Power Consumption | 7W | 27W | ~400W | ~500W | 10-17W | 10-17W | 20W | 40W | ||
Real-time Throughput Up to: | 8x 1080p30 2x 4Kp30 | 32x 1080p30 8x 4Kp30 | 80x 1080p30 20x 4Kp30 | 320x 1080p30 80x 4Kp30 20x 8Kp30 | 32x 1080p30 8x 4Kp30 2x 8Kp30 | 64x 1080p30 16x 4Kp30 4x 8Kp30 | ||||
Latency | 12.8 ms | 12.8 ms | 12.8 ms | 8 ms | 8 ms | 8 ms | 8 ms | 8 ms | ||
Encode Codecs | H.264, HEVC, YUV | H.264, HEVC, JPEG, YUV, AV1 | ||||||||
Decode Codecs | H.264, HEVC, YUV | H.264, HEVC, JPEG, YUV, VP9 | ||||||||
Audio Codecs | n/a | MP3, AAC-LC, HE-AAC | ||||||||
Features | ||||||||||
Artificial Intelligence | n/a | n/a | n/a | 150 TOPS | 15 TOPS* | 15 TOPS* | 18 TOPS | 36 TOPS | ||
New Capped CRF | ||||||||||
Flexible GOP | ||||||||||
Scaling | ||||||||||
Cropping and Padding | ||||||||||
Video Overlay | ||||||||||
YUV / RGB Conversion | ||||||||||
Configurable throughput | n/a | n/a | n/a | |||||||
Feature supported on VPU
Feature runs on host CPU
 |  |  | |
|---|---|---|---|
1st Generation VPU | |||
Modules | Server | ||
ASIC VPU
T408 | ASIC VPU
T432 | G4 VPU
Logan | |
Architecture | x86 | ||
CPU | AMD EPYC 7232P, 8/32/64-core options | ||
Server – 1RU | Supermicro 1114S-WN10RT | ||
ASIC Codensity chip | G4 | G4 (4x) | G4, T408s (10x) |
Price | $300 | $1,200 | starting at $7,000 |
Form Factor | U.2 | AIC, HHHL | 1RU Server |
Power Consumption | 7W | 27W | ~400W |
Real-time Throughput Up to: | 8x 1080p30 2x 4Kp30 | 32x 1080p30 8x 4Kp30 | 80x 1080p30 20x 4Kp30 |
Latency | 12.8 ms | 12.8 ms | 12.8 ms |
Encode Codecs | H.264, HEVC, YUV | ||
Decode Codecs | H.264, HEVC, YUV | ||
Audio Codecs | n/a | ||
Features | |||
Artificial Intelligence | n/a | n/a | n/a |
New Capped CRF | |||
Flexible GOP | |||
Scaling | |||
Cropping and Padding | |||
Video Overlay | |||
YUV / RGB Conversion | |||
Configurable tuning of
quality/throughput | n/a | n/a | n/a |
Feature supported on VPU
Feature runs on host CPU
 | | | |  | ||
|---|---|---|---|---|---|---|
2nd Generation – Smart VPUs | ||||||
Server | Modules | |||||
G5 Smart VPU
Quadra | Ampere Edition
Quadra | Smart VPU
T1S | Smart VPU
T1U | Smart VPU
T1A | Smart VPU
T2A | |
Architecture | x86 | ARM | ||||
Architecture | AMD EPYC 7232P, 8/32/64 core options | Ampre Altra Max M96-28, 96-core | ||||
Architecture | Supermicro 1114S-WN10RT | Supermicro Mega DC ARS-110M-NR | ||||
ASIC Codensity chip | G5, T1Us (10x) | G5 | G5 | G5 | G5 (2x) | |
Price | starting at $19,000 | quoted | $1,500 | $1,500 | $2,750 | |
Form Factor | 1RU Server | SODIMM-260 | U.2 | AIC, HHHL | AIC, HHHL | |
Power Consumption | ~500W | 10-17W | 10-17W | 20W | 40W | |
Real-time Throughput Up to: | 320x 1080p30 80x 4Kp30 20x 8Kp30 | 32x 1080p30 8x 4Kp30 2x 8Kp30 | 64x 1080p30 16x 4Kp30 4x 8Kp30 | |||
Latency | 8 ms | 8 ms | 8 ms | 8 ms | 8 ms | |
Encode Codecs | H.264, HEVC, JPEG, YUV, AV1 | |||||
Decode Codecs | H.264, HEVC, JPEG, YUV, VP9 | |||||
Audio Codecs | MP3, AAC-LC, HE-AAC | |||||
Features | ||||||
Artificial Intelligence | 150 TOPS | 15 TOPS* | 15 TOPS* | 18 TOPS | 36 TOPS | |
New Capped CRF | ||||||
Peer-to-peer DMA | ||||||
Cropping and Padding | ||||||
Scaling | ||||||
Video Overlay | ||||||
Audio decode / encode | ||||||
YUV / RGB Conversion | ||||||
Configurable tuning of
quality/throughput | ||||||
Feature supported on VPU
Feature runs on host CPU
Server work flows
Logan Video Server
Quadra Video Server
Blog
Brochures
Our Customers
NETINT ASIC VPUs lead the streaming technology in these markets also: