AMD Alveo MA35D media accelerator transcodes up to 32 1080p60 AV1 streams in real-time

AMD ALVEO MA350 AV1 real-time encoding card

AMD Alveo MA35D media accelerator PCIe card is based on a 5nm ASIC capable of transcoding up to 32 Full HD (1080p60) AV1 streams in real-time and designed for low-latency, high-volume interactive streaming applications such as watch parties, live shopping, online auctions, and social streaming. AMD says the Alveo MA35D utilizes a purpose-built VPU to accelerate the entire video pipeline, and the ASIC can also handle up to 8x 4Kp60, or 4x 8Kp30 AV1 streams per card. H.264 and H.265 codecs are also supported, and the company claims its “next-generation AV1 transcoder engines” deliver up to a 52% reduction in bitrate at the same video quality against “an open source x264 veryfast SW model”. AMD Alveo MA350 highlights: Auxiliary CPU – 2x 64-bit quad-core RISC-V to perform control and board management tasks AI Processor – 22 TOPS per card for AI-enabled “smart streaming” for video quality optimization Memory – 16GB […]

Year 2022 in review – Top 10 posts and statistics

CNX Software Happy New Year 2023

It’s the last day of the year and the time to look at some of the highlights of 2022, some traffic statistics from CNX Software website, and speculate on what 2023 may bring us. The semiconductors shortage continued in 2022, but things are looking brighter in 2023 with the full reopening of the world mixed with forecasts of difficult economic times that should keep the demand/supply equation in check. On the Arm processor front the biggest news of the year, at least in this corner of the Internet, was the launch of the Rockchip RK3588 octa-core Cortex-A76/A55 processor together with interesting single board computers that we’ll discuss below. Announced last year, the Amlogic A311D2 octa-core Cortex-A73/A53 was finally made available in a few SBC’s, and we finally got some news about the Amlogic S928X Cortex-A76/A55 SoC showcased in 8K TV boxes, but we have yet to see it in action. […]

AMD unveils low-cost Artix UltraScale+ AU7P FPGA and Zynq UltraScale+ ZU3T MPSoC

AMD UltraScale AU7P FPGA UltraScale ZU3T MPSoC

AMD has added two new low-cost, low-power members to its UltraScale+ family with the Artix UltraScale+ AU7P FPGA and the Zynq UltraScale+ ZU3T MPSoC. Both devices are manufactured with the 16nm FinFET process and offer entry points to the transceiver-based UltraScale+ family with features such as high I/O-to-logic density, UltraRAM, DSP, and more. AMD Artix UltraScale+ AU7P FPGA The new AU7P FPGA is the smallest from the Artix UltraScale+ family with four 12.5Gbps transceivers, up to 82K system logic cells, 216 DSP slices, 4.9 Mbit RAM, and 248 I/Os. It is offered in a 10.5 x 8.5mm InFO package. The company says the chip provides up to 50% lower static power, 20% more I/O-to-logic ratio, and twice as many 3.3V HDIO compared to the AU10P device. The AU7P is designed for space-constrained and/or power-sensitive applications such as medical imaging, machine vision, professional cameras/monitors, and automotive radar/lidar. More details may be […]

AMD launches Ryzen Embedded V3000 Zen3 processors for storage and networking applications

AMD Ryzen Embedded V3000

AMD has just launched the Ryzen Embedded V3000 family of processors with four to eight Zen3 cores, a DDR5 memory interface, twenty PCIe Gen4 lanes, and two 10GbE interfaces that make it ideal for storage and networking applications, especially since there’s no GPU at all. The new embedded processors succeed the Ryzen Embedded V2000 Zen2 family introduced two years ago, and the Ryzen V1000 processors in 2018. For some reason, AMD decided to compare the new V3000 processors against the latter and claims up to 124% greater CPU performance, 50% improved memory transfer rate, twice the number of CPU cores, and improved I/O connectivity. There are five Ryzen Embedded V3000 SKUs at launch with the V3C48, V3C44, V3C18I, V3C16, and V3C14 whose key differences can be found in the table below. The AMD Ryzen Embedded V3C18I “industrial” processor also works in sub-zero temperatures for automotive and industrial applications. Shared specifications […]

How to check TDP (PL1 and PL2 power limits) in Windows and Linux

AMD Ryzen TDP Power Limits Windows 11

A TDP (Thermal Design Power) value in Watts will usually be provided for Intel and AMD processors to help manufacturers design an appropriate thermal solution for a given processor, and it’s often used to estimate power consumption by consumers. But TDP is also often configurable, and manufacturers may decide to increase to decrease the value for higher performance or lower power consumption, so we’ll show you how to check the TDP value, or more exactly PL1 and PL2 power limits in both Windows 11 and Linux (Ubuntu 22.04). Note that TDP is being replaced by PBP (Processor Base Power) in newer processors, with PL1 (Long Duration) corresponding to BPB, and PL2 (Short Duration) to Maximum Turbo Power (MTP), at least on Intel chips. Check the TDP values in Windows 11 You’ll first need to install HWiNFO64 program, then start it leaving all options unticked (default), and go to Control Processor(s) […]

Topton N1 dual-bay NAS features AMD Athlon 3050e CPU, four Ethernet ports

Topton 2 Bay NAS N1

Powered by an AMD Athlon Silver 3050e dual-core processor, Topton N1 (aka TP-N1) NAS comes with two SATA bays for 2.5-inch or 3.5-inch drives, as well as four gigabit Ethernet ports. The system can support up to 32GB RAM, relies on M.2 SSD storage for the OS, and can provide up to 40TB of data storage through the two SATA bays. It is also equipped with an HDMI output port, several USB 3.x/2.0 ports, as well as a 3.5mm audio jack. Topton N1 specifications: SoC – AMD Athlon Silver 3050e dual-core/quad-thread processor @ 1.4 / 2.8 GHz (Turbo) with Radeon Vega 3 graphics @ 1 GHz; TDP: 6W System Memory – Up to 32GB DDR4-2400 RAM via two SO-DIMM slots Storage M.2 2280 socket for a (PCIe 3.0) NVMe SSD 2x SATA III bay for 2.5-inch and 3.5-inch driver for a maximum of 40TB data storage (based on currently available […]

AMD Ryzen Embedded 5000 Series CPUs show up in Micro-ATX motherboard

AMD Ryzen Embedded 5000 Series motherboard

Advantech has announced the AIMB-522 industrial Micro-ATX motherboard with AMD Ryzen Embedded 5000 Series Zen3 processors for AI image processing in automation and surveillance applications. I had never heard about “Ryzen Embedded 5000 Series” processors, but that motherboard comes with a choice of new SKUs such as the 12-core Ryzen 5950E, the 10-core Ryzen 5900E, the 8-core Ryzen 5800E, and the 6-core Ryzen 5600E, as well as existing Ryzen 5000 desktop processors. AIMB-522 Micro-ATX motherboard specifications: Processor (one or the other) Ryzen 5950E 12-core processor @ up to 3.4 GHz with 64MB L3 cache; TDP: 105W Ryzen 5900E 10-core processor @ up to 3.7 GHz with 64MB L3 cache; TDP: 105W Ryzen 5800E 8-core processor @ up to 3.7 GHz with 32MB L3 cache; TDP: 100W Ryzen 5600E 6-core processor @ up to 3.6 GHz with 32MB L3 cache; TDP: 65W Ryzen 5950X 16-core processor @ up to 3.4 GHz […]

Linux 5.19 Release – Main changes, Arm, RISC-V and MIPS architectures

Linux 5.19 release arm risc-v mips

Linus Torvalds has just announced the release of Linux 5.19. It should be the last 5.xx version, with Linux 6.0 coming for the next cycle: So here we are, one week late, and 5.19 is tagged and pushed out. The full shortlog (just from rc8, obviously not all of 5.19) is below, but I can happily report that there is nothing really interesting in there. A lot of random small stuff. In the diffstat, the loongarch updates stand out, as does another batch of the networking sysctl READ_ONCE() annotations to make some of the data race checker code happy. Other than that it’s really just a mixed bag of various odds and ends. On a personal note, the most interesting part here is that I did the release (and am writing this) on an arm64 laptop. It’s something I’ve been waiting for for a _loong_ time, and it’s finally reality, […]

EmbeddedTS embedded systems design