Skip to main content
  1. Home
  2. Computing
  3. News

Video encoding benchmarks on AMD Ryzen chip are good news, if they’re true

Add as a preferred source on Google

When AMD announced the new Ryzen chips, built on its Zen architecture, the excitement from the community was real. However, details on available SKUs, launch windows, and specific performance metrics were lacking. Thanks to purported benchmarks leaked by a Reddit user from a French hardware magazine, we may have a better idea of where the chips fall in terms of real-world performance. While our French is a little rusty, hardware benchmarks know no language, and the graphs shown in the article were averaged from a number of different tests, and shown as a percentage change from the very popular Core i5-6600K.

First up, we have the video encoding and rendering benchmarks. The AMD 2D3151A2M88E soars ahead, hitting 168.7 percent of the aformentioned Core i5. Only the very expensive Core i7-6900K manages to beat out the Ryzen option. AMD chips have historically suffered during CPU heavy rendering tests such as Handbrake, WPrime, Blender, and 3DSMax benchmarks, so it’s exciting to see a chip from the red team take a strong lead.

Recommended Videos

Because the benchmarks don’t mention which video card was used during the gaming tests, it’s a little less clear how the chip will perform while goofing off. The result actually falls between the Core i5-6500 and Core i5-6600K, slipping from the strong lead it took during video benchmarking.

Finally, power consumption tests reveal the chip pulls 93 Watts, just under the promised 95 Watt TDP AMD announced at the launch event. That’s much more in line with the Core i7-6600K, and should allow a little extra overhead for overclocking the chip.

As always, it’s important to take early product leaks and information with a healthy dose of skepticism. There are often a lot of misleading posts and unconfirmed rumors before hardware actually hits the market, and from everything we’ve seen, Ryzen chips are probably still a few months away at least. That being said, the benchmarks look to be in line with what we’d expect, and hope for, from a new AMD processor. With CES right around the corner, we’ll likely know sooner rather than later if these numbers hold any water.

Brad Bourque
Brad Bourque is a native Portlander, devout nerd, and craft beer enthusiast. He studied creative writing at Willamette…
Google’s Gemini can now make complex topics more fun with new interactive learning feature
Google wants to make learning more fun with interactive Gemini lessons that let you explore complex topics visually.
google-gemini-immersive-exploration

Reading about complicated subjects can get a little dry, especially when diagrams and textbook explanations leave you scratching your head. Google wants to make that experience more fun and engaging with Immersive Explorations, a new Gemini feature that turns complex topics into interactive visual experiences. You can explore everything from ancient trade routes to jet engines, uncovering details layer by layer.

https://twitter.com/GeminiApp/status/2115507092660289631

Read more
OpenAI now lets you text your ChatGPT Dot agent to get things done
ChatGPT Dot can now keep an eye on tasks and send you updates through your regular messaging app.
chatgpt-dot-agent-imessage-rcs

OpenAI is bringing its Dot agent into your regular messaging app, so you can assign tasks without opening ChatGPT. The texting feature is rolling out in beta to ChatGPT Pro users in the US, with support for Apple's iMessage and Android's RCS. You can send your Dot text, photos, and voice messages, and also receive proactive texts when it has relevant updates.

https://twitter.com/ChatGPT/status/2115496387613786374?s=19

Read more
Google’s Gemini 4 Carbon is built for complex coding and cybersecurity tasks
Gemini 4 Carbon currently ranks first on Artificial Analysis' updated Intelligence Index.
google-gemini-4-carbon

Google just dropped its latest AI model, Gemini 4 Carbon, designed to tackle complex coding jobs and cybersecurity threats. The company claims it can work through difficult engineering problems for days, finding hidden bugs and developing fixes across large codebases.

Gemini 4 Carbon also takes the top spot on Artificial Analysis' updated Intelligence Index, scoring 59 points, pushing out Claude Opus 5.5, OpenAI's GPT-6 Astra, and even its own predecessor Gemini 4 Argon.

Read more