Zen 5 dominates with Ryzen 9000 desktop CPUs, AI-focused Strix APUs, 192-core EPYC Turin and the new Instinct AI accelerator.

WhatsApp Group Join Now
Telegram Group Join Now
Instagram Group Join Now

This year's Computex 2024 saw AMD make major industry breakthroughs, with the company's CEO, Dr. Lisa Xu, showcasing the “power” of the next generation of computing.

AMD's Computex 2024 keynote showed the company means business, revealing AI-focused attractions to be at the center

AMD's progress this year hasn't been on par with its past, especially since the firm has been engulfed in AI hype since arch-rival NVIDIA managed to capture the market's attention. Computex 2024 was seen as an opportunity for Team Red to return, and it's safe to say that the firm lived up to our expectations, as it showcased not only its 2nd-generation AM5 platform, but AI Several developments focused on Here's a quick rundown of AMD's Computex 2024 key highlights.

AMD's desktop Ryzen 9000 series has finally been made official, with Zen 5 Magic getting a major upgrade

What's an AMD Keynote without a new desktop CPU lineup? That's exactly what AMD thought before unveiling the new Ryzen 9000 “Granite Ridge” series. The highlight of the SKUs is the redesigned Zen5 architecture, which not only boasts higher performance, but the platform has seen an overall increase in capabilities. Here are some of its upgrades:

  • Improved branch prediction accuracy and latency
  • High throughput with wide pipelines and vectors
  • Deep window size in design for more harmony

The Zen 5 architecture has CCDs based on TSMC's 4nm process, while the IOD (input/output die) uses TSMC's 6nm process. In terms of improvements, the Zen 5 platform offers a 2x increase in areas such as instruction bandwidth, data bandwidth (L2 to L1 and L1 to FP), and AI performance (AI & AVX512 Throughput) for front-end instructions.

Combining such changes, Zen 5 has brought decent generational upgrades with IPC growth of 15% and, in some cases, even reaching 35%, so overall, AMD has made solid gains. .

Moving on to the Ryzen 9000 series SKUs, AMD has brought a total of four different models, one of which is the Ryzen 9 9950X, coming in on the Ryzen 5 9600X. We won't go deep into the specifications as we already have a detailed post about the new models. Still, the maximum configuration in the new models is 16 cores, 32 threads, two Zen 5 CCDs, and one IOD. The initial models are targeted at the high end for now, but as the Zen 5 architecture becomes more mature over time, we'll see a drop in budget variants.

Pricing details about the new models were not disclosed, but AMD says they will be available by this July. The Ryzen 9000 series models are going to have some serious competition in the markets, and we can't wait to try them out once they hit retail. In addition, the firm also debuted the new X870 motherboard chipset for the Granite Ridge family, which you can check out here.

AMD can't seem to let go of the AM4 platform, adding new high-end models with sweet upgrades.

AMD's commitment to its AM4 socket is surprising but interesting. The firm values ​​its old customers, and the new CPU models will further elevate the platform. At Computex 2024, Team Red decided to unveil two new AM4-based processors, the Ryzen 9 5900XT and Ryzen 7 5800XT, which come with bumpy core counts and clock speeds.

Coming to the specifications, the Ryzen 9 5900XT has 16 cores and 32 threads based on the Zen 3-core architecture, an increased 72 MB cache, and a clock speed of up to 4.8 GHz in a TDP design of 105W. The other new AM4 variant, the Ryzen 7 5800XT, has 8 cores, 16 threads, 36 MB cache, and a clock speed bump of +100 MHz to 4.8 GHz. The chip is also rated at 105W TDP. So, when talking about AMD's stance with the new processors, it is only to promote the longevity of AM4 as the performance gap with the counterparts is not that big.

Interestingly, AMD compared its new AM4 CPUs to Intel's 13th generation SKUs, such as the Core i7-13700K and Core i5-13600KF, and the overall performance difference was around 10% – 15%, compared to Team Red's. I was right. AMD is still focusing on a nearly decade-old platform, and its commitment to AM4 is unprecedented.

AMD's new “Strix” AI-focused APUs have been unveiled, combining an interesting naming scheme and the latest Zen 5, RDNA 3.5 and XDNA 2 IPs.

AMD, with this year's Computex, decided to jump on the AI ​​PC bandwagon like no other, with their new Strix APU family targeting the latest AI and compute performance on mobile platforms. Team Red has dubbed the lineup the “Ryzen AI 300” series, and while it's a fresh approach, users will have to stick with processor names these days. The naming scheme evolved in such a way that AMD decided to include the keyword “AI” for industry attention or to maintain relevance in the rapidly growing AI segment.

The new Strix APUs succeeded the Phoenix, and Hawk Point series, and Team Red decided to change significantly at every level, adding new IPs such as Zen 5, RDNA 3.5, and XDNA 2, which The goal was to fight back against rivals, which so far managed to “steal” the spotlight from AMD, notably the likes of Qualcomm's Snapdragon X series.

We've discussed Zen 5 above, but RDNA 3.5 is new. As the name suggests, it's just an update of the original RDNA 3 lineup, so there's not much in terms of a generational upgrade, except for the fact that the graphics platform is much better and a higher number of compute units. Comes with a large number, which adds up. 16 for the flagship part.

The XDNA 2 NPU, however, deserves a separate paragraph. Dedicated AI engines in processors have become the next norm in the markets, and what started with Intel's Meteor Lake has turned into another race in the industry. AMD's XDNA 2 NPU offers up to 50 TOPS of AI power, which is even higher than Qualcomm's Snapdragon X Elite and Intel's Lunar Lake.

Team Red could get a breakthrough here with XDNA 2, and with the integration of the upcoming Ryzen AI 300 series APUs into Microsoft's Copilot+ AI PCs, adoption rates could increase massively.

Moving on from a slew of new APUs released, AMD announced two different models, the Ryzen AI 9 HX 370 and the Ryzen AI 9 365, both of which come with impressive capabilities. Again, we won't dive too deep into the specifications as you can see them in our detailed post here , but as far as the new variants go, they bring serious computing power. However, the interesting fact is that the new Ryzen AI 300 models stand up against market offerings, and here's how the firm compares the APUs to competitors. Overall, AMD seems to have the edge here.

In terms of availability, AMD says its first Ryzen AI 300 “Strix” Copilot+ PCs will start arriving in July 2024, and there are 100 to choose from from ACER, ASUS, HP, Lenovo, and other leading manufacturers. There will be more platforms than MSI

AMD's workstation arsenal sees the addition of new 5th Gen EPYC “Turin” CPUs and the Radeon PRO W7900 Dual Slot GPU.

At this year's Computex, AMD decided to integrate the Zen 5 magic into its latest workstation processors, the EPYC 9005 “Turin” series. These processors include different chips, including compute, cloud, telecommunications, and edge-optimized variants. Team Red hasn't given us lineup details yet, but the firm has revealed how the new Torn lineup will stack up against current Intel offerings. For this, AMD compared the “128-core” variant of its EPYC 9005 lineup of processors to the Intel 5th Gen Xeon Emerald Rapids CPU, the Xeon Platinum 8592+.

The official slides show that Team Red was able to achieve an impressive 2.5x to 5.4x increase in performance across different workloads. Since AMD has decided to keep the veil on its EPYC 9005 models, we're still unaware of specific details, such as base configurations, but we do have a list of preliminary specifications, and we've covered them in our dedicated post here. They are discussed, so be sure to check them out.

In addition to the EPYC Turin, Team Red decided to introduce a new workstation GPU variant under its Radeon Pro series, a new offering that caters to the demands of the AI ​​workstation segment. AMD unveiled the Radeon PRO W7900 workstation model, based on the RDNA 3 GPU architecture with a 2-slot compact design and the same Navi 31 GPU with 96 Compute Units and 6144 cores.

The firm claims that their Radeon PRO W7900 offers 2.01x better performance per dollar in workstation-based workloads than NVIDIA's Ada 6000 GPUs. The firm also unveiled its latest ROCm 6.1.2 suite, which includes more AI frameworks, support for Windows, and data synchronization.

The AMD EPYC Turin “5th Gen” family of servers will debut this year following the formal launch of the Zen 5 and Zen 5C architectures for desktop and client PC platforms such as Granite Ridge and Strix Point. In contrast, the Radeon PRO W7900 will be available by June 19 with an SEP of $3,499 US.

AMD and the AI ​​industry, another attempt to gain interest in the new Instinct AI accelerator market

Team Red's entry into the AI ​​markets was admittedly delayed, and the firm didn't get the same hype as rivals like NVIDIA. The company did forge ahead with options like the Instinct MI350X, which brought great value onboard, but unfortunately, it didn't garner much interest in the market.

However, AMD seems determined, as the firm has revealed an increase in its AI Instinct accelerator roadmap and revealed that it plans to launch a new AI accelerator every year, either a refresh or an entirely new product, similar to NVIDIA's strategy. will present

For Computex, AMD announced the first “refreshed” Instinct MI325X AI accelerator, which uses the same CDNA 3 architecture, 288 GB HBM3E memory, 6 TB/s memory bandwidth, 1.3 PFLOPs of FP16, and 2.6 PFLOPs of FP8 Compute performance. will do The firm also revealed a new “MI350” lineup of AI accelerators, which will be based on a 3nm process node, HBM3E memory, and next-generation CDNA 4 architecture.

In addition, the firm revealed an update to its Instinct AI roadmap, which now includes the latest Instinct MI400 AI accelerator, which will be based on the future yet unannounced CDNA architecture. Finally, AMD provided an update on the recently reported UALink platform, a new high-performance, open, and scalable AI interconnect infrastructure developed by some of the biggest names in the industry.

Well, it looks like AMD has big plans for the future of the AI ​​markets and we got just a glimpse of it at this year's Computex. The firm has shown strong commitment to rapidly develop its Instinct line-up of high-speed to ensure market relevance and competitiveness.

This is from AMD for this year's Computex. The announcements were really long and overwhelming, but this is where Team Red excels. It revolutionizes every segment of the market through its unique and compelling offering. We look forward to seeing how the aforementioned products contribute to the betterment of the industry.

WhatsApp Group Join Now
Telegram Group Join Now
Instagram Group Join Now

Leave a Comment