16 Sources
16 Sources
[1]
AMD takes aim at Intel with new 96-core Threadripper 9000 series CPU
Tom Warren is a senior editor and author of Notepad, who has been covering all things Microsoft, PC, and tech for over 20 years. AMD has announced its latest Zen 5-based Ryzen Threadripper 9000 Series of CPUs at Computex today. The 9000 Series and 9000 WX-Series are built for the demanding workstation market, and the top Threadripper Pro 9995WX will ship with 96 cores and 192 threads. This flagship Threadripper chip is designed for professionals who are working on visual effects, simulations, and AI model development. The Threadripper Pro 9995WX also has up to 384MB of L3 cache and 128 lanes of PCIe Gen 5, making it ideal to pair with multiple GPUs. AMD claims that the Threadripper Pro 9995WX is 2.2x faster than Intel's 60-core Xeon W9-3595X processor in Cinebench 2024 multi-threaded rendering. If you don't need a 96-core CPU, AMD's Threadripper 9000 Series are also targeted at enthusiasts and creators who want workstation-like performance. The Ryzen Threadripper 9980X has 64 cores and 128 threads, a base frequency of 3.2GHz, and 320MB of L3 cache. All of these new Threadripper chips, pro or not, will run at a thermal design power (TDP) of 350 watts and will work (after a BIOS update) on existing motherboards that support the sTR5 socket. Both Threadripper 9000 Series and the Pro WX-Series processors will be available from retailers in July, but AMD isn't announcing pricing just yet. Given the its high-end 7980X Threadripper CPU retailed at $4,999 in 2023, it's fair to say these next-gen equivalents will be around that price.
[2]
Computex 2025 Day Two Wrap-Up: AMD debuts new Radeon and Threadripper chips, PCIe 5.0 SSDs ramp performance
AMD Unleashes Its Performance Might with New Radeon and Threadripper Chips With Jensen Huang's keynote earlier this week, Nvidia had its day in the limelight, but AMD is now in the driver's seat with some pertinent hardware announcements. First up is the Radeon RX 9060 XT, which the company says will launch on June 5 with a $299 price tag for the 8GB VRAM version. If you want to double your VRAM to 16GB, the starting price is $349. As the successor to the RX 7600 XT, the 9060 XT uses a Navi 44 chip built on a TSMC N4P process (as opposed to TSMC N6). The 9060 XT also substantially boosts the number of transistors from 13.3 million to 29.7 million. Leaked benchmarks suggest that the 9060 XT is on average 6 percent faster than Nvidia's competing GeForce RTX 5060 Ti 8GB at 1440p, although we'll have to wait until we get actual hardware in our lab to verify those numbers. Keeping with the GPU side of things, AMD also announced the Navi 48-based Radeon AI Pro R9700. The GPU has 32GB of GDDR6 memory to tackle your most demanding AI tasks and offers 96 TFLOPS of peak FP16 performance. AMD says the launch is scheduled for June, and the usual suspects delivering cards will include ASRock, Asus, Gigabyte, and Sapphire (among others). The last major news from the AMD front pertains to processors, namely, the Threadripper Pro 9000 WX Series. The family starts with the 9945WX, which offers 12 cores, 24 threads, 64MB of L3 cache, and up to a 5.4 GHz maximum boost clock. However, the range-topping chip is the monstrous 9995WX, sporting 96 cores, 192 threads, 384MB of L3 cache, and up to a 5.4 GHz max boost clock. There are also new Threadripper 9000 HEDT processors, with the flagship in the line being the 64-core 9980X. We love the best SSDs here at Tom's Hardware, and plenty of future contenders were on display this week at Computex. Patriot was on hand with two new SSDs aimed at the consumer market. The PV563 is the world's first DRAM-less PCIe 5.0 SSD, but still delivers competitive performance with up to 14,000 MB/s reads and up to 11,500 MB/s writes. There's also the PV593, which includes onboard DRAM. While its read speeds are identical to the PV563, its write speeds are higher at 13,000 MB/s. Phison is a top vendor of SSD controllers, and the company has a new product on the horizon: the E28. This PCIe 5.0 controller should easily shoot to near the top of the benchmark charts with 14.8 GB/s reads and 14 GB/s writes. Random read and write IOPS are pegged at 2,600K and 3,000K, respectively, while consuming 15 percent less power than competing solutions using 6nm nodes. TeamGroup is a big name among SSD vendors, and the company unveiled several new SSD colors, including the T-Force AI-Flow. The T-Force AI-Flow uses thermoelectric (TEC) technology for cooling. TEC is controversial because of its high power requirements and potential for condensation buildup. Still, TeamGroup contends it has worked out these kinks with its artificial intelligence-infused cooling solution. TeamGroup also showed off a crazy-looking triple-SSD cooler called the GD120T. It's just a concept at this time, and can cool three M.2 SSDs simultaneously using a 120 mm AiO solution. Finally, the T-Force DUO360V2 CPU+SSD AiO Cooler can handle your processor and SSD simultaneously.
[3]
Computex 2025 Day Zero Wrap-Up: AI comes into focus, 500 Hz QD-OLEDs blossom, dual-GPU Arc Pro B60 workstation cards inbound
It's hard to believe that it's been a year since we last visited Computex, but here we are again with all the latest and greatest advancements coming from your favorite names in tech. We have Tom's Hardware editors on the ground in Taipei bringing you everything that we see on the show floor (and beyond), which you can peruse in our dedicated Computex 2025 hub. However, we'll give you a quick rundown of all the goodness announced during "Day Zero" of Computex 2025. As has been the case in recent years at the show, artificial intelligence (AI) weighs heavily on the minds of today's tech giants, and they are leaning heavily on incorporating the technology into nearly everything they make. Making Waves with AI When it comes to AI, Nvidia is on the bleeding edge with its accelerators. Nvidia announced its NVLink Fusion program to better use mixed-vendor hardware, allowing its customers and partners to develop custom rack-scale designs. Even though NVLink is a proprietary interlink responsible for facilitating CPU-to-CPU and GPU-to-GPU communications, NVLink Fusion will extend this capability to non-Nvidia hardware from Qualcomm, Fujitsu, Marvell, and MediaTek. In other news, Nvidia announced it would partner with Foxconn to build a powerful AI supercomputer in Taiwan. The new cluster will feature 10,000 Blackwell GPUs and will have a price tag that will easily stretch into the hundreds of millions of dollars range. Display Technology is Big Business at Computex 2025 We've just grown accustomed to Nvidia G-Sync and AMD FreeSync technologies being incorporated into nearly every gaming monitor that launches. However, there's no guarantee that you'll find this variable refresh rate technology in modern televisions. Today, Samsung announced that its 2025 lineup of popular OLED TVs will incorporate both G-Sync and FreeSync Premium Pro support. If that was enough, Samsung is also leaning into the AI side of the equation with a new AI Auto Game Mode, which can automatically scan games to make real-time adjustments to picture and sound settings based on genre. MSI is adding to a growing field of monitor manufacturers embracing 500 Hz+ panels. While we first saw these ultra-fast monitors with IPS or TN panels, MSI is one of the few using QD-OLED technology. The MPG 271QR QD-OLED X50 is a 1440p monitor with DisplayHDR True Black 500, VESA ClearMR 21000, and a 0.03 ms response time. Like Samsung, MSI is uniquely leveraging AI. The company's AI Care Sensor can detect when a person is not sitting in front of the screen so that it can power down to lessen panel wear and save power. Acer is blitzed Computex with a slew of new monitors, including several from its Predator, Nitro, and ProCreator product lines. The headliner from Acer is undoubtedly the Predator X27U F5, which features a 500 Hz 1440p panel (likely the same one used in the MSI above). New GPUs to the Rescue! It wouldn't be Computex without putting a spotlight on new gaming GPUs. In recent years, Intel has decided to rub shoulders with Nvidia and AMD in the discrete GPU market; the latest product from the company is the Arc Pro B-Series, which is aimed at workstations. The Arc Pro B50 is priced at $299 and features 16GB of VRAM, while the $499 Arc Pro B60 will feature 24GB of onboard VRAM and 197 peak TOPS. Taking things to the extreme, Maxsun is slapping two B60 GPUs onto one card called the Arc Pro B60 Dual 48G Turbo. As a result, this dual-GPU monster has 48GB of RAM and could retail for closer to $1,000.
[4]
AMD teases new Threadripper, Radeon AI Pro workstation kit
Computex AMD aims to extend its lead over Intel in the high-end desktop (HEDT) and workstation arenas with its 9000-series Threadripper workstation CPUs teased at Computex this week. Compared to previous Threadrippers (TR), the 9000 series appears to be a fairly sedate update with most of the gains coming from process improvements and the move to AMD's Zen 5 microarchitecture. Launched alongside its 9000-series Ryzen Desktop chips last summer, AMD's Zen 5 architecture delivered a 16 percent uplift in instructions per clock over Zen 4. Since then, AMD has brought the architecture to its Epyc datacenter and now its Threadripper workstation processor families. Like last gen, AMD's latest Threadrippers are available in both Pro and Non-Pro variants with up to 96 cores, 128 lanes of PCIe 5.0, eight channels of DDR5 6400 MT/s memory with ECC support, and boost clocks up to 5.4GHz across the entire lineup. Specifically, AMD's workstation-class TR Pro 9000 parts will be available in SKUs ranging from 12 to 96 cores, while the non-Pro HEDT variants will have a condensed offering ranging from 24 to 64 cores. Historically, AMD's non-Pro Threadrippers have featured fewer memory channels and PCIe lanes than the Pro counterparts. For example, AMD's TR 7000-series parts, launched in 2023, had four fewer channels of DDR5 and just 48 lanes of PCIe 5.0 connectivity compared to eight channels and 128 lanes on the Pro variant. We've asked AMD for clarification on whether this will be the case for TR and TR Pro 9000. Here's a breakdown of Threadripper parts announced today: AMD hasn't provided retail pricing for its latest crop of Threadrippers ahead of its Computex keynote this week. With that said, Threadripper is pretty much uncontested in the HEDT arena so they can charge whatever the market can bear. Intel's latest generation of workstation parts, launched last summer, are based on its now two-year-old Sapphire Rapids platform and top out at 60 cores, 112 lanes of PCIe 5.0, and eight lanes of DDR5 4800 MT/s memory. AMD's 9000-series Threadrippers are slated to make their debut in July. Launching alongside its new Threadrippers, AMD also teased a new workstation GPU which features its new RDNA 4 graphics architecture and 32GB of vRAM on board. The Radeon AI Pro R9700 appears to be based on the chip biz's previously announced RX 9070 XT graphics cards, featuring the same 128 AI accelerators, which in this case are good for 96 teraFLOPS of FP16 and 1531 TOPS of sparse INT4 performance. As we previously discussed, AMD's RDNA 4 architecture brings with it a number of major performance improvements for AI workloads, including support for hardware sparsity and FP8 datatypes. Note: At the time of writing, AMD had not yet shared retail pricing for the R9700. Compared to the Intel Arc Pro B60 we looked at yesterday, the 300-watt R9700 should deliver about twice the performance in dense INT8 workloads, such as AI inferencing, and up to 4x the perf if you're willing to give up some quality by dropping down to 4-bit precision. However, the chip still falls short of Nvidia's top workstation hardware. The GPU giant's RTX Pro 6000 workstation cards announced back at GTC in March boast 96GB of vRAM and up to 4 petaOPS of sparse INT4 performance, making them about 2.6x faster than AMD's latest generation of professional graphics. Perhaps this is why AMD has opted to compare the R9700 against Nvidia's consumer-focused RTX 5080, which it says offers up to 496 percent higher performance when running large AI models. While that may sound impressive, the chart is rather misleading and only really tells us that more vRAM lets you run bigger models without paging out to slower system memory. What AMD appears to have done is cherry-pick models too large to fit into the RTX 5080's 16GB of vRAM, forcing them to spill over into the system's slower DRAM. As you might have already guessed, this absolutely kills performance on memory bandwidth-bound workloads like AI inference. If you reran these same tests with a smaller model that does fit into the 5080's vRAM, like Qwen 3 14B Q4, we expect the results would look quite a bit different. That's because, while both cards feature a 256-bit wide memory controller, the 5080 uses much faster GDDR7 memory modules. Combined, the 5080 achieves 960GB/s of memory bandwidth. AMD hasn't shared memory bandwidth figures for the R9700, but assuming it's close to the 9070 XT's 640GB/s, we estimate the 5080 to be about 1.5x faster. When it comes to AI, fast memory won't do you much good if you don't have enough of it to fit your model. So, AMD isn't entirely without merit; there are just less misleading ways to get this point across. The R9700 is slated to hit shelves alongside AMD's Threadripper CPUs in July. ®
[5]
AMD Threadripper Pro 9000 CPUs debut with 96 cores, 192 threads, and 5.4GHz max boost
Serving tech enthusiasts for over 25 years. TechSpot means tech analysis and advice you can trust. What just happened? AMD has unveiled its Ryzen Threadripper 9000 Series "Shimada Peak" processors, featuring up to 96 Zen 5 cores and 192 threads. Targeted at the workstation and high-end desktop markets, the new lineup can reportedly deliver up to 2.2 times the performance of competing Intel Xeon chips. Designed for the xTR5 socket, the Shimada Peak platform includes two distinct families: the Threadripper Pro 9000 WX and the non-Pro Threadripper 9000 Series. The Pro lineup features six SKUs, headlined by the 9995WX, which offers 96 cores and 192 threads. It comes with a 2.5GHz base clock, a 5.4GHz boost clock, and 384MB of L3 cache. The other SKUs in the Pro lineup include the 9985WX with 64 cores and 128 threads, the 9975WX with 32 cores and 64 threads, the 9965WX with 24 cores and 48 threads, the 9955WX with 16 cores and 32 threads, and the 9945WX with 12 cores and 24 threads. Interestingly, the entry-level 9945WX features the highest base clock at 4.7GHz. All Pro models offer a maximum boost frequency of up to 5.4GHz, support up to 128 PCIe 5.0 lanes, and carry a 350W TDP. They also support 8-channel DDR5-6400 ECC memory. The Threadripper 9000 (non-Pro) family includes three SKUs: the 9980X with 64 cores and 128 threads, the 9970X with 32 cores and 64 threads, and the 9960X with 24 cores and 48 threads. The flagship 9980X features a 3.2GHz base clock, a 5.4GHz boost clock, and 256MB of L3 cache. All models support quad-channel DDR5-6400 memory and carry a 350W TDP. According to AMD, the new processors deliver up to 22 percent better performance in threaded workloads compared to the previous generation. This gain is likely due to the transition from Zen 4 to Zen 5 architecture, as well as the shift from a 5nm to a 4nm process node for the compute dies. The chips also support AVX-512 instructions, which can significantly boost performance in compatible applications. The Threadripper family is built for professional workloads such as AI development, 3D modeling, and ultra-high-resolution video editing. The processors will be available starting July 2025 for DIY builders, as well as in prebuilt workstations from leading OEMs including Dell, HP, Lenovo, and Supermicro. AMD has yet to announce pricing, but based on the MSRPs of the Threadripper 7000 Series, these chips are unlikely to be affordable for most enthusiasts or DIY hobbyists. The previous generation started at $1,400 for the entry-level SKU and topped out at a staggering $10,000 for the flagship model.
[6]
AMD's monstrous 96-core Threadripper 9000 CPUs want to stomp on Intel
AMD's Threadrippers are certainly impressive, even if they're out of reach of most wallets. AMD's monstrous Threadripper CPU is back at Computex 2025 with both the Threadripper Pro 9000 WX series and the consumer-friendly Threadripper 9000 HEDT -- both boasting a Zen 5 performance upgrade within the same socket and 350W thermal envelope. Every time AMD launches a new Threadripper generation, we have to remind readers that these are really workstation processors in desktop PC clothing, suitable for CAD work, editing applications, and so on. But that won't stop you from goggling at the specs, yet again. AMD's lineup feels familiar: the three-part high-end desktop (HEDT) lineup leads off with the 9980X, which includes the familiar 64 cores and 128 threads. The 3.2GHz base clock is identical to the older Threadripper 7980X, too, but the boost speed has increased from 5.1GHz to 5.4GHz. These are also 4nm parts, while the Threadripper 7000 family was built on an older 5nm process. AMD also has six Threadrippers within the Pro 9000 WX series, soaring to a whopping 96 cores and 192 threads. AMD's latest Threadrippers (previously code-named Shimada Peak) arrive a bit later than expected. AMD's Threadripper Pro 5000 series debuted in March 2022, followed by the 7000 series in October 2023. Naturally, the one question AMD isn't answering right now is the big one: price, though the chips will ship in July. And with tariffs and potential demand, that price could be big indeed. Here's how AMD's Threadripper 9000 (HEDT) lineup breaks down. All of the chips use the same sTR5 socket as before, with a 350W thermal envelope. That's a lot of juice. Initially, AMD is being somewhat secretive about the design of the new Threadripper processors. We might assume these new chips are built the same as the 5000 series: each of what AMD calls a "Core Chiplet Die" housed eight cores and 32MB of cache. But AMD's 9970X nudges the core count upwards without adjusting the total L3 cache. An AMD representative said that some models include disabled cores; the 32-core parts (like the 9970X) include 6 cores per CCD, and the 32-core models include 8. Each CCD still includes 32MB of cache. AMD has also advanced the Threadripper platform as a whole. The earlier 7000-series platform supported DDR5-5200 memory; the new chips support up to 8-channel DDR5-6400. AMD didn't say what chipset these new Threadrippers would be using, either. However, both the 7000 and the new 9000-series Threadrippers use PCIe Gen 5 with the same 128 lanes of I/O. AMD isn't yet disclosing any performance information regarding the new Threadrippers, however. If you want to know which Threadrippers workstation vendors will have access to, it will begin with the Threadripper Pro 9995WX, with 96 cores, 192 threads, a 2.5GHz base clock (up to 5.4GHz) and up to 384MB of level-3 cache. We've seen new mobile chips debut at Computex, and new desktop parts, too. But this year? Workstations (and AI) might be what you can expect from Asia's largest tech show this year.
[7]
Computex 2025: live from the world's biggest computing event
Computex 2025 runs from May 20 to May 23 and we'll be reporting live from Taipei The world's biggest computing tech show, Computex 2025, is back - and once again we'll be on the ground in Taipei, Taiwan to bring you all the latest news, reviews and opinions on all the coolest gadgets, laptops, components, and more. Some of the biggest names in computing, including Nvidia, AMD and Qualcomm, will be holding keynote presentations - and in this live blog we'll be reporting direct from the event, so you'll get the very latest updates on what the future holds for laptops, PCs, networking and much more. A key theme of this year's Computex event is Artificial Intelligence (AI), so we expect a lot of companies will take the opportunity to show how putting AI in their products will help improve our lives. Will it finally convince consumers that AI laptops are worth buying? Hopefully, we'll find out this week. So, keep this page bookmarked to find out all the latest consumer and B2B news and launches from Computex 2025.
[8]
AMD unveils new Threadripper CPUs and Radeon GPUs for gamers at Computex 2025
During Computex 2025, Advanced Micro Devices held a press event to introduce its Radeon RX 9060 XT graphics cards and Threadripper CPUs for next-gen gaming. With up to 16 GB GDDR6 memory, the new GPUs unlock new levels of performance while delivering a suite of new and enhanced features for next-gen gaming, the company said. AMD said it is extending its leadership in high-performance computing by unveiling new graphics and workstation products engineered to address the toughest workloads in gaming, content creation, professional industries, and AI development. The company unveiled the AMD Ryzen Threadripper PRO 9000 WX‑Series and AMD Ryzen Threadripper 9000 Series processors built on "Zen 5" architecture -- led by the Ryzen Threadripper PRO 9995WX with 96 cores / 192 threads -- deliver unmatched multi‑threaded performance, leadership efficiency, and enterprise‑grade AMD PRO Technologies, empowering professionals to bring complex visions to life faster. The Radeon AI PRO R9700 is built for AI-powered workstations. It delivers up to 4X more throughput than the previous generation, as well as expanded AMD ROCm on Radeon support, bringing high‑performance GPU acceleration to a broader range of AI and compute workloads for advanced AI development.AMD also strengthens its partnership with ASUS and will introduce the new ASUS Expert P Series Copilot+ PCs, the next-generation commercial PCs bringing AI acceleration to the enterprise. The PCs are powered by up to AMD Ryzen AI PRO 300 Series processors and features AMD PRO Technologies, offering 50+ TOPS of NPU performance for faster and more efficient AI-enhanced productivity as well as enterprise-grade security and manageability for the modern IT environment. Gaming: Radeon RX 9060 XT AMD said the Radeon RX 9060 XT GPUs unlock new levels of performance while delivering a suite of new and enhanced features for next-gen gaming.The Radeon RX 9060 XT features 32 AMD RDNA 4 compute units and doubles raytracing throughput compared to the previous generation. With up to 16GB of GDDR6 memory, these GPUs allow gamers to render the most exciting games of today and tomorrow at max settings.Additionally, the Radeon RX 9060 XT supports FP8 data types and structured sparsity, making it ready for the next-generation of AI-assisted gameplay, creative tools, and generative experiences. The Radeon RX 9060 XT 16GB variant makes a great upgrade for gamers looking to future-proof their systems with a suite of next-gen features that will keep their experiences feeling fresh for years to come. Threadrippers for workstations: Ryzen Threadripper PRO 9000 WX-Series and Ryzen Threadripper 9000 Series The AMD Ryzen Threadripper PRO 9000 WX-Series and AMD Ryzen Threadripper 9000 Series processors set new standards for high-end workstations and enthusiast desktops and empower professionals to bring their complex visions to life faster than ever.Built on the advanced "Zen 5" architecture, both processor families deliver unmatched multi-threaded performance, leading energy efficiency, and extensive platform capabilities, including expansive memory bandwidth. At the top of stack, the AMD Ryzen Threadripper PRO 9995WX offers 96 cores and 192 threads, providing extraordinary compute capacity for the most demanding AEC, M&E, and AI workloads. Every AMD Ryzen Threadripper PRO 9000 WX-Series processor comes equipped with AMD PRO Technologies, offering a robust suite of enterprise grade features including multilayered security, advanced remote manageability, and long-term platform stability -- helping professional users and IT teams achieve new levels of productivity. For enthusiasts, the AMD Ryzen Threadripper 9980X comes equipped with 64 cores and 128 threads, offering DIY customers maximum performance for the most intensive workloads including content creation, software compiling, and local AI training. Radeon AI PRO R9700 The AMD Radeon AI PRO R9700 brings next-generation on-device AI horsepower to professional workstations, pairing second-generation RDNA 4 AI accelerators with a massive 32 GB of graphics memory and PCIe Gen 5 throughput to speed local inference, model finetuning, and complex creative workloads. With up to 4X higher AI-accelerator throughput than the previous generation and full ROCm support on Linux (Windows support coming soon), the R9700 delivers high-performance AI with the control and data privacy of on-prem deployment.Engineered for scalability, Radeon AI PRO R9700 excels in multi-GPU configurations -- expanding memory and compute capacity for large-language model development, real-time rendering, and parallel simulations. AI PC AMD executive Jack Huynh was joined by S.Y. Hsu, Co-CEO of ASUS, for the on-stage announcement of the new ASUS Expert P series lineup of PCs powered by up to AMD Ryzen AI PRO 300 Series processors. With industry leading 50 NPU TOPS and AMD PRO Technologies, the new PCs are engineered to deliver lightning-fast AI compute and seamless productivity for working professionals. Designed to support the next generation of Microsoft Copilot+ experiences, Ryzen AI 300 Series processors deliver a leading peak of 50+ NPU TOPS of AI performance. Commercial systems powered by Ryzen AI 300 Series processors offer enterprises notebooks with the compute power required to support the shift to an AI-enabled workforce.ASUS notebooks powered by Ryzen AI PRO Series processors also come equipped with AMD PRO Technologies, offering enterprises and SMBs with built-in security and manageability features, as well as long-term platform stability.
[9]
AMD's press conference won Computex 2025
AMD's SVP and GM of Computing and Gaming, Jack Hyunh, at Computex 2025 in Taipei. (Image credit: Future | Madeline Ricchiuto) Of all the chipmakers at Computex, AMD had the most announcements for consumers and enthusiasts. But don't get too excited -- Qualcomm, Intel, and Nvidia set a low bar. AMD's Jack Hyunh, SVP and GM of Computing and Graphics, hosted the company's Computex press conference. While AMD's conference was held off-site, it just may have been the most interesting speech of the show. Huynh kicked off with gaming announcements, unveiling the Radeon RX 9060 GPU, new Ryzen AI Pro processors, and improvements to its FSR ("FidelityFX Super Resolution") upscaling and frame generation technology. So, what's new from AMD this summer? To set the stage for Redstone, Hyunh recapped the success of the Project Amethyst partnership with Sony to improve FSR into the fourth iteration. FSR 4 was released earlier this year with the RDNA 4 Radeon RX 9070 GPUs and will be improved with FSR Redstone when it is released later this year. It is also currently supported in 60 games, with more to come. FSR Redstone will be available on all RDNA 4 GPUs, and at launch, 40 games will support Redstone super sampling. Redstone brings machine learning enhancements like advanced radiance caching, machine learning ray regeneration, and machine learning frame generation. These advanced features will become available as an update to AMD's FSR 4 super sampling software later this year. AMD's more budget-friendly Radeon RX 9060 XT GPU features RDNA 4 architecture and offers all the advantages of FSR 4 for just $349 for the 16GB model, and $299 for the 8GB variant. If you're hoping to get your hands on the latest AMD graphics card, the Radeon RX 9060 XT will go on sale on June 5, 2025. While the Radeon RX 9060 XT is a less powerful GPU than the Radeon RX 9070 and 9070 XT, it gets all the same benefits of FSR 4, including frame generation and the upcoming Redstone improvements. Hyunh recapped the Ryzen AI 300 series' successes, though some of AMD's claims are a bit suspect. AMD claims the Ryzen AI Max+ 395 is 15% faster than Apple's M4 Pro 12-core CPU. Based on our own testing, this is true on the Geekbench 6 multicore benchmark, but the M4 Pro wins in more real-world benchmarks, like the PugetBench for Adobe suite. As for new announcements in this segment, AMD has brought the power of the new Ryzen AI Pro 300 chipsets with new devices by Asus. "If you're a fan of AMD technology," Hyunh enthused, "you're a fan of Asus innovation." Asus is expanding the ExpertBook P-series lineup to include AMD CPUs on desktops, mini-PCs, and laptops. It is also offering monitors with Free-Sync compatibility. While AMD's workstation Threadripper Pro 7000 series CPUs are relatively unmatched in terms of raw computing power, the company is upgrading the chip to a new generation on the 4-nanometer Zen 5 architecture. With up to 96 cores and 192 threads, the Threadripper 9000 series is the ultimate chip for those who believe "bigger is better." While Apple's Ultra processors and Intel's Xeon line are in a similar class, neither company has gone as heavily into the over-spec segment as AMD's Threadripper lineup. Of course, there are often diminishing returns by adding so many additional cores and threads, which is why Intel ditched hyperthreading on the Arrow Lake platform. However, there will always be exceptions. And that's the Threadripper gamble. Previous iterations of the chip have often proved AMD correct in the benchmark stakes, as the Threadripper 7980 outperforms even the Epyc 7773X data center CPU in Cinebench 2024. So if you want the most powerful workstation chip to power your special effects business, like AMD partner WetaFX, the Threadripper 9000 Series is worth the excitement. And if you need more endorsement than the facts, WetaFX's Daniel Seah told James Cameron "who is like a god," that he couldn't meet to talk about the next Avatar film because "I have to go to Taipei for AMD." AMD also announced the Radeon AI Pro 9700 GPU to further support the desktop workstation ecosystem. AMD has also announced ROCm support for the Radeon RX 9000 series and AMD Ryzen AI Max APUs, with support for additional hardware and Linux operating systems like RedHat coming later this year. ROCm is also expanding Windows support to Pytorch and ONNX-EP. AMD announced the most new chips and software support at Computex, thanks to its expanded gaming, professional, and workstation lineup.
[10]
Computex 2025: AMD News AT-A-GLANCE
During Computex 2025, AMD rolled out a suite of new hardware aimed at pushing compute limits for a variety of specialists and gurus. One of the headliners was the Radeon RX 9060 XT graphics card. With up to 16 GB of GDDR6 memory, it's designed to handle next-generation gaming workloads, delivering smoother frame rates at higher resolutions while keeping power use in check. The card includes built-in ray-tracing units and faster asynchronous compute capabilities, which means better lighting and shadows in real time, plus support for the latest display standards so you can get more out of modern monitors. On the processor side, AMD showcased updated Threadripper processors based on their Zen 5 design. There are two main families: the standard Threadripper 9000 series and the PRO 9000 WX-series. The star of the show is the Threadripper PRO 9995WX, packing 96 cores and 192 threads. That's a lot of parallel processing power, ideal for tasks like rendering 3D scenes, running large-scale simulations, or crunching big datasets. These chips also bring enterprise-level security features -- think encrypted memory and secure boot -- so IT teams can manage fleets of workstations with confidence. Improvements to the core-to-core interconnect and cache structure have trimmed latency, making it easier for multi-threaded applications to run without choke points. AMD didn't stop at traditional GPU and CPU announcements. They also introduced the Radeon AI PRO R9700, a graphics card built from the ground up for AI workstations. It claims up to four times the throughput of its predecessor, thanks to reworked compute modules and better memory handling. Importantly, it broadens support for the ROCm software ecosystem, so gurus working in machine learning environments -- whether they prefer TensorFlow, PyTorch, or MXNet -- get smoother deployment and performance. For teams focused on AI, faster model training and inference can translate into real productivity gains. Beyond individual components, AMD is partnering more closely with system builders. At Computex, they showed off the ASUS Expert P Series Copilot+ commercial PCs. These all-in-one solutions pack Ryzen AI PRO 300 series processors, which include neural processing units capable of over 50 TOPS. That extra horsepower on the edge means smarter video conferencing, instant image upscaling, and other AI-driven enhancements right on the user's desktop. Meanwhile, enterprise gurus can use centralized management tools, BIOS-level telemetry, and hardware-based cryptographic engines to keep devices secure and up to date.
[11]
AMD Ryzen Threadripper 9000 Series Launched Alongside New Radeon GPUs
AMD Ryzen Threadripper 9000 series features up to 96 CPU cores The Radeon RX 9060 XT GPU comes in 8GB and 16GB configurations The AMD Radeon AI PRO R9700 GPU will go on sale in July AMD on Tuesday unveiled the AMD Ryzen Threadripper 9000 series as its latest high-performance processors at Computex 2025. The top of the line processor, the Ryzen Threadripper PRO 9995WX, is equipped with 96 cores and 5.4GHz a peak clock speed. AMD also launched the Radeon RX 9060 XT and Radeon AI PRO R9700 GPUs with up to 16GB and 32GB of video memory, respectively. The Radeon RX 9060 XT GPU is claimed to offer double the ray tracing throughput compared to its predecessor. Pricing for the new AMD Ryzen Threadripper PRO 9000 WX-Series processors is yet to be announced, and AMD says that they will be available to purchase in the coming months, from OEMs like Dell, HP and Lenovo, and Supermicro. Meanwhile, the Ryzen Threadripper 9000 Series processors will be available via retailers in July. The AMD Radeon RX 9060 XT GPU is available in 8GB and 16GB configurations that are priced at $299 (roughly Rs. 25,600) and $349 (roughly Rs. 29,900), respectively. It will be available to purchase later this year, via GPU makers like Acer, ASRock, ASUS, Gigabyte, PowerColor, Sapphire, Vastarmor, XFX and Yeston. The chipmaker has yet to announce pricing for the AMD Radeon AI PRO R9700 GPU, which will be available from GPU manufacturers in July. AMD has introduced six models in the new Ryzen Threadripper PRO 9000 WX series, with up to 96 cores, up to 192 threads, up to 5.4GHz peak clock speed, and up to 350W TDP. The company has also introduced three new Ryzen Threadripper 9000 series processors with up to 64 cores, up to 5.4GHz peak clock speed, and up to 350W TDP. The company says that the Ryzen Threadripper PRO 9000 WX series (led by the 9995WX model) is aimed at visual effects, simulation, and AI model development. These chips are equipped with up to 384MB L3 cache, and support PCIe 5.0 connectivity (128 lanes). On the other hand, the Ryzen Threadripper 9000 series is designed for content creators and developers working on local AI training. AMD says that these processors can be used for AI training without relying on the cloud, but did not offer details about specific AI models. AMD's new Radeon RX 9060 XT GPU is equipped with up to 16GB of GDDR6 video memory, up to 32 RDNA 4 compute units, and up to 3.13GHz boost clock speed. It is claimed to offer double the ray tracing throughput delivered by the chipmaker's previous generation. The GPU supports features like FSR 4 for higher frame rates and improved image quality, and it enables AI features with support for FP8 data types. Meanwhile, the AMD Radeon AI PRO R9700 also features its second generation AI accelerators and is equipped with 32GB of GDDR6 memory, 64 RDNA compute units, up to 2.92GHz boost clock speed. It can also be used in multi-GPU setups. Support for the ROCm software stack is currently available on Linux, and AMD says Windows will also be supported in the future.
[12]
Watch The AMD Computex 2025 Presentation Live Here!
AMD will be hosting its Computex 2025 presentation in a few hours, which will focus on key products across the gaming, AI PC & Enterprise segments. AMD is known to have electrified keynotes and conferences, unveiling tons of high-performance computing products, and this year's Computex 2025 will be no different, with the company diving into the AI PC realm with an even stronger portfolio powered by the latest and greatest products. The presentation will be delivered by AMD's SVP & GM of the computing and graphics group, who will be joined by several industry leaders and partners. The presentation will be held at 11:00 AM (Taipei Time / UCT +8). The following is what AMD itself has to say about its presentation before we can add our expectations: via AMD As for what to expect, AMD is expected to introduce more products within its gaming lineup, such as the Radeon RX 9000 series. The Radeon RX 9070 series is already out on the market, and the company is expected to shift gears to appeal to more mainstream audiences, and that's where the Radeon RX 9060 series comes in. AMD has already announced the 9060 series, but we can expect more information on the upcoming graphics card, which will expand the RDNA 4 lineup. In addition to the gaming announcements, AMD is also expected to talk a bit more about its AI PC ventures and further expand the portfolio with new updates. Since Enterprise is also mentioned, we can expect some form of an update on EPYC/Instinct lines, though not in a very detailed way. With that said, the AMD Computex 2025 presentation is surely going to be exciting and something that tech enthusiasts should look forward to. AMD Computex 2025
[13]
AMD Reveals Threadripper 9000 CPUs, Radeon AI Pro R9700 GPU For Workstations
The chip designer says its upcoming Ryzen Threadripper 9000 CPUs feature up to 96 cores and AVX-512 support to boost AI workloads. The Radeon AI Pro R9700 graphics card, on the other hand, packs 32 GB of GDDR6 memory to handle larger AI models. AMD is revving up the workstation PC market again, this time with the newly revealed Ryzen Threadripper 9000 CPUs along with what it's calling the Radeon AI Pro R9700 graphics card. Announced at Computex 2025 in Taiwan on Wednesday local time, the Santa Clara, Calif.-based company said the "ultimate workstation processor just got faster" with the Threadripper 9000 series, which will become available in July. [Related: AMD Boosts Spending On Product Road Maps, Go-To-Market For Next 'Growth Arc'] Based on AMD's Zen 5 architecture, the Threadripper 9000 series comes in two variants: the Threadripper Pro 9000 WX-series, which comes with up to 96 cores and 192 threads, and the Threadripper 9000 series for high-end desktops, which sports up to 64 cores and 128 threads. All chips come with a 350-watt thermal design power. While the Threadripper Pro 9000 WX-Series is designed for professional workstations with support for the AMD Pro Technologies, the vanilla Threadripper 9000 chips are meant for prosumers who want maximum CPU horsepower in their desktop PC. All chips from both lines feature a maximum boost frequency of 5.4GHz while the base clock speeds vary from 4.2GHz for a 12-core prosumer model to 2.5GHz for the flagship, 96-core professional model, the latter of which has the largest L3 cache at 384 MB. Like the consumer-focused Ryzen 9000 processors that launched last fall, the new Threadripper chips support AVX-512. This, along with the maximum 128 lanes of PCie Gen 5 connectivity and up to eight channels of DDR-6400 memory, makes these processors "engineered to adapt to the growing needs of AI," according to AMD. Both Threadripper lines are supported by AMD's sTR5 socket, which was introduced alongside the company's Threadripper 7000 series in 2023. AMD's new Radeon AI Pro R9700 graphics card represents a departure in branding from previous Radeon Pro products due to the emphasis on accelerating AI inference workloads. Featuring 32 GB of GDDR6 memory and 128 AI accelerators, the Radeon AI Pro R9700 is capable of up to 96 teraflops of peak 16-bit floating-point math (FP16) and as many as 1,531 tera operations per second of spare 4-bit integer math (INT4). The thermal design power for the graphics card is 300 watts. The product is due out in July and will be supported by several add-in board partners, including ASRock, Asus, Gigabyte, PowerColor and Sapphire. The graphics card provides up to two times better performance than Radeon Pro W7800 from 2023 for the 8-billion-parameter DeepSeek R1 Distill Llama model, according to AMD. Compared to Nvidia's 16-GB GeForce RTX 5080 on average tokens per second, the Radeon AI Pro R9700 is 361 percent faster for the Phi 3.5 Mixture-of-Experts Q4 model and 437 percent faster for the Mistal Small 3.1 24B Instruct 2503 Q8 model, according to AMD. It's also 454 percent faster for the DeepSeek R1 Distill 32B Qwen Q6 model, 447 percent faster for the 32B Qwen 3 32B Q6 model and 496 percent faster for the same model but with a large prompt of over 3,000 tokens, the company added. AMD said up to four Radeon AI Pro R9700 graphics cards can be paired together using PCIe 5 connectivity in the same system to handle larger models such as the 123-billion-parameter Mistral Large Instruct model. The company said the new graphics cards features expanded support for its ROCm on Radeon software stack for advanced AI development.
[14]
AI takes centre stage, again: A look at what's coming at Computex 2025
Computex, hosted in Taipei's every year around the same time, is back and once again brings back the focus on future of computing. One of the largest trade shows, Computex gives us a comprehensive preview of where the PC and computing industries are headed. For the next few days, this will be the home for technology enthusiasts and industry players alike, showcasing cutting-edge innovations and setting the tone for the year ahead. This year, Computex 2025 kicks off earlier than usual, from May 19 to May 23. Monday's "media day" promises a whirlwind of press conferences and previews, headlined by Nvidia CEO Jensen Huang's keynote at the Taipei Music Center. AMD's eagerly anticipated presentation will follow mid-week, ensuring a packed agenda of announcements from leading players like Qualcomm, MediaTek, and Foxconn. Artificial intelligence is once again at the forefront of the Computex agenda. Last year's emphasis on AI's transformative potential spanned data centers, chip innovations, and localised AI applications. This year, we can expect more breakthroughs, particularly when it comes to AI-powered chips and systems. There could be announcement on a collaboration between Nvidia and MediaTek on an Arm-based system-on-a-chip (SoC) for next-generation laptops, potentially branded as the N1 and N1X. Speculation suggests Nvidia's role focuses on advanced GPU capabilities, while MediaTek contributes CPU expertise. If true, this SoC could redefine integrated graphics performance, setting a new standard for efficiency and power. There are also suggestions of an early 2026 launch, with some brands unveiling prototypes at Computex. While the industry has seen a deluge of new silicon over the past year, the shift towards AI-centric features is still in its infancy. Intel's Lunar Lake and AMD's Ryzen AI processors have started to make inroads, but many popular laptop lines are yet to embrace these advancements. Computex 2025 offers an opportunity for manufacturers to refresh their offerings, highlighting AI readiness, Copilot compatibility, and streamlined workflows. We can also expect AI integration into other electronics: from TVs to AI-enhanced monitors. These products promise to improve user experiences, automate tasks, and integrate seamlessly with existing software. However, as always, separating genuine advancements from marketing hyperbole will be key. Intel's presence at Computex 2025 marks a period of transition. The company is navigating significant challenges, including lukewarm reception to its Arrow Lake-S processors and the turbulence has led to leadership changes, with Lip-Bu Tan stepping in as CEO to steer the company forward. The news of major layoff hasn't helped company's reputation. Absent from this year's keynote lineup, Intel's focus appears to be on regrouping and planning its next moves. Updates on its Panther Lake processors are expected, offering a glimpse into the Core Ultra 300 series slated for release in late 2025 or early 2026. These processors promise enhanced performance cores, improved efficiency cores, and integrated NPUs. Also read: NVIDIA CPUs, gaming handhelds, AI in everything, AI smart rings and more at COMPUTEX 2025 Computex 2025 offers a window into the future of computing, where AI and innovative silicon solutions dominate the narrative. As the show progresses, expect breakthroughs, bold announcements, and a few surprises. Stay tuned as the world's foremost tech players unveil their visions, pushing the boundaries of what's possible in a rapidly evolving digital landscape.
[15]
AMD at COMPUTEX 2025: RX 9060 XT, Threadripper 9000, and Radeon PRO
AMD's COMPUTEX 2025 press conference wasn't about subtle nudges or incremental upgrades. It was a triple-barrelled launch, aimed squarely at gamers, creators, and AI professionals. The company revealed the Radeon RX 9060 XT for next-gen gaming, the Ryzen Threadripper 9000 series for extreme desktop workloads, and the Radeon AI PRO R9700 to stake a serious claim in the professional AI compute space. Here's everything you need to know from the press conference. Let's start with what most gamers will care about -- the Radeon RX 9060 XT. This is AMD's volume play for the RDNA 4 architecture, and while it's not a flagship SKU, it's positioned to be the new mainstream hero. Featuring 32 compute units which results in 32 ray accelerators, the 9060 XT claims to double the ray tracing throughput of its predecessor, and it includes support for FP8 data types and structured sparsity, this card is designed with both traditional gaming and AI-enhanced experiences in mind. AMD is clearly betting that future games will increasingly rely on AI workloads -- be it for procedural content generation, real-time upscaling, or dynamic worldbuilding. That's where support for structured sparsity and FP8 comes in. It's also part of a broader push to make GPUs useful beyond just pixels -- something NVIDIA has been doing for years. In terms of memory, the 9060 XT comes in two variants, one with 8 GB of GDDR6 VRAM and another with 16 GB of GDDR6. AMD didn't share detailed frame-rate numbers just yet, but it did say that the 9060 XT is aimed at gamers who want to "max settings" today's titles and be ready for tomorrow's. With DisplayPort 2.1, AV1 encode/decode, and second-gen ray accelerators, it seems like a solid update for anyone still holding on to an RX 6000 series GPU. Availability is expected later this summer, and the 9060 XT 16 GB variant will retail for USD 349 while the 9060 XT 8 GB variant will retail for USD 299. The 9060 XT will only be available as board partner cards. Alongside the RX 9060 XT, AMD also offered a glimpse into the next step for FidelityFX Super Resolution -- tentatively titled FSR Redstone. While it's unclear whether this will be classified as FSR 4.x or earn its own version number entirely, the capabilities signal a significant leap forward for AMD's image reconstruction suite. FSR Redstone introduces three key features that bring AMD closer to feature parity with NVIDIA's DLSS stack: Neural Radiance Caching, Machine Learning Ray Regeneration, and Machine Learning Frame Generation. All three rely on ML inference, something AMD's newer GPU architectures are increasingly equipped to handle thanks to their updated AI accelerators and FP8 capabilities. Neural Radiance Caching is particularly notable, as it enables more physically accurate indirect lighting and reflections -- mirroring what DLSS 3.5's Ray Reconstruction aims to deliver. Similarly, the machine learning-driven Frame Generation looks to offer smoother gameplay experiences, particularly in CPU-bound scenarios. What's impressive is the pace. FSR 4 only launched in March 2025, and yet Redstone's features are already being previewed. AMD also shared adoption stats for FSR 4 -- launched with 30 supported games in March, it's expected to double that number to 60 titles by June 5th. This is a promising sign that developers are responding well to AMD's open approach, and that the ecosystem around FSR is maturing faster than many expected. Threadripper is back, and it's had quite an upgrade. Now based on AMD's Zen 5 architecture, the Ryzen Threadripper 9000 family is going to be available as the consumer-grade 9000 series and the workstation-focused PRO 9000 WX series. The Ryzen Threadripper PRO 9995WX, sporting a jaw-dropping 96 cores and 192 threads which is the same as the Threadripper 7000 series' flagship SKU. It's targeted at professionals working in architecture, media and entertainment, AI research, and anyone else who can benefit from obscene levels of parallelism. These chips support eight channels of DDR5 memory, offer vast PCIe Gen 5 lane counts, and are designed to live in systems with high uptime and remote manageability. The PRO chips also come with AMD PRO Technologies -- meaning enterprise-grade security, remote fleet management, and guaranteed platform stability. These are the kind of features that IT admins obsess over but DIY users probably don't care much about. That's where the Threadripper 9000 non-PRO chips come in. At the top of that stack is the 9980X, which has 64 cores and 128 threads. It drops the enterprise features, but keeps the brute force performance -- making it ideal for software developers, 3D artists, and anyone compiling huge codebases or doing local AI model training on a single box. From what we've seen so far, Zen 5 delivers not only higher performance-per-watt, but also better latency scaling across cores, which should make these chips more responsive under mixed workloads. The third important announcement is the Radeon AI PRO R9700 -- a GPU designed specifically for AI workstations, not gaming. The R9700 packs 128 AI accelerators and 32GB of GDDR6 memory, backed by a PCIe Gen 5 interface and a 300W TDP. It's meant to live in high-end workstations that handle tasks like large-language model development, model fine-tuning, and on-device inference. It features the same chip as the consumer grade graphics card with more bells and whistles switch on. According to AMD, the R9700 offers up to 4X the AI accelerator throughput compared to its predecessor, the Radeon PRO W7800. Benchmarks shown during the keynote revealed significant performance gains on models like DeepSeek R1, Mistral Small, and Qwen 32B, with throughput improvements between 360-496% depending on the workload. AMD even took a swipe at NVIDIA's RTX 5080, claiming better performance across large prompt sizes thanks to its 32GB memory buffer. Importantly, the R9700 supports ROCm on Linux at launch, with Windows support coming soon. It also scales well in multi-GPU configurations -- a nod to the growing need for local compute infrastructure in AI labs and creative studios alike. For industries looking to keep AI workloads on-prem for reasons of privacy, cost, or latency, the Radeon AI PRO R9700 looks to be a compelling option. General availability begins in July 2025. While not the headline, AMD also used its time on stage to talk up the Ryzen AI 300 series and their integration into ASUS' upcoming Expert P series of commercial notebooks. These Copilot+ PCs feature up to 50+ TOPS of NPU performance and full AMD PRO Technologies support. They're aimed at professionals who want local AI capabilities without sacrificing manageability or security. Whether you're a gamer, a creator, a data scientist, or a business IT manager, AMD wants to power your next machine -- be it with GPUs tuned for ray tracing and AI-assisted gameplay, CPUs packing dozens of cores, or workstation cards designed to crunch through massive AI models. NVIDIA may still dominate the AI discourse, but AMD is clearly pushing hard to widen the narrative -- and this year's announcements show it's not just about catching up, but carving its own lane.
[16]
NVIDIA CPUs, gaming handhelds, AI in everything, AI smart rings and more at COMPUTEX 2025
COMPUTEX is where you get to see the PC computing is headed for the years to come. While CES is more about showing off concepts in technology which might be years away before you can get your hands on them, COMPUTEX is all about giving you a more hands-on experience. Yes, product launches happen at both expos but COMPUTEX is more PC-centric while CES is more tech-centric and it covers a much larger base with consumer and enterprise tech thrown into the mix. We are incredibly excited to see what COMPUTEX has in store for us this year. And as is the case with every major tech event, a few leaks do happen ahead of time and we've managed to put together a glimpse of what you can expect at COMPUTEX 2025. NVIDIA, traditionally recognized for its graphics processing units (GPUs), is reportedly set to unveil its inaugural laptop CPU at COMPUTEX 2025. Developed in collaboration with MediaTek, this Arm-based chip, potentially derived from the GB10 AI Superchip, aims to bolster Windows on Arm platforms. Benchmark leaks suggest competitive performance metrics, positioning it as a contender in the evolving laptop processor market. Does this mean we'll get to see Arm-based Windows gaming laptops? Who knows. NVIDIA is also expected to delve deeper into its AI hardware roadmap and reveal more about the upcoming Blackwell Ultra chips as well as the next-gen Rubin platform which will not only have the Vera Rubin GPUs but also the Arm-based CPU that's named Vera. Last year saw several gaming handhelds throw down with the likes of Nintendo's Switch. And while Nintendo's Switch 2 is a class apart, we've seen several manufacturers bring out Windows-based handhelds to give Valve's Steam Deck a run for its money. MSI Claw 8AI+, ASUS ROG Ally X, ZOTAC Zone were just some of the handhelds that came out last year. This year, we expect handhelds to sport more powerful handheld processors such as the AMD Ryzen Z2 Extreme. And heaven forbid if we see something like the AMD Ryzen AI Max enter the handheld segment. That's pretty much the end for everything else. The theme for COMPUTEX 2025 is "AI Next" and it encapsulates the event's focus on embedding AI capabilities into a broad spectrum of consumer electronics. From laptops equipped with neural processing units (NPUs) to wearables offering advanced health analytics, AI's pervasive influence is evident. We might even see AI in places that we least expect them to be in. AI in smart rings? Expected. AI in monitors? That's new. While manufacturers have been touting multi-day battery life in gadgets for quite some time now, it's usually measured by dropping your screen brightness to abysmal levels and using simple applications such as a word processor. Basically, you only get multi-day battery life if you use a laptop as a glorified typewriter. That might change with the kind of advancements we've been seeing with battery-tech. Manufacturers are prioritising energy efficiency, resulting in devices boasting prolonged battery life without compromising performance. This shift addresses consumer demands for longer-lasting gadgets, particularly in the realm of ultraportable laptops and wearables. While we have seen Apple and Samsung tease their smart rings for over a year, COMPUTEX 2025 might be the place where we'll see the rest of the manufacturers dive into the smart ring category. Beyond traditional computing devices, AI's integration into wearables signifies a broader application of intelligent technology. Acer has already gone ahead and announced everything that they will be showcasing this year at COMPUTEX and Acer's FreeSense Ring and AI TransBuds highlight the potential of AI in personal health monitoring and real-time communication. And since these devices don't work in isolation, we'll see a shift towards more interconnected and responsive accessories. Computex 2025 is running from Tuesday, May 20 to Friday, May 23, 2025 and Digit will be on ground covering COMPUTEX 2025. Head to Digit's Computex 2025 Hub to see the latest announcements from COMPUTEX as they happen.
Share
Share
Copy Link
AMD announces its latest Zen 5-based Ryzen Threadripper 9000 Series CPUs, featuring up to 96 cores and 192 threads, aimed at the workstation market and challenging Intel's dominance in high-performance computing.
AMD has unveiled its latest Zen 5-based Ryzen Threadripper 9000 Series CPUs at Computex 2025, marking a significant advancement in high-performance computing for workstations and enthusiasts
1
. The new lineup, which includes both Pro and non-Pro variants, is set to challenge Intel's dominance in the high-end desktop (HEDT) and workstation markets4
.Source: NDTV Gadgets 360
The flagship Threadripper Pro 9995WX boasts an impressive 96 cores and 192 threads, making it a powerhouse for demanding professional workloads
1
. With up to 384MB of L3 cache and 128 lanes of PCIe Gen 5, this CPU is designed to excel in visual effects, simulations, and AI model development1
4
.AMD claims that the Threadripper Pro 9995WX is 2.2x faster than Intel's 60-core Xeon W9-3595X processor in Cinebench 2024 multi-threaded rendering
1
. This performance leap is attributed to the transition from Zen 4 to Zen 5 architecture and the shift from a 5nm to a 4nm process node for the compute dies5
.The Threadripper 9000 Series offers a range of options to cater to different user requirements:
4
5
.1
5
.All models in the lineup support DDR5 memory, with the Pro variants offering 8-channel DDR5-6400 ECC support, while the non-Pro models support quad-channel DDR5-6400
5
.The Threadripper 9000 Series introduces several key improvements:
5
.5
.1
5
.Related Stories
AMD is positioning the Threadripper 9000 Series as a solution for professionals working in AI development, 3D modeling, and ultra-high-resolution video editing
5
. The processors will be available starting July 2025 for DIY builders and in prebuilt workstations from major OEMs like Dell, HP, Lenovo, and Supermicro5
.While pricing details have not been announced, it's expected that these high-end processors will command premium prices, potentially ranging from $1,400 for entry-level models to around $10,000 for the flagship 96-core variant
5
.Source: CRN
Alongside the new CPUs, AMD also teased the Radeon AI Pro R9700, a workstation GPU featuring the RDNA 4 graphics architecture and 32GB of VRAM
4
. This GPU is designed to complement the Threadripper processors in AI-focused workstations, offering 96 teraFLOPS of FP16 and 1531 TOPS of sparse INT4 performance4
.Source: LaptopMag
As the high-performance computing landscape continues to evolve, AMD's latest offerings demonstrate the company's commitment to pushing the boundaries of what's possible in professional-grade hardware, setting the stage for intense competition with Intel in the workstation and HEDT markets.
Summarized by
Navi
[2]
[3]
[4]
1
Business and Economy
2
Technology
3
Business and Economy