The Hot Take: We dearly need this competition...
Intel's Arc Pro GPU journey began with the first-generation Alchemist A-series products, and last year, the company introduced its Battlemage B-Series products. The first generation of products was aimed at the budget segment, offering good perf/$, and while the positioning continues with the Battlemage lineup, it looks like Intel is slightly moving towards a higher-end segment with its Arc Pro B60, B65, and B70 series. This move comes at a time when AI is the talk of the town, and local AI agents are becoming more and more popular. Also, Intel's recent workstation lineup, the Xeon 600 series, makes getting […]Read full article at https://wccftech.com/review/maxsun-intel-arc-pro-b70-32g-graphics-card-hands-on-impressions/
Read the full article
The Hot Take: Ai will consume ALL THE THINGS!!
The massive AI gold rush has a new bottleneck set in its sights, CPUs. But what's driving the demand? We interview industry experts to find out.
Read the full article
The Hot Take: Ai sucking everything up.
AMD's next-gen Radeon GPUs based on the RDNA 5 architecture are still far away from launch as memory shortages grip the PC segment. Memory Shortages & Rising Component Prices Are The Reason Behind AMD's Push Back on Radeon "RDNA 5" Gaming GPUs The Radeon RX 9000 GPUs based on the RDNA 4 graphics architecture launched last year. This year, AMD launched the Radeon RX 9070 GRE for gamers, still based on the RDNA 4 architecture. While the new card aims to provide gamers with a good 1440p solution, the majority of those who have been waiting for next-generation solutions from […]Read full article at https://wccftech.com/amds-radeon-rdna-5-gaming-gpus-slip-to-late-2027-or-early-2028/
Read the full article
The Hot Take: We'll see, they keep teasing it. But i feel they don't feel they have milked that Ai money cow enough to drop new hardware yet.
For almost a year, the RTX 50 Super series has been part of the rumor mill, but with the AI boom snatching production lines, causing memory prices to skyrocket, hype for the lineup had died down. Now, a potential RTX 5060 Super with 12GB of VRAM is apparently in the works, with the 50 Super series as a whole allegedly getting "back on track."
Read the full article
The Hot Take: The more GPUs the merrier, consumers win and get options and price points.
Intel has once again signaled that graphics processors remain a strategic part of its PC business, despite lingering uncertainty surrounding the future of its Arc desktop graphics card lineup.
Read the full article
The Hot Take: Yeah looks like that CZ market is going to get a new avenue of revenue soon enough. Hoping the prices can be kept down is my only concern.
MSI has shown off new cooling and power tricks for future Nvidia RTX graphics cards.
The outfit has said that since the company is not expecting a major gamer GPU launch this year, it used the gap to show what might land on future Nvidia RTX cards and what it will have to come up with to match.
MSI said that it is working on three main areas for its next designs: cooling, power delivery and the PCB. The cooling work includes new fans, heat pipes, thermal pads and baseplates, all dressed up as an advanced thermal architecture. The fan design uses ultra-thin metal blades rather than the usual plastic jobs.
MSI says the seven-blade all-metal design can deliver up to 40 per cent better airflow. The trick is a high-rigidity metal structure packed into a 0.8mm blade, which should resist deformation at higher speeds.
Thinner blades give more effective airflow area, while wider paths reduce resistance during high-speed operation.
MSI is also working on advanced spiral-groove heat pipes. These increase contact area compared with conventional heat pipes, which should help shift heat away from the GPU more efficiently.
The company has added diamond-composite thermal pads for memory modules to improve heat dissipation. There is also a diamond-copper composite baseplate, with a diamond-copper layer stacked between two copper layers.
MSI says this creates a high-conductivity path from the GPU to the heatsink. All these parts come together in a fully integrated cooling module.
One early design was shown as a next-generation Gaming Trio graphics card. MSI displayed it on an existing RTX 5090 32GB GPU, although it remains a prototype rather than a finished retail card.
The final version is expected to arrive with future Nvidia GPUs, so for now, it is more engineering peep show than shopping list.
MSI is bringing its Safeguard technology directly to high-end graphics cards. The feature first appeared on the company’s MPG power supply line.
The same protection and control, handled through software and hardware, will now work from the 16-pin connector on the graphics card. That means users will not need a compatible PSU to get the feature working.
MSI is still validating the technology with more power supplies and is expected to give it a new name. The company is also adding server-grade reusable fuses, called eFuse, to future GPUs. These are designed to protect the card from electrical damage using an internal gate-based reset mechanism.
The company says the fuses are resettable and reusable, with a short-circuit response of about 200ns. That should help long-term reliability, although anyone who has watched 16-pin connector drama will know reassurance is doing plenty of work here.
MSI showed the design on its RTX 5090 SUPRIM Safeguard card. As with the cooling module, these technologies are intended for future graphics cards rather than the shelves today.
It all points to GPU makers preparing for more power, more heat and more expensive lumps of hardware that need not cook themselves while running the next AI-slathered benchmark.
Read the full article
By ckasprzak | TkOut | May 8, 2026 |
Hardware
The Hot Take: Ai & GPUs need that bandwidth.
Rambus wants to address ongoing AI bandwidth problems with its new PCIe 7.0 Switch IP that features Time Division Multiplexing. Rambus Introduces PCIe 7.0 Switch IP with Time Division Multiplexing for Scalable AI and Data Center Infrastructure Press Release: Rambus, a premier chip and silicon IP provider making data faster and safer, today announced the Rambus PCIe 7.0 Switch IP with Time Division Multiplexing (TDM), a new addition to its advanced interconnect IP portfolio designed to address the rapidly escalating bandwidth, latency, and scalability requirements of AI, cloud, and high-performance computing (HPC) systems. As AI infrastructure grows in scale and architectural complexity, […]Read full article at https://wccftech.com/rambus-bets-on-time-division-multiplexing-to-fix-pcie-7-0-for-ai-workloads/
Read the full article
The Hot Take: Getting confusing with weather they are dropping their GPU's.
Benchmarks of Intel's upcoming and fastest gaming handheld SoC, the Arc G3 Extreme, have been leaked, surpassing the Ryzen Z2 by 25%. Intel Packs Its Strongest Battlemage GPU, & 14 CPU Cores Inside the Arc G3 Extreme Gaming Handheld SoC We recently covered Intel's first Arc G3 gaming handheld, which has been listed by online retailers. While the retailer listing was void of details for the SoC itself, we now have more specs and even benchmarks of the upcoming chip & they look phenomenal. Starting with the CPU, the Intel Arc G3 Extreme is going to be the top offering […]Read full article at https://wccftech.com/intels-arc-g3-extreme-handheld-chip-crushes-ryzen-z2-extreme-benchmark-leak/
Read the full article
The Hot Take: Not good, but then again would need to happen if Nvidia is planning on buying intel. We all loose a third wheel market option and we're going to continue having high prices.
New leaks claim that Intel's upcoming Xe3P graphics architecture won't feature any discrete gaming GPUs, and even the next-gen Xe4 lineup isn't confirmed to. Intel is instead prioritizing the datacenter and workstation segments for new graphics IP, and is featuring them on mobile parts.
Read the full article
The Hot Take: The more the merrier!
Elon Musk's SpaceX set to produce 'own GPUs' at its own multi-billion fab as the company warns that it may be unable to purchase all the silicon it needs to meet its goals.
Read the full article