Processors - Page 32

The Radeon 890M GPU or 'RDNA 3.5' is the revamped and overhauled integrated graphics found in AMD's Ryzen AI 9 HX 370 mobile chip. YouTube outlet TechEpiphany has posted a new video showcasing Radeon 890M gaming benchmarks (captured via an ASUS Vivobook 16) in ten games, pitting it against NVIDIA's entry-level GeForce GTX 1650 desktop gaming GPU. And, impressively, it comes out on top.

Released in 2019, the GeForce GTX 1650 is an entry-level desktop 75W GPU introduced by NVIDIA that uses the first-generation 12nm Turing 'RTX' architecture without the dedicated AI or RT bits. Sure, it's now pretty old as a desktop gaming GPU, but it's still a viable card. It's great to see integrated graphics reach a point where you can play most games with decent performance and visual fidelity settings on a low-powered APU.

The ten titles include Devil May Cry 5, God of War, Hellblade: Senua's Sacrifice, Cyberpunk 2077, Resident Evil 6, RoboCop: Rogue City, The First Descendant, Forza Horizon 5, Tomb Raider (2013), and Shadow of The Tomb Raider.

The rumors earlier this year about NVIDIA and MediaTek co-developing an Arm-based AI processor continue marching on, with the new Arm-based AI processor for Windows systems expected in the first half of 2025.

In a new report from MyDrivers, we're learning that the new Arm-based SoC co-developed between MediaTek and NVIDIA will debut this year, and aim directly for Qualcomm's new Snapdragon X series SoCs (as they're Arm-based chips). The new AI processor will also compete against AMD and Intel, with their respective AI processors based on the x86 architecture.

NVIDIA and MediaTek's new AI processor will reportedly be made on TSMC's new 3nm process node, based on the Arm architecture, and feature RTX everywhere (of course). The new chip is reportedly in the design phase, with verification and sampling expected to occur in Q4 2024, which leads directly into a 2025 launch.

After issuing an official statement on the ongoing issues surrounding its 13th and 14th Gen CPUs, confirming that a 'microcode' update will "fix" the problem of potential hardware failure, instability, and degradation - many motherboard makers have begun rolling out new firmware with the '0x129 microcode' update as new beta BIOS releases.

With countless motherboards covering Intel's Socket 1700 Series of processors, including a wide range of Z790, B760, B660, Z690, W680, and other boards, ASUS has reached out to us to confirm that in addition to its comprehensive BETA rollout, the company is working hard to ensure that it provides timely UEFI BIOS updates to all affected customers.

ASUS has already updated around 211 different Intel motherboards from its wide range of ROG Maximus, ROG Strix, ProArt, TUF Gaming, Prime, EX, TX Gaming, and Pro lineups with beta BIOS release. ASUS notes that the best way to see if your Intel board has a '0x129 microcode' update available is to check the motherboard's support page.

Earlier, we reported on a new major security flaw that was discovered to be affecting millions of AMD processors, including the popular Ryzen desktop CPU range. The vulnerability gives unwanted users access to and the ability to run malicious code virtually undetected. On the plus side, this 'Sinkclose' vulnerability, as it's called, requires access to the system kernel, so it is rather sophisticated and technical to execute.

Another plus is that AMD has been quick to respond since Wired made the widespread news of the vulnerability public. AMD's security bulletin lists the vulnerabilities and outlines the complete list of AMD EPYC and Ryzen processors that will be given firmware updates.

However, it looks like AMD isn't going to be fixing the exploit for all of its affected CPUs. In a statement provided to Tom's Hardware, which noticed the absence of older Ryzen CPUs from the early days of Zen, the company says that these CPUs exist "outside our software support window." In that they're too old to continue supporting, even after a major security flaw was discovered.

AMD Ryzen and EPYC processors have been exposed to a new vulnerability called "Sinkclose" which could affect millions and millions of AMD processors worldwide.

In a new report from WIRED, we're learning that the new "Sinkclose" vulnerability allows intruders to run malicious code on AMD processors when they're placed in "System Management Mode" which is a sensitive mode that holds crucial firmware files used for operations.

But, hackers will need to insert a piece of code that would provide them with "deep access to an AMD-based PC or server". Once the hackers have this access, they can install a malware called a bootkit, which is undetectable by anti-virus software, which is means to protect your PC.

Intel is still yet to launch its new Core Ultra 200 series "Arrow Lake" CPUs, with teases of its next-gen Nova Lake-S being the successor to Arrow Lake on desktop.

The news on Nova Lake is coming from leaker "Jaykhin" on X, who said that Intel's next-gen consumer desktop platform is Nova Lake, and that Panther Lake will be mobile only. This is a similar play to Intel using Meteor Lake on mobile only, and Arrow Lake on desktop.

Intel's upcoming LGA 1851 platform will launch with the Core Ultra 200 series "Arrow Lake" CPUs, with Nova Lake-S arriving after and fitting into the same socket. Dell teased this back in May through leaks, that Panther Lake laptops will arrive in 206, and Nova Lake CPUs would follow in 2027... which is still years and years away at this point.

A few days I reported that Intel had postponed its Innovation 2024 event into 2025, but the company has just reached out to me to clarify its next-gen Arrow Lake CPUs are still on track for 2024.

The email I received said that Intel saw my piece about Innovation 2024 being postponed, and that "while Innovation was postponed to 2025, our next-gen desktop platform, codenamed Arrow Lake, remains on track for Q4 2024 launch and we will be sharing more details as we get closer to launch".

I see that VideoCardz is reporting that OC3D reached out to Intel for clarification on the Innovation 2024 event being postponed, with an Intel spokesperson telling OC3D: "There are no changes to Intel's launch plans, timing, or product readiness. We are excited about the new product announcements and momentum we have in store, including significant announcements this fall. We'll share more details on our next-gen desktop processor products, codenamed Arrow Lake, later this year".

AMD's new Zen 5-based Ryzen 9000 series "Granite Ridge" desktop CPUs are slowly finding their way out into the wild, and now we've got the first delidded Ryzen 9000 series CPU... which didn't go as expected.

ASUS China manager Tony Yu delidded the AMD Ryzen 9000 series CPU, which involves removing the IHS (integrated heat spreader) so you can use higher-end forms of thermal paste like liquid metal. Not just that, but you can reduce CPU temperatures when stressing out your CPU under stronger workloads, but how did it go?

Tony Yu shared his Ryzen 9000 series CPU delidding story on social media platform Bilibili, where he said he had the "world's first" Zen 5 delidded CPU that was a complete failure. The on-board I/O die was cracked, which means the processor was useless afterwards. Parts of the cracked die were found on the IHS as well as on the chip itself, showing just how big the mistake was for Tony.

AMD's new Ryzen AI 300 series "Strix Point" APUs are floating out into the market inside of new laptops, but Mini-PCs based on Strix Point APUs won't be unleashed until October.

AOOSTAR has said that it will be as early as October until we see Mini-PC systems based on the AMD Ryzen AI 300 series "Strix Point" APUs. The manufacturer already has the motherboard ready to handle Strix Point, but they're still waiting for the APU itself for testing.

An AOOSTAR representative said: "Strix Point's mini computer will not be available until October at the earliest. Although the motherboard is ready, we have not yet provided the CPU for testing".

Intel is reportedly outsourcing more of its chips to TSMC on its new 3nm process node, with 3nm backend design outsourced and increased advanced packaging and testing by Taiwanese companies.

In a new report from UDN, we're hearing that Intel is outsourcing more of its chips to TSMC as well as Taiwanese companies including Egis, Shichixin, and KYEC. UDN reports that none of the Taiwanese companies commented on the rurmors, but Intel has been expanding its outsourcing to TSMC fo a while now.

Intel has been struggling with its own advanced processes and is still a year away from its Intel 18A node coming online, with its GPUs and new CPUs being outsourced to TSMC in the next 12-18 months. Intel's new Core Ultra 200 series "Arrow Lake" CPUs are being outsourced to TSMC, with its next-generation Falcon Shores AI chip also being fabbed at TSMC and will be entering mass production on TSMC's 3nm process node in 2025.

Update: I received an email from Intel PR saying that even with Innovation 2024 being postponed, that its next-gen Arrow Lake CPUs are still on track for a Q4 2024 release. You can read the updated version with the full comments here.

Intel has postponed its upcoming Innovation 2024 event from September, and not just by a few days or weeks, or even months, but "until 2025" the company explained in an email.

The company explained: "after careful consideration, we have made the decision to postpone our Intel-hosted event, Intel Innovation in September, until 2025. For the remainder of 2024, we will continue to host smaller, more targeted events, webinars, hackathons and meetups worldwide through Intel Connection and Intel AI Summit ev ents, as well as have a presence at other industry moments".

Intel and its partners have begun rolling out the microcode update that will fix stability and degradation issues with 13th and 14th Gen Intel Core desktop CPUs. The update arrived slightly ahead of the 'mid-August' window, with new BIOS updates with the 0x129 microcode available for several motherboards.

ASUS motherboard owners were the first to be informed of the microcode's availability, and a BIOS update (still listed as BETA - via the ROG Forum) is available for Z790 motherboards. This covers the company's enthusiast and high-end boards in the ROG MAXIMUS and ROG STRIX line-ups, with ASUS noting that the BIOS update also improves 'system performance.'

MSI has also announced the first batch of BIOS updates for a range of its Intel boards, covering Godlike, Ace, Carbon, and Tomahawk models, with the new 0x129 microcode available now. ASRock is also listing BIOS updates for some of its motherboards.

Hindsight is 20/20, and all that, but a new report at Reuters indicates that the Intel of 2024 could have been a very different company if it hadn't passed on the opportunity to invest in OpenAI. Citing "four people with direct knowledge," Intel discussed acquiring a 15% stake in OpenAI in 2017 and 2018.

Discussions and talks took place over several months, with the report stating that the then-CEO Bob Swan didn't see generative AI as becoming a significant thing in the market anytime soon, so it wouldn't be able to recoup its investment. That investment is rumored to have been worth $1 billion for 15% of the company.

Recent reports state that OpenAI is worth around $80 billion, so you can guess how this decision to turn down OpenAI is being viewed today.

Intel participated in a recent event in China with ASUS, where it discussed issues surrounding the current Raptor Lake generation of desktop CPUs and provided some new information on its next-gen Arrow Lake products, which will arrive later this year.

To get the current-gen stuff out of the way, Intel addressed the upcoming BIOS update for its 13th and 14th Gen Core CPUs, which will arrive in a matter of days. Designed to resolve stability issues and prevent damage to hardware, a representative from the company stated that the fix would not affect the K series' "turbo frequency" or overclocking capabilities.

For those who attribute Intel's recent CPU issues to models like the Cire i9 13900K being power-hungry and expecting the BIOS update to 'nerf' peak performance, this announcement does indicate that the problem isn't simply about the amount of power these CPUs draw. However, when discussing next-gen Arrow Lake, we learn these new CPUs will draw at least 100W less power when maintaining higher frequencies.

Intel has reached quite an important milestone today with its most advanced node -- Intel 18A -- which will be used for client CPUs (Panther Lake) and server CPUs (Clearwater Forest).

The company has now confirmed both Panther Lake and Clearwater Forest have both left the lab and have been successfully turned on, booting into an operating system and hitting the important milestones in less than 6 months after they were taped out.

What does this mean? It means that the silicon development is on track, with Intel confident that production for Panther Lake and Clearwater Forest to start sometime in 2025. The new Intel 18A process node has also been deployed for an external customer, with the first chips for this customer to tape out in 2025 featuring RibbonFET and PowerVIA technologies (which are available to all customers).

Remember the rumors about Intel's Beast Lake processors, silicon designed to be highly performant gaming chips, which sounded all very exciting? Well, that excitement has been well and truly quelled with the news that Beast Lake is apparently canceled.

That's the latest from YouTube leaker Moore's Law is Dead (MLID), who tells us that Beast Lake - which he first aired info on, and apparently represented the "full realization of Intel's Royal Core project" - is now canned.

The flagship for Beast Lake was, in theory, a chip with 4 'extra big' cores (very high performance) plus 32 efficiency cores, a mixture which could have been seriously potent in terms of driving gaming frame rates - or that was certainly the idea.

ASUS is reportedly preparing a new ROG Z13 Flow 2025 compact gaming tablet, featuring AMD's heavily beefed-up Strix Halo APU.

The leaks are coming from thermal specialist and designer Sam Jiun-Wei Hu, who published some documents covering the ROG Flow design for 2025, which is when AMD's new Strix Halo APU will be released. The engineer was not shy in saying that the ASUS ROG Z13 Flow 2025 tablet was made for Strix Halo, with some huge details leaked.

The thermal design schematics show us how much space was saved by replacing the CPU and GPU with a single Strix Halo APU due, with both platforms using LPDDR5X memory, which already has a large vapor chamber covering it. Strix Halo on the new ASUS ROG Z13 Flow 2025 will have an even bigger one. We're to expect an improved cooling design that weighs less, but the overall design is thicker. Engineers have described this as less portable and holdable.

A business that sells a mixture of custom and preconfigured computer systems for desktops, workstations, and servers has released a report revealing its failure rate for CPUs in the wake of the recent Intel fiasco.

The CPU industry is currently on fire, with recent damaging reports about the instability issues with Intel's 14th and 13th Gen CPUs. There has been quite a lot of ambiguity around the fix for the problem, which has been linked to a microcode issue that resulted in excess voltage requests and general instability. Intel has said it's issuing a microcode update in mid-August that will remedy the problem. Unfortunately, Intel still hasn't said how significant the issue's impact is and how many CPUs are affected.

The general concern is the breadth of the impact across SKUs and generations, which was exacerbated by developers coming out and saying they have seen failure rates of up to 100% across 14th and 13th Gen CPUs. Now, Puget Systems, a custom and preconfigured computer system business has published a report revealing its internal failure rate estimations.

AMD's new Zen 5-based Ryzen AI 300 series "Strix Point" APUs are finally out in the wild, with some fresh leaks on the beefier "Strix Halo" APU that will be released in 2025.

The new Strix Halo APU will feature up to a higher 16 cores and 32 threads of Zen 5 processing power, and a massively upgraded RDNA 3.5 GPU system over the current Strix Point APUs that will deliver discrete GPU levels of gaming performance.

In the new leak, we're seeing the new Strix Halo APU huge package: 37.5 x 45.0 mm and using the same FP11 package. The APU itself will be much smaller, with the biggest die on the Strix Halo APU being the RDNA 3.5 GPU that will consume 307mm² die space, while the 2 x Zen 5 CCDs use 66mm² die space, with a total die space (Strix Halo is multi-chiplet) of 439mm².

AMD's new Ryzen AI 300 series "Strix Point" APUs have had their new RDNA 3.5-based Radeon 890M integrated GPU tested at 18W of power, compared to 28W+ power and the Arc GPU on Intel's current Meteor Lake processors.

The gaming tests between the integrated GPUs was conducted by TechEpiphany, comparing the 17W Strix Point APU against the 28W+ Meteor Lake CPU. On the AMD side of things, the Ryzen AI 9 HX 370 "Strix Point" flagship APU was used, with 16 compute units of RDNA 3.5-based GPU with clocks of up to 2900MHz.

It is inside of the ASUS VivoBook S16 OLED which is equipped with 32GB of LPDDR5X-7500 memory, and tuned down to 17W as the APU itself is set to 60W by default. For comparison, the Intel Core Ultra 7 155H "Meteor Lake" processor was used, with 8 Xe-Cores and a 2.25GHz GPU clock (which is 100MHz slower than the Core Ultra 9 185H integrated GPU). This chip also featured 32GB of RAM, but LPDDR5X-6400 (slightly slower).