close
  • chevron_right

    Early-adopter tax is in full force for the first batch of AM5 motherboards

    news.movim.eu / ArsTechnica · 4 days ago - 17:51 · 1 minute

The MSI MEG X670E Godlike raises interesting questions, like, "could God make a motherboard so expensive that even He could not afford it?"

Enlarge / The MSI MEG X670E Godlike raises interesting questions, like, "could God make a motherboard so expensive that even He could not afford it?" (credit: MSI)

Building a PC around a new processor is expensive at the best of times, and that's triple-true of AMD's new Ryzen 7000 chips. AMD has started with its $300-and-up high-end chips , leaving mid-range options until next year. The CPUs only support DDR5 RAM, which is still more expensive than DDR4 at the same capacities. And the first round of motherboards that include the new AM5 CPU socket are here, and they're pretty expensive.

The cheapest motherboard currently available from the likes of Newegg and Micro Center is the ASRock X670E PG Lightning , which, despite being the least expensive motherboard available, is an X670E board that will support PCIe 5.0 GPUs when they eventually arrive (even the newly announced GeForce RTX 4000-series still uses PCIe 4.0). The motherboard is missing a few features we like to see—no built-in Wi-Fi and Bluetooth, limited audio outputs, relatively small heatsinks for the voltage-regulator modules (VRMs) and other components—but it does have four M.2 SSD slots of varying speeds and plenty of hookups for case fans and front USB ports.

If it's something you care about, the cheapest X670E board with Wi-Fi is also one of ASRock's, the X670E Pro RS , available for $280 at Newegg and Micro Center .

Read 5 remaining paragraphs | Comments

  • chevron_right

    20-year-old Linux workaround is still slowing down AMD systems

    news.movim.eu / ArsTechnica · 5 days ago - 20:51 · 1 minute

A second-generation Epyc server chip from AMD, one that may have been running 2002-era Linux code slowing it down.

Enlarge / A second-generation Epyc server chip from AMD, one that may have been running 2002-era Linux code slowing it down. (credit: Getty Images)

AMD has come a long way since 2002 , but the Linux kernel still treats modern Threadrippers like Athlon-era systems—at least in one potentially lag-inducing respect.

AMD engineer Prateek Nayak recently submitted a patch to Linux's processor idle drivers that would "skip dummy wait for processors based on the Zen microarchitecture." When ACPI support was added to the Linux kernel in 2002—written by Andy Grover, committed by Linus Torvalds—it included a "dummy wait op." The system essentially read data with no purpose other than delaying the next instruction until the CPU could fully stop with the STPCLK# command. This allowed for some power saving and compatibility during the early days of ACPI implementation when some chipsets wouldn't move to an idle state when one would expect it.

But today's Zen-based AMD chips don't need this workaround, and, as Nayak writes, it's hurting them, at least in specific workloads on Linux. Testing with instruction-based sampling (IBS) workloads shows that "a significant amount of time is spent in the dummy op, which incorrectly gets accounted as C-State residency." The CPU, seeing all this low-effort dummy work, can push into deeper, slower C-State, which then makes the CPU take longer to "wake up," especially on jobs that require lots of switching between busy and idle states.

Read 3 remaining paragraphs | Comments

  • chevron_right

    Is Moore’s law actually dead this time? Nvidia seems to think so

    news.movim.eu / ArsTechnica · Thursday, 22 September - 17:08

The green lines make it faster.

Enlarge / The green lines make it faster.

When Nvidia rolled out its new RTX 40-series graphics cards earlier this week , many gamers and industry watchers were a bit shocked at the asking prices the company was putting on its latest top-of-the-line hardware. New heights in raw power also came with new heights as far as MSRP, which falls in the $899 to $1,599 range for the 40-series cards.

When asked about those price increases, Nvidia CEO Jensen Huang told the gathered press to, in effect, get used to it. "Moore's law is dead," Huang said during a Q&A, as reported by Digital Trends . "A 12-inch wafer is a lot more expensive today. The idea that the chip is going to go down in price is a story of the past."

Sorry, how expensive?

In justifying Nvidia's price increases, Huang defended the raw power of the 40-series cards compared with past offerings. "The performance of Nvidia's $899 GPU or $1,599 GPU a year ago, two years ago, at the same price point, our performance with Ada Lovelace is monumentally better," he said. "Off-the-charts better."

Read 10 remaining paragraphs | Comments

  • chevron_right

    Raptor Lake : Intel annonce des chiffres hallucinants pour ses futurs CPU

    news.movim.eu / JournalDuGeek · Tuesday, 13 September - 14:30

fgh-158x105.jpg

Intel parle d'un CPU capable d'atteindre les 8 GHz une fois overclocké.

Raptor Lake : Intel annonce des chiffres hallucinants pour ses futurs CPU

  • chevron_right

    AMD is making laptop CPU model numbers simultaneously less and more confusing

    news.movim.eu / ArsTechnica · Wednesday, 7 September - 17:10 · 1 minute

AMD's new naming scheme for Ryzen 7000 laptop CPUs, broken down.

Enlarge / AMD's new naming scheme for Ryzen 7000 laptop CPUs, broken down. (credit: AMD)

Even for someone who pays a lot of attention to them, processor model numbers can be hard to parse. Ideally, each model number would communicate information about the chip's underlying technology and capabilities, making it easy to quickly tell the difference between a new chip and an old chip or a fast chip and a slow chip. But these model numbers also serve a marketing purpose, both for AMD and for PC makers who want to advertise that their systems are using the latest and greatest chips.

AMD is making an effort to resolve this tension with a revamp of its laptop CPU model numbers , which will go into effect in 2023, when Ryzen 7000-branded laptop processors begin shipping. Here's how AMD breaks it down:

  • The first digit now indicates the year in which the CPU was launched, with Ryzen 7000 CPUs coming in 2023, Ryzen 8000 in 2024, and Ryzen 9000 in 2025.
  • A higher second digit indicates better performance. It will no longer have anything to do with the underlying CPU architecture, as it sometimes does for older chips (Ryzen 5 5500U is Zen 2-based, for example, while Ryzen 5 5600U is Zen 3-based).
  • The third digit will now indicate the CPU architecture being used. A "1" means either the original Zen or Zen+, a "2" denotes Zen 2, a "3" denotes either Zen 3 or Zen 3+, a "4" denotes Zen 4, and a "5" will refer to the as-yet-unannounced Zen 5.
  • The fourth digit is another loose performance indicator. CPUs ending in "0" will be slower, and CPUs ending in "5" will be faster (as of this writing, those are the only two numbers in use).
  • The suffix will denote the TDP of the chip, as it currently does. HX-series chips start at 55 W TDPs, Hs-series chips start at 35 W, U-series chips range from 15 to 28 W, e-series chips are 9 W parts targeted at fanless systems, and C-series chips are just U-series chips in Chromebooks instead of Windows PCs.

AMD also provided this slide, showing how the new model numbers will work in practice for the Ryzen 7000 series (which will encompass at least four distinct CPU architectures, from Zen 2 to Zen 4).

Read 5 remaining paragraphs | Comments

  • chevron_right

    Puissance, date de sortie, prix… tout savoir sur les CPU AMD Ryzen 7000

    news.movim.eu / JournalDuGeek · Tuesday, 30 August - 15:00

7000-158x105.jpg

Tout indique que les troupes de Lisa Su sont encore sur le point de frapper un grand coup. Intel est prévenu !

Puissance, date de sortie, prix… tout savoir sur les CPU AMD Ryzen 7000

  • chevron_right

    AMD will announce Ryzen 7000 CPUs August 29. Here’s everything we know about them

    news.movim.eu / ArsTechnica · Wednesday, 17 August - 17:36 · 1 minute

AMD's Ryzen 7000 chips will be unveiled in late August.

Enlarge / AMD's Ryzen 7000 chips will be unveiled in late August. (credit: AMD)

The final phase of AMD's Ryzen 7000 CPU rollout is approaching, nearly nine months after initially teasing them at CES. The company will livestream their formal unveiling at 7 pm Eastern on Monday, August 29, alongside more details about the AM5 processor socket and 600-series chipsets. Expect to hear more specific news about performance, plus pricing and availability, for the first of what will presumably be many Zen 4-based processors.

AMD has been releasing a steady drip of details about the new CPUs since January, and various leaks and rumors have filled in some of our other knowledge gaps. Let's briefly summarize what we know (and what we think we know).

Faster CPUs, same number of cores

Compared to the nearly 2-year-old Ryzen 5000 processors and the Zen 3 architecture, AMD says that we can expect at least a 15 percent improvement in single-threaded performance , thanks to both clock speed increases and an 8-10 percent increase in instructions-per-clock (IPC). The company also promises performance-per-watt improvements, in part thanks to a new 5 nm manufacturing process (Zen 3 CPUs are 7 nm parts).

Read 11 remaining paragraphs | Comments

  • chevron_right

    Intel’s loss is AMD’s gain as EPYC server CPUs benefit from Intel’s delays

    news.movim.eu / ArsTechnica · Wednesday, 3 August - 16:59 · 1 minute

AMD's EPYC server processors are benefitting from Intel's delays.

Enlarge / AMD's EPYC server processors are benefitting from Intel's delays. (credit: AMD)

Earnings reports for tech companies this quarter have been mixed at best. Apple , Microsoft , Alphabet , and others have managed to eke out a little growth, while the likes of Meta and Nintendo shrank a little, and most companies' projections for the next quarter are also less-than-optimistic.

One company that has been hit particularly hard is Intel , which saw its revenues decline from $19.6 billion in Q2 of 2021 to $15.3 billion in 2022. The company's earnings presentation ( PDF ) showed weakness across the board for a variety of reasons: weaker demand for consumer PCs, money invested in getting the Arc dedicated graphics products off the ground, and "competitive pressure" in the server CPU market.

That competitor is AMD, whose EPYC line of server processors was just one bright spot in a record quarter for the company . Revenue increased from $3.9 billion in Q2 of 2021 to $6.6 billion this year, with $673 million of that additional revenue coming from EPYC processor sales and the company's data center division. This is a big deal for AMD, which had some success with its Opteron server CPUs way back in the mid-2000s but had mostly ceded that ground to Intel throughout the 2010s.

Read 3 remaining paragraphs | Comments