Neki Intelovi evangelisti bi rekli da je to naša krivica jer nismo odmah otišli u BIOS i mijenjali voltaže, jer eto, to moramo raditi s proizvodima out of the box, tvorničke postavke su vrlo nesigurne.
To, a i mogli bi se nadovezati da veći dio krivnje ipak spada na AMD, jer do 2017. je sve bilo OK. Iz godine u godine lijepo 350$ quad-core_high-end_gaming_crown CPU i nitko se nije žalil.
I onda odjednom Intel mora nešto kemijati po inače provjerenom receptu samo da dodaju za gaming nebitne jezgre i eto problema di ga ne treba.
__________________
AMD Ryzen 7 Pro 4750G | be quiet! Pure Rock 2 Black | MSI B450 Tomahawk Max II | 32GB G.Skill DDR4-2666 Value | 256GB AData SX8200 Pro NVMe | 2x4TB WD Red Plus | Fractal Define 7 Compact | Corsair CX450M
AMD Ryzen 5 7600 | Noctua NH-U12A chromax.black | MSI MAG B650 Tomahawk Wi-Fi | 128GB Kingston FURY Beast DDR5-5200 | 256GB AData SX8200 Pro NVMe | 2x12TB WD Red Plus | Fractal Define 7 Compact | eVGA 650 B5
Intel says it hasn’t developed a reliable tool yet to detect if a Raptor Lake or Raptor Lake Refresh processor has been affected by the instability issue that can permanently damage a chip. Nevertheless, the company told Tom’s Hardware that it “continues to investigate the possibility of a detection tool” and “will issue an update if one becomes available.” The company also clarified its stance on power limits with the new microcode that can mitigate the crashing errors with chips that haven't experienced instability yet. Several 13th- and 14th-gen Intel Core chips have had instability issues that led to system crashes, but it took a few months for the company to find the root cause. The company finally released its findings in late September, plus a microcode update that is thought to fully address the issue once and for all—at least for processors that a degradation of the clock tree circuit hasn't yet impacted.
Unfortunately, any damage done to your processor cannot be reversed by the microcode update. So, even if you install it as soon as it comes out for your motherboard, you might still experience some instability. Thankfully, Intel released an extra two-year warranty on top of the existing three years for all affected processors from the Intel Core i5, i7, and i9 families, giving them five years of coverage. So, even if you have a problem a couple of years from now, you could still RMA the affected chip and get a new one without the issue (or maybe even a free upgrade). However, it’s still inconvenient to find out that your Intel Core processor is dead or dying as it’s happening. After all, getting a replacement chip via the RMA process could take a few days, so you’ll be without your computer within that time unless you have a backup device. That’s why a tool that would detect if your CPU has suffered degradation and is about to fail would be nice.
Aside from the microcode code updates, the chipmaking giant has advised everyone to stick to the Intel Default Settings as part of the mitigation against the Vmin shift causing the instability issues. However, there are a few caveats. “The microcode update 0x12B (which includes previous microcode updates 0x125 and 0x129), in addition to Intel Default settings, is the full mitigation for the Intel 13th and 14th Generation Desktop Processor Vmin Shift Instability issue,” Intel said. However, Intel says that even though using Intel’s default power profile is still recommended even after applying patches, users are free to increase the PL1 and PL2 power limits beyond the 'recommended values' and still remain in warranty (you can raise them to 4096W). However, users should still follow the safety settings, such as the IccMax and other settings listed at the top of this table, to remain within warranty (table also here).
As such, Intel says that its warranty will still cover processors that use higher power delivery profiles that remain within the IccMax threshold of the particular processor. So, users can still adjust the PL1/PL2 settings of their Intel Core CPUs without losing warranty coverage. However, the company also says, “Users who desire to overclock or utilize higher power delivery settings than recommended can still do so at their own risk as overclocking may void warranty or affect system health.”
Intel may have one less worry on its very full plate: it believes it has now fully addressed its desktop CPU instability woes with 13th and 14th Gen “Raptor Lake” chips. While the company says there’s still no fix for processors that are already damaged, it confirms to me that the “Vmin Shift Instability” is the root cause of the unusual instability and crashes that these desktop CPUs have been demonstrating for many months. And, Intel says, the four sets of mitigations it’s now shipped are the fix for the problem. “Yes, we’re confirming this is the cause and that it is fixed,” Intel spokesperson Thomas Hannaford tells The Verge.
That’s good! It also wasn’t clear until now. As of August 30th, Intel was still telling us that “Vmin Shift Instability” was not necessarily the root cause or the only cause of the unusual crashes we’ve seen with 13th and 14th Gen chips. So last Wednesday, when Intel announced it had diagnosed the root cause of Vmin Shift Instability, it was not clear whether it had diagnosed why these chips were failing, period, or just one potential cause. I spent a week going back and forth with Intel about this, and the company initially provided answers that didn’t quite hit the mark — but it’s now willing to say that Vmin Shift Stability is the root cause.
The short version is that Intel’s chips and their paired motherboards were asking for too much voltage, and that voltage was damaging parts of these chips, prematurely aging them. Intel is still not publicly providing the batch numbers or serial number ranges of Raptor Lake chips that suffered from oxidization during manufacturing — it claims it fixed that issue long ago — and it does not have an update on a tool to let you easily test your chip to see if it’s aged prematurely. Intel is also still not providing an estimate of how many chips are likely to be irreversibly impacted, or how lenient it will be with warranty support.
26.09.2024., 15:06
