I have seen reports of problems encountered with my new Ryzen 7000X3D, which in some almost sounded literally bruciati danneggiando anche la scheda madre. Nelle scorse ore sono emerse molte nuove informazioni and, fortunately, sows che the origin of the problem has not been identified And che AMD ei brand siano già al lavoro su una soluzione.
AMD has not yet released an official communication and many information arrived from Tom’s Hardwareche avrebbe a sua volte ricevute da diverso fonti interne, che preferiscono remain anonymous.
La prima novità è che The problem if I can present it with i Ryzen 7000X3D if with i Ryzen 7000 “lisci”, It is also chipped with 3D V-Cache sombrano più sensibili. One other detail is that I can take over with any mother schedule: ASUS, MSI, Gigabyte, ASRock and Biostar are all related to the problem in question. While all the information is clear from Tom’s Hardware (and confirm a statement from ASUS to der8auer), the problem arose from a voltaggio del SoC troppo alto, oltre la soglia di sicurezza, And it can be caused by manual overclocking (the voltage of the SoC is reset for a few more memory overclock margins) if the voltage is pre-programmed by the EXPO profile. In some almost, a high voltaggio troppo damage and temperature sensors of the chip and thermal protection mechanisms, Rendo di fatto impossibile al chip proteggersi in caso di surriscaldamento.
Not knowing its own temperature, the processor will not only continue to work as it feels hot, but Continue also to absorb energy, walking further and safer limits; This behavior leads to a surriscaldamento che, alla fine, causa danni come quelli visibili nelle foto, or a dissaldatura come accaduto all’overclocker der8auer.
Igor’s Lab It has been scoped, covering the photo of the damaged chip to the diagrams of the new Ryzen 7000, which may damage it if it creates a correspondence to the VDDCR (CPU Core Power Supply). Subsequent investigations have confirmed that the chip has power from the socket and that its last quest and danni maggiori were in correspondence with the vCore pin, but if they were not also on the CPU SoC, CPU_VDDCR_SOC and CPU VDD MMSIC pin; Qui sono menos evidenti, ma only perché non richiedono tanta corrente quanto i vCore.
The safety limit for the SoC is 1.25V The fonts have not confirmed Tom’s Hardware che spingersi to 1,4V and other non fa che will increase the risk say this problem. Obviously it’s not mathematical, it’s not that carrying your chip at 1.4V or maybe if it’s damaging, it will increase the probability that you’ll hit a lot. Pur essendo oltre il value consigliato, 1.35V sembra invece “sicuro”, ma provatelo a vostro rischio e pericolo.
AMD sows yes to work its a solution, che dovrebbe concretizzarsi in a pad or a tetto massimo al voltaggio, inserted in the firmware. In this way, if the EXPO profile modifies the parameters in the BIOS, it will only be subject to a limit that the finer state was not imposed. Purtroppo, the fonts quoted from Tom’s Hardware stated that AMD can not completely prevent the change of the voltage of the SoC: From the moment that comes the gesture of the VRM, i produttori di schede madri potrebbero permettere agli utenti di superare i limiti imposti da AMD. If you think you won’t care about the first volta that hits anything from the genere, it’s enough to think about the external clock generators that allow overclocking of the Intel “non-K” processors on their Madrid schedule with B660 chipset.
Unfortunately, in response to the fix from AMD, some manufacturers (with ASUS and MSI) have failed in the BIOS that did not solve the problem, limiting the possibility of using hand tools with the voltage.
For the time being, a platform with Ryzen 7000 on board is expected, I would suggest at least one I will control the values of your EXPO profile from the BIOS or with a software such as HWiNFO, for assicurarvi di essere sotto 1,4V; If a DDR5-6000 kit is used, the EXPO profile should operate all’interno dei limiti, ma controllate comunque per sicurezza. If you have an even voltage or higher, please keep it warm, turn off EXPO and use the memory at standard frequency; Lo stesso consiglio vale se avete eseguito dell’overclock manuale sulle RAM e vi siete spinti en alto con il voltaggio.
