Lesson from the tale: Avoid purchasing from Asus. Issues with data integrity and 12900K.
Lesson from the tale: Avoid purchasing from Asus. Issues with data integrity and 12900K.
Well I guess I would have to disagree with you on this point. We can all see that getting a refund after owning a product for 2 years then it breaks under warranty doesn't make sense. So they have to cut off the time frame you can get a refund and most companies do 30 days as a standard. This is enough time to reasonably see if you like the product or not and if you have any issues with it. After that you can still get a replacement for the part which for most people is enough. I mean parts fail and we have warranties for that so we can get a replacement part be it them repairing the original part or a totally new part. Also a product failing early on is pretty typical as that is usually when they would do so. They either break early or start to fail around the time when wear and tear gets to them and that is usually after the warranty period for obvious reasons.
However, this isn’t true. There’s a limit in place, and even Intel provides a three-year warranty for their chips (Source: https://www.intel.com/content/www/us/en/...ssors.html). On the other hand, Asus clearly states they cover their boards for three years too, but unlike Intel, they don’t include a refund option in their terms. So consider this: why would Intel permit a refund within the stated period while most board makers refuse to? Or should we ask the same question differently: why let manufacturers deny refunds entirely when Intel does allow them?
The logic behind your point assumes the system remains reliable after about three months. But in reality, data shows that a complete failure can occur within or near three months. For the Intel Core i9-12900K and Asus Z690 Prime-P D4 board, both tested under stress, experienced total breakdowns. This suggests the memory controller on the 12900K may have failed, while the motherboard itself showed signs of instability—like memory errors detected by memtest86.
On the board side, I managed to run four unbuffered DDR4-3200 sticks from Crucial without issues before switching to Intel. The AMD Ryzen 9 5950X and Asus X570 TUF Gaming Pro handled them fine earlier. Yet, with the same sticks on Intel, the test stopped after just a minute. This points to a problem with the motherboard’s support for these specific memory types.
At least on the Intel side, the first set of DIMMs passed memtest86 for over ten hours before errors forced it to stop. The second set failed completely within seconds. The warranty promise seems to hinge on stability after thirty days, but here, stability doesn’t appear to follow that rule.
It’s crucial to reflect on why these assumptions exist behind the warranty language. The reasoning often hinges on stability claims, yet real-world failures contradict them. If you’ve experienced a platform that stops working after a short period, why accept a policy that says it should last longer?
The core issue is clear: if a product can fail catastrophically within months, shouldn’t the warranty reflect that risk? Or are we simply accepting that manufacturers will not always deliver on their promises?
I believe you're misunderstanding my point. The failure curve for most products typically starts with many early failures due to manufacturing issues, followed by a higher rate of breakdowns over time from wear and tear. Usually, manufacturers include warranties that cover these early failures before they become widespread. This doesn't mean you can expect your product to fail randomly between these stages—it's just less likely. If you're keeping it past 30 days, the main problem seems to be the product breaking unexpectedly, which I find a bit unfair for both the company and consumers. Products often fail sooner because of defects, so calling a faulty item a bad product isn't entirely reasonable. Additionally, the situation with warranties can be confusing; getting a full refund after three years might feel like you're being taken advantage of when you were just starting out.
I’m sorry you’re experiencing issues with your new setup. By the way—are the RAM modules being evaluated on the QVL boards? I’ve always relied on QVLs for system builds because they provide some assurance from the manufacturer. While QVLs shouldn’t control RAM or storage performance, knowing they’ve been tested helps narrow down troubleshooting options.
The issue with a product malfunctioning quickly is puzzling to me. What exactly counts as "unreasonable" here? My concern is that the device begins to show issues within three months of use, even though the motherboard comes with a three-year guarantee. It’s strange to accept something that could start failing so early, especially when the warranty covers up to three years.
The concern lies in how you interpret reliability trends. You seem to think the failure point is near the end of the warranty period (around day 31), but what does "early" really mean? Are we considering 30, 60, or 90 days? Your definition matters.
Additionally, you assume the manufacturer will only repair the motherboard if the CPU is also working, which isn’t guaranteed. The logic behind sending a product back for replacement assumes the other parts are fine, but that’s not always the case.
Your reasoning also overlooks the fact that once a defect is identified, the manufacturer expects you to keep using the product despite the repair. This creates a risky situation where you might be stuck with a faulty item even after a fix.
If you’ve dealt with similar problems before, it’s clear why this would seem unfair. You’re questioning whether fairness should focus on the company’s actions or your own experience. The real issue is whether the warranty terms are realistic and transparent enough to protect consumers.
I don't recall whether the memory I bought was included in the QVL list on the motherboard when I purchased it. To be honest, there are many more memory modules available than what Asus can realistically check and add to their QVL list. Another issue with QVL lists is that they expand or change over time. If a module isn't listed, it might be because it hasn't been tested yet—or maybe it failed their tests. Removing a module doesn't clarify whether the reason is lack of testing or actual failure. Also, a motherboard must meet JEDEC standards, so if a RAM module also complies with that standard (not just XMP), it should work with the board since they both adhere to the same standard.
I just returned to them after experiencing issues with Gigabyte. My Z690 hero has performed well. It seems unlikely the problem is with the platform itself, but rather a faulty part like the motherboard or CPU. I ran identical tests on my original build and didn’t encounter this issue. If it was a broader platform concern, there would likely be more reports.
Essentially any system will have issues because catching every defective item during quality checks is nearly impossible. Assuming one faulty product after thorough testing is unrealistic. If you kept the item beyond 30 days, it’s reasonable to think you were happy with it afterward. Also, I regret that manufacturers don’t offer refunds as part of their warranty for clear reasons. If this were a standard practice, more companies would follow suit.
They're a bit of a mixed case for me. Around 2000-2007 they worked well with their motherboards. Then from about 2008 onward, some Asus boards started having issues, and it seemed like others had similar problems too. But I've been using Asus boards again since around 2015 (with my Core i7-3930K). I don’t update my computers often, and for a while I used Tyan boards instead. I’ve also tried Gigabyte and MSI consumer boards—they function just fine.
Statistically, the chance of a warranty failure isn’t extremely high, but what matters is how severe those failures are. It’s not just one bad product that caused the issue; it’s about patterns and severity. You mentioned reading my post, but I wonder if you actually considered these points.
Your explanation ignores key details: the CPU (possibly a faulty memory controller) and the motherboard itself. In your case, I tested four RAM modules in a Ryzen system, ran memtest86 for 17 hours and 9 minutes, and it passed before switching to another system. That’s significant evidence.
The situation with the 12900K/Z690 is even worse—your system couldn’t run memtest86 for 27 seconds, which is a clear red flag. You’re not just dealing with one bad part; it’s two major issues. The video you shared was the second time this happened, and it got me to record it.
Your excuses don’t explain why this occurred or how often it should happen. It’s not about quality control—it’s about a flawed platform. If you can’t run memtest86 for even a short time, the system is fundamentally broken.
As for QA, your response doesn’t address why they missed it. The problem isn’t just oversight; it’s a systemic failure. And when it comes to sampling rates, it’s about whether companies truly test thoroughly or just rely on assumptions.
This isn’t just a technical issue—it’s a matter of trust and accountability. If you’re not willing to acknowledge these facts, then your concerns are valid.