Just an update, as it looks like for now Intel has every intention to accept and substitute boxed units of 13th and 14th gen CPUs that have shown instabilities due to degradation caused by high voltage. So for these boxed units it looks like the class action will not be needed and actually may fail as Intel is honoring its warranty. The keyword here is "boxed", as this applies only to CPUs that have been bought in packaged units.
On the contrary, as it appears, if you bought a pre-built PC with a 13th or 14th gen CPU from a system integrator (even big ones like HP or Dell...) then you may be in a bit more trouble, as it looks like the SI will pass the ball to Intel, and Intel will pass the ball back to the SI, but none of them will actually take responsibility and substitute your faulty hardware. This appears to happen because Intel have different warranties with system integrators, and also because there may be a mechanism for which Intel can pause or deny claims if further investigation is required (such as the SI designing systems where the CPUs are either run out of spec, or placed in an out of spec environment). So here the class action may have more ground.
Intel is in equal trouble regardless of whether the product is boxed or not.
First, RMAing does nothing to detract from the loss in revenue from the downtime caused by Intel CPU failures. Intel replacing the CPU will not reduce the damages as a result of that failure and downtime. It's one thing if their CPUs are failing at a normal rate (around 1.2%) but given the 50% failure rate claimed by multiple sources it is vastly beyond normal customer expectations of the product and as a result does not qualify for protection from a product defect tort. Any reasonable observer would come to the conclusion that a 50% failure rate for a CPU is ridiculously out of line given the normal failure rate. Please reference 2nd link below for more information.
Second, Intel has not made a broad public announcement informing customers. Comments on reddit and on the down low messages do not come anywhere close to meeting their requirement to inform customers of this issue. It's almost irrelevant that Intel will RMA CPUs if customers are not aware of the issue in the first place. The CPU is naturally one of the last parts a customer will blame and Intel is absolutely using that as part of a game of waiting out the warranty or tiring customers with it's lack of disclosure. It is actively generating more liability by not replacing every potentially affected CPU on the market as the harm to parties increases in the form of additional downtime and computer troubles.
Third, degradation is permanent. As Intel stated, every CPU 65w and higher is impacted. What is the remedy for customers who have an impacted CPU unknowingly who later experience issues as a result of damage caused by Intel yet the CPU is now out of warranty? None. It's nonsensical for Intel to approve RMA's one by one when they've made it clear that every CPU 65w and higher is effected. They are trying to reduce their liability at the cost of end users, end of story.
Fourth, Intel was denying RMA's prior to this becoming public. Alderon games has publicly commented on how many RMAs Intel denied despite it being Intel's issue as the root cause. Since Intel's post regarding the voltage issue, they've come out and questioned why Intel was denying their RMA's when they've made a statement acknowledging that they were aware of the oxidation defect. I'm not even sure you can trust that Intel will even properly uphold the warranty for each individual RMA. To me it seems like a play to sweep this under the rug with RMA's that only Intel will know about and divide customer power so they can silently deny warranties on a case by case basis once they think media attention has died down. I absoltely would not put that past a company like Intel.
Fifth, Intel knew of oxidation issues for a year and only just released a statement. Just for this problem alone, completely aside from the voltage issue, Intel knowingly let customers suffer the issue and I have to wonder just how much liability in regards to RMAs they dodged by doing so. I imagine no small amount.
Product defects are part of torts law, wherein a company's acts or omissions result in injury or harm (physically, financially, or otherwise) to another. There are three kinds of torts: intentional, negligent, strict.
www.law.cornell.edu
Product defect and torts law are well defined and it could very well be argued that Intel qualifies for all three. Intel could be sued for strict liability for it's product defect and separately for it's neglegent failure to disclose the oxidation issue for an entire year. It's failure to disclose oxidation and it's current failure to disclose ongoing issues to all customers could be considered an intentional tort as well.
www.law.cornell.edu
Just to raise the mood a bit anyways, if your CPU has not yet shown the signs of critical instabilities, it appears that Buildzoid in
this video may have just found a way to keep the newest Intel Default Settings, while also limiting the CPU to less than 1.4v (so to not cause degradation) and get a nice performance boost (compared to having only the standard Intel Default Settings) and lower temperatures. So now Intel may just have to do the same tweak, or something similar, in their mid-August patch and this issue may be solved without incurring in reduced performances. Also, you can just follow the Buildzoid video and have it now.
That video is only intended to show you how to get as much performance as possible with safety features turned on and the Intel default profile.
It's never implied that it is guaranteed to stop degradation. We don't know which particular voltage is the problematic one, at what level it becomes problematic, how well adjusting voltage manually interacts with the claimed buggy micro-code, or even if Intel is telling the truth. Level1Tech's gaming server provider data showed that turning down CPU clocks and voltage only delayed the issue from appearing, it did not stop it. It could very well be that Intel's micro-code update merely delays degradation outside of warranty.
.
