alpha754293

Not really, only because the systems are just in old Antec full tower cases that are probably somewhere around 15 years old now (for the case itself), so there's really not that much that's exciting nor interesting to look at. It is. LTT doesn't really seem to something akin to like a "build log" section of the forum, and therefore; being that this pertains to networking, ergo; why I posted it on this section of said forum. No, the "question" that would've came about previously would be the whole "why is my 100 Gbps Infiniband card only getting around 14 Gbps?" when it was plugged into the secondary PCIe slot. But the moment that I removed the discrete GPU, and then plugged said Mellanox card into the primary PCIe slot, then it was able to obtain close to the 100 Gbps speeds again. The more INTERESTING question could be can I force BOTH of those motherboards to accept said discrete GPUs in the secondary slot because I think that the last time that I tried it, after POST, an error message came up saying that the GPU was in the secondary slot and wouldn't boot. So, if there is a way to FORCE it to boot with it in the secondary slot, that would be great. And no, neither of the motherboard manuals cover that. (In fact, the motherboard manual for the X570 TUF Gaming Pro WiFi doesn't even have a table for multi-GPU operation because I don't think that the board supports it and/or that there aren't enough PCIe slots for that. As a result, I don't remember, but I don't think that there is even the option to be able to run both slots at x8/x8 being as it is possible that the secondary PCIe slot is an x4 slot going through the chipset, and I don't know if there is a way to FORCE the system to accept that as the slot for the discrete GPU and FORCE the system to proceed with the boot sequence.) (On the flip side, it works now as a purely headless system, so I'm not super worried about not having a discrete GPU in there. However, if there are problems that involve the Mellanox Infiniband card operating at full speed, and I need console access, I'd likely need to spend $300 on the TinyPiloy iKVM to be able to get console access without said discrete GPU installed.)

Not really. it's just a function of money. I actually bought the NICs and cables because of a video that LTT made a few years ago where you can buy the network cards off eBay for about half the price of retail. So it seemed like it was "good enough" of a deal for me. And then if you calculate the cost efficiency based on $/Gbps, 100 Gbps would be cheaper per port than even 10 GbE. It saves you money if you're able to make use of the bandwidth. Mellanox ConnectX-4 dual VPI ports 100 Gbps per port, Infiniband network card. (MCX456A-ECAT). It's a PCIe 3.0 x16 card. With two 100 Gbps port and they're Virtual Protocol Interconnect meaning that I can set the ports as running either in Infiniband (IB) mode or in Ethernet (ETH) mode or have one port for each. (Although I'm only using one port per card per system right now, because whilst I would like to deploy 100 GbE, the switches still aren't quite cost effective enough for me to use it. Plus the PCIe 3.0 x16 is only able to support a max of 128 Gbps which would be shared between both 100 Gbps ports, so it actually is better for me to only use the one port for now.) I use my cluster for HPC/CAE/CFD/FEA applications. This is an example of the type of stuff I run on my micro HPC cluster: This specific simulation actually ran on my 4-node cluster system (where each node was dual socket, each socket was 8 cores, with HTT disabled, and each node also had 128 GB of RAM (for a total of 512 GB for the entire 4-node cluster).) previously. This is another video of another type of simulation that I ran previously as well: This simulation is designed to simulate what would happen if you were about to be in a head-on collison and then both cars swerve away from each other, but not enough and end up hitting each other anyways. Stuff like this. Yes and no. I think that when I first set up my first node/system, I did try and see if I can install the discrete GPU (I think it was either the GTX 660 or the GTX 980) into the secondary PCIe slot and have the Mellanox card in the primary slot. And as I also vague recall, the POST screen complained about the discrete GPU not being in the primary slot and refused to boot. But now that both systems seems to be running relatively stable, they're completely headless now. And because the 5950X doesn't have an iGPU, it means that I literally have no video output coming from the nodes now, at all. (Not unless I spend $300 per node to get the TinyPilot iKVM.) So right now, I just remote in over ssh and/or VNC. re: 14 Gbps Yeah, I don't know what was going on with that either. I forget how to tell/check what the PCIe link rate is in Linux (although I'm pretty sure that I could've googled it, but I think that would have just been "information only" i.e. I don't know if there was a way for me to FORCE it to have a specific link rate). And even if I could, that might end up causing system instabilities, which wouldn't really have fixed the problem properly anyways. One of my hypothesis is that it actually dropped the link rate from PCIe 3.0 x4 down to either PCIe 3.0 x2 or PCIe 2.0 x4. Hard to tell at this point (since I've put the Mellanox card in the primary PCIe slot now). (The Mellanox cards are only PCIe 3.0 x16 cards anyways, so PCIe 4.0 won't really help.) As a 100 Gbps card, I've never gotten 100 Gbps exactly. The closest that I've gotten was about 97 Gbps in Windows and 96.58 Gbps in Linux (CentOS 7.7.1908). In an actual application, it was testing at around 89 Gbps. So, if it had dropped the speed down from PCIe 3.0 x4 to PCIe 3.0 x2, then it would've only been delivering a theorectical max of 16 Gbps, which if it was hitting 14 Gbps, would mean that it was getting 87.5% of the PCIe 3.0 x2 theorectical bandwidth. If it was PCIe 2.0 x4, then that should've been capable of 20 Gbps, which 14 Gbps would be 70% of the theorectical bandwidth. Either way, something was off. And I couldn't tell if it was "firing the signals" on the rest of the connectors, but because there might not have been an electrical x16 connection in the x16 physical slot, the signals that were being "fired" to the part of the card that didn't have closed contact/connection (due to the slot being x16 physically, but could've been only x4 or x8 electrically), the card could've been expecting a response to those signals, which it never got back. So...who knows. It's a pity that the Mellanox ConnectX-5 Ex (MCX556A-EDAT) cards are still too expensive for me, even on eBay, because otherwise, the Ryzen system can actually use it with those cards being a PCIe 4.0 x16 card, which means that you would actually be able to support both ports running at 100 Gbps each, out of the shared 256 Gbps interface that the PCIe 4.0 x16 slot would afford.

I have two AMD Ryzen systems where both are with the Ryzen 9 5950X processor, but one as an Asus X570 TUF Gaming Pro WiFi motherboard whilst the other has an Asus ROG STRIX X570-E Gaming WiFi II motherboard. Earlier today, I was trying to diagnose an issue with my 100 Gbps network connection between the two Ryzen nodes and the microcluster headnode, where upon running ib_send_bw, I was only getting around 14 Gbps. Now that I took the discrete GPUs out from each of the systems, I'm getting 96.58 Gbps on my micro HPC cluster now. Yay!!! (I can't imagine there being too many people who have Ryzen systems with 100 Gbps networking tying them together.)

I dunno. I guess that depends on how much footage their systems is trying to ingest/render simultaneously. I guess that's also the advantage of having enterprise-grade NVMe SSDs such that at 30 TB of raw capacity, even 1 DWPD still means that you're writing 30 TB/day.

***RAID IS NOT A BACKUP.*** Depending on how much data Dream writes to the NVMe headnode, this can potentially be a TERRIBLE idea. The Kioxia CM6-R 30 TB drives has a write endurance of of 1 drive writes per day (30 TB). If they have set up the storage headnode as a raidz2 (5 drives + 2 for parity), that means that distributed across those 5 data drives, they can be writing 150 TB/day. This means that if the sustained data rate, over the course of a day, is higher than 1736 MB/s, they will wear out the drives by the time the warranty expires. Any faster than that, and they will wear out the SSDs before the warranty expires. I'm not sure if there's a way to pass the NVMe drives as L2ARC for the hard-drive-based server, but if you could have passed the NVMe drives over fabric so that the hard-drive-based server would be able to see it and use it as L2ARC, I would have done that instead, because the Kioxia CM6-R 30 TB NVMe SSDs are designed as read-intensive SSDs (per their product brief: https://business.kioxia.com/content/dam/kioxia/shared/business/ssd/doc/eSSD-CM6-R-product-brief.pdf) and NOT write-intensive drives. The Kioxia CM6-V 12 TB are for mixed enterprise use, which means that it has a write endurance of 3 drive writes per day. Dual 25 GbE. mehhhh....... My Ryzen system is rocking a Mellanox ConnectX-4 dual 100 Gbps Infiniband NIC.

Quick/brief update on this saga: So, after about roughly a month of arguing with Asus, Asus agreed to replace my Asus Z690 Prime-P D4 motherboard with an Asus ROG STRIX X570-E GAMING WIFI II motherboard. I actually originally requested the Asus X570 TUF Gaming Pro WiFi motherboard instead (because that's what my other Ryzen 9 5950X system was already using, so I know that I don't have any issues with that system), but they denied that request and instead, sent the more expensive motherboard as its replacement. By this point, I was just happy that they were sending ANY X570 motherboard as a replacement for the clearly defective Asus Z690 Prime-P D4. Course, also by this point, I had also already filed a lawsuit against them in court, and was in the process of serving them the papers when they sent me the email with the replacement offer. In my email exchange with them, I told them that if they couldn't find a resolution that we can mutually agree on, that I will have to pursue the matter in court, they said that due to the fact that I had indicated that I was looking into pursuing legal action, that the Asus Corporate Customer Care team was supposed to contact me, and to the best of my knowledge (or at least nobody whom I've been in contact with, has ever identified themselves as being from the Asus Corporate Customer Care team), so said Asus Corporate Customer Care Team never contacted me like they said they would/like they said they were supposed to. I'm currently in the process of putting the new-to-me Asus ROG STRIX X570-E GAMING WIFI II through its paces, and checking for system stability issues. But given my previous experience where I might not see the problems until approximately 3 months in service, so I am keeping my plaintiff's claim open as I have 6 months to serve them. From a cost perspective, because they sent me a more expensive motherboard, I netted out being roughly even. So....depending on the situation (as I still contend that the fact that the board was able to run memtest86 for almost 11 hours in one test, but less than 27 seconds in another), if you're willing to put in a little bit of time to research the laws and statutes in regards to consumer protection, and depending on your specific situation, you MIGHT actually be able to get a remedy if your motherboard catastrophically fails like this, and Asus (or any other motherboard manufacturer for that matter) refuses to issue a refund to you and you are outside of the refund window. Doesn't necessarily mean that this would ALWAYS work.

They're a bit of a mixed bag for me. From like 2000-2007-ish, their motherboards worked well for me. And then from like 2008-ish or so, some of my then Asus motherboards started acting up and ancedotally, it seemed like that I wasn't the only one with issues with Asus motherboards, but I've been back to using Asus boards since maybe ca. 2015-ish timeframe (with my Core i7-3930K). (I don't really update my computers all that often and for quite some time, I was using Tyan motherboards instead.) (In between that, I've tried Gigabyte and MSI for consumer boards and they're "ok" as well in that they function.) And I know that statistically, the warranty failure rate (for motherboards), I would surmise, isn't super high to begin with, but what the warranty failure rate doesn't tell you is that if and/or when it fails, the degree of said failures. But it's not JUST one faulty product (that went into determining that there is a problem with the platform). Did you even READ my post at all??? You are explicitly told that both the CPU (which I am guessing could be the on-die memory controller that may be faulty) PLUS the motherboard that results in my saying that the Z690 has a problem, or so it would appear. Nowhere, in any of your replies, have you ever accounted for these facts. And this is further implicated by the fact that I tested four DIMMs of RAM in my Ryzen 9 5950X/X570 system, spent 17 hours 9 minutes running memtest86 on that system, which it passed, BEFORE I took the memory out of that system and put it into the 12900K/Z690 system, where it then failed to run memtest86 to such a degree that it resulted in the video (where said 12900K/Z690 system spontaneously resetted itself). You aren't anywhere remotely CLOSE to addressing these facts. The platform is so bad that I can take WORKING DIMMs of RAM, transplant them from a WORKING system, and said Z690 platform/system can't even run memtest86 for TWENTY-SEVEN seconds, (according to memtest's own timer), as shown in the video. So, no, it's not ONE faulty product. It's TWO faulty product. And the DEGREE that it has failed (if you have a system that can't even run memtest86 for TWENTY-SEVEN SECONDS, I'd say that's a pretty significant issue with the system/platform). it's quite literally an #epicfail. Think about it this way: when was the last time you've had a system that CANNOT RUN memtest86 for TWENTY-SEVEN SECONDS? (The video was actually the SECOND time that it happened. It actually happend immediately prior to that and normally, you should NEVER expect a system to spontaneously reset itself, but it happened the first time, and that's what got me to take out my phone and record it on video (which I purposely recorded the entire boot sequence in order to show the mode of failure/what I'm seeing). All of your excuses FAILS to account for why this happened in the first place and/or if this should be happening at all. I don't really care that QA didn't catch it. The video clearly shows "it happened". Now the question is what is Asus going to do about the fact that it's happened and QA missed it. Oh...and to your point about QA not being able to catch every defective product - that depends on two things: 1) Your QA regiment and 2) your sampling rate/sampling frequency. As Steve from GamersNexus has shown, some companies DO test 100% of their products through their automated QA process. But if your QA process fails to catch a faulty product, is it because you aren't testing for that fault (either because you don't know that fault CAN exist or because you are making a conscious and deliberate decision NOT to test for that fault). If they don't know, well, now they do (thanks to my reporting). If they are making a conscious and deliberate chose NOT to test for this fault (usually due to time, which equals cost/money), then that's their bad/their problem. Case in point: The 2009-2011 Toyota vehicle recalls totaled to around 16.2 million vehicles which were triggered by upto 37 deaths. This means that the number of deaths / total number of vehicles recalled is around 0.00023%, which, statistically speaking is generally insignificant. However, because they were deaths (i.e. the degree of the mode of failure), they garnered a LOT of public media attention and eventually, Toyota was found to be responsible for those deaths based in part, on the confidential Barr report. (Source: https://en.wikipedia.org/wiki/2009–2011_Toyota_vehicle_recalls) In other words, just because you didn't test for it, doesn't remove your liability should it fail. And whilst a 12900K/Z690 not working isn't a life-or-death situation, but the fact of the matter is that it really is only going to be a matter of time before you will see this kind of processing power inside your car (if it isn't there already in a segregated manner with all of the various control modules that are in a modern vehicle). Therefore; today, it's just my computer not working. If that computer was in your car, it'd be the whole Toyota vehicle recall thing again. Oh and by the way, Toyota settled the recalls for $1.1 BILLION. Based on WHAT? Once again, you have failed to provide an answer to the question "why is 30 day the threshold for the customer acceptance criteria?" And perhaps more importantly, what are the assumptions you are making behind WHY you are picking 30 days instead of 90 days? Why not 180 days? Why 30 days? You have failed to provide an answer in regards to this very basic and fundamental question. If the ECU in your car died after 90 days, can I use your logic here and go "too bad. So sad. You passed the 30 day mark, so it must mean that you liked your brand new car."??? "...then it's a fair assumption that you were satisfied with the product" That's an asinine assumption to make. On what grounds or basis are you making this assumption? You have failed to provide an answer in regards to this very simple and basic question. As I have previously stated, the only way that you would assume this whole "30 day thing" that you seem to be on about, assumes that 30 day should be the length of time for the customer acceptance criteria without providing any data to say WHY 30 days should be the length of time for the customer acceptance criteria. Again....what is the basis and/or on what grounds are you basing this assumption off of??? This is an asinine assumption to make and you have failed to provide any supporting data/evidence that shows WHY 30 days should be the length of time for the customer acceptance criteria. Or said in a slightly different way - why 30 days? Why not 60 days? Why not 90 days? Why not 180 days? You are arbitrarily making the decision that "okay. 30 days is good (enough)" with NO data, NO evidence that you are using to base your decision off of. That's insane to make such an arbitrary decision without any data nor evidence to back up the question "why 30 days?". Yes. And the reason is money. Their warranty policy is LITERALLY written which shows that they care more about money than they care about you as a customer. It's ALL about money. When they see you, they don't see you as a customer. They LITERALLY have the cartoon thing where their eyes turns into dollar signs. You can have a high warranty annualized failure rate and still be popular, and make lots of money because what customers care about is that if there is a problem, what are you (as the company responsible for the manufacturing and sales of the product) going to do about dealing with such problems. Case in point: Dell has like about a 26-27% failure rate. (Sources: https://www.ruggedmobilityforbusiness.com/2009/10/what-laptop-manufacturers-dont-want-you-to-know/, https://www.geckoandfly.com/6311/the-most-reliable-laptop-survey-best-netbook-reliability-comparison/) And yet, despite this, so many businesses use Dell computers enough for Dell to make $94B in revenue for FY21. So whereas motherboard manufacturers see warranty refunds as a negative (loss of money), Dell has seen and built their entire business model around, quite literally, fixing one out of every four product they make and sell AND they can remain highly profitable despite having one of the industry's highest failure rates. Therefore; to your point about why companies don't offer refunds as a part of their warranty - it's literally because they care more about money than they do about keeping you as a customer. If Asus had refunds as a part of their warranty policy, I would stick with Asus motherboard above all other motherboard OEMs because I know that if I have a problem with their products, I will be taken care of. But alas, (and you still haven't answered why), you are literally arguing for their shitty behaviours and I can't really seem to figure out why (other than to surmise that you either work for them or you work with them via the supplier base or something). Beyond that, I can't see why anybody would be defending a company's shitty warranty policy in an industry that ALL, collectively, have the same shitty warranty policy. If Asus changed their warranty policy that they would offer refunds upto the entire duration of their warranty period, I would ONLY buy Asus (for consumer boards at least) from now on if that were the case. But instead, you're here, defending their shitty policies and once again, having failed to provide an answer in regards to the question that I asked you "why do you tolerate a company's shitty warranty policy when you know it's shitty?"

I don't remember if the memory that I purchased were on the motherboard's QVL list at the time when I bought it or not. To be fair and frank though, there are a lot more memory modules that exists in the market than what Asus can realistically test and put on their QVL list. The second problem with QVL lists is that it grows and/or changes over time and that if a module isn't listed on said QVL list, is that because they didn't test it (yet) (or ever), or is it because they tested it and it failed their QVL tests? The exclusion of a module doesn't tell me which of the two potential reasons why a module isn't on there. Further, a motherboard should be compliant to the JEDEC standards, which also means that if a RAM module is also compliant to said JEDEC standard (i.e. NOT XMP); then it should qualify with said motherboard given the fact that they're both compliant to the same JEDEC standard.

Why would a product breaking seem "silly" to you? I'm not sure what would be "silly" about a product breaking. The problem that I have is you have a product that started to fail within 3 months-in-service. Even if the motherboard was warrantied for 3 years, why would you accept a product that would or can start to fail, pursuant to the stated policies, on day 31 of ownership? THAT is what, I would find, would be "silly" to me. The problem with this statement is that you are assuming that the shape of the reliability curve approaches is global minimum threshold at around the 30 day mark (per your comment about how "Product can fail especially earlier on it life due to defects...", but you failed to define what "earlier" means. Are we talking 30 days? 60 days? 90 days? What's "early" according to your definition? Further, the assumption that the warranty also makes is that if the motherboard fails, then you can send it back for a RMA repair, which means that other components haven't also failed as well. But again, that's an assumption on the part of the motherboard manufacturer. (i.e. board fails, but your CPU is fine). The motherboard manufacturer assumes that if your CPU is fine, then they can repair the motherboard under RMA, send it back to you, you pop your CPU back in, and then you're back up and running. That assumption doesn't hold/isn't true here. The CPU has already been authorized for a refund. For all I know, it can be the on-die memory controller that failed on it. Either way, once you know that there is a problem with a product, the motherboard manufacturer then makes a further assumption that you are going to continue using that problematic product even after said RMA repair. And as I've told Asus, I've already RMA'd the CPU back to Intel for a refund, and therefore; even if Asus were to repair the motherboard under a RMA, I would no longer have a way to verify that the RMA repairs were successful because I won't have a CPU anymore. If you have already experience a PLATFORM failure like this, where both the CPU and the motherboard are being RMA'd, which means that you KNOW that there is a problem, why would you continue to buy something that you have already experienced a problem with? You've already experienced a problem with a product. So why would you keep buying the same product over and over again? That makes no sense to me whatsoever. Sounds like somebody is getting all butthurt over this. Since when is this about being "fair"? if this was about "fairness", then the motherboard manufacturers should've adopted a warranty policy similar to Intel's where they stand by their products for upto 3 years, during which, you can get a refund if it fails. That would be fair, because at least then the warranty policies of Intel and Asus would be aligned. But as it stands, in its present state, it's not. (And to your earlier point, none of the motherboard manufacturers allow for the provision of getting a refund from their documented warranty policies. Fair would be if they actually allowed and enabled that. That is what would be fair. You write about "...isn't really fair to the company or anyone else." Where is the fairness to the customer? More importantly, why aren't you talking/writing about that. And/or why do you accept the fact that none of the motherboard manufacturers have a documented provision in their warranty policies which allows them to give a refund for a defective product for the entire duration of the warranty - 3 years? None of the motherboard manufacturers having the refund provision in their warranty policies is crap. Why would you accept crap?) And also further to your statement: "...isn't really fair to the company or anyone else." Are you arguing that it isn't fair to the company because I am documenting my experiences with their product for all to see and read about? Because to me, it sounds a LOT like rather than talking about what the COMPANY is doing, what you're talking about instead, is the person who found out what said company is doing. Why is the focus placed on the messenger rather than on the company? That's not even REMOTELY CLOSE to being true and representative of the facts. You can't just send the CPU back to Intel on day 1094 of ownership, claim that there's a problem with it, hoping you'll get a refund. I don't know why you would make that up. (It amazing what the brain will pass on as knowledge when a gap in knowledge exists.) You have to be able to proof/provide the data and the evidence that abides by and is aligned to the documented warranty policy from Intel, which means that in order to be able to send it back for a refund, you would have to be able to provide documented data/evidence/proof that there is a problem with it in order to even QUALIFY for said refund; of which, refund is only one of THREE remedies available to said Intel (and by extension, to you.) Your statement: "you just got a 3 year lease on a product for free" is patently false and is literally NOT what Intel's Boxed Processor Limited Warranty states. I don't know why you would make stuff up like that, but there you have it. (You want to talk about "fair" - if you want to talk about "fair", why would you make stuff up like that which you should know or ought to know that your statement is knowingly fraudulent misrepresentation (or at minimum, neglegent misrepresentation)? What's "fair" about you making stuff like that?)

But this isn't the case here. I agree with you that there is a cutoff, but even Intel, for example, has a 3-year limited warranty on their processors (Source: https://www.intel.com/content/www/us/en/support/articles/000005862/processors.html) Conversely, Asus' own warranty policy also state that they warranty their motherboards also for 3-years, except that Intel has an explicit statement/clause in the terms of their warranty for a refund (and/or repair and/or replace) whereas Asus (and pretty much all other motherboard manufacturers) do NOT have the refund clause in their warranty policy. So think about it: why would Intel allow you to get a refund for upto 3 years which is within their limited warranty documentation whilst motherboard manufacturers don't allow you to get a refund at all? Or asking the same question in a slightly different way: why would we allow motherboard manufacturers to get away with NOT being able to issue a refund (EVER), whilst Intel offers a refund for upto 3 years, as stipulated in the terms of their warranty policy? The underlying assumptions in your statement above assumes that the platform is stable and operative once you've hit > 30 days/1 month in service (for motherboards, where MOST retailers allow for 30 day refund window). But that's not the case here. The data and evidence with the failure of this Intel Core i9-12900K and this Asus Z690 Prime-P D4 motherboard shows that you can have a full, complete, and catastrophic PLATFORM failure within or close to 3 months in service. (I think that I'm actually a little over 90 days because I started building the system around Christmas 2021 and this failure really reared its head maybe about a week ago?) It is entirely possible that the on-die memory controller on the 12900K has failed which is a part of the contributing factor in regards to memtest86 showing the memory errors during the course of the test. But on the part of the motherboard, the fact that I was able to test an entire different set of four DIMMs of Crucial DDR4-3200 unbuffered, non-ECC memory, (same Crucial P/N), which PASSED on the AMD Ryzen 9 5950X/Asus X570 TUF Gaming Pro WiFi system/platform before I popped those same four sticks into the Intel Core i9-12900K/Asus Z690 Prime-P D4 platform, which resulted in the video shown above, clearly shows that there is also a problem with the motherboard. At least on the Intel platform, the first set of four DIMMs, memtest86 was able to run for 10 hours 52 minutes before memtest self-aborted "due to too many errors". But on the second set, as you can see in the video, it couldn't even run the test for a minute before the system spontaneously reset itself. As I mentioned, I don't use XMP. The expectation that I have (from the motherboard) is that I should be able to pop in the other set of four DIMMs of memory, and have it run memtest on it, at least for the same 10 hours 52 minutes as the first set of four DIMMs. But on the second set, it couldn't even run it for 60 seconds. The assumption with the warranty policy is that the system and/or the platform and/or the CPU is stable after the first 30 days. I think that it is important to think about the assumptions that went on behind the scenes and into why and how the warranty policy is drafted (and approved/authorized by the company). But here, in this case, this is clearly NOT the case. Asus has repeatedly offered to repair the motherboard under RMA. And I repeately tell them, that even if they performed said repairs under RMA, because I have already sent the CPU back to Intel for a full refund, therefore; I would have no way of testing and verify that the repairs have been performed correctly and that it fixed the issue. And you can say "well...it's your choice whether you want to buy another CPU or not". To which, I say: "If you have experience with a PLATFORM that doesn't work/perform as you would expect, why would you give the PLATFORM another opportunity to make a fool of yourself?" As the saying goes "Fool me once, shame on you. Fool me twice, shame on me." So why would you avail the PLATFORM the opportunity to fool you again? If you know, for example, that a product has a problem with it, why would you keep buying that product? That makes no sense. Like I said, I agree with you that there should be a cut-off period. Both Intel and the motherboard manufacturers typically have a 3-year limited warranty on their products. The only difference is that Intel EXPLICITLY state that refund is a potential remedy for warranty claims whilst none of the motherboard manufacturers offer that. So why would Intel offer said refund under their 3-year limited warranty policy whilst we allow and accept the fact that motherboard manufacturers do not offer the same? Why should be accept that as the status quo, if you have one of the key components that is required for the operation of the motherboard CAN be refunded within the 3-year limited warranty period, per policy? (i.e. why one but not the other?) Why do and should we accept the fact that motherboard manufacturers don't offer ANY kind of refund at all? Pursuant to your point, even if the motherboard manufacturers had a 1 year policy, that's still better than the status quo, which is none at all. Do the motherboard manufacturers have an issue in regards to the confidence of their products which is why they don't/won't offer even a 1-year refund warranty policy?

Perhaps a more interesting and important question that should be asked in light of this is "why DO we put up with companies who don't or won't offer a refund, say within the first year of purchase or something along those lines?" I am of the opinion that if you can't boot into memtest86 and stay there for more than 90 seconds, you should be able to get a refund. And if you can't, then why do we, as consumers in the industry, are willing to put up with this fact that we can't get a refund for a defective product, for a system that's only been in service for 3 months? The industry's collective unwillingness to stand by their products says a LOT about the companies, their products, and the industry as a whole.

As I mentioned, I was updating my original post when you had replied. I don't use XMP.

Well, I showed them the pictures and video that you see here. It would've been on the retailer if I was still within the 30-day refund window. This is 3 months in service (3MIS). It's hard to say whether it's the motherboard or the CPU or both. And I say that because the memory controller is on-die, on the CPU, so if there is a problem with the memory controller which may be contributing to the over 10,000 errors that memtest86 originally found, then that would be a CPU issue (and/or more specifically, the on-die memory controller issue). As for the motherboard, the fact that I was at least able to run the test for a few passes in the 10 hours and 52 minutes that memtest86 was running, shows that it CAN test the memory. But then, the moment that I swapped the memory from my 5950X system to the 12900K/Asus Z690 Prime-P D4 system and it spontaneously resets the system as shown in the video within 90 seconds; clearly that shows that there is also a problem with the motherboard as well. Either way, I'm not going to subject my wallet to testing this over and over and again, and enriching the parcel delivery services company, where I am constantly shipping stuff back and forth in order to try and figure out what is the root cause of the issue. At this point, I have deemed that the processor and the platform/motherboard is not reliable and untrustworthy. I am glad that Intel was willing to offer a full refund of the purchase price of the processor. I think that Asus should be doing the same for a motherboard that's only been 3 months in service. (As a warranty issue, Asus should be taking a LOT of interest as to why a motherboard is failing 3MIS.) Apparently, Asus doesn't care enough about their product nor their customers or they don't care enough to be willing to offer a refund on a motherboard that has failed in such a catastrophic and spectacular way. (The fact that I can't even stay in memtest86 for > 90 seconds is VERY telling that there is CLEARLY an issue with their motherboard. If I were Asus, I'd DEFINITELY want to issue the refund to buy the board back to try and figure out why.) memtest86 is not a super complicate program to run (compared to an entire OS). And if I can't get through 90 seconds of that, it's no wonder why I am having data corruption issues on that system. (My 5950X, on the other hand, has been super solid/stable.)

Yes, the system spontaneously reset itself very early in the test. To your point regarding the current capability, that's the thing though: The original four DIMMs that I had installed in the system, at least it was able to run the test to the point where memtest86 self-aborted "due to too many errors" (apparently once memtest logs > 10,000 errors, it thinks that's too many, and stops testing). But at least it was able to get that far. When I swapped the memory for the four sticks from my 5950X system (both system are using Crucial 32 GB DDR4-3200 unbuffered, non-ECC memory, Crucial P/N: CT2K32G4DFD832A), that's when this spontaneous reset occurred. And the weird/crazy thing is that on my 5950X system though, the four sticks that came from said 5950X system passed memtest86 in said 5950X system. It was only after they were moved to the 12900K/Asus Z690 Prime-P D4 system where it would spontaneously reset the system like this. So clearly, there is something wrong with the 12900K in conjunction with this Asus Z690 Prime-P D4 motherboard.

Yup. Sorry -- I was editing the post with more details/information when you replied. I ran memtest86 thrice already: Once on the 12900K with the original 4 DIMMS of memory, which produced this: And then I ran memtest86 on my AMD Ryzen 9 5950X system (also four sticks of Crucial 32 GB DDR4-3200), which produced this: And then I popped the four DIMMs that were in my 5950X system into the 12900K system, which produced this: