Just in case you were wondering if we ever looked into this, well, now you know! BTW, we just posted a big Patreon written blog update on our plans for this year! www.patreon.com/posts/76702217?pr=true (finally back to posting regularly on Patreon! Thanks for your patience while we tried to get back ahead of the main content and months of unending reviews!)
You didn't use a cheap ARGB hub loaded up with fans and rgb lights.... did you? I have had a feeling this had something to do with it.... no one ever asks exactly WHICH rgb products you are using, and we KNOW the standards for rgb products are ALL OVER THE PLACE!! (as in.. THERE ARE NO STANDARDS!!! lol) Within the first MONTH of by my FIRST computer with all brand new parts and out of my own pocket some cheap argb hub fried the sound on my ASROCK B550 (the system recognizes it just fine with zero errors, it just puts out zero sound) and ASROCK even said they would RMA it with no questions asked but I told the guy that I plugged in an ARGB hub from some no-name fan manufacturer and the sound just stopped and he said he has seen argb hub do all kinda of crazy stuff to their mobo's .... AND I was chatting with a NEWEGG rep about a different RMA and brought up ARGB hubs frying stuff and he said essentially the same thing..... so MAYBE it would have been worth looking into... (or reading your comments, because I kept leaving comments about this in all the vids you made about the backwards caps... but I understand reading is hard lol :P nah I know you get A LOT of comments and you have MANY MORE things to do) - so I just wanted to FYI you guys about it, in case it's ever helpful :)
"We torture tested the equipment. We have a result. The result is "Nothing"." That just happens sometimes. But yeah, it's ESPECIALLY annoying when you're EXPECTING a flashier result.
I love it when things go wrong for you, you guys screw up, or you get weird results, or unusual hardware, and then you come to explain it. Its the most interesting content. Least interesting: Generic benchmarks, boring.
a negative result is still a result. its a big problem in the scientific community that people only want to publish findings when they actually find something. so we end up not having a lot of data that DISPROVES links between two things etc. glad you're getting these things on the record even if theyre not "exciting" enough for the main channel
@@hunterscooters5101 this isnt a "fixed" board. it had the capacitor installed backwards just like the ones that failed. it doesnt show that asus fixed the issue it shows that the issue was rare and that not all boards with the backwards cap would fail
You're very right. Publish or parish leads to a lot of bullshit research from higher education. This is also why researchers will collect data from many different variables at once and just hope that at least two of the variables have a relationship. That's not how you're supposed to test. The chance of a false positive goes dramatically up when you test in this way. One small note is that you can't really disprove a link between variables with a single (probably relatively small) sample. You either observe a relationship or you fail to observe a relationship. So maybe there is a weak relationship but you just didn't find strong enough evidence.
@@jonlaw16 yes but false negatives are also a thing. Just because you didn’t get a result from testing doesn’t mean that there is no relationship. You just didn’t have the right conditions for the event to occur. Publishing a negative result would not be the same a disproving that relationship. Unless you specifically set out to disprove it, it’s not worth publishing.
In case you guys are ever wondering "Man, we talked about X thing X long ago and said we were doing something with it, and we're *still* not done/not resolved it, I wonder if anyone still cares about it or remembers it?" Hi, that person is me. Legitimately and not being facetious those kinds of long term *actual* genuine tests are seriously interesting to me, and updates like this are incredibly satisfying and fulfilling as they bring closure to a question. I often wonder about long term projects that seem to have faded away on channels, and though they don't occupy my mind all the time I do find myself thinking in passing about those types of things somewhat frequently, and I do sincerely enjoy getting updates on them when there's something to share- including the finding that there was nothing significant! That's an important part of the scientific method!
Could not agree more brother. Those long term projects that are said will be returned to, like 9/10, are not. Wheter it be RU-vid, DIY type TV programs, or a freakin magazine seldom do I see a project returned to a year later bc the results are uninteresting.
I think they should mention it in their other videos, like the news recap. Just give us a quick update on what projects they have going on in the background.
me too tbh, i hate niot having closure no matter how small something is, especially if its something long term like... the findings either way are legitimate and interesting and come from a real intensive testing
That's some fairly wholesome content. We've have a team that spent thousands of dollars on a project and a year of their time, and the primary concern is simply to update the community and regain their table space. I love how the testing is complicated but the humanity behind it, is still simple. Humble.
The nice thing about having the community's support is that we can occasionally try to explore something like this and walk away knowing we tried, rather than trying to turn it into something bigger than it is. Thank you for the kind words and support!
@@gnextras You didn´t try. You DID find out what happens. Nothing. That is still information which is useful overall and for some owners of the motherboard maybe something that puts their mind at ease. I am here for exactly this. Information. Not for spectacle. You all provide us with trustworthy information over the last years. And i really hope you know how much this is worth these days.
Glad to see that you’re still keeping us updated on stories that either didn’t come to plan or just a time/space conflict I appreciate these smaller updates on things that some people may have forgotten
I know what you need to do to make it pop. Use the board for something you rely on, like a stream PC for a event. If your luck is like my luck it will go.
Sounds like a good home here and more extra content to me so that’s awesome! It’s cool to see all those man/ equipment hours get out somewhere since I’m sure it’s annoying with stories like this one
I think its cool that you are putting out your "negative results" somewhere. In scientific research, publishing negative results is critical but way to overlooked and underappreciated. I know what you are doing is not really scientific research but I think negative results are still important. Glad to see you are not letting them go to waste.
What I like about this is that you're showing the work and data behind the kickback of "could not replicate condition" that you often see from a shop (especially w/cars, OMG), which is perfect for the extras channel, IMO!
Nice, been wondering about this every now and again. Either it was being worked on behind the scenes or, well, this. Awesome that you *do* spend the time on these sort of things though.
Back in the day I bought a first run modmat just because I thought it was cool, but I'm glad the money went to a team that truly has their heart in it. Like you could not pay 99% of reviewers enough to do this kinda stuff and here you are putting thousands into it and only getting a "behind the scenes" vid.
Great job, Steve and crew! You did get a very reasonable and thorough conclusion. We were just expecting a different answer. Again, another fine T&E. Please continue this type of reporting.
I could absolutely see a confirmation bias issue showing up in cases like this. By the time you've spent $1400 on motherboards and lots of time and resources testing them to try and make them explode, you _really_ want to be able to produce some content based on it. I could see less scrupulous people doing some crappy things.
I appreciate any GN content. It’s still interesting to see the results (even if it’s not the ‘exciting’ result). Love y’all’s thoroughness - and will watch any update videos. Appreciate your work
There’s a running concern in academia called publication bias where only positive results are published. Releasing these tests that don’t work out instead of burying them may help reduce the general overreaction from the public whenever some random issue blows up on the internet. Perhaps people will be more patient and wait for more conclusive testing after seeing enough negative results.
Very awesome to have quick little updates like this on longer term things in the works / lack of results / other plans for the office etc. Really interesting to see how you all work.
This is def main channel content. Make a video featuring the guys who did the testing and give a recap of their methodology and their findings. Even though they didn’t blow up it’s still interesting.
Thank you for taking the time to do this video, you had the right idea: all that money and work needs to be recognized, and it doesn't matter it didn't yield alert-worthy results.
This has probably already been said but I think it's still super important to talk about when an issue isn't happening despite outside claims. Even if it won't get the most views I love the idea of mentioning these things, though I feel if it's only mentioned in the hardware news it could be glossed over when someone is looking for it
I actually found this quite interesting. In particular, I appreciate the commentary on the approach you took to testing, as well as how you recorded data and attempted to record evidence of the failure.
Sometimes you gotta know when to throw in the towel and stop giving in to the sunk cost fallacy. However it is commendable that you guys stuck with it for this long, trying to get it to fail. Even if the issue is long resolved it’s still interesting to see what went wrong in the first place and what the potential failure mechanism was. As other people have said, a negative result is still a result, and it’s super interesting that your sample did not fail after all these months where other people’s sample failed very quickly. Thank you for updating us on this project even though the board is still working, which is almost as interesting as if the board had failed. (Also if one of those gosh dang bots instantly replies to this comment…)
Long time ago my old FM2+ board was NOT actually compatible with the hyper 212 OG. I forced it too work, and bent that Board so badly it made cracking noises and couldn't mount stand offs anymore. It was a very noticeable C shape. It still works today. If I ever unmount it's gonna give up I'd bet.
As a customer that bought 2 Asus motherboards for my office and home rig that I assembled myself, this fills me with confidence and relief knowing that I'll have good millage out of it for years to come. Keep the good news coming 👍
The old saying in electrical engineering is simply "5-5" it means faulty electronics fail in either 5 minutes or 5 years with no inbetween with experience its more like 3 days or 3 years, but that might be worth keeping in mind when testing things for failure i think this problem is the capacitors that were factory tested exploded, the ones that were not factory tested were fine. Electrolytic caps go through a process called forming over 100 hours or so and in the process of forming the cap is polarized, this cap was simply never powered on the right way around and formed backwards if a backwards formed capacitor has any disadvantages im not sure, i dont think anyone ever cared to try
I would love if stuff like this could get a small mention in a HW news once it's over as well, and then basically keep it to saying that nothing happened, and to check GN Extras for context / the full non-story.
But this Is a good thing to post as well as your other stuff because uploading information on when products are not good vs when they are good & when they fix an issue gives consumers a better scope on whats a good product & what they should & shouldnt buy.
Thanks for all your effort GN Team! Even if the results did not pan out in this instance, we always look forward to your journalistic integrity raising the standard for information for us, the average consumers!
Thank you very much for your hard work, time, and money spent on investigating this The data gathered is still valuable! It very possibly could have been a multiple points of failure issue that most motherboards wouldn't have issues with I'm glad you're going to be publishing stuff like these now so we can at least see the fruits of your work which is still very much valid
Sensational or not a result is a result. Thanks for all the hard work you guys put in. I like seeing videos like this that just talk about the testing done.
Fascinating that a solid cap placed backwards can withstand that usage. If they were still using regular electrolytics I'm sure it wouldn't have lasted.
I think it would be a good idea to release a semi-regular (for example, quarterly or twice a year) video about your lab updates. You could talk about what new long-term projects there are, if there are any updates on other long-term projects, show what's new in the lab, talk about testing methodology et cetera. That way it is still a decent content piece even if the testing didn't produce any sensational results. It's still a very valuable information, and I'm sure your viewers would find it a very engaging.
On a similar note i'd love to see a video on how temperature / voltage actually impact the lifespan of parts. So much effort is spent investigating how we can lower temps by a few degrees but virtually nothing investigating whether it actually matters.
This is actually good findings and reassuring that the Asus Z690 board is actually pretty rock solid despite the component placement fault that some had. I wonder if it just prevents the quick USB charging? As in, it won't deliver above USB spec power current as a quick charge port should. (BTW, as Steve mentioned, Asus will still fix your board if it fails due to the flipped cap until the warranty period is up.)
I bought a Maximus Xi Formula years back and still have yet to take the shield off the board knowing there are TONS of bugs under the shield, bought it from a dude who lived in grasslands so he had tons of bugs even around his door 😅
>Ultra high-end mobo doesn't spectacularly fail. Well good! Also shows like the 4000 series power connector that failures like this are extremely rare and not something the end user should honestly be concerned about.
Some outlets do private / paid testing for that reason, but we just don't accept money for it because the ethical complications become difficult to navigate.
Having no exciting result here is actually a really good thing so I like to know about it. Hey this board survived all of our testing for over a year. Pretty good stuff.
Sometimes with power circuit issues like these running a long term test simply doesn't create the failure effect. For instance capacitors in power circuits are generally stressed the most when you first turn on power. So instead of running a long term test you need to run a series of tests constantly cycling the power off and on, say on for 5 minutes and off for five minutes (Basically you need to test to see how long it takes for the capacitor to discharge after power off to determine the proper off cycle time). I would be willing to bet that nearly all the boards that blew up did so when the power was turned on. Although it's a different scenario anyone that has worked restoring old tube guitar amps that have sat for years without being powered up need to hook them up to an isolated variac and slowly bring up the voltage over a period of several hours so the caps can reform (Not as big of a problem with modern caps) or the caps will literally blow up. I've had amps brought in where when you open them up it looks like someone had a party inside with a lot of confetti as a lot of older caps where made from paper
Maybe you could put up a live board cam for certain tests on here. That'll show people what you're doing and still sounds more interesting than half the content on RU-vid. Obviously you'd mute the audio but it would be entertaining to hear Steve arguing with motherboards in the middle of the night.
It's possible that the cap was printed/marked backwards and that someone somewhere on some assembly line was aware of the bad batch so installed them "backwards" and it only became a problem for the few that made it through with truly backward caps when they repeated the quick fix on a good batch of caps. It would be a mess to prove that, but it would also explain why it seems to be very limited in the number that were affected vs the number that were available with the defect.
I work in automotive electronics, I have seen reverse caps instant fail functional test, I have seen them limp along hundreds of hours in a thermal chamber (you need a tenny Jr btw, its 120 volt and can fit most mobo's) and years down the road as a warranty recall hanging off the ass end of a car
Great job! I am probably not alone in saying that lot of the daily driver/productivity crowd keep their PCs for many years and find this info VERY useful. Unfortunately both motherboard manufactures and PC RU-vidrs require a constant flow of new products and built-in-obsolesce to survive. Also Big Kudos for testing USB!. Although people like to get over-excited about VRMs I find that onboard peripheral chips fail far far more often than anything CPU related. My old MSI board is still running but the front panel USB chip has died, the built in Ethernet has died etc... P.S This is also why we need more PCIe slots and not just a bunch of M.2 slots....
This is actually quite interesting to see and think about. The capacitor that was affected is an SMD aluminium polymer electrolytic right? Aluminium electrolytic capacitors come out of assembly stages nearly entirely symmetrical, and are then "formed", where thousands of capacitors get clamped to the same voltage rails and their voltage is gradually slowly over hours pumped up to the test limit which is a little higher than the desired voltage rating. It's at this forming stage that they gain both their polarity and their final capacity. Slow forming lowers defect rate. When the capacitors are subsequently reverse biased, a few things can happen. They can and often do instantly explode, but that obviously hasn't happened to shipped units since the PCBs were factory tested. They might fail silently and become open-circuit, and the rest of capacity present in the net takes up the slack with some undesirable stray currents. It's possible that they survive with moderate damage and still provide some capacitance. Or the capacitors can even get re-formed, so the final capacitor manufacturing step got repeated in reverse and now you potentially have a nearly normal reverse-labelled capacitor. It shouldn't always work, as prior to re-forming, dielectric absorption needs to be deep depleted, and good forming is very slow, but by all reason, on the boards that shipped out, it's a possibility that they got rapidly reformed somewhat successfully. Indeed gradual voltage ramp-up is a recommended one-time post-soldering factory procedure for some power components. It can help components recover from previous shock and reduce in-field failure rate. Additionally the startup mode of many PC power supplies may be considered - in order to limit inrush instability, they ramp up the voltages over the not inconsiderable fraction of a second while holding the system in forced reset, taking it out of reset with PS_ON signal. Also capacitors formed in reverse from the factory happen as well, and that happens more often than some capacitor manufacturers are willing to admit.
It's good to know the result of this so in the future when we choose products we can see these videos about quality, endurance, performance of a product.
Given the focus here on standardised measurements and units, I'm surprised to see the American date format displayed on screen, rather than the ISO standard to equally annoy/please all viewers whilst using a format that actually makes sense.
tbh caps being backwards is way more common than expected, first thing that comes to mind is the first 3DO console, basically had a cap backwards, and only now we're seeing a lot of them dying and all of them bc of that (which is great if u know the infomration and are a bit skilled at soldering), same with i think the sega cd i think? or some 90s cd based system i dont remember and i think even a mac of some kind too? Caps usually dont really care if they're backwards at first but after many many many hours of usage they start to and might leak or whatnot easier, some take weeks, others months, other years and other might never have a problem, i think it might be also related with the production quality too, if one is already barely into spec theyre probably more prone to leak bc of this, and also temperature probably? Tbh if yall want to keep these for test benchs and whatnot and reuse them long term, it might be a good idea to change the cap for a new one and put it in the correct orientation, tho one suddenly blowing up on camera without anyone expecting might be good content so... Either way, cool to see this kinda behind the scenes content and appreciate how much more there is to this channel
After 2 MSI boards died with 2 years, and no ASUS ever has -- I'm sticking with ASUS. I do live in what some would call a sht-hole country... but still, ASUS all the way!
