|
BlankSystemDaemon posted:Part of it is that it's perfectly statistically likely that a lot of people who've habitually bought Asus have ended up with boards that've had no systemic problems - which effectively sets up selection and frequency bias. BUT If I build a system, 1-9 months later a component dies, I replace and it go on to use the system for a much longer period of time with no issues that also tells me Asus isn't worth it. There is almost certainly some confirmation bias in that, but I can also say that since I switched from ASUS as my preference, I haven't had any non-personally caused hardware failures.
|
#
?
Apr 22, 2023 14:40
|
|
|
|
| # ? May 16, 2024 13:05 |
|
|
i'm genuinely reluctant to take the stuff people say about asus gpu's at face value for the above reasons, but i also don't think the biases are as strong with products outside the gpu range. mobos are pretty much these boring things people just have to get for the whole build, much like ram, and you'll rarely be able to see any of them anyway beyond basic rgb asus is to gpu's what ekwb is to watercooling if you will, with the same caveats
|
|
#
?
Apr 22, 2023 14:43
|
|
|
Enos Cabell posted:my Opteron 165 build oh man, that sounds rad. A Sempron or a Duron build would also be a cool retro project imo
|
|
#
?
Apr 22, 2023 14:48
|
|
|
repiv posted:and then ASUS started quietly deleting all the old AM5 BIOS downloads from their site, leaving only the newest version I was on version 1406. I noticed a version 1408 came out recently, and IIRC it had the line "Improved X3D compatibility" or some such in the release notes. I like how they rephrased it into "Optimum performance" in the latest 1409. Also, does Ryzen Master stop bitching about Hyper-V nowadays? I'd like to play with the curve optimizer.
|
|
#
?
Apr 22, 2023 15:31
|
|
|
I run the Asus X670E plus Wifi with a 7800X3D and it's been an absolutely fantastic board so far. The component quality for the price is crazy (https://www.youtube.com/watch?v=zAAqUk15ryM), overclocking ram is rock solid, and you can often get it for a huge discount on Amazon Warehouse. This is because people buy it, aren't tech savvy enough to do the BIOS flashback (a low bar, but one that many can't clear). I got mine for like $200, which is unbeatable, and I highly recommend this route anyone who is looking for AM5 on a budget and is ok with rolling the dice a little. Very impressive components, PCIE5, SSD gen 5, it's great. There's no Warehouse stock at this exact moment but sign up for inventory emails on a place like camelx3. Thanks for the heads up about the firmware though, that's scary. I'm updating right now. slidebite posted:If I build a system, 1-9 months later a component dies, I replace and it go on to use the system for a much longer period of time with no issues that also tells me Asus isn't worth it. There is almost certainly some confirmation bias in that, but I can also say that since I switched from ASUS as my preference, I haven't had any non-personally caused hardware failures. Consumers can and do have preferences but I've never understood this line of thought. Your personal experience is irrelevant even if the component failed twice, beyond saying "I will personally not buy this". This is something of a point of frustration to me because you get this a lot in my general line of work. People getting crazy mad and leaving reviews all over the internet because their personal component failed. We put out an excellent product, but you'll never get people to understand that no manufacturing process is going to yield 0% errors. Some will slip through QA, and it's ok; by all means note your issues online but the length some people will go to tarnish a tech product just because they personally got a faulty one is kind of unreal. Tech, especially on the high end, is absurdly complicated. It's actually a minor miracle that the failure rates are as low as they are in this sector. Regardless I'm sorry you had that experience and no one should have to deal with that friend.
|
|
#
?
Apr 22, 2023 15:37
|
|
|
All good man. We all have preferences on what to/not to buy whether it's "legit" or not. I won't buy an Asus similarly to how I'll never buy a Chrysler product.
|
|
#
?
Apr 22, 2023 16:02
|
|
|
Ryzen Master is resetting my RAM overclock back to DDR5-4800. loving hell what even. —edit: Meh, not worth the hassle.
Combat Pretzel fucked around with this message at 16:08 on Apr 22, 2023 |
|
#
?
Apr 22, 2023 16:05
|
|
|
slidebite posted:Yeah, I totally understand a selection/frequency bias and it's absolutely a thing. There's simply no statistical significance to it. It could maybe be crowdsourced, but it's a hell of a lot of work - and even then, having people keep track of specific revisions (assuming PCNs are even done, which isn't a given) is going to be a nightmare. EDIT: Maybe a consumer protections bureau could handle ingesting user-submitted data, but are users interested in submitting data? But in order to control for different markets (Americas, EMA, Asia, et cetera), you'd have to have multiple consumer protections bureaus working together across countries. BlankSystemDaemon fucked around with this message at 18:13 on Apr 22, 2023 |
|
#
?
Apr 22, 2023 17:32
|
|
|
Personally I have massive doubts about pretty much all blanket generalizations about brand quality, at least once you go above the garbage level. I've seen far too many times where people gaslight themselves or others into believing that component X is or is not responsible for their problems based on internet reputation. Dopes out there still saying "AMD drivers bad" if someone has any type of issue and happens to have an AMD GPU, even if the issue is unrelated. And I've seen too many "tier lists" that are junk in one direction or another. Right now I'd put a lot of weight for AM5 problems on the first-gen memory platform thing, rather than any brand. DDR4 and 3 were the same way. And because that's going to have CPU, mobo, and ram all involved it's easy to get "this one works fine, this one doesn't" and have that be an interaction rather than one particular component. BlankSystemDaemon posted:It could maybe be crowdsourced, but it's a hell of a lot of work - and even then, having people keep track of specific revisions (assuming PCNs are even done, which isn't a given) is going to be a nightmare. IIRC there was at one point way back when a user-submitted hard drive reliability survey that was tied to some old enthusiast site. IIRC it was extremely difficult to get unbiased results because most people only reported when they had failures. So NAS & server grade drives looked extremely good, better than they really were, because they were more likely to be used by hyper-enthusiasts who updated their reports every year regardless. I think the only really good conclusion it could ever show was "yep, IBM deathstars fail a whole lot". Which we probably didn't need a survey to figure out. 
|
|
#
?
Apr 22, 2023 19:19
|
|
|
Klyith posted:Personally I have massive doubts about pretty much all blanket generalizations about brand quality, at least once you go above the garbage level. And yeah, there's a very consistent and almost knee-jerk to take broad generalizations of a decade ago and apply them to the current situation. As for tier lists, I think the only ones that're worth using are the ones that're well sourced with trustworthy reviewers as the source - and at that point, it's no different than using Wikipedia for its references section (provided you check the references). Absolutely there are problems with my proposal, I wasn't actually making it with any kind of hope of getting anything out of it - more to illustrate, poorly perhaps, how hard it is to even begin to get right - and even once you've got the bureaucracy up and running, you're still relying on user-reported data.
|
|
#
?
Apr 22, 2023 19:28
|
|
|
Klyith posted:Personally I have massive doubts about pretty much all blanket generalizations about brand quality, at least once you go above the garbage level. There are also products where a particular model or generation are just plain bad, it happens to every manufacturer, sometimes it is impossible to predict a flaw or failure mode that won't show up in the normal testing and validation. Like take the Seagate ST3000DM001, I had one fail at just over 3 years and that particular model was bad enough there was even a class action lawsuit about it. I keep crystal disk info running in the background scanning every drives SMART stats once a day and it caught the impending failure of mine early so I was able to back up everything and replace the drive, and I replaced it with another Seagate drive that is still here working totally fine. I've had basically every brand hard drive, and I've seen them all fail in one way or another enough to know there is no substitute for backups and health monitoring. Unfortunately motherboards and video cards and stuff like that typically don't have as graceful or predictable failure modes as mechanical hard drives and are a much lower volume product so it is harder to tell the difference between a manufacturing defect of a single board or a straight up design flaw that affects all of them to varying degrees because you won't have blackblaze publishing statistics showing something obvious like "47% of these failed during the third year of operation".
|
|
#
?
Apr 22, 2023 19:46
|
|
|
I'm on the one Asus board (proart b650-creator) without a new BIOS update for this overvolting issue. On the other hand, I checked all my voltages and unlike the reddit users reporting 1.3-1.4 CPU/CPU SoC voltages in both bios and hwinfo, I see completely normal 1v and 1.1v values. I still want to see a new BIOS update but, to my best understanding, it seems like my board isn't overvolting the CPU. The x670 proart board did get a bios update so I can't speak for that, but I would guess the proart boards tend to be one step more "workstationy" than all the other consumer/gamer boards so aren't chasing benchmarks as much. Will still be watching for bios updates over the next couple of days but I might be in the clear. This might explain my lacklustre memory results compared to other people able to get 4800mhz stable at 128GB, if my SoC was running at 1.1v and I was comparing it against SoCs running at 1.35v. e: and a bunch of users reporting memory overclocks no longer working after updating their boards, which also tracks. Desuwa fucked around with this message at 20:14 on Apr 22, 2023 |
|
#
?
Apr 22, 2023 20:12
|
|
|
Indiana_Krom posted:There are also products where a particular model or generation are just plain bad, it happens to every manufacturer, sometimes it is impossible to predict a flaw or failure mode that won't show up in the normal testing and validation. Oh absolutely. Nothing we can really do about it other than watch closely to make sure it's not a pattern. Also, sometimes even companies with a decent reputation decide they don't give a poo poo and make something that should fail testing. I dunno profits are down this quarter or something. If you'd told me a few years ago that a major brand was gonna put out a PSU that was pretty much a time bomb, I wouldn't have picked Gigabyte as my first guess. And then they went hog wild on burning down their reputation.  Indiana_Krom posted:Like take the Seagate ST3000DM001, I had one fail at just over 3 years and that particular model was bad enough there was even a class action lawsuit about it. Heh, I had an IBM Deathstar and one of those. And the hosed up bit is that both times the crazy failure rates were a known fact that I was aware of. But mine kept trucking for so long that they lulled me into a false sense of security. Both times I lost data. The Deathstar I lost almost everything, the only backups I did back then was burning a CD of my docs folder a couple times a year. (We all need that one experience.) The seagate was just some minor stuff and downloads, but made me get more serious about my strategy because I'd been slacking off. Klyith fucked around with this message at 21:31 on Apr 22, 2023 |
|
#
?
Apr 22, 2023 21:27
|
|
|
Backblaze provides reliability reports for the drives they use: https://www.backblaze.com/blog/backblaze-drive-stats-for-2022/
|
|
#
?
Apr 22, 2023 21:47
|
|
|
Malloc Voidstar posted:Backblaze provides reliability reports for the drives they use: https://www.backblaze.com/blog/backblaze-drive-stats-for-2022/ Would be nice to see the total number of each drive per manufacturer in that graph, too. e: currently agonizing over which drives to get for a new NAS build. gently caress, it helps to actually click the link. Prescription Combs fucked around with this message at 01:23 on Apr 23, 2023 |
|
#
?
Apr 23, 2023 01:20
|
|
|
Prescription Combs posted:Would be nice to see the total number of each drive per manufacturer in that graph, too. Click the link. Backblaze are extremely comprehensive. The extremely interesting thing is that often the Prescription Combs posted:e: currently agonizing over which drives to get for a new NAS build. redundancy > the difference between a 1% AFR and a 4% AFR Buy the thing that's cheaper and put the money in the bank, now you can replace the drive that dies when it's one day past warranty.
|
|
#
?
Apr 23, 2023 01:31
|
|
|
Prescription Combs posted:Would be nice to see the total number of each drive per manufacturer in that graph, too. Also the steady climb in Seagate and Toshiba drive failures is mostly explained by those also being the oldest drives in service, so increasing failure rates are expected as many of them are just aging out. That chart is not exactly a good one to post out of context, although the whole article is filled with charts much like it, basically you have to read the whole thing to make any sound conclusions from it.
|
|
#
?
Apr 23, 2023 02:04
|
|
|
Malloc Voidstar posted:Backblaze provides reliability reports for the drives they use: https://www.backblaze.com/blog/backblaze-drive-stats-for-2022/ maybe seagate were secretly trying to bring down Huawei from inside with their lovely spinning rust
|
|
#
?
Apr 23, 2023 02:07
|
|
|
Indiana_Krom posted:Also the steady climb in Seagate and Toshiba drive failures is mostly explained by those also being the oldest drives in service, so increasing failure rates are expected as many of them are just aging out. That chart is not exactly a good one to post out of context, although the whole article is filled with charts much like it, basically you have to read the whole thing to make any sound conclusions from it. AFR is normalized for drive age. (What can produce a spike on those charts is when they buy a crappy model in 2018 and a bunch have failed out by 2022, that's when the overall manufacturer AFR will be highest.) SwissArmyDruid posted:maybe seagate were secretly trying to bring down Huawei from inside with their lovely spinning rust A thing to note is the backblaze uses a mix of consumer and enterprise-grade seagates, while for the others they use pretty much only enterprise. (Also note that backblaze continues to buy lots of seagates despite the AFR difference.) Klyith fucked around with this message at 02:35 on Apr 23, 2023 |
|
#
?
Apr 23, 2023 02:18
|
|
|
I've had quite a few MBs in the last 20+ years. Most of them were Asus (including the one I have right now, X670E-PLUS), 1 Gigabyte and 2 MSIs. The Gigabyte one, I had it back in 2001, and swore to never buy another again. MSIs are fine, but Asus were always just better. And right now, while I am running bios version 1408 didn't have any problem with the ones before, just upgraded because the internets said so. And with hard-drives, most of my spinning disks were Seagate over the years (Quantum Fireball -> Maxtor -> Seagate). Never had one die. Bought 2 WD drives (was around 2010s) , both died within a year. Volguus fucked around with this message at 02:38 on Apr 23, 2023 |
|
#
?
Apr 23, 2023 02:35
|
|
|
before they crashed and burned by the mid-00s Abit mobos were by far the best for Slot 1/Socket 370, Socket A and 478
|
|
#
?
Apr 23, 2023 03:00
|
|
|
Abit was good but EPoX was one of the interesting standouts from that time to me. Couldn't believe at the time they had to leave the market along with others like Iwill. PCChips enjoyed the coke dusted vibe that Asrock seems to put out these days. I don't remember them working well but they sure tried some odd stuff and were usually pretty cheap.
|
|
#
?
Apr 23, 2023 04:16
|
|
|
Palladium posted:before they crashed and burned by the mid-00s Abit mobos were by far the best for Slot 1/Socket 370, Socket A and 478 the NF7 and BH6 are probably two of the most legendary boards of all time. rip
|
|
#
?
Apr 23, 2023 04:16
|
|
|
Cygni posted:the NF7 and BH6 are probably two of the most legendary boards of all time. rip Oh, I would have thought it would be this one: ABIT AT7 . It was a bold move at the time, had a friend who bought it.
|
|
#
?
Apr 23, 2023 05:12
|
|
|
How about good ol' DFI? Their boards were pretty high end back in the day
|
|
#
?
Apr 23, 2023 07:59
|
|
|
HalloKitty posted:How about good ol' Diamond Flower International? DFI boards were pretty high end back in the day only had their brief heyday in the socket 939 days by the time core 2 rolled out hardly anyone is gonna spend $$$ on AMD K8s and pretty much every P965/P35 mobo had massive FSB OC headroom
|
|
#
?
Apr 23, 2023 08:10
|
|
|
Volguus posted:Oh, I would have thought it would be this one: ABIT AT7 . It was a bold move at the time, had a friend who bought it. "As you can guess all of these ports on the rear board don't really conform to any known ATX specification so ABIT ships a custom I/O panel with the board for the back of your case." Amazingly forward thinking.
|
|
#
?
Apr 23, 2023 09:43
|
|
|
Arivia posted:"As you can guess all of these ports on the rear board don't really conform to any known ATX specification so ABIT ships a custom I/O panel with the board for the back of your case." nothing that special when the back I/O plate is removable right from the very start of the ATX spec
|
|
#
?
Apr 23, 2023 09:51
|
|
|
Speaking of Asus reliability, I don’t know why but my 7900X + Asus B650E-F sometimes randomly hangs for a few seconds when I’m doing something as simple as opening links in Edge or opening/closing Calculator (Win10). It happened just now a few times; I had task manager openeing and the CPU wasn't loaded or anything. I just have stock EXPO timings so I have no idea what the issue could be. Really annoying on what should be a top end system. Josh Lyman fucked around with this message at 10:50 on Apr 23, 2023 |
|
#
?
Apr 23, 2023 10:44
|
|
|
Chiming in on ASUS reliability - I've owned a reasonable amount of their stuff and not had anything fail on me yet. This is anecdotal, of course, but if you don't habitually keep up with Computer News that's how personal preferences form. I used to build boxes more often than I do now a couple of decades ago, and I do remember being wowed by the packaging on an Asus mobo compared to the few previous builds. Black insert, gold letters. Very sleek. May have been a high spec board on discount. Came with all the doodads it needed to. Booted up perfectly with an Asus backround image from the get-go. This is all ridiculous in therms of "reliability" of course, but it left a good impression. If you're cutting all the corners you can, you might not ostensibly go through the trouble to do this poo poo. At least until you've built up an unshakeable reputation and are willing to burn it - a trick you should only be able to do once. With the same batch I had ordered a Sapphire video card, which cost more than the mobo, came in the equivalent of a paper bag and turned out to be DOA. I returned it and said I don't want another one, I'll pay you the difference for the (I think first-gen) DirectCU Asus version. That box worked flawlessly for a decade until it was tackled by a toddler. I've had a couple of tiny Asus (Atom) laptops, which are idiotic for having just 32gb of storage that can and will literally be completely used up by Windows if you leave it too long, but otherwise had magnificent battery times and have now "just worked" for near a decade. My wife's recent desktop has an Asus (AM4) mobo and I think GPU too, no problems in a year. Bottom line - I found a manufacturer that made things that worked. So long as they kept making things that worked, I wouldn't have cared if other manufacturers also made things that worked. I would have been happy paying 50 dollaridoos premium building a box every eight years to not have to do research on which of the other manufacturers are not rubbish in this hardware generation for this particular component, because besides them just working I don't give a poo poo. Now I'm finding out that Asus currently sucks at DDR5 memory management and might have a processor smoking habit. Not thrilled. Because I have a TUF B650 sitting in a drawer waiting for the rest of the components to arrive in a week or so. Edit: Similar feelings regarding the NVMe Samsung SSDs falling from grace. Thankfully found out before buying in. Hoping Kingston Fury is not poo poo but I guess I'll see. Duuk fucked around with this message at 11:47 on Apr 23, 2023 |
|
#
?
Apr 23, 2023 11:21
|
|
|
PC LOAD LETTER posted:Abit was good but EPoX was one of the interesting standouts from that time to me. Couldn't believe at the time they had to leave the market along with others like Iwill. epox was my first mobo with the boot code display unfortunately they got hit really really bad by the bad capacitor batches lol
|
|
#
?
Apr 23, 2023 12:20
|
|
|
Hey so has anyone been overclocking their X3D chip? I want to get the most out of the 7800X3D but it's difficult to understand what's actually possible. Some people even say you can't overclock it, but other guides you clearly can. I don't know what I'm doing, could anyone lay out some basic steps here? I do run a 6000mhz 30cas latency expo overclock. That's about it. Temps are not a concern because I have the 7800X3D connected to a three fan AIO heatsink monstrosity that couldn't be more excessive vs. the heat output of the chip. Given temps should be stable, how much room does that allow? I'm reading that the 3D caching somehow makes things a lot harder? Klyith posted:Personally I have massive doubts about pretty much all blanket generalizations about brand quality, at least once you go above the garbage level. If you think about it, a huge part of society's problems are due to people judging something (an idea, product, situation) based on an inherently limited personal interaction with it. Very few people will go around the internet actively promoting a product or idea, but people will sure as poo poo run around trashing a product on every single informational outlet if they have a bad experience. The route to progress is never straight, ya know? By being so harsh on even small failures (see: the public perception of the lower-end card AIBs) we solidify the status quo. For example most people stop seeking out, understanding and enjoying new music fairly early in their lives. Every generation thinks that most or all music that came after their formative years is poo poo. They're just not able to appreciate new things anymore and we view that as a normal part of becoming an adult. You're expected to close your mind to the point where it's socially weird to be into "young people music" as an adult. Extend that general sentiment to literally everything and it's easy to see how much it's hosed us over.  Sir this is a Dollar General
|
|
#
?
Apr 23, 2023 13:01
|
|
|
Taima posted:Hey so has anyone been overclocking their X3D chip? I want to get the most out of the 7800X3D but it's difficult to understand what's actually possible. Some people even say you can't overclock it, but other guides you clearly can. I don't know what I'm doing, could anyone lay out some basic steps here? Here's a video: https://www.youtube.com/watch?v=90UBUq1mLGY BlankSystemDaemon fucked around with this message at 13:14 on Apr 23, 2023 |
|
#
?
Apr 23, 2023 13:06
|
|
|
I still smile when I think of all the FiC va-503+ systems I built for my friends.
|
|
#
?
Apr 23, 2023 13:08
|
|
|
Taima posted:Hey so has anyone been overclocking their X3D chip? I want to get the most out of the 7800X3D but it's difficult to understand what's actually possible. Some people even say you can't overclock it, but other guides you clearly can. I don't know what I'm doing, could anyone lay out some basic steps here? I'd watch this: https://www.youtube.com/watch?v=90UBUq1mLGY I'd look at option #3 that starts at 14:43. Its not too hard to do and it still can get you some gains. The other options either don't get you much or are fairly difficult AND require a decent degree of luck (ie. ECLK overclocking) to pull off. Taima posted:I'm reading that the 3D caching somehow makes things a lot harder? Die stacking can have an effect, but it looks like AMD nerfed the overclocking on the 7800x3D to make the 7900/7950x3D more attractive to buy. The pros have been able to get the 7800x3D to 5.3-5.4Ghz or so on good air or AIO's via ECLK so there is some head room above 5Ghz (which is what they're limited to by AMD) there. The problem with ECLK increases is that it tends to make the system unstable quite quickly since mobo's it'll effect the PCIe bus. The PCIe bus is really touchy and normally won't budge much at all. Like 1-3Mhz increase is typical. Some AMD mobo's let you adjust just the CPU's bus clocks (ECLK 1 mode) alone though which is how they're pulling off 12-14Mhz+ increases on it and getting to 5.3-5.4Ghz with air or AIO's. edit: doing it all manually does suck but Ryzen Master has a auto curve optimizer and I think Hydra still works too. Just takes a long time to run. Like 8+ hours or something. PC LOAD LETTER fucked around with this message at 13:24 on Apr 23, 2023 |
|
#
?
Apr 23, 2023 13:11
|
|
|
stability testing curve optimizer tuning is a huge pain in the rear end because normal stress tests won't find all possible issues. To be honest, I'd just set expo on and forget it. it's not worth all the hassle just to get 5% more performance in a small handful of games.
|
|
#
?
Apr 23, 2023 13:13
|
|
|
Thanks yall really appreciate it. This is exactly what I wanted; it can be hard to separate the wheat from the chaff on youtube videos especially in scenarios like this where there are a lot of opinions.
|
|
#
?
Apr 23, 2023 14:44
|
|
|
3d cpu's perform pretty well with negative co offsets, but definitely run corecycler and stuff like that. i can easily run -25, but -30 breaks when ... my computer completely idles. which only happened because i had guests over that day. definitely not something easy to test for. for 5800X3D, -20 should basically be free, and then you can bump it up if you're feeling bold. overclocking's another matter god help anyone with more than eight cores to run tests on though
|
|
#
?
Apr 23, 2023 14:51
|
|
|
My 5800X3D had issues at -30 but runs fine at -25, which is good because mine appears to be an absolute silicon lottery loser that runs crazy hot no matter the cooling, even compared to other 5800X3Ds.
|
|
#
?
Apr 23, 2023 15:03
|
|
|
|
| # ? May 16, 2024 13:05 |
|
|
sounds like ours are cut from the same scuffed wafer
|
|
#
?
Apr 23, 2023 15:12
|
|