Vendor-specific issues are the worst

[cst1992]

I don't know if this is the correct forum for this post (Hardware/System Builder's Advice just didn't seem appropriate as it's not strictly hardware and/or system; mods feel free to move to another forum as appropriate) but here I go anyways.

This is the second time a vendor-specific issue has caused me to pull half my hair out.

First, I've been dealing with unexplained crashes in a new AMD AM4-based system that has an Asus motherboard in it plus a Ryzen 5 5600X.
After a month of running okay, it just froze solid in the middle of games without even spitting out a bluescreen.
Troubleshooting went from changing out the RAM which happened to be a non-QVL non-kit, system corruptions leading to multiple OS reinstalls, multiple memtests to no avail, CPU checking software, etc etc.
The problem turned out to be (most likely) PBO Fmax Enhancer - an Asus-specific feature that is not even supposed to be used on 5000 series processors, because they would misbehave!
I was on the verge of throwing out my $350 motherboard and getting a similarly-priced AsRock one - which definitely would have fixed the (non-AsRock) issue, leading me to junk a perfectly working board!

Second one is when I tried to install Grub on my new Windows 10 laptop running Intel Optane (this was back in 2019 when Intel was peddling Optane to gullible layman Intel loyalists). Try hard as I might, I just couldn't load the bootloader! It just went straight to Windows!

A couple hundred lost hair later, I realized that Intel had TWO copies of the EFI - one on the Optane drive and one on the hard drive, and all I had done was replace the C: drive one with the bootloader, which was getting BYPASSED BECAUSE THE OPTANE ONE WAS THE ONE GETTING LOADED UNLESS YOU JUST WENT OUT AND DISABLED OPTANE ENTIRELY! AAAAAAAARGH!

 

[Macro Device]

I also had a little bit of wombo combos with ASUS. Two of their LGA1155 motherboards failed to update their BIOS (with two USB sticks, they considered the new firmware invalid; with one stick, they pretended to flash it; with the last stick I tried, they didn't even see the file). All I wanted is to be able to make my 3770K working properly (OC didn't work; power management didn't work). One of these mobos is P8P67, don't remember another's name.

Other brands had silly misbehaviours as well but not so ridiculous and game breaking.

Like, why? Why don't you test your wares before selling them? Most vendors seem extremely allergic to doing so. Especially the game developers...

 

[Frick]

Asus motherboard also had over voltage protection thing that ... I never knew if it worked or not, because I can't physically monitor voltages 24/7, but it did trigger seemingly randomly.

 

[cst1992]

Like, why? Why don't you test your wares before selling them? Most vendors seem extremely allergic to doing so.

Because management wants to SELL MOAR STUFF and QA wastes time...

I'd be seriously skeptical of going for Asus boards in the future. The era of them "just working" like my good old Z97-A is sadly over...

 

[Bill_Bright]

Asus motherboard also had over voltage protection thing

I believe you are thinking of their "Anti-Surge" feature. This feature is "in theory" supposed to protect the motherboard (and mounted components) from excessive surges from the power supply. Well, fine but why would there be excessive surges coming from a power supply? Yes, power supplies can fail catastrophically, and take out everything connected to them, but that is a very VERY rare event. Even no-name generic PSUs are pretty good at quickly terminating output in the event of such a failure.

IMO, the ASUS Anti-Surge feature causes more problems than it is worth. I don't understand why it is still around.

It should be noted that virtually all motherboards, even budget ones, have pretty robust regulation circuitry which includes components that are designed to protect themselves from surges. Plus, PSUs are specifically designed to isolate the output voltages from incoming surges and spikes. And if the computer is powered through a decent surge and spike protector, or better yet, a “good” UPS with AVR, this feature just become even more redundant, at best.

I think we need to understand that typically when a motherboard maker comes up with a new good idea, the other major makers soon follow suit. ASUS motherboards with ASUS Anti-Surge have been around at least since 2011 yet Gigabyte, ASRock, MSI and the other major manufacturers have not implemented their own versions of that feature. You have to wonder, or at least I do, why? It can't be a patent infringement thing because it is easy to create slightly different versions of a feature that does the same thing without violating or infringing on patented designs. So, to me, there can just be no other reason; the competing motherboard makers don't see the feature as needed. And I certainly agree.

If the ASUS Anti-surge feature proved itself to be worthy suggesting anti-surge protection at the motherboard level was needed, consumers and the IT tech sites would be demanding other motherboard makers implement it too. But that is not happening.

So, I recommend you disable that feature and see what happens. If you see no difference, then fine, enable it again.

But I will note, there are many reports in the forums where the ASUS Anti-Surge feature shut down a computer and reported an excessive surge. Yet no other evidence of surges existed and disabling the feature resulted in the computer working fine thereafter.

 

[freeagent]

IMO, the ASUS Anti-Surge feature causes more problems than it is worth. I don't understand why it is still around

I think its an Intel thing. It was on my last Asus Z77 board, but in none of my ROG AM4 boards.

 

[Bill_Bright]

I think its an Intel thing.

Nah! The ASUS Anti-Surge feature is on ASUS AMD boards too, and has created problems for those folks as well. I was able to google and find several. I will say this - just looking through the ASUS website, there does not seem to be as many current models with the feature. Maybe ASUS got the message!

 

[bonehead123]

Well, practically every mobo maker (and all other parts too) has or has had certain models/versions/features that have or have had certain issues from time to time, AsSus included....

I've used just about every brand of computer component out there in the past 15 yrs or so, and this has been my experience... sh*t happens, be it from lack of testing, sloppy/non-existent QA/QC, a bad batch of components, damage in transit, or any number of other potential causes

And yea, crashes are probably the most frustrating & hard to pinpoint of them all, but it's not the end of the world, so keep on keeping on & be happy

 

[ShrimpBrime]

Asus motherboard also had over voltage protection thing that ... I never knew if it worked or not, because I can't physically monitor voltages 24/7, but it did trigger seemingly randomly.

The infamous "USB over current detected, shutting down in 15 seconds" ......

This usually happened because of an open fuse or other possible issue on the circuit that would cause an open.

In some cases I remember the resolve was not actually on the board in others it was the board.

Since 99.8% of motherboard users cannot diagnose and repair these circuits, the boards went to the trash. Anything thats broken can be repaired.

___________________

Back in the day, DFI lan-party boards where known for 4 LEDs of death. I've had several simply give up the ghost for no apparent reason other than I was overclocking. RTC chip probably the culprit if I where to gamble a guess.

There was a very specific way to clear cmos on these boards if the regular way didn't work. I cannot recall all the steps, but it involved holding the insert key or something or other.

 

[Bill_Bright]

sh*t happens, be it from lack of testing, sloppy/non-existent QA/QC, a bad batch of components, damage in transit, or any number of other potential causes

Very true. And until Man can create perfection 100% of the time, sh*t will always happen.

But that is different from a "feature" that has been around for years that consistently demonstrated problems and/or where the buying public just didn't accept it (and often, where the manufacturer refused to admit it).

Don't get me wrong. ASUS is certainly one of my preferred brands. Bonehead is right - every company has had their Edsels.

 

[chrcoluk]

I believe you are thinking of their "Anti-Surge" feature. This feature is "in theory" supposed to protect the motherboard (and mounted components) from excessive surges from the power supply. Well, fine but why would there be excessive surges coming from a power supply? Yes, power supplies can fail catastrophically, and take out everything connected to them, but that is a very VERY rare event. Even no-name generic PSUs are pretty good at quickly terminating output in the event of such a failure.

IMO, the ASUS Anti-Surge feature causes more problems than it is worth. I don't understand why it is still around.

It should be noted that virtually all motherboards, even budget ones, have pretty robust regulation circuitry which includes components that are designed to protect themselves from surges. Plus, PSUs are specifically designed to isolate the output voltages from incoming surges and spikes. And if the computer is powered through a decent surge and spike protector, or better yet, a “good” UPS with AVR, this feature just become even more redundant, at best.

I think we need to understand that typically when a motherboard maker comes up with a new good idea, the other major makers soon follow suit. ASUS motherboards with ASUS Anti-Surge have been around at least since 2011 yet Gigabyte, ASRock, MSI and the other major manufacturers have not implemented their own versions of that feature. You have to wonder, or at least I do, why? It can't be a patent infringement thing because it is easy to create slightly different versions of a feature that does the same thing without violating or infringing on patented designs. So, to me, there can just be no other reason; the competing motherboard makers don't see the feature as needed. And I certainly agree.

If the ASUS Anti-surge feature proved itself to be worthy suggesting anti-surge protection at the motherboard level was needed, consumers and the IT tech sites would be demanding other motherboard makers implement it too. But that is not happening.

So, I recommend you disable that feature and see what happens. If you see no difference, then fine, enable it again.

But I will note, there are many reports in the forums where the ASUS Anti-Surge feature shut down a computer and reported an excessive surge. Yet no other evidence of surges existed and disabling the feature resulted in the computer working fine thereafter.

I am always wary of vendor specific features now.

On my board is a few ASRock specific options, one related to some kind of special optimised ram profile, and another related to the CPU, I ignored both of them. There was a 3rd one also related to ram timings, out of curiosity I tested it, and there was no adjustments to ram timings at all in Windows.

Not a fan of MCE either, something Intel doesnt support and CPU's are out of spec with it on. I see it different to voltage and clock speed adjustments as some bios has it on by default, and I feel it needs a big warning when enabling it.

 

[Bill_Bright]

On my board is a few ASRock specific options

Me too - especially if they require me to download some vendor specific (read: proprietary) software (other than basic drivers) to make it work or to enable those features.

 

[cst1992]

If the ASUS Anti-surge feature proved itself to be worthy suggesting anti-surge protection at the motherboard level was needed, consumers and the IT tech sites would be demanding other motherboard makers implement it too

I don't know if this fits the bill, but my new AM4 motherboard has a USB-C front-panel port. Once I plugged in a faulty USB-A-to-C dongle on it to connect my mouse to, and it disabled the port, and Windows popped up a "device is drawing more power than the port can supply" message. Obviously this must have been due to a short in the dongle.
Motherboards these days are pretty good overall already. Once I splashed water on my phone and Android disabled the USB port saying there was liquid or debris in the port, and that the port was disabled until the issue resolved. I blew some hot air on it for a few minutes and the port was automatically unblocked by the OS. Pretty impressive.

 

[Bill_Bright]

and Windows popped up a "device is drawing more power than the port can supply" message.

Are you sure it was a "Microsoft Windows" error message? I don't see how the OS would know unless the HW told it so.

 

[cst1992]

I'm pretty sure, as it popped up from the system tray. Some sensor must have told Windows so.

Almost cut-to-cut similar to this:

https://www.reddit.com/r/simracing/comments/wx1n00

 

[Bill_Bright]

Okay. I've heard of such errors, just never seen one myself.

I agree, s single USB device should not exceed the capability of the port. A port should be able to support a hard drive, which to me, would be the most demanding device that didn't need its own power supply. So unless you were using a port powered hub with multiple devices connected to the hub, then I agree, it must have been a faulty dongle - assuming the port still works fine otherwise.


Edit comment: fixed a type and added a clarification - b_b

 

[Klemc]

ASUS :

Extreme Engine Digi+
- 5K Black Metallic Capacitors
ASUS Q-Design
- M.2 Q-Latch
- PCIe Slot Q-Release
- Q-DIMM
- Q-LED (CPU [red], DRAM [yellow], VGA
[white], Boot Device [yellow green])
- Q-Slot
ASUS Thermal Solution
- M.2 heatsinks
- M.2 heatsink backplate
- VRM heatsink design
ASUS EZ DIY
- BIOS FlashBack™ button
- BIOS FlashBack™ LED
- Clear CMOS button
- ProCool
- Pre-mounted I/O shield
- SafeSlot
- SafeDIMM
Aura Sync
- Aura RGB header
- Addressable Gen 2 headers

 

[cst1992]

it must have been a faulty dongle - assuming the port still works fine otherwise.

Yup, I'm using another dongle in the same port with no issues.

 

[cst1992]

First, I've been dealing with unexplained crashes in a new AMD AM4-based system that has an Asus motherboard in it plus a Ryzen 5 5600X.
After a month of running okay, it just froze solid in the middle of games without even spitting out a bluescreen.
Troubleshooting went from changing out the RAM which happened to be a non-QVL non-kit, system corruptions leading to multiple OS reinstalls, multiple memtests to no avail, CPU checking software, etc etc.
The problem turned out to be (most likely) PBO Fmax Enhancer - an Asus-specific feature that is not even supposed to be used on 5000 series processors, because they would misbehave!
I was on the verge of throwing out my $350 motherboard and getting a similarly-priced AsRock one - which definitely would have fixed the (non-AsRock) issue, leading me to junk a perfectly working board!

An update to this: crashes happened regardless of PBO Fmax Enhancer. If processors are misbehaving because of it, the symptoms must be different than what I experienced.

The real cause of my issue turned out to be a bad WiFi/Bluetooth card (my motherboard is a WiFi variant) which started to fail more and more often and was causing system crashes/corruption as a result. I'm not using it now, and the system has not frozen once in 3 days of continuous use.

I still absolutely blame Asus for this though. Maybe I got a lemon, but they should have been using more reputable manufacturers' products from the get-go.

 

[tabascosauz]

An update to this: crashes happened regardless of PBO Fmax Enhancer. If processors are misbehaving because of it, the symptoms must be different than what I experienced.

The real cause of my issue turned out to be a bad WiFi/Bluetooth card which started to fail more and more often and was causing system crashes/corruption as a result. I'm not using it now, and the system has not frozen once in 3 days of continuous use.

I still absolutely blame Asus for this though. Maybe I got a lemon, but they should have been using more reputable manufacturers' products from the get-go.

Fmax enhancer the Asus "feature" has never worked for me on past few CPUs. Pretty much insta-crashes under load. It has negligible value anyway for whatever one might want to do with PBO, whether it's all-out OC or undervolting. It's not like it magically unlocks Fmax headroom beyond +200 either, so as far as I'm concerned it's pretty much worthless.

Intel AX200/201/210 have generally been solid on all boards AMD/Intel, but Asus put in a new MediaTek card in your board (which is a "new" AM4 product). It might necessitate taking the VRM shroud off to access but you might still be able to replace it with an AX200. These days the M.2 wifi card is usually encased in a metal box in the rear I/O, but there might be screws to open the box.

I think it's the same AMD/MediaTek collab RZ60x card that replaced AX200 in all AM5 boards, as well as some Rembrandt/Phoenix APU laptops. Basically, it wasn't to improve performance, it was because AMD wants to be seen having control over all aspects of its platform. Honestly, it's a travesty. It sucks, and has been proven to suck in countless laptop reviews. I can live with mine in the Zephyrus as I don't use my laptop often, but even so it's frustratingly slow to connect after sleep.

As to vendor specific issues, I was bit by Gigabyte recently, seems many if not all of their AM4 ITX (at the very least) boards are incapable of pushing the 75W spec through their x16 slot to power a RTX A2000. Dunno if it'll ever get enough traction to be brought to Gigabyte's attention.

 

[cst1992]

Asus put in a new MediaTek card in your board (which is a "new" AM4 product). It might necessitate taking the VRM shroud off to access but you might still be able to replace it with an AX200. These days the M.2 wifi card is usually encased in a metal box in the rear I/O, but there might be screws to open the box.

I think it's the same AMD/MediaTek collab RZ60x card that replaced AX200

Right in one on both. It's a rebranded MT7921 card - RZ608. I have ordered an AX210 to replace it.
The shroud has screws, but it requires taking the motherboard out of the case.

but even so it's frustratingly slow to connect after sleep.

I could live with it if it were limited to that, but taking down the whole system? That card is going straight to the bin.

 

[tabascosauz]

Right in one on both. It's a rebranded MT7921 card - RZ608. I have ordered an AX210 to replace it.
The shroud has screws, but it requires taking the motherboard out of the case.

I could live with it if it were limited to that, but taking down the whole system? That card is going straight to the bin.

Actually I'm not sure I would go for a AX210 at this point (even though it is the direct contemporary to the 6E RZ608), since Intel's BE200 is now out and seems to be a straightforward swap for most AX210 systems

The other issues with RZ608 seem to be platform dependent but the slow reconnection is a universal issue with it. Once it gets connected it tends to hold the connection fine, but this whole AMD/MediaTek venture has been forever soured in my mind

For RZ608 did you use Asus drivers or windows update fetched ones?

 

[cst1992]

I tried the ones on Mediatek's site, but they had issues so I tried one version older and that worked fine, apart from the freezes.

The drivers are not the issue - the card itself failed to the point the drivers can't even be installed anymore.

 

[OliverQueen]

Nah! The ASUS Anti-Surge feature is on ASUS AMD boards too, and has created problems for those folks as well. I was able to google and find several. I will say this - just looking through the ASUS website, there does not seem to be as many current models with the feature. Maybe ASUS got the message!

Cannot remember seeing it on my old ROG Maximus X Hero Z370 board either, nor the TUF Sabertooth Z87 board that just retired after 10yrs 24/7/365 on-time. Been a long time since even looked at full features though of those boards so if was present, I never realised or needed it obviously.

 

[Bill_Bright]

so if was present, I never realised or needed it obviously.

Pretty sure it is enabled by default. However, for most users, and I will assume for you too, your PSU didn't throw off any excessive surges that triggered the feature (no surprise with a quality EVGA - my preferred brand too! ). But even basic/generic PSUs should be more stable with decent, regulated output in compliance with the ATX Form Factor requirements and be able to suppress and isolate excessive transits sufficiently enough for most motherboard regulator devices to deal with easily. This is even more so if the computer is running through a decent UPS with AVR or even a decent and properly functioning surge and spike protector.

So, if yours never triggered (which is typical) you likely didn't and never would be aware of it. I guess that's a good thing but it begs the question. If not needed, do we really need it? And, as is apparently and frequently the case, if it throws off false positives that cause problems and unnecessarily alarm unsuspecting users, do we even want it?

All I can say for certain is that I am personally aware of several ASUS board owners who received an ASUS anti-surge warning message, yet everything tested just fine. They disabled the feature and their computers continued to give them years of quality service.

I will say, if you get such an error, don't just ignore it. Test your PSU, swap in a spare if you have access to one and test your wall outlet to ensure it is properly wired AND grounded to Earth ground. If all checks good, then you might consider disabling the feature.