Caporegime
Its getting late, i'm getting tired, take a stab at it, i want to end this circular argument.
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
AMD Ryzen 7 7800X3D CPU Burns Up
- Thread starter Jamin280672
- Start date
Caporegime
You are, you are responsible for whatever happens to your GPU when you use third party overclocking tools, you use it, you break the GPU, its your fault. Not Nvidia and not the creators of the tool.
Can you at least tell me how that works?
Can you at least tell me how that works?
It's only going around and around because you make ridiculous comparisons. ASUS, MSI, Gigabyte etc aren't random third party nobodies they are AMD board partners under their direction if they don't follow AMD's direction then it's on the board partner for making rubbish software AND AMD for not QAing it, neither should get off looking good.
Seeing as your all obsession with random software if the latest Candy Crush total Pokémon warfare Go hit the app stores and has a trojan I blame the software house for being rubbish but also Google and apple for letting it on their platform without QAing it.
everyone gets their fair share of you suck, and nobody should get to point their finger saying it's that other guy's problem.
Seeing as your all obsession with random software if the latest Candy Crush total Pokémon warfare Go hit the app stores and has a trojan I blame the software house for being rubbish but also Google and apple for letting it on their platform without QAing it.
everyone gets their fair share of you suck, and nobody should get to point their finger saying it's that other guy's problem.
Last edited:
Caporegime
You're using Asus Software to overclok your AMD CPU whom both say that voids your warranty.
So who is at fault here?
Now, can you think a little more outside of whatever box constraint you have going on and explain why Asus, not AMD are getting all the flack for this?
So who is at fault here?
Now, can you think a little more outside of whatever box constraint you have going on and explain why Asus, not AMD are getting all the flack for this?
Last edited:
At fault for what ? what CPU overclock ?
you're just making random stuff up now,
Who is at fault for the Actual problem here, not some made up hypothetical, that voltages were set too high on many brands of partner boards,
if AMD gave instructions to the board partners that were wrong. AMD.
If AMD gave instructions to the board partners that were correct and the board partners screwed up implementing them, then the board partners but also AMD for not having systems in place to check their instruction were being followed. business 101 when outsourcing anything.
you're just making random stuff up now,
Who is at fault for the Actual problem here, not some made up hypothetical, that voltages were set too high on many brands of partner boards,
if AMD gave instructions to the board partners that were wrong. AMD.
If AMD gave instructions to the board partners that were correct and the board partners screwed up implementing them, then the board partners but also AMD for not having systems in place to check their instruction were being followed. business 101 when outsourcing anything.
Last edited:
Caporegime
Irrelevant.
The IMC for the RAM is rated to run at 4800MT/s, enabling EXPO or XMP or DOCP is overclocking.
Do i need to explain why Asus are getting all the crap for it.
The IMC for the RAM is rated to run at 4800MT/s, enabling EXPO or XMP or DOCP is overclocking.
Do i need to explain why Asus are getting all the crap for it.
Last edited:
You might have a point somewhere, but that’s not AMD validating memory on its partners motherboards.
How much do you think AMD should charge Asus for being lazy and outsourcing it’s work to AMD? However, if AMD is attached to Asus so closely maybe AMD’s brand is at risk from Intel and Nvidia **** ups so should QA those too?. Where would you stop?
AMD never suffered from Spectre and Meltdown after all yet Intel and Nvidia still haven’t worked out a fix for those. Maybe AMD should become a testing and QA firm?
Last edited:
Caporegime
We do not want AMD locking this stuff down, these CPU's are unlocked because they are quite happy to let us play, they do this knowing that some of us will break them doing that, when you do they may ask "did you enable EXPO?" You say i don't know what that is, they don't question it and just send you a new one, in fact AMD have agreed to replace all the CPU's effected by this.
Its almost a gentleman's agreement, we let you play, don't take the pee.
With that we have to trust people like Asus to get the settings right when we use their software to overclock AMD's CPU's, we don't expect AMD to also do the job Asus should be doing in protecting that CPU when we use the tools they provide, we don't expect AMD to allow us to play, even turn a blind eye so long as we don't take the pee.... and on top of that pay for Asus QA.
The reason people got upset with Asus is not only because they faild in that, mistakes can happen, but also because they tried to weasel their way out of their responsibility.
Its almost a gentleman's agreement, we let you play, don't take the pee.
With that we have to trust people like Asus to get the settings right when we use their software to overclock AMD's CPU's, we don't expect AMD to also do the job Asus should be doing in protecting that CPU when we use the tools they provide, we don't expect AMD to allow us to play, even turn a blind eye so long as we don't take the pee.... and on top of that pay for Asus QA.
The reason people got upset with Asus is not only because they faild in that, mistakes can happen, but also because they tried to weasel their way out of their responsibility.
Soldato
AMDs initial figures are all based around their reference design [motherboard]. It’s mentioned in their review kit. Just to put what you’re saying into perspective, you’re suggesting AMD hand over ES to partners having not plugged them into anything at all themselves. Requirements need to be met for PCIE and regulatory purposes and board partners are quite obviously given these designs.
Actually the point of failure has yet to be determined, other than it’s inside AMD’s CPU. Two points of contention here were the CPU’s thermal protection wasn’t working correctly prior to 1.0.0.7 and at least some select CPU are potentially prone to failure under high VSOC. Both of these conditions were present on all motherboards, not only ASUS.
Last edited:
Caporegime
Third party overclocking tools are not a part of that platform design, in the same way MSI Afterburner is not a part of Nvidia's platform design, if that Russian fellow who made MSI Afterburner gets something wrong and it melts your GPU whose responsible for that, the argument you're making here that would be Nvidia's responsibility.
You know it isn't.
You need to watch the GN video again.
The CPU receives power from externally, ie the motherboard, it relies on the motherboard to stop feeding it power when: in this case the CPU gets too hot, no one has suggest that in-of-its-self does not work, if you overheat these CPU's they will first throttle and eventually shut off, your screen will go black as the system reboots.
The motherboard reads thermal sensors located physically on the die, there are 64 of them on Ryzen CPU's, they are also critical the AMD's dynamic boost algorithm.
Again that thermal shut off depends on the interphase between the CPU its self and the thing that's feeding it power, that is the firmware, where you're using the Asus Republic Of Gamers Hero Extreme 670 Extreme BIOS that is not something AMD made or have any control over.
Steve Burke mocked Asus for having extra protection chips on the motherboard, various motherboard circuit sensors like you would find in a high end PSU, that did absolutely nothing.
PS: Just for info, the point of failure was at the base of the Through Silicon VIA's for the 3D chache. Again you need to whatch the GN video.
Last edited:
Associate
Aren't AMD responsible for the AGESA code that used in Asus/MSI/Gigabyte's motherboard firmware's ?
It is AMD that give the specifications of the platform to the board makers, so yes AMD is responsible if they give them the wrong voltage specs.
It is AMD that give the specifications of the platform to the board makers, so yes AMD is responsible if they give them the wrong voltage specs.
Associate
- Joined
- 28 Sep 2018
- Posts
- 2,317
An easy question to ask of AMD is why was vsoc tolerance first increased to 1.4v and then decreased to 1.3v?
For those not aware, these ranges provided end up being part of the auto rules by mobo manufacturers. So when AMD increases or decreases voltage tolerances, the auto rules applicable in the associated mobo firmware/bios are adjusted for it.
Last edited:
Caporegime
Yes, again tho you need to watch the GN Video, all of it and carefully, Steve is a clever guy, he knows what he's talking about and he's identified everything about this.
That's his job as a competent and knowledgeable tech journalist.
In your Asus BIOS there are two overclocking sections for both the CPU and the RAM, i don't have and Asus motherboard, mine is Gigabyte, i can see the two menus in my BIOS, one is quite simple, just does the job kinda thing, the other is more Gygabyte'ish and is more advanced, the former is the AMD AGESA menu, if you use it it will will use AMD's own firmware, if you use the more fancy one, in your case probably called something like Republic Of Gamers Extreme Tweaking Utility you're using the one Asus made and relying on the Asus firmware.
Steve Burke addressed this. if you use the basic one, the AMD AGESA utility it will set the correct SoC voltages, there is no problem with that. If you use the ROG utility that's when the CPU goes bang.
Last edited:
Soldato
1) Can you explain how third-party overclocking tools are related with anything we are discussing?
2) PROCHOT Control & PROCHOT Deassertion Ramp Time changes can be found in recent ucode revisions under Amdcpnpkg. These changes are provided by AMD. The changes state specifically that ramp time had no effect, meaning the function was not working correctly. The initial GN video even covers the fact PROCHOT doesn't appear to be working as intended.
3) The sensor locations aren't that relevant unless it's possibly attributed to why the mechanism wasn't working i.e. if they weren't located in the correct place on die - which is for AMD to ascertain. You're still after all this time conflating the OCP threshold with another issue. At this point, the CPU is already dead. If you watch the videos carefully, it’s possible the cooler was lifted from the CPU to instigate the CPU failure. If that were indeed the case, it would be related to thermal mechanisms. That shouldn’t be conflated outright with failure due to overvoltage. After all, Roman applied 1.45v VSOC and the CPU did not fail. There isn't really any evidence for you to claim otherwise, as GN doesn't include this information leading up to the failure.
4) The video lists dozens of potential reasons for failure. The SEMS analysis only really explains the consequential damage and origin.
Read my synopsis of that video a few pages back. Steve was wrong on many fronts, including all of his measurement information.
Last edited:
Associate
The AGESA code does all the CPU initialisation, it isn't just an alternative overclocking menu.
Caporegime
You're being deliberately obtuse at this point, it should be obvious to you by now ^^^^^^^
Read the post above you.
----
You answered your own question @Silent_Scone the CPU was already dead when Steve lifted the cooler, Steve made this point in Asus continuing to slam an already dead CPU more more voltage after killing it by over-volting it in the first place, the dead CPU reached 213c before Steve cut the power, mocking that even the Asus external custom protections did absolutely nothing to prevent power being feed to a CPU that had long since short circuited.
I'll address the 1.4c thing as well, AMD allowed 1.4v to the SoC, that was the limit cap, they did this so extreme overclockers could have some fun with it. that's not to say Asus or you should set that, unless you're doing LN2 overclocking, the AMD AGESA never did.
AMD have now reduced that cap to 1.3v, because if you can't play nice with your toys we will take your toys away, refer my point in post #1,188.
Read the post above you.
----
You answered your own question @Silent_Scone the CPU was already dead when Steve lifted the cooler, Steve made this point in Asus continuing to slam an already dead CPU more more voltage after killing it by over-volting it in the first place, the dead CPU reached 213c before Steve cut the power, mocking that even the Asus external custom protections did absolutely nothing to prevent power being feed to a CPU that had long since short circuited.
I'll address the 1.4c thing as well, AMD allowed 1.4v to the SoC, that was the limit cap, they did this so extreme overclockers could have some fun with it. that's not to say Asus or you should set that, unless you're doing LN2 overclocking, the AMD AGESA never did.
AMD have now reduced that cap to 1.3v, because if you can't play nice with your toys we will take your toys away, refer my point in post #1,188.
Last edited:
Caporegime
I'll add this too, what's really cynical about that is some motherboard vendors, not just Asus are framing that in their wording as if this was part of the problem, as if AMD allowing 1.4v was AMD saying use 1.4v, its in this way that these people, not just Asus.... shirk thier responsibilities by trying to throw AMD under the bus for letting us play.
Last edited:
Soldato
If you go to 12:00 on the failure video (where Steve also incorrectly concluded the CPU was receiving 50mV over set point whilst not accounting for plane resistance, a mistake which was made in both initial videos) you will see the system tripped, subsequently leading to what he calls a partial shutdown and 00 Q-Code. Without cycling the AC the cooler is then removed (something an end user would never do). Moreover, a set point of 1.45v was used to rapidly simulate a failure. A discerning mind would be able to differentiate a high OCP trip point being a different issue here, it has nothing to do with the CPU failing in the first instance. In any case, the corrections for PROCHOT can be seen quite clearly in the Igorlabs piece some days ago, so moving the goal posts doesn't quite work here. The protective mechanisms were updated by AMD.
AMDs limit of 1.4 isn't actually a limit? if you have a copy of any whitepapers or voltage guidelines feel free to share them here! Nobody is expecting for you to be swayed at this point as it's clear in your opinion AMD can do no wrong, but it's still good to correct weird pseudoscience. If we can't rely on the tech media to correct their own data there's not much point in pressing end users to be objective.
Last edited:
Caporegime
You can trick the firmware in to thinking the CPU is running at a lower voltage than it actually is, its why you get readings from the software that are lower than the readings from the circuit its self through a multimeter.
Asus, and others have been accused of doing this in the past.
If you do that you overide the physical cap, one might argue its why they do it, its how you get 1.46v from a 1.4v cap.
Asus, and others have been accused of doing this in the past.
If you do that you overide the physical cap, one might argue its why they do it, its how you get 1.46v from a 1.4v cap.
Soldato
SVI3 is AMD's internal telemetry.
It's simple ohms law. Socket and power plane losses...I've seen this mistake made many times by the written media, no need to delve deeper or try to spin it.
Last edited: