Dr. DroBut if that PR stunt doesn't shake consumer trust for you, then nothing will.

I did express concern on what AMD would've done with RDNA3 and RDNA4 support before the correction was made, but I'm not going around saying AMD GPU's are doorstops now because they won't get day 1 updates for the latest AAA turd that needs updates from GPU vendors because the developer and publisher didn't choose to optimize the game. They either walked back or clarified the statement, but the continued outrage even though there's nothing worth to be mad at now is expected.
And I'm not the one using that news as ammo against AMD at every chance to be a shield for the brand who designed the connector, but like I said I'm not surprised at the bias and double standards. If people still choose to bash one side while being dismissive of a dangerous power connector than nothing will shake your trust, definitely for those who decided to hand over money for such a flawed power connector.

AusWolfWhy not improve efficiency instead of churning out 450-600 W cards?

Compare the “power consumption” and resulting graphics performance of former multi-GPU setups with a single high-end model such as today's RTX 3090/4090/5090. Efficiency has more than improved.

Greetings from a former multi-GPU enthusiast,
Fred.

Hecate91I did express concern on what AMD would've done with RDNA3 and RDNA4 support before the correction was made, but I'm not going around saying AMD GPU's are doorstops now because they won't get day 1 updates for the latest AAA turd that needs updates from GPU vendors because the developer and publisher didn't choose to optimize the game. They either walked back or clarified the statement, but the continued outrage even though there's nothing worth to be mad at now is expected.
And I'm not the one using that news as ammo against AMD at every chance to be a shield for the brand who designed the connector, but like I said I'm not surprised at the bias and double standards. If people still choose to bash one side while being dismissive of a dangerous power connector than nothing will shake your trust, definitely for those who decided to hand over money for such a flawed power connector.

They've literally blown their own socks off, mate. Defend them when they deserve to be defended, professionalism and transparency clearly mean nothing to them and they've treated all of us as if we were 5. That's what bothers me: and straight to the corporate box of lies they went as a "remedy".

That "statement" meant nothing, it walked nothing back nor did it commit to anything.

TheinsanegamerNA SECOND GPU has burned! Oh noes!

?Meanwhile, if you go by steam's dismal card counts, there are 26,400 9070xts out there. That is a failure rate of.....0.007%.

And I feel like this reminder is needed: 8 pins burned too.



Still waiting on hard numbers that 12v2 fails more than the 8 pin BTW.

It's on AMD's AIBs for not including load balancing.


The connector can handle 600w just fine. That is what it is designed for.

What it CANNOT DO is run 600w, or 300w, or 200w over 1 pin. 1 pin can only safely handle 100w.

There's barely a day that goes by where I don't see a burned 12VHPRWR / 12V2X6 connector on reddit.

Anyone still claiming it isn't an issue or that it's failure rate is equal to or lower than 8-pin is in complete denial.

None of these connectors have load balancing, the spec doesn't call for it. Heck it likely makes it impossible given even the Astral doesn't have it.

Waldorfwhile it is Nv design, no one forced amd/gpu makers to use it.
to me there is no difference between someone designing crappy stuff, or others that decided to use it instead inhouse solutions.
just shows all corporations are greedy to the point of where they make products you technically dont want to buy/use.

Sapphire iwas thinking it could get away with it. At the end of the day it's just an extra pitfall now even AMD users have to worry about. I'd avoid any card that have the connector like a plague. Your build looking 1% prettier ain't worth the hassle.

FR3D1I quote as follows:

Source ComputerBase (de)

Which is correct, but often concealed.
PCI-SIG

With a friendly smile,
Fred.

Do you know how PCI-SIG works?

It's a body of industry players wherein the companies themselves sponsor and then write standards the body produces. The only part that is handled by non-sponsors is the interoperability.

In the case of 12VHPWR and 12V2X6, the sponsors were Nvidia and Dell with Nvidia taking the lead on writing the standard.

This has got to be the 18th time I've pointed this out, even JohnnyGuru himself points out the sponsors: jongerow.com/12VHPWR/

PCI-SIG made a statement years ago pointing out that was / is Nvidia's fault: arstechnica.com/gadgets/2022/12/pci-standards-group-deflects-assigns-blame-for-melting-gpu-power-connectors/

And an wiki article pointing out that it was introduced by Nvidia: en.wikipedia.org/wiki/12VHPWR

Your "source" is nothing more than a random on the internet spreading misinformation. People really should know better than to use such a source.

Dr. DroThey've literally blown their own socks off, mate. Defend them when they deserve to be defended, professionalism and transparency clearly mean nothing to them and they've treated all of us as if we were 5. That's what bothers me: and straight to the corporate box of lies they went as a "remedy".

That "statement" meant nothing, it walked nothing back nor did it commit to anything.

They said day-one game support for RDNA 1/2 will be uninterrupted. They switched drivers to a different development channel, because they're stable as they are. That's the end of the story for me.

Anyway, why bring this up under a totally unrelated article? Isn't it enough to discuss this where it should be?

This 12-pin connector has to go, that much is certain.

AusWolfThey said day-one game support for RDNA 1/2 will be uninterrupted. They switched drivers to a different development channel, because they're stable as they are. That's the end of the story for me.

Anyway, why bring this up under a totally unrelated article? Isn't it enough to discuss this where it should be?

This 12-pin connector has to go, that much is certain.

Yeah, sure. Because noncommittal word salad statements from multi-billion dollar corporations interested in selling you their new products are indeed to be interpreted in the most positive and charitable manner, with unreserved trust because they have a proven track record of keeping their word here. Once this blows over all they have to do is package an RDNA 2 driver with the same marketing name as the new one every 3 months and you've already forgiven them for it.

"they said it's stable, so it is!", do you believe in Santa Claus too? And I'm still waiting for "le big evil" power connector to burn my house down. 2 generations and 3 years with the same cable, I don't even have a little toast to show for it. so, in the midst of all the doomposting, why not bring up something tangible.

Excuse me for not trusting a word AMD says, even if I genuinely believe this is going to be a net positive in the future.

Dr. DroYeah, sure. Because noncommittal word salad statements from multi-billion dollar corporations interested in selling you their new products are indeed to be interpreted in the most positive and charitable manner, with unreserved trust because they have a proven track record of keeping their word here. Once this blows over all they have to do is package an RDNA 2 driver with the same marketing name as the new one every 3 months and you've already forgiven them for it.

I never cared for up-to-date drivers whether it be AMD or Nvidia or even Intel to begin with, so that word salad was never meant for me.

It was meant for those who thought that Battlefield 45923675 or CoD 359735 won't run on their 6700 XTs. Well guess what, it will.

Dr. Dro"they said it's stable, so it is!", do you believe in Santa Claus too?

No, I believe my eyes. I never had any issue with any driver for RDNA 2, and I don't have a reason to believe that it'll change in the future.

You still haven't answered my question. Why talk about this here? There are more appropriate places to vent your unhinged hatred towards anything bearing the colour red, you know.

FR3D1Maybe they'll try a little less “bending and breaking” at first, right? It's not as if, for example, the 8-pin PCIe connector has never burned out since it was invented. Wrong! Difference? Nowadays, people are too quick to blame others.

And you seem quick to kiss Jensen's butt.
Nvidia designed/overspecified this safety margins lacking ATX12VHPWR/12V-2x6 mess.

With contact pins being significantly smaller their real current carrying capacity is below pins of 6/8 pin PCIe connectors, not higher.
Further tighter spacing of pins and hence heat being more focused would need some derating.
And with far more separate pins on connector without load balancing overcurrent becomes fast far higher compared to real current carrying capacity of pin.

As result it doesn't have real tolerance for things going wrong without approaching melting/burning risk.

TheinsanegamerNStill waiting on hard numbers that 12v2 fails more than the 8 pin BTW.


The connector can handle 600w just fine. That is what it is designed for.

Your leatherjacket Messiah won't ever admit numbers.
And graphics card makers/brands again don't want to make themselves look worser than others by admitting anything.

600W is too much for these tiny contact pins and their tight spacing to have proper safety margins.
Even with proper load balancing in specification proper safety margins should be there.
Now it's like Chinese Tofu construction.

EsaT600W is too much for these tiny contact pins and their tight spacing to have proper safety margins.

This. Even 3-400 W is too much, it seems. This connector is not good.

I'll never have it in my PC if I can help it, but I built a PC for a colleague once, using this new connector. Its pins are ridiculously tiny.

AusWolfI never cared for up-to-date drivers whether it be AMD or Nvidia or even Intel to begin with, so that word salad was never meant for me.

It was meant for those who thought that Battlefield 45923675 or CoD 359735 won't run on their 6700 XTs. Well guess what, it will.

Even when I had Nvidia cards I only updated drivers every 5-6 months or when a game got some random patch/update that wouldn't let me launch the game without the latest driver.

I don't understand the obsession with needing the latest driver every month, then again I don't buy the latest AAA un-optimized crapfest that probably needs up to date drivers.
But as I said people keep being dismissive and defending their favorite company with the "it doesn't melt for me so the connector is perfect" while doubling down on making sure everyone knows how much they hate the other brand.

ShadowizedAm I wrong in thinking you don't get that many burnt pins unless the adapter was defective?

I mean we did see from the tech YouTuber JaysTwoCents an example of MSI's cable from a native ATX 3.0 or 3.1 power supply had clasps inside the connecter that were pushed in further then the rest. So it's possible their manufacturing tolerances and QA sucks, so the 6 clasps were pushed in and not fully connecting with the pins in this example.

@Hecate91
my 2080 doesnt need new drivers for years now (nor my games), yet virtually every year there were drivers that improved/fixed things in one form or another,
not just limited to games.
i dont update to any (new) driver, unless release notes show i should.

ignoring that there is more than one game that was "optimized", that still got driver "updates", as not everything show every time you run the game, or to everyone testing it,
or that many games will get changes/updates/fixes, that might even require a driver update, to work.
not everything update wise is always about un-optimized AA.

and dont say anything about me defending anyone, company wise.

Waldorf@Hecate91
my 2080 doesnt need new drivers for years now (nor my games), yet virtually every year there were drivers that improved/fixed things in one form or another,
not just limited to games.
i dont update to any (new) driver, unless release notes show i should.

ignoring that there is more than one game that was "optimized", that still got driver "updates", as not everything show every time you run the game, or to everyone testing it,
or that many games will get changes/updates/fixes, that might even require a driver update, to work.
not everything update wise is always about un-optimized AA.

and dont say anything about me defending anyone, company wise.

I've never seen any game update that needed a driver update as well. Improvements are nice, but not a must by far. We're talking about extremely edge cases here.

If you don't update your drivers without good reason yourself, then you know.

But again, this isn't the thread. Please, back to the connector.

On an aside, the funny thing about the original PCIe connector was that the only difference the 8-pin connector had was an additional ground and a sense pin. And the presence of the sense pin (which is basically grounded) told the GPU it could draw more power from the existing three pins used for the 6-pin connector. So the thing was overengineered from the get go.

My guess with 12VHPwr was to use more active monitoring and load balancing so the cable could be less overengineered (or "just enough engineered"), but most implementations avoided that.

AusWolfI have one problem with this.

Why not improve efficiency instead of churning out 450-600 W cards?

I strongly feel that if the consumer market had been 'efficiency focused' these last few generations, there'd have been virtually no performance improvements; every release would've been no more exciting than the GTX1080ti->RTX2080ti.
4K, RTX, etc. would be mere thoughts.

AusWolf"Oh look, our new 9989 XT Ultra Pro Ti is so fast!" - Yeah, right! :rolleyes:

If hardware makers stayed in the land of sanity with power consumption on their cards, we would have never needed a new connector to replace 3x old ones in the first place.
2x 8-pins should be more than enough for any GPU.

Tired old argument at this point but, not really wrong.

shuffy_losysOn an aside, the funny thing about the original PCIe connector was that the only difference the 8-pin connector had was an additional ground and a sense pin. And the presence of the sense pin (which is basically grounded) told the GPU it could draw more power from the existing three pins used for the 6-pin connector. So the thing was overengineered from the get go.

Pretty much, yup.

PCIe 6-pin is electrically-equivalent in current-carrying ability to a 'P4' 4-pin +12V CPU Power connection (2x+12V 2xGND)
The P4 and EPS 8-pin are rated 8A/pin, 150W-192W. The 4+4/8-pin EPS is 235W-384W.

Given what we've seen 'pushed' on the 2x6, there's an insane amount of overhead left in Old-ATXspec PCIe 6-pin.

TBQH, I feel like the easiest and cheapest fix for this entire debacle, is revising the OG 6-pin to be "half a 2x6".
-No more sense pins (or, add them externally, like +12V-2x6; to maintain Gen1-6pin compatibility)
-6-pin gets 3x8.333A (300W) nominal current rating. 1:1 +12V:GND pinning.
-6+2/8-pin gets retired or reserved for the +1 'halo tier' AIB cards, as 2 extra ground pins may improve voltage stability when exceeding rating.
-Standardize the same high-temp conductors as +12VHPWR/+12V-2x6

Stack 2 of these 'ATX 3.2 +12V2x3' receptacles on a card ala GTX 680, and just about every issue is solved.

Even, in the odd case that someone plugs an ancient PSU w/ only PCI-E 6-pins in, most PSUs already were built 3x+12V and 3xGND on the 6-pins.
In the even-rarer case that someone 'blows something up' from using an ancient ATX 2.x PSU with an 'RTX 6090', that's 100% on the end-user.
-As-is, you can 100% follow spec, have the newest 1st-party kit, and it still burns up and the end-used is blamed :shadedshu:

shuffy_losysMy guess with 12VHPwr was to use more active monitoring and load balancing so the cable could be less overengineered (or "just enough engineered"), but most implementations avoided that.

I'm really not exactly sure what-all happened.
IIRC, The rumor was that the extra pin-block on the +12V-2x6connector was for active current monitoring, just like the XBOX 360's power brick's monitored and maintained stabile voltage.

Turns out, it's just a passive 'hi-lo' pull-to-ground that indicates to the GPU/PSU whether there's a 450W limit or 600W (which, seems to do/mean jack-diddly-squat, anyway.)

LabRat 891TBQH, I feel like the easiest and cheapest fix for this entire debacle, is revising the OG 6-pin to be "half a 2x6".
-No more sense pins (or, add them externally, like +12V-2x6; to maintain Gen1-6pin compatibility)
-6-pin gets 3x8.333A (300W) nominal current rating. 1:1 +12V:GND pinning.
-6+2/8-pin gets retired or reserved for the +1 'halo tier' AIB cards, as 2 extra ground pins may improve voltage stability when exceeding rating.
-Standardize the same high-temp conductors as +12VHPWR/+12V-2x6

Stack 2 of these 'ATX 3.2 +12V2x3' receptacles on a card ala GTX 680, and just about every issue is solved.

Even, in the odd case that someone plugs an ancient PSU w/ only PCI-E 6-pins in, most PSUs already were built 3x+12V and 3xGND on the 6-pins.
In the even-rarer case that someone 'blows something up' from using an ancient ATX 2.x PSU with an 'RTX 6090', that's 100% on the end-user.
-As-is, you can 100% follow spec, have the newest 1st-party kit, and it still burns up and the end-used is blamed :shadedshu:

I just can't understand why they didn't simply go with a 2x6 pin configuration and make the keys different with a Mini-Fit unless cable manufacturers were like "we ship millions of these and we need to save pennies on smaller gauge cables pls" or something

LabRat 891I'm really not exactly sure what-all happened.
IIRC, The rumor was that the extra pin-block on the +12V-2x6connector was for active current monitoring, just like the XBOX 360's power brick's monitored and maintained stabile voltage.

I was thinking active monitoring more on the card's end, since the PSU should be doing it's own monitoring. And the PSU shouldn't really care how much the other end draws as long as it's not exceeding the current rating.

Hecate91We're on the 3rd or 4th known case of an AMD card with a melting connector, on Nvidia cards it is probably into the hundreds now, and on the third revision the connector still melts.

Numbers, how do they work?

The 9070xt, using pessimistic 9070xt sales numbers from Steam, that would be a 0.0014% failure rate. Oh no, the horror!

That's nothing. RMA rate for any given GPU is usually between 1-2% for good cards, upwards of 5% for cheap cards from cheap brands.

Same with nvidia. "hundreds". Uh huh. nvidia sells MILLIONS of geforce GPUs per year. Even a few thousand burnt cards would be a rounding error.

Hecate91And I don't expect any PSU or GPU company in the industry to admit just how many connectors failed because it would hurt their sales.

So there is no factual evidence for any of these claims, just pure speculation.

As I linked earlier, the 8 pin can also melt. Not just on nVidia cards either.

Hecate91The connector is too fragile compared to the 8 pin molex PCI-e GPU connector, there's no reason why any full size GPU needs the 12 pin connector, especially when 8 pin PCI-e molex can handle more than 150W per cable.

this is just wrong, in so many ways. If GPU manufacturers started using the 300w capability of 8 pins, you'd be crying about melted 8 pins instead. there are plenty of GPUs that need more power then an 8 pin can provide.

Having a greater than 100% safety margin is great to prevent user error from damaging hardware, but it is also wasteful of space and material.

Hecate91AMD didn't design 12vhwpr or 12v2x6.
It's on Nvidia and Dell for not including load balancing in the design specs.

Spoiler

www.jongerow.com/12VHPWR/index.html

PCI-SIG ratified the design. AMD is a member of PCI-SIG.

Hecate91It can't given that it melts at half of the rated power with nothing in the design to prevent unbalanced loads.

So in a failure state, the connector fails?

Well color me shocked!

The 8 pin would fail under the same conditions. All multi pin connectors need a design that balances the load being drawn. for some reason some manufacturers assume they could get away with not doing so with the 12v2, and are now failing in small numbers.

evernessinceThere's barely a day that goes by where I don't see a burned 12VHPRWR / 12V2X6 connector on reddit.

Anyone still claiming it isn't an issue or that it's failure rate is equal to or lower than 8-pin is in complete denial.

Unless you have actual hard numbers, claiming anyone else is in denial is hilarious irony.

evernessinceNone of these connectors have load balancing, the spec doesn't call for it. Heck it likely makes it impossible given even the Astral doesn't have it.

Oh my goodness, does nobody read

the connector NEVER DOES THE LOAD BALANCING. That is the job of the device receiving power!!!!

the 8 pin didnt do load balancing either. The cards did. How do you think those 3x8 pin cards functioned?

EsaTAnd you seem quick to kiss Jensen's butt.
Nvidia designed/overspecified this safety margins lacking ATX12VHPWR/12V-2x6 mess.

With contact pins being significantly smaller their real current carrying capacity is below pins of 6/8 pin PCIe connectors, not higher.
Further tighter spacing of pins and hence heat being more focused would need some derating.
And with far more separate pins on connector without load balancing overcurrent becomes fast far higher compared to real current carrying capacity of pin.

As result it doesn't have real tolerance for things going wrong without approaching melting/burning risk.




Your leatherjacket Messiah won't ever admit numbers.
And graphics card makers/brands again don't want to make themselves look worser than others by admitting anything.

600W is too much for these tiny contact pins and their tight spacing to have proper safety margins.
Even with proper load balancing in specification proper safety margins should be there.
Now it's like Chinese Tofu construction.

You are mad. And "worser" isnt a word.

I'm curious, do you thin that Jensen made the 6800xt? How about the 5800x3d? Should probably check the sig of someone before accusing them of worshiping Jensen....

600w is not too much, when balanced correctly. USB C is now pushing 240w over 4 pins, all WAY smaller then what is being used here, and you dont hear about those melting. Of course, the standard mandates proper load balancing and handshakes, so perhaps PCI SIG eeds to get more invasive and mandate more circuitry in the space.

shuffy_losysThat'd require bumping up the voltage, but I'm sure all the PSU manufacturers would love to offer say 24V.

Problem with going 24v, or any other voltage, is that you break all backwards compatibility with everything. GPUs, motherboards, Blu ray drives, fans, the list goes on.

To do so, you have to get the whole industry on board and convince them to move in lockstep to the new standard. Intel couldnt do that in their heyday with BTX.

TheinsanegamerNProblem with going 24v, or any other voltage, is that you break all backwards compatibility with everything. GPUs, motherboards, Blu ray drives, fans, the list goes on.

To do so, you have to get the whole industry on board and convince them to move in lockstep to the new standard. Intel couldnt do that in their heyday with BTX.

The emphasis was meant to convey sarcasm. But even if someone were to seriously consider it, they could limit it to just GPUs since that's the only thing that really needs a lot of power in a system.

shuffy_losysThe emphasis was meant to convey sarcasm. But even if someone were to seriously consider it, they could limit it to just GPUs since that's the only thing that really needs a lot of power in a system.

GPUs also pull power from motherboards and use 12v for signaling.

It's not that easy. You need to upgrade the backend bus architecture to the same voltage. Or you need to iclude an expensive VRM system to convert voltage back and forth, which takes up a lot of space and makes a lot of heat.

If it were that easy, we'd have made GPUs 24v ages ago, back in the Thermi days.

TheinsanegamerNNumbers, how do they work?

The 9070xt, using pessimistic 9070xt sales numbers from Steam, that would be a 0.0014% failure rate. Oh no, the horror!

That's nothing. RMA rate for any given GPU is usually between 1-2% for good cards, upwards of 5% for cheap cards from cheap brands.

Same with nvidia. "hundreds". Uh huh. nvidia sells MILLIONS of geforce GPUs per year. Even a few thousand burnt cards would be a rounding error.

Strawman arguments, how do those work?

The point is the connector shouldn't be failing in the way it is, and this "but the sales numbers bro" nonsense is probably a reason why no one cares enough to make a better power connector when marketing and sales managers tell the engineers no, and why make it better when you can just hide the real issue with telling people they're plugging it wrong and they should buy a new PSU every time they buy a new GPU.

TheinsanegamerNSo there is no factual evidence for any of these claims, just pure speculation.

As I linked earlier, the 8 pin can also melt. Not just on nVidia cards either.

Have any hard evidence on how many 8 pin connectors fail then? Or do you just want to a defend mega corporation for a shitty design?

TheinsanegamerNthis is just wrong, in so many ways. If GPU manufacturers started using the 300w capability of 8 pins, you'd be crying about melted 8 pins instead. there are plenty of GPUs that need more power then an 8 pin can provide.

Having a greater than 100% safety margin is great to prevent user error from damaging hardware, but it is also wasteful of space and material.

This makes no sense at all, and the 8 pin connector is so overbuilt it wouldn't melt because the pins are thicker, it doesn't take an EE degree to figure out thicker metal pins and wires equals more durable.
And no that's not how a safety margin works, having a safety margin means the connector is built better than it needs to be in an amount over the rated specs, not just functioning to the bare minimum of the spec. It's like expecting seatbelts alone to protect you when blind spot warning, auto braking, seatbelts and airbags are significantly better protection.
The wikipedia article points out how the safety margin on 12vhpwr is much lower. 1.68 for 8 pin molex, 1.14 for 12v2x6.
en.wikipedia.org/wiki/12VHPWR

TheinsanegamerNPCI-SIG ratified the design. AMD is a member of PCI-SIG.

The connector was sponsored, in other words paid to be approved and pushed to market by Nvidia and Dell.
I'm reposting this article that @evernessince posted above, since people still want to ignore it. PCI-SIG admits themselves that members are responsible for testing the standard themselves, it absolutely baffles me how a standards organization responsible for these standards doesn't bother to verify the sponsored connector for safety and quality vs. existing standards.
arstechnica.com/gadgets/2022/12/pci-standards-group-deflects-assigns-blame-for-melting-gpu-power-connectors/

AMD isn't forcing their partners to use 12v2x6, but for some reason Sapphire thought it would make their card look better, or assumed it would be fine at 350W but of course it's not.

TheinsanegamerNSo in a failure state, the connector fails?

Well color me shocked!

The 8 pin would fail under the same conditions. All multi pin connectors need a design that balances the load being drawn. for some reason some manufacturers assume they could get away with not doing so with the 12v2, and are now failing in small numbers.

You mean in a failure state that shouldn't even be happening if the protections to prevent it from becoming unbalanced were apart of the standard, not needing to buy a $100 monitoring device to know when the connector is about to melt.
8 pin connectors didn't fail in the same way unless using a dumpster fire of a PSU, and yet reports of 12 pin connectors is happening almost every day, despite we're told the connector is better than the old 8 pin connector because it's smaller and more convenient to plug in than 2 or 3 8 pin connectors.

LabRat 891I strongly feel that if the consumer market had been 'efficiency focused' these last few generations, there'd have been virtually no performance improvements; every release would've been no more exciting than the GTX1080ti->RTX2080ti.
4K, RTX, etc. would be mere thoughts.

There are no performance improvements anyway. Please no one tell me that a 5070 Ti vs 4070 Ti battle doesn't make you yawn, meanwhile AMD's performance improvements are more like price decreases.

TheinsanegamerNOh my goodness, does nobody read

the connector NEVER DOES THE LOAD BALANCING. That is the job of the device receiving power!!!!

the 8 pin didnt do load balancing either. The cards did. How do you think those 3x8 pin cards functioned?

This isn't correct, having multiple 8-pin connectors does in fact balance the load across multiple cables even if they were to be combined into a single 12v blob on the card side (although the vast majority that use separate 8-pin don't combine them into a single 12v on the card side).

Mind you, that wasn't what I was saying. I was referring to the spec as a whole, not just the connector. Something I specifically pointed out in my comment. I could have worded that better but you likely would have found a reason to disagree anyways. People read, you are just being obtuse. Intentionally misinterpreting things because you refuse to admit the standard / connector is trash.

TheinsanegamerNUnless you have actual hard numbers, claiming anyone else is in denial is hilarious irony.

You sure as heck didn't provide hard numbers for your 8-pin claim. I'm just going to point out that the evidence of constant reddit posts of melting 12VHPRW / 12V2X6 is far greater than the single instance you can muster. Reddit posts aren't the highest order of evidence for sure but their consistency and the issue's longevity lend credence to the idea that the issue is very real and impacting a far greater number of people than 8-pin melting ever has.

If you are going to call other claims out for lack of evidence, at least make sure your own are not significantly more deficient. If all you have is one example of 8-pin issues over decades of the standard being out, that just proves how reliable it is in comparison to 12VHPWR / 12V2X6. God knows why people defend this terrible standard, any increase in failure rate is bad.

ChaitanyaOnly Astral series of cards have load balancing for input power and even that apparently is breaking the specification for this fire hazard.

I don't think the ROG Astral has load balancing on the 12V-2x6 connector, but it does have sensors though, which is why you can monitor it in their GPU Tweak III as Power Detector+:

CheeseballI don't think the ROG Astral has load balancing on the 12V-2x6 connector, but it does have sensors though, which is why you can monitor it in their GPU Tweak III as Power Detector+:

More vendor-specific bloatware for something that should just work? Oh god no! :fear:

AusWolfMore vendor-specific bloatware for something that should just work? Oh god no! :fear:

I do not consider GPU Tweak III as bloatware as all the functions in the software are only for monitoring or tweaking ASUS graphics cards. I would consider Armory Crate as bloatware.

GPU Tweak III may actually be better than MSI Afterburner/RivaTuner Statistics Server in this regard.