NV 12VHPWR issues revisited

Page 18 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

MrTeal

Diamond Member
Dec 7, 2003
3,749
2,145
136
33% more cores and 33% wider bus on the same process and clocks probably won't regress. Maybe they won't reach 600W, but it'd be pretty surprising if there wasn't a bump in TDP.

That brings some interesting questions for "OC" partner cards. Everyone's done with one 12VHPWR connector so far other than that one Galax HOF with two, but if the default TDP creeps up near 600W we might see more of them with two. Actually finding PSUs for that will be fun though, most people would probably be looking at one native and one adapter?
 

DeathReborn

Platinum Member
Oct 11, 2005
2,786
789
136
5090 TDP raised to 600W.

Who wants to bet that some OEMs will ship 450W cables in some of their boxes? Or even some 600W cables will get burned when using Furmark on these GPUs? Can Nvidia manage to release the most expensive consumer GPU with zero drama?
Haven't the 600W "rumours" been around for months?
 

DrMrLordX

Lifer
Apr 27, 2000
22,370
12,177
136
@WelshBloke

Just an fyi but 12VHPWR is mostly deprecated in favor of the newer (but still eerily similar) 12V-2x6 which ups the power limit per cable to 675W (75W can be drawn though the PCIe slot). That being said, 12v-2x6 is still likely to be less proportionately-robust than 8-pin PCIe, and pushing 600W through one of those things would probably not be a great idea.

 

MrTeal

Diamond Member
Dec 7, 2003
3,749
2,145
136
The only changes between 12VHPWR and 12V-2x6 is on the male pinned connector on the GPU; sense pins are shorter and power pins a bit longer. Otherwise the cable and the power rating is the same. It'd be weird for such a high power card to actually pull much power through the slot. Generally you only see much PCIe slot draw on slot powered cards, or ones trying to skirt a medium power card to a single connector like the RX480. GPU-Z screenshot I found of a 4090 showed it pulling 446W board power and only 7W of that through the slot.

Still way less proportionally robust than equivalent 8pin connectors though. Three 8pin connectors (450W rated) have effectively 1.26x higher contact rating than a 600W 12VHPWR connector.
 

reb0rn

Senior member
Dec 31, 2009
250
76
101
I will defo not buy 5090 if its rated 600w so i need to limit it to 450w as this shit gone burn, this last week info about less power might be those nvidia morons seen what we see all the time that 12V-whatever connector is SHIT
 

WelshBloke

Lifer
Jan 12, 2005
32,128
10,148
136
I will defo not buy 5090 if its rated 600w so i need to limit it to 450w as this shit gone burn, this last week info about less power might be those nvidia morons seen what we see all the time that 12V-whatever connector is SHIT
I can push my 6950xt so it goes up to 350w and that heats up my room a fair bit, I can't imagine one item pumping out >600w into my case!
 

reb0rn

Senior member
Dec 31, 2009
250
76
101
The powa is not issue if its safu, 7900 xtx had no issue pushing 450w as it has 3x 8 pin that is 2x 12 wires to push 12V over it and more robust connector, while this ^&%*(& of crap is just 2x 6 wires with small and weak connection
 

coercitiv

Diamond Member
Jan 24, 2014
6,957
15,595
136
All these nice little changes fail to address the main issue, it's still a flawed design with too little room for error. I would not be surprised if we see yet another change to the "standard" once the 500W+ GPUs reach the wild.

The good news is everyone with < 400W GPUs is safe using it by now. That's where the rating should have stopped imho, any 400W+ card should carry two connectors.
 

MrTeal

Diamond Member
Dec 7, 2003
3,749
2,145
136

That's good design. As long as you see that color, it's not plugged in properly. Not sure what colorblind people would see though. Hope the bright whatever color they see stands out from the black.
Most people who are colorblind just see one color for another. Even people with achromatopsia still see shades of grey instead of colors. I'm sure it'll be fine unless they see all colors as black, which is a whole other issue.
 

coercitiv

Diamond Member
Jan 24, 2014
6,957
15,595
136
Don't they? Not the light, but isn't the main feature of the revised design a change in pin lengths so it won't start the card unless the connector is properly seated?
And even first gen cards had a light indicator and a protection mechanic that did not power the card under certain circumstances (probably coupled with the sensing mechanic). The "feature" was poorly documented though, and if we search for something like 3090 red light we get plenty of results from forums, some folks didn't even realize they were supposed to change their adapters with better cables.
 

DrMrLordX

Lifer
Apr 27, 2000
22,370
12,177
136
Are any of these protection methods going to help if the card momentarily pulls too much current over a properly-seated cable? There isn't much wiggle room, unlike PCIe 8-pin.
 

Heartbreaker

Diamond Member
Apr 3, 2006
4,653
6,113
136
Are any of these protection methods going to help if the card momentarily pulls too much current over a properly-seated cable? There isn't much wiggle room, unlike PCIe 8-pin.

Actually there is lot of wiggle room for that. Cable/Connector Power ratings are based on long term heating thresholds. It's not about sudden failure based on exceeding a certain value.

It's spec'd to deliver 600 watts endlessly and not heat up too much. A short term spike is not a problem. It's the average over time that cause too much heat.

You could pump 800W through it for minutes in a row, if you let it cool a bit after before resuming load. But if you could make it draw 800W for hours, then you probably would end up causing problems, but you aren't doing that with out power shunt mods, and if something goes wrong you only have yourself to blame.
 

reb0rn

Senior member
Dec 31, 2009
250
76
101
They gona burn, as soon test bench and games start to use full power for prolonged time... all rest is shilling broken design
you can never/ever have perfect as lab connection there is dust, moisture, prolonged heat use damage and so on, only morons working for NV do not get that
 

poke01

Diamond Member
Mar 8, 2022
3,040
4,031
106
They gona burn, as soon test bench and games start to use full power for prolonged time... all rest is shilling broken design
you can never/ever have perfect as lab connection there is dust, moisture, prolonged heat use damage and so on, only morons working for NV do not get that
I’m still of the opinion that above 450 watts this cable should not be used. Below 450 watts it’s fine.

I can’t fathom spending $2K with that plug, it’s truly for people with professional needs or money to 🔥
 

DrMrLordX

Lifer
Apr 27, 2000
22,370
12,177
136
Actually there is lot of wiggle room for that. Cable/Connector Power ratings are based on long term heating thresholds. It's not about sudden failure based on exceeding a certain value.

It's spec'd to deliver 600 watts endlessly and not heat up too much. A short term spike is not a problem. It's the average over time that cause too much heat.

You could pump 800W through it for minutes in a row, if you let it cool a bit after before resuming load. But if you could make it draw 800W for hours, then you probably would end up causing problems, but you aren't doing that with out power shunt mods, and if something goes wrong you only have yourself to blame.
PCIe 8-pin is specced at 50% of its capacity, while 12VHPWR and 12v2x6 are at around 90%+ of capacity. If you push 12v2x6 to rating constantly and then have spikes on top of that, that is what I'm talking about. If you do that to PCIe 8-pin connectors, it laughs it off with extra capacity unless something has gone really wrong (which does happen).
 

Heartbreaker

Diamond Member
Apr 3, 2006
4,653
6,113
136
PCIe 8-pin is specced at 50% of its capacity, while 12VHPWR and 12v2x6 are at around 90%+ of capacity. If you push 12v2x6 to rating constantly and then have spikes on top of that, that is what I'm talking about. If you do that to PCIe 8-pin connectors, it laughs it off with extra capacity unless something has gone really wrong (which does happen).

Again, this is simply you being ignorant of that connector ratings mean. Spikes are a complete non issue. The card would need a spike bigger than it would take to melt it's silicon, to harm the connector.

These are standard Microfit 3 connectors, rated for 10.5 Amps x 6 pairs x 12v = 756Watts in 12 pin config.

Plus, again that how much it could handle continuously over long periods, and that will have it's own safety margin built in, so it could probably handle more beyond that indefinitely.

It has a safety margin (GPU spec) on top of safety margin (for the standard Connector itself).

But sure live in fear of a connector because you don't understand how ratings work.
 

KompuKare

Golden Member
Jul 28, 2009
1,221
1,570
136
These are standard Microfit 3 connectors, rated for 10.5 Amps x 6 pairs x 12v = 756Watts in 12 pin config.
I don't know about the earlier PCIe 8pin's 80% figure or the 90% claim for 12VHPWR, but surely 600W / 756W is about 80% which is far closer to 90% than 50%.

To this day, I can't see any possible advantage to these new connectors yet Nvidia stubbornly stick to them in some Apple-like "we know best" attitude.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |