GTX 1050TI PCI-E Card: Understanding Additional Power Input and Consumption

Hello,
I plan to buy a graphics card and I have a big dilemma, because what most people consider an advantage, I consider experience as a disadvantage. It is about GTX 1050TI cards and the lack of additional power supply in the vast majority of these cards, only the most powerful and expensive ones, exceeding TDP> 75W, have an additional input. Personally, I would like to have an additional power input, even when theoretically the graphics power consumption is within the PCI-E specification, because I have seen (including on my own board) an overheated, even melted power supply plug and a 24-pin motherboard input working with graphics , which had no additional power. So the question is - if the graphics have an additional power input, then it gets the max from the PCI-E bus first, and if it lacks power, then does it take from an additional source? In such a situation, an additional input is not really needed in

And what will change if you find out that it takes evenly? I think xbitlabs used to make such measurements once but I'm not sure but it used to be.

But what will change the fact that it charges (I assume) mainly from additional inputs as such a card has? After all, it either does it without additional power or not. Well, were those constructions where you saw the power supply something solid and were the mobaskes themselves baked by neglecting to clean or ventilate the interior of the housing?

I guess the pci-e version also matters as to the amount of amperage passed on the 12V line through the motherboard.

WidgetSW wrote:

(...) because I have seen many times (including on my own board) an overheated, even molten plug of the power supply and a 24-pin input of the motherboard working with graphics that did not have additional power.

And I've never seen (maybe the quality of power supplies and / or motherboards). Not only that - curly 750Ti / 1050Ti I kept on the power supply with a 20pin plug, so only one yellow wire was responsible for 12V. Nothing was done brown / black, nothing burned ... For the sake of conscience, I added a second yellow wire + ground in the 4pin plug.

Quote:

So the question is - if the graphics have an additional power input, then it gets the max from the PCI-E bus first, and if it lacks power, then does it take from an additional source?

It is different, but often it is more or less even (for cards with 1x6pin, so 150W max).
The tomshardware service measures this, here you have the GTX 1060:

I am an electronics engineer and I assure you that the overheated 24-pin ATX input and power supply plug were not caused by human error or neglect, but by the long-term high-current demand of the motherboard due to the lack of direct connection of the + 12V line to the graphics. The same problems were on the last motherboards that did not yet have the ATX12V processor power input. For me, this is an important issue, so I would like to know what the power balance looks like in cards with additional input. It will depend on whether I decide to pay extra PLN 50 to e.g. Gigabyte GTX 1050Ti WindForce, which has an additional 6-pin connector.

EDIT: thanks to Rome for the screen, you added a post at the time I wrote. From the IDLE posts it is clear that the card does not pull in the first place with PCI-E, only the ridiculous 10W divides roughly in half. But because the card pulled on one 12V wire, I'm really shocked. The PCI-E specification shows that the card from the bus can suck 3A on the + 3.3V line and 5.5A on the + 12V line. The + 5V line is not there. Even a good, thick copper wire would have a hard life.

Maybe there are constructions that first take the current from additional sockets - but you need electronics on the graphics and how it gets from where it may look like above.

So it turned out that even if it was continuous from additional ports, you would have to pay extra if I missed something? After all, you are looking for with additional power inputs and not whether it has regulations where it draws electricity from?

enhanced, oh you missed it, you probably didn't get it completely, but never mind. His friend's screen explained a lot, and I hope that the same principle also applies to other cards. Theoretically, therefore, the GTX 1050TI with an additional power input will pull about half of what the version without an additional plug from the PCI-E bus. And that was what I meant.

You think so wrongly. If the card is as above 120W then it will pull 120W (as if it could of course) all pci-e. 75W is probably the limit for pci-e v3. Have I not seen anything yet?

Why should the card take everything from the disc if it has an additional slot?
Anyway you have on the graph pasted by me - it draws (more or less) half of the current from the PCIe socket on the board and the other half through an additional 6pin ...

enhanced, rome is right, see that even a funny 10W in Idle divides in half. Of course, the other thing is with cards> 150W at full load, then with PCI-E pulls around the max anyway, and the rest with additional power. If I wanted to buy such a strong card, I would not bore the topic, because I know that the rail next to them is loaded at maximum load. I was talking about cards with TDP max 75W. I think the topic is exhausted.

rzymo wrote:

Why should the card take everything from the disc if it has an additional slot?
Anyway you have on the graph pasted by me - it draws (more or less) half of the current from the PCIe socket on the board and the other half through an additional 6pin ...

I was talking about a card with no additional power input. That it would take the pci-e limit if there is collateral on the disc.

The author is looking (concluding from the description of burned discs etc) graphics with a dedicated power supply - since he is looking for 75W and the pcie already has a 75W limit, I think this topic was more aimed at knowing that he should choose a card with additional power supply as he saw disc burned by graphics that were continuous under the pci-e limit and yet such 75W is under the pci-e limit.

It is rare for such a client but now I do not know if he will look under the limit without a dedicated connector or rather that he should have peace of mind he should look for a weaker Watt card or with a dedicated connector but not where he draws more than 150W (and even less so that he does not download again from the max limit pci-e slot, although then the next connector should come).

So that the topic without much sense because you have to take poor graphics and those approaching under the limit of the pci-e slot already with additional power supply or otherwise mobo will not live long.

"That he would take under the pci-e limit if there is disc protection." - the disc is not a protection. The limits are saved in the card's BIOS. This can be seen nicely on services measuring the power consumption only by the card (and not the whole set) - TechPowerUp or TomsHardware. It can be seen that when the card has a TDP set at e.g. 120W, it hits so much and it does not exceed this value (in such Furmark it dumps the clocks and voltage strongly to fit within the power limit).

When the 24pin plug is properly pressed, any card without an additional socket will work properly - unless you overclock it, in the controllers you will increase the power limit and for long hours you will load it with programs generating 100% load all the time (e.g. Furmark / OCCT / MSI Kombustor) - then in fact one of the elements in the power path may not withstand thermally.
In games, consumption will be lower, and fluctuating, so on average the card will consume less power than the socket / motherboard limit.

As I wrote before, I have been using cards without an additional slot for many years (reasonably overclocked) and I have not been able to beat or even brown the ATX or PCIe slot on any motherboard ...

Hello, I bought a Gigabyte GTX 1050TI 4G WindForce for peace of mind, which has an additional 6-pin power input, but after looking how energy-efficient these Pascale are, I would stresslessly take 30 PLN of the cheaper Gigabyte 4G OC, which also has 2 valves 1cm smaller diameter), semi-passive mode and is 4cm shorter. My WindForce in games at full details barely reaches 60% TDP, even light OC does not change it, so you can safely take and use without stress. I consider the topic exhausted.