[Partol]

It's not so important what is the exact definition of TDP.
What's important is (1) how did they measure TDP? and (2) is that measurement accurate?
Would not surprise me if some companies simply measure power usage and call it TDP.
Better to read independent product reviews, than to assume the manufacturer's numbers are correct.

Edit:
Example of how power usage and TDP can be very different.
http://cryptomining-blog.com/1141-a-bit-more-on-the-actual-power-usage-of-geforce-gtx-750-ti/

"TDP" of GTX 750 Ti = 60W
power limit (in vbios) = 38.5W
actual tested power usage = 28 to 35W

Edit 2:
To confuse you even more ....
http://www.geforce.com/hardware/desktop-gpus/geforce-gtx-750-ti/specifications
Nvidia's site says "Graphics Card Power (W) = 60W"
It does not say "TDP".

 

[TwoCables]

Quote:

Originally Posted by JTHMfreak

@TwoCables
As said earlier, I only upped this psu I have now because I had bought sli 480s, didn't think my 750 watt corsair would cut it for those, pretty sure it would not.
I don't know if I plan on adding a third 970 (which all three would be overclocked, of course), or going with two of the newer gpus.
Money isn't much of an object, pretty soon I should have around $400 extra money to play with monthly consistently, so budget is fine as long add the product is of good quality.
The psu I have now I caught on a great sale for only $90 if I remember right, through newegg, so no money was wasted there. Although, I don't mind spending extra on a component that is so vital to the rest of the system, and will last at least five years.

First, just a humble request: please don't "mention" me when I am already participating in a thread because I have my subscriptions preferences set so that I am automatically subscribed to every thread I reply to (and every thread I create). So "mentioning" me only does one thing: it sends me an e-mail telling me what I already know. So, it's annoying. The only time "mentioning" me is useful (for anyone, really) is when they aren't already in the thread and you want their attention.

Anyway, I wasn't wondering why you purchased that PSU. I was only saying that it's still a good PSU and should have a few years of life left in it and it should be more than plenty for your plans. I would still buy a new one just as a backup. That way, when you finally need to replace your current PSU, all you have to do is unbox the new one and install it.

Quote:

Originally Posted by JTHMfreak

That's not bad at all.My first corsair psu was a dud early on in its life, which as someone who plays with electronics often I know it will happen. Exchanged it for the same unit and had years of use until I sold it.would buy corsair again.
Is there a reliable way to test how my psu is performing, I'm guessing it would be to get one of those pull-a-watt things, if any.
Just curious to see how well my current psu is doing.

It's not Pull-a-Watt, it's Kill A Watt (it's a play on the word kilowatt). However, it's just a wall meter and a wall meter will not be able to tell you how your PSU is performing. You would need fancy PSU testing equipment for that. My recommendation is to just relax. As I said, a true PC Power & Cooling unit like yours is quite good and you've been very kind to it.

Quote:

Originally Posted by Klue22

Also, Kill-a-Watts suck:

No they don't. Whoever burnt that Kill A Watt pulled too much power out of it. It is only capable of handling 1875VA (15A at 125VA). In other words, around 1800W, give or take - depending on the actual voltage being supplied by that wall outlet.

Regarding TDP:

TDP is simply how much heat in watts any cooling must be able to dissipate in order to keep the chip from overheating. That's it. There's nothing more to it. It has absolutely nothing to do with the card's maximum or average power consumption, but it can SEEM related if the card's maximum is very close to the TDP.

 

[Klue22]

Thats my own kill-a-watt that I used to own. And I only ever put 1300w of load on it. Granted it was 24/7 but the idea is kind of that you can monitor total usage. I would NOT trust ond to hold 1800w.

 

[TwoCables]

Quote:

Originally Posted by Klue22

Thats my own kill-a-watt that I used to own. And I only ever put 1300w of load on it. Granted it was 24/7 but the idea is kind of that you can monitor total usage. I would NOT trust ond to hold 1800w.

Well, I would call yours defective because like you said, it's designed to monitor usage and so it should be capable of handling that much power even 24/7.

However, you also have to remember that it's just a ~$20 wall meter that was made in China.

 

[Mega Man]

Nah that is from a loose connection.

Odds are it was super old. And worn out.

You can tell as the damage is limited to one side of the plug to much Wattage would of just broke the unit not burned one prong

It may of been new. And still loose

 

[candy_van]

We had a PSU calculator, at one point.
Probably the only one to ever hit the internet that didn't wildly over-estimate what people needed.

 

[Chargeit]

I just used mine the one time and it held up fine. I'm not sure I'd trust it for 24/7 usage though.

 

[TwoCables]

Well, mine is still just like new and I've had it plugged into my wall since the day I bought it in March of 2011.

 

[TTheuns]

I'm all for disproving faulty methods, but it would be nice to have an alternative that doesn't mean reading 5000 reviews. TDP is not the way to go, so the only way is to wait untill someone has tested that particular component?

 

[ASUSfreak]

If you REALLY want to know your REAL power draw... use a Fluke (not other brand crap, unless descent brand)

Make/use an extention cord (cut an existing one open without touching the inner insulation

)

Hold your clamp around ONE wire (not the ground one ofcourse) and measure the AC amps.

Now measure the AC voltage of that wall socket and multiply it

E.g. 227V x 4.3A = 976.1W --> always correct, heat losses included...

EDIT: seems I cannot insert picture atm so I included it in the attachement

stroommetingLarge.jpg 100k .jpg file

 

[TTheuns]

Quote:

Originally Posted by ASUSfreak

If you REALLY want to know your REAL power draw... use a Fluke (not other brand crap, unless descent brand)

Make/use an extention cord (cut an existing one open without touching the inner insulation

)

Hold your clamp around ONE wire (not the ground one ofcourse) and measure the AC amps.

Now measure the AC voltage of that wall socket and multiply it

E.g. 227V x 4.3A = 976.1W --> always correct, heat losses included...

EDIT: seems I cannot insert picture atm so I included it in the attachement

stroommetingLarge.jpg 100k .jpg file

Very helpful information for the future, but I'd like to know what my future components will pull before I buy them and the power supply.

 

[Mega Man]

Quote:

Originally Posted by ASUSfreak

If you REALLY want to know your REAL power draw... use a Fluke (not other brand crap, unless descent brand)

Make/use an extention cord (cut an existing one open without touching the inner insulation

)

Hold your clamp around ONE wire (not the ground one ofcourse) and measure the AC amps.

Now measure the AC voltage of that wall socket and multiply it

E.g. 227V x 4.3A = 976.1W --> always correct, heat losses included...

EDIT: seems I cannot insert picture atm so I included it in the attachement

stroommetingLarge.jpg 100k .jpg file

i have owned fluke ( now i use field piece, they are tailor made for HVAC which is what i do ) and you would be surprised how accurate the cheap DMMs are

you can also just buy this tool instead of modifying a cord ..wow, it is now discontinued
feildpiece als1
works great ! if you can find it i recommend it !

the downside is it DOES NOT tell you what your real wattage is, it tells you what you pull from the wall

 

[TwoCables]

Quote:

Originally Posted by TTheuns

Very helpful information for the future, but I'd like to know what my future components will pull before I buy them and the power supply.

Just for now, ask us. Make a thread on Overclock.net. That's the best option right now. I'm being serious. The world simply doesn't have a better choice right now (other than reading a professional review that properly tests power consumption). I mean really, what other options are there? We can't use eXtreme Outer Vision's PSU calculator anymore because their current one is horrible, so we're stuck at the moment.

However, with PSU prices these days, you could buy a high-end 750W PSU for just one video card and you could even never overclock and that would be perfectly fine. You see, the only thing bad about buying an overkill PSU is, you can end up wasting money that could've been spent on something else (for me, that would be groceries). So with PSU prices the way they are, this is not really much of an issue anymore - except of course for extremely tight budgets. Even then, we have high-powered options that aren't that bad at all.

Quote:

Originally Posted by Mega Man

i have owned fluke ( now i use field piece, they are tailor made for HVAC which is what i do ) and you would be surprised how accurate the cheap DMMs are

you can also just buy this tool instead of modifying a cord ..wow, it is now discontinued
feildpiece als1
works great ! if you can find it i recommend it !

the downside is it DOES NOT tell you what your real wattage is, it tells you what you pull from the wall

Exactly.

It doesn't matter how accurate the reading is, you still have to estimate how efficient your PSU is and then multiply your meter's readout by that estimated efficiency. So, if you're estimating 80% efficiency, then multiply by .80. For 85%, multiply by .85. For 90%, multiply by .90. Just put a decimal in front of your percentage and multiply your meter's readout by that number.

If your calculator has a percent button, then you can also just multiply the meter's readout by the estimated efficiency percentage. So, you just enter the meter's readout, press the multiply button, and then you enter the percentage (no decimal) followed by the percent sign and press the equals button.

 

[ASUSfreak]

Quote:

Originally Posted by Mega Man

i have owned fluke ( now i use field piece, they are tailor made for HVAC which is what i do ) and you would be surprised how accurate the cheap DMMs are

After a quick google I now know what HVAC means

Well I work on the Belgium electrical grid and we always use Fluke's. Some "clients" of us complained about haveing bad voltage at their home. When we come and measure it, it shows a nice result of 230/400V but when they use their "cheap DMM's" from the local store it does result in 10-20V difference in some cases

Nevertheless I don't mean all people should buy a Fluke

way to expensive for normal use

I don't have a fluk;e at home, I also have a cheap one, heck it's still from when I went to school

but if I'm to lazy to get to the toolbox in the cellar and get my cheap DMM, I just open the door and walk to my work-vehicule and take the flukle hehehe (it's a shorter route than going to the basement

)

Back on-topic: efficiency means A LOT in electrical stuff

it's like cos-phi (or how you guys call it in English???)