Is there any way to measure PCI Power Usage?

[minifridge1138]

Hello Everyone,

I'm wondering if there is any way to measure the amount of power being used by the PCI rows on the LB.

I' thinking of adding a GT 640 for an additional 65W. I know it isn't a powerful card, but it's a cheap way to get 10X the number of cuda cores I have now (GT 120 has 32 cuda cores). I don't need to power a display with it.

I already have:
Apple 5870 = 75W (6 pin) + 75W (6pin) + 38W (PCI) = 188W (AMD Spec)
Apple GT 120 = 50W (PCI) = NVidia Spec

The GT 640 would only use 65W from the PCI lane, so i'm good on paper. I'd just like to see what I'm actually using.

thanks!!!

 

[Thessman]

Hello Everyone,

I'm wondering if there is any way to measure the amount of power being used by the PCI rows on the LB.

I' thinking of adding a GT 640 for an additional 65W. I know it isn't a powerful card, but it's a cheap way to get 10X the number of cuda cores I have now (GT 120 has 32 cuda cores). I don't need to power a display with it.

I already have:
Apple 5870 = 75W (6 pin) + 75W (6pin) + 38W (PCI) = 188W (AMD Spec)
Apple GT 120 = 50W (PCI) = NVidia Spec

The GT 640 would only use 65W from the PCI lane, so i'm good on paper. I'd just like to see what I'm actually using.

thanks!!!

1 istat
2 Hardware Monitor

 

[ThisIsNotMe]

1) Buy Card
2) Install Card
3) Test Card
4a) Profit
4b) Return Card For Full Refund

 

[Inconsequential]

Hello Everyone,

I'm wondering if there is any way to measure the amount of power being used by the PCI rows on the LB.

I' thinking of adding a GT 640 for an additional 65W. I know it isn't a powerful card, but it's a cheap way to get 10X the number of cuda cores I have now (GT 120 has 32 cuda cores). I don't need to power a display with it.

I already have:
Apple 5870 = 75W (6 pin) + 75W (6pin) + 38W (PCI) = 188W (AMD Spec)
Apple GT 120 = 50W (PCI) = NVidia Spec

The GT 640 would only use 65W from the PCI lane, so i'm good on paper. I'd just like to see what I'm actually using.

thanks!!!

What are you worried about?

You've got 75W from the slot irrespective of what else is plugged in to other slots.

istat menus shows power draw (version 3 NOT 4). Only accurate to +/- 10% I'd imagine.

 

[Tesselator]

Only accurate to +/- 10% I'd imagine.

Why would the accuracy be lower than "near exactly" or "spot on"?

 

[666sheep]

istat menus shows power draw (version 3 NOT 4). Only accurate to +/- 10% I'd imagine.

v4 does that too. I use this one. Accuracy seems pretty good, it shows 2 digits after comma. Reads amps, but you can quickly convert them using e.g. this: http://www.rapidtables.com/calc/electric/Amp_to_Watt_Calculator.htm

 

[Inconsequential]

v4 does that too. I use this one. Accuracy seems pretty good, it shows 2 digits after comma. Reads amps, but you can quickly convert them using e.g. this: http://www.rapidtables.com/calc/electric/Amp_to_Watt_Calculator.htm

Has he put power draw back then? It's one of the reasons why I went back to 3.

Also, just because it goes down to 2 decimal places does not mean it is an accurate value, just that it has a high resolution.

 

[666sheep]

Yes, power draw reading works in 4.05 I have.

I understand that iStat reads values from hardware sensors, correct? If so, it should be as accurate as sensors are.

 

[Inconsequential]

Yes, power draw reading works in 4.05 I have.

I understand that iStat reads values from hardware sensors, correct? If so, it should be as accurate as sensors are.

The sensors arnt going to be accurate to much more than 5%, or even if we are lucky 1%.

 

[666sheep]

It's still 200% better than 10%