idle mode

[cj3waker]

I finally got a kill a watt meter and i found sometheing pretty interesting

only used on a 775 system so far.
750i ftw
q9550 @ 3.5
2x2gb ddr2
8800gt @ 700mhz
(2) sata hdds and sata Optical
(4) sff21e
ocz modstream 500w

at complete idle the sytem runs at 144w, any usage jumps it up to 158w, and it will stay there until an actual load seems to be put on. game menus loading screens w.e im still ~158. gaming for a few hours puts me at a consistent 250w. and under furmark and prime 95 I only get up to 285. Obviously this is an older system, but the q9550 and 8800gt are much less efficient than any 1156 offerings or newer midrange gpus.

Ill test out my mini itx 1156 gaming rig with its gtx 570 next. and i cant wait to see how low sandy bridge is. Ive been feeling that people value wattage on PSUs way more than quality, On this 775 system even this 500w could be considered overkill. At max load I cant not pull over 300w from the wall. Meaning a 400w PSU could pull this entire system as less than 75% load. Obviously a bigger psu means higher upgrade possibilities, but I think most people buy much more wattage than they need and perhaps a quality lower wattage PSU would be a better purchase.

Neweggs PSU calc put this system at 535

ill update with the 1156 shortly

 

[tiraides]

You should treat PSU wattage like RPMs on a car, to a certain degree: the longer and the more often you push it towards the max, the shorter lifespan it will have. In your case, a good 500 watt power supply should last you years as your system would never use more than 60% of its max power. (I recall several PSU experts here noting that PSUs generally run at their best while using 40% to 60% of their rated loads.)

 

[cj3waker]

I disagree, With a strong PSU you should be able to run much higher than 40%. with an 80% efficient PSU at complete max load my system should only be using 228w, the remaining power is lost due to inefficiency. So with a 500w psu im running less than 50% load.

Update: I finally added an ssd to my mini itx rig so ive gotta wait for steam to re dl some games before I can test. also everything is stock because of some bios upgrades

setup
h55n-usb3
i3 530
2x2gb ddr3
intel x-25v
2tb samsung 5400
(2) scythe GT
EVGA GTX 580

idles at 80w. And i believe I can shave some voltage off. Ill have gaming and full load wattage soon

edit: *per killawatt meter* which i learned may not be as reliable as I had hoped. Im looking at about the same
~250 gaming ~290 full cpu and gpu load

 

[Dangman]

Here's Paul Johnson's post, PSU reviewer of HardOCP.com, about the inaccuracy of the Kill-A-Watt:
http://hardforum.com/showpost.php?p=1032190998&postcount=7

Another Paul Johnson's post:
http://www.jonnyguru.com/forums/showpost.php?p=27745&postcount=26

In addition, three other PSU experts backs up Paul Johnson's statement:
Oklahoma Wolf of JonnyGuru.com:
http://hardforum.com/showpost.php?p=1034843536&postcount=21
http://hardforum.com/showpost.php?p=1034555880&postcount=17

JonnyGuru of JonnyGuru.com and the former senior PSU engineer over at BFG acknowledge the inaccuracies of a KAW (Post #7 in regards to Post #2):
http://www.jonnyguru.com/forums/showthread.php?t=5977
http://www.jonnyguru.com/forums/showpost.php?p=29907&postcount=2

Redbeard of Corsair also acknowledges the inaccuracies:
http://hardforum.com/showpost.php?p=1032811067&postcount=22

Related discussions:
http://hardforum.com/showthread.php?t=1447774
http://hardforum.com/showthread.php?t=1509210
http://www.jonnyguru.com/forums/showthread.php?t=3102
http://www.jonnyguru.com/forums/showthread.php?t=2695
http://www.jonnyguru.com/forums/showthread.php?t=2284

And this is the quote from Paul Johnson, again the PSU editor here at HardOCP discussing power supplies in general:

Yes and a quick search would turn up this topic a million times over. Here is the recap:

1) APFC can fool Kill-A-Watts into giving you abnormally low readings (some times giving better than 100% efficiency)

2) Power supplies derate with temperature anywhere from 2w/c above a nominal rated at value to 10w/c.

3) Kill-A-Watt's and most power meters sample too slowly to catch transient loads (the Transient load from our tests is 117w and is COMPLETELY missed by Kill-A-Watts).

4) Power supplies last longer if you stay in the 40% to 60% range of their output.

5) power supplies are quieter if you stay in the 40% to 60% range of their output.

6) Power supplies are cooler if you stay in the 40% to 60% range of their output.

The power meters in UPS software are just as bad. You have to spend some change before you get anywhere near an accurate power meter when your PSU has APFC.

Also note that Tiraides said 40% TO 60% load. You've must've thought he said only 40% or something.

 

[cj3waker]

220 vs 229? that doesnt seem like much a difference

 

[Dangman]

220 vs 229? that doesnt seem like much a difference

Huh? I can figure out where that 229 figure comes from but not the 220. Elaborate please.

Look, all I'm saying is that you shouldn't trust those KAW numbers that much. They are sometimes ballpark accurate. But unfortunately the chance of them not being ballmark accurate is really high as evident by the above posts, threads, and statements. So while your KAW may say that you're only pulling 285W at max full load, considering at how inaccurate they can be sometimes, that figure may be dramatically lower than what's actually happening. Look at the very first Paul Johnson link I provided: The KAW reported the load was 215W. However the significantly more accurate power meter to the left of that KAW showed the actual load of 353W. Thats a big difference right there.

In addition, as clearly stated in the post I quoted, PSUs do last longer, are quieter, and run cooler when in the 40% to 60% range of their total power output. Can you run PSUs at higher loads? Why yes you can. But it wouldn't exactly be an optimal solution as said above. Now that is a general guideline and not hard law so people will and still run their PSUs at 70% to even 80% of the total capability. So when I personally make PSU recommendations, I try to keep those percentages in mind as well as the possibility of future upgrades.

However, while I do agree with you in general that people put too much focus on the wattage and sometimes buy more than they actually need, I have to disagree with you with making a blanket statement that people should get higher quality but lower wattage PSUs. Here's the deal: There are many situations where the higher quality but lower wattage PSU costs just as much as the higher wattage but still decent quality PSU. Considering that more than likely that the average enthusiast honestly wouldn't notice the difference between the two PSUs in terms of quality. So from a price to performance standpoint, considering that you're gonna get more headroom for future upgrades, closer to the optimal PSU load conditions, and get roughly the same noticeable performance with that larger PSU for the same price as that lower wattage but higher quality PSU, it isn't always a good idea to just say "go with the lower wattage PSU".

In addition, there are a lot of people who tend to stick with the same PSU over many years through multiple PC upgrades. So while that lower wattage PSU may have been fine for their current system that point in time, that probably won't be the case 2-3 years down the line with they upgrade their PCs with newer parts. So you should not gloss over that particular fact that when buying a PSU, you have to keep future PC upgrades in mind.

 

[cj3waker]

those links you posted show him testing a killawatt meter vs the actual wattage displayed on the bigger unit correct? in all the screens its off by a slight amount. ie 220 on the killawatt but really 229? that was the only thing i could discern from the posts.

 

[Dangman]

those links you posted show him testing a killawatt meter vs the actual wattage displayed on the bigger unit correct? in all the screens its off by a slight amount. ie 220 on the killawatt but really 229? that was the only thing i could discern from the posts.

Ok referring to the first link I posted: the big unit is display the actual DC wattage. PSUs uses and are rated at DC, not AC power. As such, a computer power supply has to covert AC power into DC power and thats where a bit of power is usually lost during the conversion. Thats what efficiency generally refers to. A KAW tries to read/display the AC wattage at the wall, not DC at all. So that picture shows that the actual DC load on the PSU is 219W. The KAW is showing 215W AC power at the wall. That would mean that the PSU has 102% efficiency. See something wrong there?

Now that green box next to the KAW is the Digital Power Meter: It's significantly more accurate that a KAW when it comes to measure AC wattage being converted to DC wattage. In that device, it's showing 2.94A. Multiply that by 120V and you get 353W, the actual AC power usage. That's 62% efficiency there.

Now referring to the second link: The actual DC load is 691W. The KAW reads the AC load as 638W. That would mean that PSU is 108% efficient according to the KAW. Again, see something wrong there?

So KAW can be off by a large margin as shown above.

 

[cj3waker]

interesting. ive got a buddy whose an electrical engineer who swears by the unit i have. ill have to dig some answers out of him.

 

[Dangman]

interesting. ive got a buddy whose an electrical engineer who swears by the unit i have. ill have to dig some answers out of him.

Yeah look at the discussions above: No one knows exactly why the KAW is so bloody inaccurate with modern PSUs. By all accounts it shouldn't be happening but it is.

 

[cj3waker]

how do we know the wattage is incorrect and not the amperage?

 

[Dangman]

how do we know the wattage is incorrect and not the amperage?

They're both related dude: Wattage is determined by VA or voltage x amperage. So if the amperage is incorrect, the wattage is still incorrect.

Again, the Digital Power Meter can read amperage and determine wattage just fine most of the time.

 

[cj3waker]

I realize that, but the power meter was showing wattage near the killawatt meter, correct? so what if the error isnt with the wattage reading on both units, but the amperage reading?

 

[MrWizard6600]

The kill-a-watt's are really designed to measure true ohmnic (aka resistive) devices, they expect 0° separation between voltage and current, meaning all apparent power is true power with no reactive power.

The issue of keeping all apparent power (or all Volt-Amps) used is known as Power Factor Correction (PFC), and is the vector ratio of true power : apparent power, meaning 1 is ideal. A lower PFC (meaning a higher reactive power load aka a higher inductive load) doesn't end up putting a charge on your bill but if you end up putting huge industrial-sized inductive loads on your PowerCo's lines they might well end up sending somebody by asking wtf you're doing, because it screws up their power delivery for the whole neighborhood.

Similarly a bad power factor by say the power supply used by your monitors or speakers will push the voltage wave off the current wave for all the devices drawing power from the same circuit, meaning a kill-a-watt's measurements for a PC are actually effected by your monitors and speakers, if they're on the same circuit.

So yeah, long story short, Kill-a-watt's are bad, or at least, they're what you expect when you try to get $20 worth of hardware to do $1000 worth of hardware's job. The only way to really get a good picture of what's going on is to fire whats coming out of your wall under a scope. AC is great because we can go up to the MegaVolts for long distance travel and bring it back down to a nice usable 120V for local use, with all the transformation being done at 99% efficiency, but when you need to pull a DC current out of that sinusoidal wave, what you end up with can be really problematic. The actual technical explanation for what powers going where at any point on the voltage period requires imaginary numbers, matrices, and annoyingly hard calculus.

 

[cj3waker]

thankyou mrwizard, so is there a way to discern what types of electronics will read incorrectly? or is the killawatt meter basically useless in every regard?

 

[Dangman]

I realize that, but the power meter was showing wattage near the killawatt meter, correct? so what if the error isnt with the wattage reading on both units, but the amperage reading?

The power meter was showing the amperage not actually wattage. You have to multiply that amperage reading by 120V to get the true wattage. And an error with amperage reading is not probable.

Also, basically any PSU or device has a PSU that uses APFC will have pretty bad inaccurate readings with the KAW. With that said, if you had actually read some of the links I posted, even multiple KAWs will disagree with one another with something as simple as household light bulb.

 

[MrWizard6600]

Also, basically any PSU or device has a PSU that uses APFC will have pretty bad inaccurate readings with the KAW. With that said, if you had actually read some of the links I posted, even multiple KAWs will disagree with one another with something as simple as household light bulb.

now that just sounds like cheap sh!t to me.

ps, congrats on becoming a mod.... now I've gotta edit all those posts where I called you names....

 

[Dangman]

now that just sounds like cheap sh!t to me.

ps, congrats on becoming a mod.... now I've gotta edit all those posts where I called you names....

LOL! I can still see the original edits ya know

Thanks for the congrats!

 

[dogbert_2001]

The power meter was showing the amperage not actually wattage. You have to multiply that amperage reading by 120V to get the true wattage. And an error with amperage reading is not probable.

Also, basically any PSU or device has a PSU that uses APFC will have pretty bad inaccurate readings with the KAW. With that said, if you had actually read some of the links I posted, even multiple KAWs will disagree with one another with something as simple as household light bulb.

Shouldn't APFC make the KAW more accurate, since it's trying to make the PSU look more resistive?

 

[Dangman]

Shouldn't APFC make the KAW more accurate, since it's trying to make the PSU look more resistive?

Not sure about that.