Why do i need an 850w PSU to power SLI/CF

[Pandamonia]

OP: Don't OC and that PSU should work fine actually. You will notice one definite disadvantage to run at capacity and that's noise. I overbuy my PSU not because I honestly think I need it but because running with a lot of headroom keeps the PSU fan off. If noise doesn't bother you, you don't live in a hot climate, and you are fine with not OCing for now, it really should work as long as there's nothing wrong with your unit.

i have my eyes on a AX860 for £140.00, i have some big commission coming in February so i might just have to wait until then

 

[ChaiBabbaChai]

you see this makes no sense to me. i would say that running at 620w is exactly what its designed to do or its not a 620w PSU.

It sounds like sales talk to me.

LOL - no dude... You sound like you don't know anything about electronics. He was right. You need the headroom. A 620w PSU isn't designed to run at 620w output constantly AND last for several years. Also, there are efficiency sweet spots for every different PSU. You're wasting your money (power bill) if you run it out of that range. While you may be drawing 620w from the PSU, the PSU is drawing much more than that from the wall. You can possibly find a good review on your specific model of PSU and see what the efficiency is in certain ranges that they tested...

If you know anything about cars, you would know that redlining the engine for a long time is going to break it very quickly. Same goes for power supplies, but for different reasons. It isn't marketing BS at all. If anything, the max rating of the PSU is marketing BS and running it at that peak for extended durations might prove fatal to the PSU. Ever heard of RMS and Peak output for speakers and power amps?

 

[coffeejunkee]

Hmm, comparing cars and pc's usually doesn't work well. If I buy a 620W psu I actually expect it to be able to put out 620W for at least the duration of the warranty. Do you have any examples of psu's dying prematurely from running at their max load, preferably in a controlled testing environment?

About efficiency, yes most psu's run most efficient around the 50% mark. But in the case of a general purpose/gaming rig (which we are talking about here) having a lot of headroom only means you will have really bad efficiency during idle.

 

[spat55]

i have my eyes on a AX860 for £140.00, i have some big commission coming in February so i might just have to wait until then

I got a Evga Supernova g2 1000w for £130, 80+ Gold and fully modular. Excellent value for money

 

[lehtv]

Also, there are efficiency sweet spots for every different PSU. You're wasting your money (power bill) if you run it out of that range. While you may be drawing 620w from the PSU, the PSU is drawing much more than that from the wall. You can possibly find a good review on your specific model of PSU and see what the efficiency is in certain ranges that they tested...

Actually, efficiency at 100% load is typically only about 2 percent lower than at the optimal 50%-60% load. This difference alone is in most cases enticing enough to justify spending considerably more on a unit that would operate closer to the optimal load. You are absolutely correct though that running a unit at full load is not a good idea in the long term

 

[UaVaj]

if you "clearly" know the amount of constant power draw.

simply buy a psu that is "verified" to deliver that constant power draw.

not rocket science.



as for getting a psu with head room. how many of you insist on paying $2 for for $1 hamburger?

 

[lehtv]

as for getting a psu with head room

Yes, go on.

how many of you insist on paying $2 for for $1 hamburger?

Huh?

 

[Pandamonia]

LOL - no dude... You sound like you don't know anything about electronics. He was right. You need the headroom. A 620w PSU isn't designed to run at 620w output constantly AND last for several years. Also, there are efficiency sweet spots for every different PSU. You're wasting your money (power bill) if you run it out of that range. While you may be drawing 620w from the PSU, the PSU is drawing much more than that from the wall. You can possibly find a good review on your specific model of PSU and see what the efficiency is in certain ranges that they tested...

If you know anything about cars, you would know that redlining the engine for a long time is going to break it very quickly. Same goes for power supplies, but for different reasons. It isn't marketing BS at all. If anything, the max rating of the PSU is marketing BS and running it at that peak for extended durations might prove fatal to the PSU. Ever heard of RMS and Peak output for speakers and power amps?

I guess the PSU inside the PS4 is a 450w then? It uses 300w so did they build in the extra headroom?

I doubt they did personally i bet they have a PSU that is rated for 300w. The difference is most PC users have no idea how much their system needs.

 

[tolis626]

I guess the PSU inside the PS4 is a 450w then? It uses 300w so did they build in the extra headroom?

I doubt they did personally i bet they have a PSU that is rated for 300w. The difference is most PC users have no idea how much their system needs.

I don't know about the PS4,but I had 2 PS3s die on me (The first one while I was setting a personal record in CoD MW2 ),both due to "Power delivery system failure".I think they meant the PSU,but understand whatever you want.Plus,the consoles use a lot less energy than a PC.Really almost an order of magnitude less than some extreme systems around here.

Another sad story involving a bad PSU was my father's previous PC.I tried to persuade him to go with a larger PSU than he initially had in mind,but he insisted that it was "Money thrown out of the window".He didn't listen and went with a cheap-ish (Don't remember which brand,it's been 4 years since then) PSU that was just enough.He also didn't overclock,so he felt safe.6 months after,the PSU went kamikaze and took the GPU,the mobo and a hard drive with it.It also smelled really bad,but that's another thing.

My personal experience,therefore,pushes me towards a "Don't cheap out on the PSU!!!" line of thought when buying a PC.The decision is yours.Just don't cry if something similar happens.I know I almost cried.

 

[Torn Mind]

I really have my doubts that degradation is a compelling reason to not run a PSU at full load. If the voltage is in the ATX spec and the waveform is within spec at a given amperage, the PSU should then function properly.

However, not running a PSU at full load in order to make sure that the components do not accidentally operate out of spec and thus cause failure at least has basis in reality.

Running at full load might be dangerous if the PSU is exposed to transients such as surges, undervoltage, etc, and without a line conditioner upstream, yes, the PSU is vulnerable to "shocks" in the AC power delivery.

It would be interesting for those with an oscilloscope to test an old PSU can see what its waveform is like at 100% load.

 

[Slomo4shO]

I guess the PSU inside the PS4 is a 450w then? It uses 300w so did they build in the extra headroom?

I doubt they did personally i bet they have a PSU that is rated for 300w. The difference is most PC users have no idea how much their system needs.

Not sure where you are getting the idea that the PS4 consumes 300W on load...

 

[Torn Mind]

I don't know about the PS4,but I had 2 PS3s die on me (The first one while I was setting a personal record in CoD MW2 ),both due to "Power delivery system failure".I think they meant the PSU,but understand whatever you want.Plus,the consoles use a lot less energy than a PC.Really almost an order of magnitude less than some extreme systems around here.

Another sad story involving a bad PSU was my father's previous PC.I tried to persuade him to go with a larger PSU than he initially had in mind,but he insisted that it was "Money thrown out of the window".He didn't listen and went with a cheap-ish (Don't remember which brand,it's been 4 years since then) PSU that was just enough.He also didn't overclock,so he felt safe.6 months after,the PSU went kamikaze and took the GPU,the mobo and a hard drive with it.It also smelled really bad,but that's another thing.

My personal experience,therefore,pushes me towards a "Don't cheap out on the PSU!!!" line of thought when buying a PC.The decision is yours.Just don't cry if something similar happens.I know I almost cried.

Yes, cheap out enough and you will find garbage. Diablotek, Logisys, and whatever old PSUs out there that will run but totally lose voltage and/or ampere regulation when something in the PSU gives up the ghost, most likely a capacitor in the secondary and no proper feedback circuit to disable the PSU when it is dead.

But there are degrees to cheaping out. A Corsair CX430 might result in a higher incidence of reliability issues, but blowing up downstream components is much less likely than the utter garbage tier. Then there are the "high-quality OEM" PSUs from Seasonic floating around for about $45-$50, which have lower grade caps but won't blow things up. And then, at least for the U.S market, we find the Rosewill Capstone 450 and Seasonic G-360, which have Japanese caps and quality manufacturing processes.

However, shelling out big bucks is only worth if there are tangible returns. $60 for a Seasonic G-360 or Rosewill Capstone 450 and $100-$200 for upstream devices such as a UPS or line conditioner that grants actual extra protections is a better buy for systems that pull less than those wattages than going big and buying an AX860.

 

[Makaveli]

Hmm, comparing cars and pc's usually doesn't work well. If I buy a 620W psu I actually expect it to be able to put out 620W for at least the duration of the warranty. Do you have any examples of psu's dying prematurely from running at their max load, preferably in a controlled testing environment?

About efficiency, yes most psu's run most efficient around the 50% mark. But in the case of a general purpose/gaming rig (which we are talking about here) having a lot of headroom only means you will have really bad efficiency during idle.

Coffeejunkee he is correct.

And that other guy is clearly showing his inexperience in building.

You do not run a PSU at maximum wattage if you want it to last.

This is common knowledge.

Use google to find the information its out there!

 

[Pandamonia]

Coffeejunkee he is correct.

And that other guy is clearly showing his inexperience in building.

You do not run a PSU at maximum wattage if you want it to last.

This is common knowledge.

Use google to find the information its out there!

I have never tried running it at full but it also will only be full during load when gaming i expect.

Its also a HX620w which is a brilliant PSU, just not enough for 2x780 GTX

 

[TemjinGold]

This thread has made me curious: Does anyone actually have any posted proof that running a QUALITY PSU (not a crap brand) at max rated wattage will kill the PSU or lower its lifespan? Sure I hear plenty of people chanting that and plenty more continue to chant it because others do. I'm not advocating anything here. I'm just curious if anyone actually recalls ever seeing real life proof of this happening. Because if not, that's one seriously powerful trick that marketers have played on the market.

There are TONs of people scared to not overbuy their PSUs now and lots of them going, "Oh you'll be sorry if you do!" or "I buy 1k watt to be safe!" etc. I'm not going to judge anyone but I am really curious to find out the truth. How many of you have ACTUALLY (first-hand experience or seen it, not "heard it from someone") witnessed this? I will be the first to say I've never seen it. I've seen plenty where someone buys like Diablotek crap and it ends badly but never seen a quality unit like a corsair/seasonic/antec/etc. end badly because someone didn't overbuy.

 

[Pandamonia]

This thread has made me curious: Does anyone actually have any posted proof that running a QUALITY PSU (not a crap brand) at max rated wattage will kill the PSU or lower its lifespan? Sure I hear plenty of people chanting that and plenty more continue to chant it because others do. I'm not advocating anything here. I'm just curious if anyone actually recalls ever seeing real life proof of this happening. Because if not, that's one seriously powerful trick that marketers have played on the market.

There are TONs of people scared to not overbuy their PSUs now and lots of them going, "Oh you'll be sorry if you do!" or "I buy 1k watt to be safe!" etc. I'm not going to judge anyone but I am really curious to find out the truth. How many of you have ACTUALLY (first-hand experience or seen it, not "heard it from someone") witnessed this? I will be the first to say I've never seen it. I've seen plenty where someone buys like Diablotek crap and it ends badly but never seen a quality unit like a corsair/seasonic/antec/etc. end badly because someone didn't overbuy.

I think it used to be that bronze PSU's lose efficiency after 75-80% but todays Platinum dont. So maybe things have changed?

 

[Slomo4shO]

This thread has made me curious: Does anyone actually have any posted proof that running a QUALITY PSU (not a crap brand) at max rated wattage will kill the PSU or lower its lifespan?

The most common PSU failures are related to capacitors and fans. Both of these two factors are stressed at higher temperatures. Models running at near or above 100% capacity will run hotter than a model running at 50-80%. In turn, the rate of failure of the fan or capacitors will increase due to higher the fan running at maximum RPMs for extended periods of time or the life span of capacitors diminishing as temperatures rise.

Here are good reads:

http://www.jonnyguru.com/modules.php?name=NDArticles&op=Story&ndar_id=8
http://www.jonnyguru.com/modules.php?name=NDFAQs&op=FAQ_Question&ndfaq_id=6
http://www.gigabyte.com/microsite/111/tech_070112_ud_whyusesolid.htm
http://www.illinoiscapacitor.com/tech-center/life-calculators.aspx

 

[TemjinGold]

Yep, I get that and appreciate you offering up something real. However, none of this points directly to quality units dying from running AT max rated capacity for extended periods. Note that quality units can actually run at levels beyond what they are rated and I have no doubt what you are talking about would definitely happen should this be the case for extended periods.

 

[Torn Mind]

This thread has made me curious: Does anyone actually have any posted proof that running a QUALITY PSU (not a crap brand) at max rated wattage will kill the PSU or lower its lifespan? Sure I hear plenty of people chanting that and plenty more continue to chant it because others do. I'm not advocating anything here. I'm just curious if anyone actually recalls ever seeing real life proof of this happening. Because if not, that's one seriously powerful trick that marketers have played on the market.

There are TONs of people scared to not overbuy their PSUs now and lots of them going, "Oh you'll be sorry if you do!" or "I buy 1k watt to be safe!" etc. I'm not going to judge anyone but I am really curious to find out the truth. How many of you have ACTUALLY (first-hand experience or seen it, not "heard it from someone") witnessed this? I will be the first to say I've never seen it. I've seen plenty where someone buys like Diablotek crap and it ends badly but never seen a quality unit like a corsair/seasonic/antec/etc. end badly because someone didn't overbuy.

I doubt it. The best most people can do is send their old units to a hardware review site like JonnyGURU or Hardware Secrets and ask the review(OklahomaWolf and Gabriel Torres, respectively) to analyze the output electrical waveform at various levels of amperage because obtaining an oscilloscope and load tester is an expensive endeavor for an individual.

If the voltage of the rails stay within ATX specs, that is one test passed.
If the ripple/noise stay within ATX specs and no other funny stuff shows on the oscilloscope, that is another test passed.
If the above works at the max rated wattage, then the PSU should be good to go, for the most part at least. Perhaps some strange inconsistencies might happen with the 5V standby rail and thus some sleep issues might occur, but if nothing is detected on the scope and everything is in spec, the main thing to worry about are when the capacitors will finally breath their last and transients that cause the PSU components to run out of spec.

 

[TemjinGold]

I think it used to be that bronze PSU's lose efficiency after 75-80% but todays Platinum dont. So maybe things have changed?

Efficiency has nothing to do with lifespan.

 

[tolis626]

This thread has made me curious: Does anyone actually have any posted proof that running a QUALITY PSU (not a crap brand) at max rated wattage will kill the PSU or lower its lifespan? Sure I hear plenty of people chanting that and plenty more continue to chant it because others do. I'm not advocating anything here. I'm just curious if anyone actually recalls ever seeing real life proof of this happening. Because if not, that's one seriously powerful trick that marketers have played on the market.

There are TONs of people scared to not overbuy their PSUs now and lots of them going, "Oh you'll be sorry if you do!" or "I buy 1k watt to be safe!" etc. I'm not going to judge anyone but I am really curious to find out the truth. How many of you have ACTUALLY (first-hand experience or seen it, not "heard it from someone") witnessed this? I will be the first to say I've never seen it. I've seen plenty where someone buys like Diablotek crap and it ends badly but never seen a quality unit like a corsair/seasonic/antec/etc. end badly because someone didn't overbuy.

Well,run any electronic device at 100% for extended periods and witness its downfall.

If a PSU's circuits are constantly running at capacity,it will wear down much sooner.It's inevitable.Not saying that quality PSUs will cause disasters like those I mentioned in my previous post,but still,it's a risk sometimes.

Also,I don't think many people will suggest a 1000W PSU for someone who will use a 4670(Non-k) and a 280x at stock clocks,but using a 600-650W PSU for CF/SLI is cutting it too close and asking for trouble,if it's enough at all.So I'd say exaggerate a little when buying PSUs.Just a little.

 

[Slomo4shO]

Yep, I get that and appreciate you offering up something real. However, none of this points directly to quality units dying from running AT max rated capacity for extended periods. Note that quality units can actually run at levels beyond what they are rated and I have no doubt what you are talking about would definitely happen should this be the case for extended periods.

Sorry, there is no case study to refer you to for your piece of mind. Nevertheless, the principles of physics do apply and they have been proven time and time again via experimentation. Temperature and high voltage/current deteriorate electronic components. You can piece the rest together for yourself.

 

[Makaveli]

I have never tried running it at full but it also will only be full during load when gaming i expect.

Its also a HX620w which is a brilliant PSU, just not enough for 2x780 GTX

It is a good psu.

 

[Torn Mind]

The most common PSU failures are related to capacitors and fans. Both of these two factors are stressed at higher temperatures. Models running at near or above 100% capacity will run hotter than a model running at 50-80%. In turn, the rate of failure of the fan or capacitors will increase due to higher the fan running at maximum RPMs for extended periods of time or the life span of capacitors diminishing as temperatures rise.

Here are good reads:

http://www.jonnyguru.com/modules.php?name=NDArticles&op=Story&ndar_id=8
http://www.jonnyguru.com/modules.php?name=NDFAQs&op=FAQ_Question&ndfaq_id=6
http://www.gigabyte.com/microsite/111/tech_070112_ud_whyusesolid.htm
http://www.illinoiscapacitor.com/tech-center/life-calculators.aspx

The fan would, or at least should, be running full blast if the PSU is running at max load while a bigger unit would take longer for the fan to kick in. Of course, there is the matter of how much air is being moved in an out of the unit by the different fans.

 

[YBS1]

i bet you £10 that your system is pulling nowhere near 1200w from the wall.

Power Consumption GeForce GTX 780 Ti 2-way SLI

System in IDLE = 129W
System Wattage with GPU in FULL Stress = 617W
Difference (GPU load) = 488W
Add average IDLE wattage ~20W
Subjective obtained GPU power consumption = ~ 508 Watts

"Our test system is based on a power hungry six-core Intel Core i7-3960X Extreme Edition Sandy Bridge-E based setup on the X79 chipset platform. This setup is overclocked to 4.60 GHz on all cores"

You would be wrong....Groove's system and mine are very similar and you can search for the thread on mine shutting down a 1300 watt platinum rated psu during 3DMark runs. A 1600 watt Lepa solved that. Also, the power a 4.6GHz SB-e pulls is miles away from what a 5.0+GHz will be pulling.