Eight power supplies compared

I‘VE EXTOLLED the virtues of quality power supplies so many times that saying they’re the most often overlooked PC component feels like a cliche. But it’s true. In my experience, poor quality generic-brand power supplies are often the root of PC stability problems. Also, as new processors and graphics cards demand more power than ever before, quality power supplies are becoming more important.

But what makes a good power supply, anyway? Most assume that power supplies carrying higher wattage ratings are superior, but as the collapse of the so-called “megahertz myth” illustrates, higher numbers aren’t always better. First and foremost, a power supply should deliver clean, consistent power to system components. Power supply efficiency is also important; an efficient power supply can save you money on every electricity bill, especially if you have your system running 24/7. Environmental variables like temperatures and noise levels matter, too.

Armed with a couple of test systems, temperature probes, noise level and power consumption meters, an oscilloscope, and an all-important bathroom scale, we can test all those metrics, plus a few others, ourselves. There’s no need to rely on manufacturer spec sheets, marketing claims, or wattage ratings. Are cheap, generic power supplies really that much worse than high-end models that cost twice as much, or more? And among those high-end power supplies, is there really much difference from manufacturer to manufacturer? Let’s find out.

Comparing specs
Here’s how the power supply units (PSUs) we’ll be looking at compare in terms of specifications. By the way, the SH-ATX465P4 is a no-name unit I picked up at a local shop for $40 for the sake of comparison.

PSU Wattage PFC Fans Warranty Price
Antec NeoPower 480W Active 1 120mm 3 years $110
Enermax EG485P-SFMA24P 485W Passive 2 80mm, 1 blower 3 years $129
OCZ PowerStream 520W Passive 2 80mm 5 years $130
SH SH-ATX465P4 465W NA 2 80mm NA $40
Silverstone SST-ST30NF 300W Active None 3 years $155
Ultra X-Connect 500W None 2 80mm 1 year $99
Vantec Ion 2 350W None 1 120mm 3 years $45
Zalman ZM400B-APS 400W Active 1 80mm 1 year $87

I should mention that we also invited RaidMax, Thermaltake, and PC Power & Cooling to participate in this comparison, but they either declined or didn’t respond to our requests.

We’re looking at a full range of power supply wattages—from 300W all the way up to 520W. Conventional wisdom would probably suggest that more wattage is better, but as we’ll soon see, the PSUs with lower wattage ratings hold their own.

In addition to a range of wattages, we also have several power factor correction (PFC) implementations. Power factor (PF) is the ratio of true power to apparent power where true power refers to power that’s actually being consumed and apparent power is calculated by multiplying the voltage by the current being drawn. If voltage and current are out-of-phase, apparent power will be greater than real power. This is a common condition for circuits that contain capacitors and inductors—power supplies being a perfect example—and it results in a PF less than 1. PFC aims to bring voltage and current back into phase, raising the PF towards a goal of 1, where real power equals apparent power.

In their quest for a more ideal power factor, the Antec, Silverstone, and Zalman power supplies employ dedicated circuitry to correct out of phase voltage and current. This is referred to as active PFC. The Enermax, and OCZ power supplies use passive components like capacitive filters to correct power factor, which is appropriately referred to as passive PFC. In general, active PFC results in a higher power factor than passive PFC, which in turn results in a higher power factor than no PFC at all. But does having a higher power factor really do you any good?

Yes and no.

If you’re a large business running banks of machines, the power company may charge you for the apparent power draw you present to the grid rather than your real power consumption. In this case, a lower power factor can cost you money over time. If you’re running an uninterruptible power supply (UPS), a lower power factor can also result in shorter run times off battery. However, if you’re a regular home user with no UPS and no power factor-related charges on your electricity bill, PFC isn’t going to do anything for you. It will, however, make your local power company happy; higher client power factors allow for more efficient distribution of power across the grid.

Speaking of efficiency, overall power supply efficiency is also something to consider. In this case, efficiency refers to the percentage of real power drawn by the power supply that’s actually delivered to system components. More efficient power supplies will draw less real power to run a system, saving even home users on electricity costs. More efficient power supplies should also run cooler since the percentage of real power drawn that isn’t delivered to system components is radiated as heat. Not all power supply manufacturers publish efficiency specs, but we’ll be able to compare the relative efficiency of the power supplies we’re looking at by measuring real power consumption at the outlet.

Our stack of power supplies offers plenty of variety on other fronts, like cooling. At one end of the spectrum, there’s a passively cooled unit from Silverstone. At the other, we have an Enermax unit with two fans and a blower. And, of course, there are a number of single and dual-fan implementations in between. We’ll see how different cooling designs impact system temperatures and noise levels in a moment.

On the warranty front, three years seems to be the consensus. There are a few deviations, though. OCZ deserves extra praise for offering a generous five-year warranty, while Ultra and Zalman should be scolded for their stingy one-year warranties. Since a quality power supply can potentially handle a few years worth of system upgrades, a good warranty is essential.

Finally, we come to price, where we have quite a spread. There’s a $115 gap between our cheapest and most expensive PSUs and plenty of models to fill in the gap. It’s interesting to note that our two lowest wattage PSUs are so far apart in price. Vantec’s 350W Ion 2 retails for only $45, while Silverstone’s passive 300W PSU sells for more than three times that.

Wait, there’s more. Each of the PSUs we’re looking at also has a different mix of power connectors. Rather than go through them all individually, I’ve listed them in a handy chart below.

PSU Main power 8-pin EPS 6-pin PCI-E 6-pin AUX 4-pin P4 4-pin peripheral SATA 4-pin floppy
Antec NeoPower 24-pin 0 1 0 1 6, 2 2, 2 0*
Enermax EG485P-SFMA24P 24-pin 0 1 0 1 7 4 2
OCZ PowerStream 24-pin 1 0 1 1 6, 2 2 2
SH SH-ATX465P4 20-pin 0 0 1 1 6 2 2
Silverstone SST-ST30NF 20-pin 0 0 1 1 6 2 2
Ultra X-Connect 20-pin 0 0 1 1 8 0** 1
Vantec Ion 2 20-pin 0 0 1 1 6 2 2
Zalman ZM400B-APS 20-pin 0 0 1 1 7 2 2

*The NeoPower comes with a splitter that yields two four-pin floppy connectors, but it will use up one of the PSU’s four-pin peripheral connectors
**Ultra X-Connect owners are entitled to a free SATA splitter that serves up two SATA power connectors at the expense of one Molex peripheral connector
Note that the OCZ PowerStream is the only power supply of the lot with an eight-pin EPS power connector for high-end server and workstation boards. The Antec and Enermax PSUs are the only ones with six-pin PCI Express power connectors. The Antec, Enermax, and OCZ power supplies are also the only ones with 24-pin main power connectors. 24-pin power connectors are common on dual-processor motherboards and many new LGA775 boards, but for more common 20-pin boards, the 24-pin power supplies ship with 24-to-20-pin adapters.

I’ll explain why the Antec and OCZ PSUs have two different connector counts listed for some of their power connectors when I discuss those power supplies individually.

Rail by rail
Although a power supply’s total wattage can be helpful, it’s more interesting to look at how current and wattage is spread across each voltage rail. I’ve summarized current and power maximums for each power supply below.

Unfortunately, manufacturers don’t necessarily provide max current and power ratings for the same AC input voltage. Antec, Enermax, OCZ, and Zalman’s numbers correspond to the full range of AC input, but Vantec only provides numbers for 115V AC, and Ultra only for 230V AC input. Silverstone provides numbers for both, and there’s no telling what our generic SH power supply’s maximums correspond to.

Maximum output current (A)

DC Output +3.3V +5V +12V
Antec NeoPower 30 38 18, 15
32
Enermax EG485P-SFMA24P 34 40 16, 15
OCZ PowerStream 28 40 33
SH SH-ATX465P4 38.8 44.5 22
Silverstone SST-ST30NF (100-120V) 21 20 17
30
Silverstone SST-ST30NF (200-240V) 25 23 18
35
Ultra X-Connect (230V) 28 30 34
Vantec Ion 2 (115V) 14 30 15
Zalman ZM400B-APS 28 40 18

With some of the PSUs listing max output current for different input voltages, it’s hard to compare numbers across the board. There are a few things worth noting, though. First, both the Antec and Enermax PSUs sport dual 12V rails. Antec’s dual 12V rails combine for a max output current of 32A—one amp lower than the sum of their individual max currents.

Also note that current across the Silverstone PSU’s +5 and +12V rails is capped between 30 and 35A, depending on the input voltage. Components won’t be able to draw maximum current from both rails at the same time. Overall, the Silverstone’s +5V max currents are also pretty low. Notice, also, the relatively low +3.3V max current on the Vantec Ion 2. That PSU’s 12V rail tops out at a low 15A, too. The Zalman PSU’s +12V rail looks a little weak, as well, in comparison with its relatively strong +3.3 and +5V rails.

Finally, check out the monster +3.3V and +5V rails on our generic SH PSU. Perhaps they should have diverted some of that current to the +12V line, which looks weak by comparison.

With a little help from Ohm’s law, it’s easy to translate maximum currents to maximum output power. However, there are some additional limits on what the PSUs can handle with combined +3.3V and +5V loads. Let’s look at those limits.

Maximum output power (W)

DC Output +3.3V +5V +12V
Antec NeoPower 100 190 216, 180
384
460
Enermax EG485P-SFMA24P 113 200 192, 180
280
480
OCZ PowerStream 92.4 200 396
500
SH SH-ATX465P4 128 222.5 264
350
465
Silverstone SST-ST30NF (100-120V) 70 100 204
140
250
Silverstone SST-ST30NF (200-240V) 83 115 216
180
300
Ultra X-Connect (230V) 93 150 408
200
480
Vantec Ion 2 (115V) 47 150 180
180
323
Zalman ZM400B-APS 93 200 216
235
380

The table above may be a little confusing, so allow to me illustrate how to read it with an example. All the numbers you see are max wattages, and some of the cells in the table extend across multiple columns because of shared maximum capacities. The Enermax unit can supply a maximum of 113W on the +3.3V rail and 200W on the +5V rail, as the table shows. However, the two rails cannot together exceed 280W of total output. The Enermax also has two +12V rails, one of which can output 192W while the other peaks at 180W. Combined, the four rails of the Enermax PSU can supply a maximum total of 480W. The bottom, shared cell for the Enermax that extends across all three columns delineates this total shared max capacity.

Clear as mud?

You should be able to read the specs for the other PSUs using the same basic methods. Shared cells mean shared max capacities, and not every PSU is arranged the same, so not every entry in the table is organized the same. Note that the Antec and OCZ PSUs are distinguished by their lack of shared capacity between the +3.3V and +12V rails; both can max out their +3.3V and +5V rails simultaneously.

Antec’s NeoPower 480W

Manufacturer Antec
Model NeoPower 480W
Price (street) $110
Availability Now

Antec gets modular Antec power supplies have won awards and accolades in both of our previous PSU round-ups, so our expectations were high for the company’s new NeoPower 480W. The NeoPower is a somewhat radical design in that its peripheral power connectors are detachable. One may connect and disconnect them from the PSU as needed. This modular design makes for much cleaner system internals than usual, provided that not all the power leads are occupied. Gone are the days when a bundle of unused power connectors must be zip-tied together and tucked away into a corner of the case.


Look ma, no extra power cables


One fan to rule them all

Although its lack of peripheral power leads will draw stares, the NeoPower isn’t much to look at otherwise. The PSU’s flat titanium-grey finish is, well, dull. The muted aesthetic may blend well with aluminum cases, though.

Rather than use multiple fans to keep the NeoPower cool, Antec relies on one internal 120mm fan. The fan is temperature controlled and there’s plenty of venting at the rear, so it should do a reasonably good job of keeping things cool without making too much noise. The NeoPower also has a couple of “fan only” four-pin Molex connectors that can control system fan speeds in step with the power supply fan.

The NeoPower’s peripheral cables all come separately, which is really quite snazzy. There’s nothing particularly special about the cables themselves, though.

What is special is what you can do with to the cables. Antec bundles the NeoPower with extra four-pin Molex and SATA connectors—two of each—that can be clipped onto available power leads—right on to the wires. The extra connectors are a great idea, especially since their judicious use can result in extra-tidy cable routing.

Coolergiant/Enermax’s EG485P-SFMA24P 485W

Manufacturer Coolergiant/Enermax
Model EG485P-SFMA24P
Price (street) $129
Availability Now

Say that three times fast The Coolergiant/Enermax EG485P-SFMA24P is easily the most awkwardly named power supply in this comparison. The unit comes from Enermax and Coolergiant, but apparently neither company could come up with a better name. For the sake of brevity, I’ll be referring to this PSU as an Enermax unit for this comparison.

In addition to having the lengthiest name, the Enermax power supply is also longer than most of the PSUs in this comparison by about an inch. The extra length makes squeezing Enermax unit into smaller mid-tower cases a bit difficult, but once you shoehorn it into place, the extra inch isn’t a big deal.

With more gold than a rapper’s smile, the Enermax PSU’s bling value is off the charts. Between the brushed gold casing and polished fan grills, the bling-bling might be a little excessive, but there’s no accounting for taste.

Gold finish aside, the Enermax unit is dialed when it comes to cooling. The PSU is equipped with two 80mm cooling fans and a neat little blower that pulls air from the case. Fan speeds are controlled by an analog knob at the back of the unit, making the Enermax easy to fine-tune for more cooling or lower noise levels.

The Enermax PSU has a massive array of cables that not only includes a six-pin PCI-E connector, but also four SATA plugs out of the box.

If those four SATA connectors aren’t enough, Enermax also throws in a SATA splitter to power an additional two devices. The PSU also comes with a 24-to-20-pin power adapter for motherboards with 20-pin plugs.

OCZ’s PowerStream 520W

Manufacturer OCZ
Model PowerStream 520W
Price (street) $130
Availability Now

Redemption? The PowerStream is OCZ’s first foray into the power supply business, and the company isn’t being shy about offering a high-end unit right off the bat. Everything about the PowerStream screams enthusiast, which is exactly what I like to see.

For better or worse, the PowerStream has the distinction of being the largest power supply in this comparison. It’s about half an inch longer than the Enermax unit and a major pain to squeeze into smaller mid-tower cases. Some of the pain comes from trying to squeeze the PSU into the case, but even more comes from the fact that it’s impossible not to scratch the PSU’s gorgeous mirror finish in the process.

The PowerStream’s mirrored pewter finish manages to be sexy without being gaudy. There’s some green internal lighting that comes on when the PSU is powered up and even a little bling with polished gold fan grills, but nothing too garish.

The gold fan grills cover a couple of 80mm fans that channel air across the top of the case. Neither fan faces the system’s processor, which could result in higher CPU temperatures. We’ll test that in a moment.


Voltage control: even sexier than the mirror finish

Borrowing an idea from Antec’s TrueControl power supply, the OCZ PowerStream lets users adjust the voltage of each of the rails. Voltages options range from 2.8 to 3.8V, 4.5 to 5.5V, and 10.8 to 13.2V for the +3.3V, +5V, and +12V rails, respectively. When you’re within 5% of a rail’s intended voltage, an indicator light will change from orange to green. If 5% isn’t close enough for you, it’s easy to drop into the system’s BIOS or use Motherboard Monitor to fine-tune voltage further. A word of caution, though: Motherboard Monitor and motherboard BIOSes don’t always display voltages correctly.

The PowerStream is loaded with cables, each of which is either neatly sheathed or twisted to avoid tangles and clutter. Notice that two of the PSU’s four-pin Molex connectors are sheathed in a special casing. According to OCZ, this is to provide cleaner power to sensitive hard drives and graphics cards.

A 24-to-20-pin power adapter and bundle of zip ties round out the PowerStream package. The zip ties are a particularly nice touch, since I’ve never set up a system without using at least a couple to tidy up power supply cables.

Before I wrap things up on the OCZ PSU, it’s worth reiterating the fact that the PowerStream carries a five-year warranty—two years more than any other PSU in this comparison. When you’re dropping $130 on a power supply, extra coverage makes it much easier to justify the expense as an investment.

SH’s SH-ATX465P4 465W

Manufacturer SH
Model SH-ATX465P4 465W
Price (street) $40
Availability Now

Generic, baby We’ve had several requests to include a generic power supply in our comparisons, so this time around, I picked up an SH brand 465W unit from a local shop. For the equivalent of $40, 465W of “P4 approved” power seems like a deal. I could have opted for an even cheaper power supply with a lower wattage rating, but we’ll be testing with some pretty high-end hardware, and I didn’t want to risk frying anything.

Unfortunately, information on the SH-ATX465P4 is scarce; the PSU doesn’t even come in a box. Here are a couple of photos for reference, though. I’d say more, but there’s really nothing special about this PSU that a couple of captioned pictures can’t illustrate.


There’s a fan speed switch with low, medium, and high settings, except it doesn’t work


Standard cables


High Quality? We’ll see

Silverstone’s SST-ST30NF 300W

Manufacturer Silverstone
Model SST-ST30NF
Price (street) $155
Availability Now

0dB How is it that the most expensive power supply in this comparison also offers the lowest wattage? Silence. No, I’m not talking about special low-noise fans. I’m talking about absolute silence that can only be delivered by passive cooling. Silverstone’s SST-ST30NF is the first passively cooled PSU to grace the Benchmarking Sweatshop, and given the recent trend towards silent computing, I’m betting that it won’t be the last.

With lots of internal venting, cooling ridges, and fins, the Silverstone unit is definitely dialed for passive cooling. To be honest, though, I was expecting taller fins with more surface area. The PSU will certainly heat up under load, but with so many cooling fins radiating heat inside the case, more aggressive system or processor cooling may be needed to compensate.

How hot does the Silverstone unit get? Hot enough to cause extreme discomfort if you touch it for only a second. When the PSU is running at full tilt, you could probably cook an egg on it and maybe a side of sausage or bacon. Thankfully, Silverstone equips the PSU with a handy temperature warning light that warns you when it’s too hot to handle.

Silverstone doesn’t offer anything special in the cabling department, which is disappointing for a $155 power supply. Motherboard power connectors are sheathed, but that’s about it. The leads are also a little on the short side, which could create problems in mammoth full tower cases.

I should mention that Silverstone claims the power supply’s 300W rating is very conservative. The PSU apparently packs a deep reserve that offers over 400W of peak power. We’ll be putting that assertion to the test shortly.

Ultra’s X-Connect 500W

Manufacturer Ultra
Model X-Connect 500W
Price (street) $99
Availability Now

Modder’s delight Although I can only ever remember seeing Ultra-branded products in the Tiger Direct catalog, the company is branching out, and its new X-Connect line of modular PSUs is available from a number of different retailers. A friend of mine raves about his X-Connect 500W, and after seeing the PSU, I just had to include it in this comparison.

What can I say? I’m a sucker for gorgeous modular PSUs. Like the Antec NeoPower, the Ultra X-Connect’s power cables can be detached from the PSU as needed. Ultra takes things one step further than Antec by making the motherboard power connectors removable, as well.

If you’re into bright, shiny things, the X-Connect will definitely catch your eye. The PSU has a mirrored blue finish that’s looks gorgeous against the PSU’s silver fan grills and accents.

Speaking of fans, the X-Connect’s cooling setup is pretty standard: one 80mm fan above the system processor and another at the rear. I’m a little worried that fancy grill on the internal fan will restrict airflow, though.

The X-Connect’s big claim to fame is its modular cables, which are all neatly sheathed in a UV-reactive casing that glows under black light. Groovy. If you have a case window and UV cold cathode, this PSU will definitely add some visual flair to your system. Ultra even wraps its own casing around the power plugs to give the cables a more polished look. Unfortunately, that casing makes the plugs a little thicker than normal. The slightly thicker plug casing shouldn’t be a problem in most systems, but I had to hack it away to get the power connectors to fit on the admittedly tight Abit KV8-MAX3 motherboard.

Gorgeous looks and swanky cables aside, the Ultra X-Connect isn’t all roses. The PSU’s one-year warranty is disappointing, especially given the $99 list price and the fact that virtually very other PSU in this comparison carries at least a three-year warranty.

Update 11/11/2004 – Ultra now offers a three year warranty extension for its X-Connect power supplies.

Vantec’s Ion 2 350W

Manufacturer Vantec
Model Ion 2 350W
Price (street) $45
Availability Now

How much? The Vantec Ion 2 power supply is only $5 more expensive than the generic SH unit I picked up for testing. The Ion 2 is also the lightest PSU in the comparison, and it’s only packing 350W of power. That doesn’t give me a lot of faith in the Ion 2’s capabilities going into this comparison, but I’ve always liked underdogs. At the very least, the Ion 2 has the potential to be a surprising low-cost contender.

At least the Ion 2 doesn’t look like a cheap, generic PSU. It actually looks a lot like the Antec NeoPower, minus the dull grey finish and modular cables.

The Ion 2 uses a massive 120mm fan and plenty of venting at the rear to keep things cool. The fan is temperature-controlled, but unlike the Antec unit, the Ion 2 doesn’t offer any “fan only” power connectors for temp-controlled system fans.

Vantec scores big points for the Ion 2’s cables, though. Every cable is sheathed, from the motherboard connectors to all the floppy plugs. On a $45 PSU, that’s incredible. Like the Silverstone PSU, power leads are a little too short for use in monster full tower cases, but there’s plenty of length for mid-towers.

As if sheathing every cable weren’t enough, the Ion 2’s four-pin Molex connectors also come with “EZ-Grip” plugs. As the name implies, the plugs are easier to handle than standard four-pin connectors. Unplugging devices is effortless, which is admittedly a minor issue, but it’s one more little thing that Vantec has done right with the Ion 2.

While EZ-Grip is handy, it’s not perfect. The thicker power plugs don’t sit flush on graphics cards like the Radeon 9800 XT, but trimming the offending clip takes all of a couple of seconds with a pair of scissors.

Zalman’s ZM400B-APS 400W

Manufacturer Zalman
Model ZM400B-APS 400W
Price (street) $87
Availability Now

Active PFC under the hood Although Zalman was displaying a passively cooled external 400W PSU at Computex not long ago, that PSU isn’t ready for prime time yet. Zalman does have a new internal model, though. The ZM400B-APS packs active power factor correction and 400 watts of power, so it could be a dark horse in this comparison.

And I meant dark horse literally. The Zalman PSU comes dressed in black, which could become the new beige for PC cases and components.

The flat black finish is decidedly conservative, and so is the power supply’s cooling. Zalman relies on a single 80mm exhaust fan to keep the PSU cool, although there’s some extra venting on the inside to help draw warm air out of the system. Still, the single 80mm fan has me wary of the PSU’s ability to keep a system cool while maintaining low noise levels.

The Zalman’s cabling is a simple affair. After seeing every cable sheathed on the Vantec Ion 2, which costs half as much as the Zalman unit, it’s disappointing that none of the Zalman’s power cables are sheathed in any way.

There’s a little redemption lurking for Zalman inside the box, though. The PSU comes with a fan splitter that adapts a four-pin Molex connector to power three-pin fans. Zalman also includes four Velcro cable ties, which are a little classier than zip ties.

Unfortunately, redemption is short-lived. Zalman’s one-year warranty for the PSU is really not enough, especially considering the $87 price tag and the fact that just about everyone else in this comparison offers three years of warranty coverage or more.

Our testing methods
All tests were run three times, and their results were averaged, using the following test systems.

System Athlon 64 Dual Opteron
Processor AMD Athlon 64 3200+ 2.0GHz AMD Opteron 246 2.0GHz (2 processors)
Front-side bus HT 16-bit/800MHz downstream
HT 16-bit/800MHz upstream
Motherboard Abit KV8-MAX3 Tyan Tiger K8W
North bridge VIA K8T800 AMD-8151 AGP tunnel
South bridge VIA VT8237 AMD-8111 I/O hub
Chipset drivers Hyperion 4.53 AMD chipset driver pack 2.10
Memory size 512MB (1 DIMM) 2GB (4 DIMMs)
Memory type Corsair XMS3500 PC3000 DDR SDRAM Corsair CM72SD512RLP-3200/S Registered PC3200 DDR SDRAM
Graphics ATI Radeon 9800 XT NVIDIA GeForce 6800 Ultra
Graphics driver Catalyst 4.9 ForceWare 61.76
Optical drive

Liteon 16X DVD-ROM

SCSI Controllers

NA

Adaptec 29320-R

SCSI Driver

NA

Adaptec 3.00S4

Storage

Hitachi Deskstar 7K250 SATA 250GB

Maxtor 740X-6L ATA/133 40GB
Maxtor Atlas 10K IV SCSI 147GB
Seagate Cheetah 10K.6 SCSI 147GB
Western Digital Raptor WD360GD SATA 37GB
Western Digital Raptor WD740GD SATA 74GB

Operating System Windows XP Professional SP2 with DirectX 9.0c

I put together a couple of test systems for this comparison. The first is a moderately taxing Athlon 64 3200+ with a Radeon 9800 XT and few other accessories. The second is a monster of a dual Opteron workstation with a GeForce 6800 Ultra, 2GB of RAM, and a stack of 10K-RPM hard drives. The DC voltage and AC ripple of each PSU was tested at idle and under load for both tests systems. I also used the Athlon 64 system for our temperature and noise level tests inside an Antec Lanboy enclosure with a Vantec Stealth 80mm exhaust fan.

Systems were tested at idle and under a load consisting of Folding@Home, HD Tach’s read transfer test, and a looping 3DMark03 Mother Nature demo at 1600×1200 with antialiasing and anisotropic filtering cranked all the way up. I used the following versions of each application:

The test systems’ Windows desktop was set at 1280×1024 in 32-bit color at an 85Hz screen refresh rate. Vertical refresh sync (vsync) was disabled for all tests.

All the tests and methods we employed are publicly available and reproducible. If you have questions about our methods, hit our forums to talk with us about them.

Weight
There’s a general consensus that the heavier a power supply is, the better it is. I’m not sure if there’s a proper rationale for that consensus. Perhaps the assumption is that heaver PSUs have more goodness inside. More likely, it’s the fact that we generally perceive heavier objects to be more sturdy and reliable.

To kick things off, I dragged the PSUs into the bathroom to weigh them on the scale. This is a pretty generic bathroom scale, and although it has a fancy digital display complete with one digit after the decimal point, I still had to weigh the PSUs while standing on the scale to get a reading at all. After subtracting my weight, I was able to approximate the weight of each PSU.

The Ultra and Silverstone PSUs are the heaviest of the lot, but I’m not sure it’s because of more goodness inside for either. The Ultra’s fancy cable sheathing appears to add some weight, and the Silverstone PSU’s reliance on passive cooling could account for some of its heft. At the other end of the spectrum, the Vantec Ion 2 wins the featherweight crown. It’s lighter than our generic SH PSU and a little more than half the weight of our heaviest PSUs. Of course, we won’t know if this weight theory pans out until we look at some other performance metrics.

Temperatures
I used Motherboard Monitor to measure the CPU and system temperatures of our Athlon 64 test rig in an Antec Lanboy case. To best simulate a real-world system, I added a Vantec Stealth 80mm exhaust fan and an aftermarket Thermaltake Silent Boost K8 processor cooler. Since many enthusiast-oriented motherboards are sporting temperature-controlled processor and system fans these days, I let the KV8-MAX3 manipulate system and processor fan speeds based on system temperatures. Fan speed temperature thresholds were consistent for each PSU. Ambient room temperature was maintained between 26.5 and 27 Celsius.

Measurements were taken after 10 minutes idling at the Windows desktop and after a further 10 minutes under our Folding@Home/HD Tach/3DMark03 load. The system was allowed to cool for 20 minutes with the case panels off between each PSU test.

At idle, there isn’t much variation in temperature between the power supplies. The Enermax PSU’s high fan speed setting manages a slightly lower processor temperature, but only by a degree. The Silverstone and SH PSUs, on the other hand, produce higher system temperatures by a degree.

Under load, the most we can manage is a two-degree spread in CPU temperatures and a three-degree spread for system temps. As expected, the Silverstone PSU leads the way with the highest temperatures on both fronts. Zalman and Ultra’s cooling solutions don’t look so great, either.

At the other end of the spectrum, the Antec and Vantec PSUs produce the lowest CPU and system temperatures of the lot. Considering that both use a single 120mm internal fan and ample venting at the rear, it’s not surprising that they perform similarly in this test.

Noise levels
Noise levels were measured 1″ from the front and rear of the test system using an Extech 407727 digital sound level meter. The meter was placed out of the direct path of airflow.

The Antec, Enermax, and Silverstone PSUs round out our quietest top three at idle. Given its passive cooling, it’s a little surprising that the Silverstone PSU doesn’t result in lower noise levels. I suspect that the system’s temperature-controlled fans have to work harder and louder to compensate for the extra heat radiating from the PSU.

Our generic SH power supply shows its first sign of real weakness in our noise level tests, at least at idle. What happens under load?

The SH power supply makes our system even louder. Silverstone’s passive PSU loses some luster here, too, as system and processor fans spin faster to expel heat radiating into the case from the power supply’s cooling fins.

The Antec, Enermax, and OCZ units are among the quietest PSUs under load. Interestingly, the Enermax PSU is quietest with its high fan speed setting, perhaps because its blower can more quietly vent warm air from the case than our temperature-controlled system fan. The Ultra and Zalman PSUs also contend for the lowest noise levels under load, but they’re either too loud from the front or the rear to consistently stay in the top three.

System power consumption
I used my trusty watt meter to measure power consumption at idle and under the same load as our temperature and noise level tests. The watt meter measures the power consumption of the entire system, sans monitor, at the outlet.

On the efficiency front, Silverstone, Enermax, and Vantec round out the top three at idle and under load while the Antec NeoPower sucks power like it’s going out of style. Well, maybe it’s not that bad, but at idle the Antec PSU does pull 10W more than the nearest competitor and a little under 30W less than the frugal Silverstone unit.

With our more demanding dual Opteron system, the Silverstone PSU continues to pull fewer watts than the competition. Enermax’s PSU also continues to score well in our power consumption tests.

DC voltages – Athlon 64
Voltage tests were conducted over a 100 second time interval with a Pico ADC-212/3 digital oscilloscope. There was very little voltage deviation over the time interval, so the results are averages.

There’s quite a bit of range in idle voltage across the power supplies, but everyone’s within a few percent of their respective rails. There are a couple of interesting things to note, though. First, the SH power supply has the strongest 3.3V rail. That jibes with the PSU’s spec sheet, which boasts the strongest maximum 3.3V current of the bunch.

It’s also interesting to see the OCZ power supply’s rails off the mark at all. I took the time to dial them in using the motherboard’s BIOS, but apparently the BIOS readings were wrong. I’m more inclined to believe our swanky digital oscilloscope than a motherboard BIOS that gets updated on an almost monthly basis, but in any case, the OCZ’s voltages are well within acceptable tolerances.

Perhaps the biggest surprise is the Antec PSU’s low 12V rail, especially since it has two independent 12V lines. We’ll be keeping an eye on that as the tests unfold.

Under load, the overall picture doesn’t change much at all. The power supplies stack up much like they did at idle.

DC voltages – Dual Opteron

Our dual Opteron system is quite a bit more demanding than our Athlon 64 test bed, but you wouldn’t know it looking at idle voltages. Not even the relatively low-wattage Silverstone PSU bats an eyelash. The 350W Vantec’s 3.3V rail’s a little lower than I’d like to see, but it’s within Vantec’s 5% tolerance.

With everyone else within 3% of target voltages on a loaded dual Opteron system, I’m pretty impressed. Still, Antec’s in-spec but comparatively low 12V rail irks me.

Again, not much changes under load. I was half expecting one of the lesser PSUs to burst into flames once I had the dual Opteron system humming at full tilt, but it didn’t happen. Even our generic SH power supply stays within tolerances, though its 12V line drops a little.

AC ripple voltage – Athlon 64 – 3.3V rail
AC ripple, the AC component of what should otherwise be DC power, was measured over the same 100 second interval as our DC voltage tests. Since there’s all sorts of variance, I’m busting out some line graphs so show ripple at idle and under load across the entire time interval. The less AC ripple on each DC rail, the better.

Unfortunately, with eight PSUs, it’s impossible to graph them all together on a readable graph. I’ve graphed each PSU separately, combining idle and load ripple voltage on the same graph to make things a little denser.

There isn’t much ripple on any of the 3.3V rails, at least not on our Athlon 64 test system. The Zalman unit appears to have less ripple than any of the others, though.

AC ripple voltage – Athlon 64 – 5V rail

Not much changes moving to the 5V rail. We’re still not seeing a whole lot on the ripple front, and that’s a good thing.

AC ripple voltage – Athlon 64 – 12V rail

And this, ladies and gentlemen, is why I don’t recommend generic power supplies. Compared to the rest of the field, whose 12V DC rails have hardly any AC ripple, the SH power supply is all over the map under load. AC voltages are still within plus or minus 0.2 volts, but they’re much higher than the competition.

AC ripple voltage – Dual Opteron – 3.3V rail
Moving to our dual Opteron system…

Not even our dual Opteron system can coax excessive AC ripple from our power supplies on the 3.3V rail. Silverstone and Zalman look particularly tight here.

AC ripple voltage – Dual Opteron – 5V rail

Not much changes moving to the 5V rail, where everyone’s pretty close. Silverstone and Zalman continue to exhibit a slight edge, though.

AC ripple voltage – Dual Opteron – 12V rail

This is why generic PSUs don’t belong in high-end systems. The SH power supply’s 12V AC ripple is a mess under load, much worse than anyone else’s. I’m a little concerned about the Vantec and Zalman PSUs, too. They’re not exhibiting nearly as much ripple as the SH PSU, but there’s a definite jump in AC ripple from idle to load.

I have to admit that I’m a little shocked that our passively cooled, 300W Silverstone PSU is showing the least 12V ripple under load. It’s not by much, but Silverstone has been consistently ahead of the rest of the pack, whose ripple profiles share roughly the same amplitude.


Coolergiant/Enermax EG485P-SFMA24P
OCZ PowerStream 520W
October 2004


Antec NeoPower 480W
Silverstone SST-ST30NF
Vantec Ion 2 350W
October 2004

Conclusions
It’s not easy to come to come up with sweeping conclusions about the power supplies we’ve seen today. I’m not convinced that adage that heavier power supplies are better holds water anymore, though. Here’s a quick summary of my thoughts on each unit: Antec NeoPower — The NeoPower’s modular design is a huge improvement over the mess of unused cables that plagues most systems, and I suspect we’ll see all high-end power supplies eventually jump on the bandwagon. Antec also scores points for the NeoPower’s remarkably low noise levels, consistent voltages, and low AC ripple. Unfortunately, our NeoPower suffers from a low 12V rail and high power consumption. Neither are deal breakers, especially since our 12V readings are within Antec’s +/- 3% tolerances, but they hold the NeoPower just shy of Editor’s Choice gold.

Coolergiant/Enermax EG485P-SFMA24P — Despite its garish looks and awkward name, I’m quite impressed with the EG485P-SFMA24P. This power supply’s voltages are consistent, ripple is low, and efficiency is quite good. The Enermax’s noise levels and cooling potential are easily adjusted, too. Quality doesn’t come cheap, or in this case pretty, but Enermax unit earns a shared Editor’s Choice Gold award nonetheless.

OCZ PowerStream — I loved adjustable voltage rails when Antec put them on its TrueControl power supply, so it’s no surprise that I love them on the PowerStream. The PowerStream combines that with a classy mirror finish; an array of cables that are either shield, sheathed, or neatly wound; and great performance in our voltage tests. Throw in a five-year warranty and eight-pin EPS support for server and workstations boards, and we have our other Editor’s Choice Gold recipient.

SH SH-ATX465P4 — Although the SH power supply we picked up for testing probably isn’t the worst generic brand out there, its shortcomings are immediately apparent when one looks at AC ripple on the 12V rail. Even on our relatively conservative Athlon 64 system, the generic unit had trouble keeping up with the competition—competition that includes several power supplies with lower wattage ratings. Your system deserves better.

Silverstone SST-ST30NF — If there were an award for cojones, Silverstone would get it hands down. Don’t let this power supply’s 300W rating deceive you. It had no problem handling our dual Opteron system under a punishing load, although I’m not sure I’d want to rely on the power supply’s deep reserve on a day-to-day basis. Overall, the Silverstone PSU boasts some of the most impressive AC ripple and power consumption scores of the lot. Not bad for passive cooling. Unfortunately, passive cooling doesn’t come cheap. $155 is a lot to drop on something that says 300W, but if you’re on a quest for the quietest system on the block, the Silverstone’s SST-ST30NF is the place to start. Editor’s Choice Silver for Silverstone.

Ultra X-Connect — If aesthetics are your main concern, Ultra’s X-Connect is the swankiest power supply of the group. With a mirror blue finish and gorgeous UV-reactive modular cables, the X-Connect is a case window’s dream. Performance is pretty decent, too—nothing spectacular, but no major problems either. Unfortunately, the X-Connect’s chances at stardom are ultimately stymied by the power supply’s relatively high noise levels and stingy one-year warranty.

Vantec Ion 2 — Although the Ion 2 doesn’t have the horsepower to handle systems like our dual Opteron monster, this lightweight power supply is a heck of a deal at $45. With cable sheathing throughout, low noise levels and power consumption, and consistent performance on our Athlon 64 system, it’s easy to recommend the Ion 2 for less demanding systems with limited budgets. Incredible value gets the Ion 2 an Editor’s Choice Silver.

Zalman ZM400B-APS — Although the Zalman’s performance is pretty decent across our exhaustive battery of tests, AC ripple on the 12V rail was a little flaky with our dual Opteron system under load. That has me worried about this 400W power supply’s ability to handle monster systems. More importantly, I’m unimpressed with the power supply’s one-year warranty. Zalman can surely do better for an $87 power supply, especially when Vantec supports its $45 Ion 2 for three years.

In the end, there’s a clear difference between generic power supplies and name brand units from major manufacturers. There’s also quite a bit of variation from one name brand power supply to another, so long as you know where to look and what’s important.

Comments closed
    • Stiletto One
    • 15 years ago

    About the Neopower: I wonder if the modular design’s extra connection (mechanical instead of solder) was introducing extra impedance and thus causing voltage drops across the board. After all, all three (four?) rails were a hair low, but had good ripple.

    • Wintermane
    • 15 years ago

    A simple rule of thumb on all support equipment be it psu or ups or anything else for anything else…. the more you spend on the thing its gona connect to the more you better spend on it relative of course to the general spendyness of what your dealing with.

    Thus if your puter costs norm plus 200 then your psu prolly should be norm plus 25% in quality and likely price.

    The trick of course being finding out what the heck is normal;/

    • espetado
    • 15 years ago

    Uhm, I have a little ‘Elix’ Powermeter and I thought, why not measure the voltages on my newly bought OCZ Powerstream. First problem; how do you measure? Just stick it in a molex? I know the basics of electronics, so 5v =red and black, 12v = yellow and black, 7v= read and yellow, but how do you measure the 3v rail? Is it even right to stick the ends of the meter in a molex or do I have to measure somewhere else?
    So far I got 5.24 v out of the red and black and 12.36 out of the yellow and black, are these readings even close to reality, because thy seem so far compared to the article.

    • spworley
    • 15 years ago

    One power supply line I’d love to see compared is PC Power & Cooling’s units. I actually have one in my current system, but it’s likely overkill. But I admit I don’t know how MUCH overkill it is and whether it was necessary. [or maybe PP&C is actually nothing special but seems to imply it is, like Monster Cable or Prada].
    Kudos to Damage for a great article! We just want more! 🙂

    • Buub
    • 15 years ago

    Here is MY experience. There are many crappy PSUs out there. Some better-built units use decent parts, but still try to be affordable. Antec and Enermax are primary examples of this second group. They are well made but still inexpensive. Then there are the units that are made well first, then made to cost second. An example of the third group would be PC Power and Cooling.

    My experience is that Antec and Enermax PSUs will last forever if they are kept under reasonable load. If you run them for an extended period at closer to max load, they will fail. I have had it happen to multiple units from both companies. The difference between these and the cheapo units are that they are stable and non-spiky until the point where they really start to go. Moreover, when they go, you have plenty of warning because one or more voltage rails slowly drifts below spec over time. When it starts getting alarmingly low, time to replace the PSU. A cheapo might get spiky and give you inconsistent power, or it might just go in a flash.

    The third group, typified by PC Power and Cooling, costs more than your average geek is willing to spend, but the component standards are even higher. These can generally “handle the heat” better than the second group.

    Here is an example. I have a dual-processor Athlon server. It has an AMI MegaRAID controller with a bunch of SCSI drives plugged into it, a SCSI DAT library (6-tapes), and three IDE drives RAIDed. It has lots of fans to keep air flowing. This machine has already burned out two Antec TruPower 550 EPS (server) power supplies. I got tired of replacing them, so spent the extra bucks and went with PC Power & Cooling’s ~500w EPS server PSU.

    In the Antec case, the failure could be seen coming miles away. The 5v rail would continue to fall slowly over time. Eventually, it would get to the point where the machine became unstable, but lots of warning was given beforehand.

    The PC P&C power supply has been absolutely rock-solid stable since it was installed. Every voltage rail is as precisely at the same voltage as the day it was new. There has been no measurable degradation, and the machine has been up for months at a time between reboots.

    Such is my saga…

      • Buub
      • 15 years ago

      Note related recommendations…

      – According to my experience, over-specing a PSU actually has merit. If a 400W PSU won’t last for very long at 390W, and that’s what your system consumes, you might consider going 550W so the PSU has some “room to breathe”. It will probably last longer as a result.

      – I would not hesitate recommending Antec, Enermax, and other companies who build PSUs to similar standards for home desktop systems. Just don’t under-spec them (see above).

      – I will ALWAYS hesitate recommending a generic or off-brand PSU to anyone, except my worst enemy. Too many ways for it to f*ck up your system, and the second-tier PSUs are not that expensive.

      – From now on, I will always use something like PC P&C in my big server(s) and my heavily over-clocked machine. For the lighter boxes, the second-tier stuff will work OK.

    • fyo
    • 15 years ago

    I wish there would have been some tests with a system geared towards low noise. (Clearly, I’m not expecting to fit a dual Opteron system with a pile of 10k disks in a small, low-noise box, but still)

    The noise measurements seem to indicate that the main source of noise in this case is not the PSU fan, but rather the CPU or case fan.

    Even using the same setup, it would probably be possible to tweak the fan speeds (in other words, allow a higher case temp).

    -fyo

    • Delphis
    • 15 years ago

    I’d have loved to have seen the SeaSonic SuperSilencer 400AGX represented, also something from ThermalTake too in their Active PFC range too. I’m going to want to buy a new PSU for an Athlon64 server system I’m planning on building. So far the SeaSonic 400AGX is winning in my view ($90 from DTT computers too) since it’s very quiet, very efficient, doesn’t have a lot of jazzy crap that you’re paying for.

    • SpotTheCat
    • 15 years ago

    Coolermaster is about to release a new power supply. Maybe you can suppliment this review with other manufacturers too.

    • SpotTheCat
    • 15 years ago

    damn, I was wondering if I should even bother changing out my antec truepower with the PC power and cooling 510 watt I have sitting idle 🙂

      • hmmm
      • 15 years ago

      If your computer works, then I don’t see why’d you change it out.

    • Dposcorp
    • 15 years ago

    Awsome review.
    Those of us with a dual Opteron or any dual system thank you for using it to test.
    95% of the time I get Antec, but I am looking at an Enermax now, so I am glad it did well.

    Thanks again for the hard work.

    • HammerSandwich
    • 15 years ago

    Active PFC is important for any system that runs on a UPS.

    As #46 hinted, “q[http://www.silentpcreview.com<]§ and Lee Garbutt at §[<http://www.systemcooling.com<]§.

      • muyuubyou
      • 15 years ago

      This is probably the first big modern PSU round-up here. I expect more and better to come.

      I reckon UPS+PSU pairs do twice the AC/DC conversion, right?. Would be nice to see those DC PSUs with central UPS feeding DC to them. In my office we’re probably wasting a lot.

        • just brew it!
        • 15 years ago

        If they are run-of-the-mill “standby” UPSes, the inverter doesn’t kick in until the power actually goes out. So in reality the inefficiency of the double conversion only comes into play when you are running off of the battery power during a power outage.

      • Dissonance
      • 15 years ago

      That statement refers to Silverstone’s claim that the PSU’s 300W rating is very conservative. I’m not in the 200-240V AC input range, so that 300W rating corresponds to only 250W max across the 3.3, 5, and 12V lines.

        • HammerSandwich
        • 15 years ago

        Do you know the actual DC load from the dual Opteron rig? Do you know the PSU’s efficiency? If not, I don’t see how 323W AC consumption tells us anything about the PSU’s true output capability.

          • Dposcorp
          • 15 years ago

          All I am really looking for is a sign that one of the power supplies I would buy or have bought failed in the dual Opteron system.

          According to Diss, the dually system was loaded with:

          NVIDIA GeForce 6800 Ultra
          ONE (1) Optical drive
          2GB of ram (4 DIMMs)
          Five hard drives (1 7,200 RPM and 4 10K RPM)

          The Antec and Enermax *[< DID NOT FAIL.<]* That, along with their good warranties, are why I pick them.

          • just brew it!
          • 15 years ago

          Even if you know the power draw of the mobo and the efficiency of the PSU, that /[

      • Edward Ng
      • 15 years ago

      Adam, you beat me to the punch!

      The UPS situation is the first one that comes to mind, but I am personally a victim to a situation where PFC is more than just useful for saving money on UPS purchase costs.

      When a PSU has PFC, and more in particular, Active PFC, the VA drawn/consumed is nearly identical to watts power consumed; as such, it cuts down on the amount of current being drawn by the system in general. As a reviewer, I often have multiple rigs and/or test platforms running at once, and the current drawn adds up. For example, let’s say I use my Power Angel and determine the draw from three active machines, at once, to be 2.4amps, 3.1amps and 3.3amps, respectively. This adds up to ~8.8amps (not counting hundredths). Now, I also happen to live in a a top floor apartment in NY. It’s hot in the summer, so I run my a/c; that thing draws hefty current whenever its compressor is running, and I use it to acclimate my ambient temps for proper heatsink testing. Now, consider that I share my circuit breaker with the kitchen, which includes the refrigerator and other items that use lots of current. My circuit breaker only allows me 20amps current maximum, this being an old apartment building with faily ancient electrical infrastructure. I had the breaker trip on me a few times during testing, due to excessive current draw when the A/C and the fridge compressors both clicked on in conjunction with multiple machines running at once.

      Since switching to all Active PFC power supplies, I have only had this occur once, and that was with the A/C, refrigerator, 3 machines AND the microwave on.

      This is also the reason why in Europe, PFC is REQUIRED. It dramatically eases the load on the general electrical power infrastructure.

      However, PFC isn’t directly tied in with EFFICIENCY of PSUs; the efficiency tested for in the review is a measure of the AC-DC conversion efficiency, and that is not directly affected by the power factor correction features of the PSU.

      I believe this PSU comparison was excellent, and am in no way trying to knock the comparison down a notch. I love the in-depth analysis of the rails. I just wanted to clarify this one item, which, I think, would’ve been quite easily tested by simply utilizing one more tool, a Power Angel or Kill-a-Watt meter.

      -Ed

    • slymaster
    • 15 years ago

    Excellent review. There may have been a few points of confusion regarding power factor and efficiency – but hey, that is what the discussion is for – to bring these issues to light, and thus improve the testing in the future.

    This is the best review of power supplies that I have ever read. I have only read one other of similar quality, but it is a few years old. It is good to see a thorough update, as modern high-end systems become power hogs.

    Question: If the dual opteron (with several HD’s) only draws 359 W (with the least efficient power supply) at the wall under full load, than what justification is there for a 550 W supply ? Large servers often have multiple power supplies to handle different components. I think the PSU manufacturers have brainwashed the masses into buying bigger (and more expensive) power supplies than are needed.

    If the least efficient power supply draws 359 W at 75% efficiency, than that translates into 269 W. I thought power supplies are rated by their output capcity, not how much they are able to draw from the socket. If I am wrong, please let me know.

    • notfred
    • 15 years ago

    I have to say I agree with #13 about confusion in the article over Power Factor and Efficiency:
    Power Factor doesn’t matter if you are a residential power customer, you get billed by how much power you use. Power Factor matters if you are a commercial customer, because you get billed by power used and power factor of your load, which is why you sometimes see banks of capacitors outside industrial plants to correct the power factor. They are cheaper than paying the penalty for the really bad PF.

    I am concerned on the wattage readings – some meters read true power consumed, others are thrown by the power factor. What was this meter?

    Efficiency is just power out / power in. You can get power in off a true power meter and then you need to measure power out to the board.

    I also don’t think the graphs are too helpful in just displaying the voltages. I would have liked to see how much the voltage differs as the load increases so you can easily see the difference between a supply that is just calibrated low but keeps the same voltage regardless of power and one that is calibrated better but sags when you start drawing power from it.

    I also think the supplys should have been tested on a proper test rig and pushed to the limits, if they say they will deliver x amps on a rail or y watts on a combination, then test them to that limit and beyond.

    A comment on the 12V ripple on the generic supply. Originally 12V was only used for drive motors and supplying the serial ports. If your computer only uses this then that supply is fine as the motors don’t care about the ripple.. P4s introduced the concept of using the 12V supply for additional power for the CPU. In this case you are dependent on the motherboard for the quality of the CPU power conversion circuitry which may deal fine with that ripple, then again it may not and may cause problems with the CPU. Again it would be interesting to graph magnitude of ripple against load.

    The weight of a PSU being a sign of quality comes from the idea of a quality PSU handling a lot of power and so needing big heatsinks. With differing fans this becomes less relevent.

    Sorry for the length, but I’m a hobby electronics geek with a physics degree 😉

      • Dissonance
      • 15 years ago

      The watt meter we used measures true RMS watts.

        • just brew it!
        • 15 years ago

        IIRC “true RMS” is not the same thing as taking power factor into account.

        True RMS simply means that the true RMS voltage and true RMS current are both measured, then multiplied together to calculate the wattage. In order to account for power factor, the relative phase of the voltage and current (which is non-zero if the PF != 1.0) must also be taken into account.

        AFAIK, just because a meter claims to measure “true RMS” does /[

          • Dissonance
          • 15 years ago

          The meter professes to measure what the power company is going to charge you for. It can also measure power factor by taking RMS and dividing by apparent power, but there doesn’t seem to be a way to lock it to displaying apparent power.

            • just brew it!
            • 15 years ago

            Ahh, OK… that helps clarify things.

            If the meter manufacturer’s claim is correct, then the meter should be measuring actual power consumed, since that is what your electric meter measures.

      • muyuubyou
      • 15 years ago

      notfred said: /[

      • yokem55
      • 15 years ago

      I’m hoping to build a dually Opteron soon (as soon as pci-e becomes available for it) and while my Antec 350 has served me pretty well, I’m nervous about it being up to the task of handling such a substantially larger system. This review has both squashed those fears (the great performance of the lower wattage models) and given me a fair amount of interest in what I might want to get anyway…..

        • just brew it!
        • 15 years ago

        (and #46 too), do keep in mind that a PSU with more reserve capacity will not be pushed as hard, will probably run cooler, and should last longer as a result.

    • just brew it!
    • 15 years ago

    Great review! A few minor points/nits:

    – What exactly is the point of having adjustable voltages, anyway? Wouldn’t it be better if the PSU makers just improved their QC to the point where all of the rails were spot-on when they leave the factory?

    – Re your concerns about the fan grill on the Ultra X-Connect restricting airflow: How hard would it be to remove the existing grill and replace it with a standard wire grill?

    – You might want to invest in a small digital postal scale, if you’re going to make a habit of weighing components. 😉

    – I am a little bit puzzled by your concluding comments on the Silverstone. How can it be suitable for those who are “on a quest for the quietest system on the block”, when it placed 3rd in the front-of-case noise test at idle, and dead last under load? I think your own noise level tests are evidence that radiating the heat from the PSU directly into the case and relying on case ventilation to remove that heat is /[

      • Dissonance
      • 15 years ago

      -Adjustable voltages rule because if you have a rail that’s off by a little, or a lot, you can correct it to exactly 3.3, 5, or 12V. I just wouldn’t necessarily rely on the motherboard’s voltage monitoring to tell you where the sweet spot is.

      -The grill can be removed or replaced with a screwdriver.

      -The Silverstone is the only PSU you could truly build a silent system with. You’d probably need water cooling to do it, a Reserator perhaps, but every other PSU is going to make more noise than the Silverstone.

        • just brew it!
        • 15 years ago

        – Given that most people don’t have a voltmeter, wouldn’t a more useful feature be rails that simply aren’t “a little off”? Most quality PSUs already use “remote sensing” wires (look for pins on the ATX connector that have two wires running into them instead of one) to correct for the voltage drop across the ATX wiring harness, so making a PSU with voltages that are “spot on” out of the box should not be rocket science.

        – How are you going to build a completely silent system with the Silverstone? Even with watercooling you will still need case fans, because all of the heat from the PSU is being dumped into the case. I suppose you could watercool the PSU too…

          • flip-mode
          • 15 years ago

          Um, if you’re water cooling with a Reserator as he said (gpu, chipset, cpu) what do you need fans for? HDDs would need something I suppose. Any watercooling components for HDDs available?

            • just brew it!
            • 15 years ago

            /[

            • flip-mode
            • 15 years ago

            heh, maybe you could find a method of watercooling the PSU exterior. Could that be done? Would the Reserator make the grade with that much added heat?

    • espetado
    • 15 years ago

    I actually just ordered an OCZ 520 a week ago, going to pick it up in an hour or so. That was based on another (equally good) psu article, don’t remember where but the OCZ came out on top there.

    • flip-mode
    • 15 years ago

    Nice review TR, thank you. Great community response. Good looking out Jackson (#13).

    Still like to see the $56 Aspire 520watt in there. Affordable, attractive. I wish I had an oscilliscope and the knowledge to use one.

    • muyuubyou
    • 15 years ago

    I’m basing my next PSU buy on this (in fact I think I will replace mine). Had the TR a store… 😉

    • wierdo
    • 15 years ago

    That was good stuff, hope to see more of this again, very helpful. VERY!

    • Logan[TeamX]
    • 15 years ago

    Nice work!

    I guess I’ll be considering an OCZ PSU alongside my usual Thermaltake fanboyisms when I build my Winchester box this winter.

    Thanks for the heads up guys!

    • 5150
    • 15 years ago

    Good to see something other than the usual reviews around here. Thanks for listening!

    • Anomymous Gerbil
    • 15 years ago

    Deleted – what rubbish I wrote 🙂

    • FireGryphon
    • 15 years ago

    Great review, but there’s one thing I’d like to see: how long will a system run Prime95 with each PSU before choking? I’d like to see just what the consequences of these numbers are.

      • hmmm
      • 15 years ago

      I don’t think there is much reason to believe that any of the PSUs reviewed (except perhaps the generic) are noticably less capable of running a system at full load for extended periods of time.

        • eckslax
        • 15 years ago

        Yeah, unless you specifically get a lemon, you’ll probably be sitting there for quite a while waiting for the system to crash. 😉

        Oh course, all bets are off on the generic PSU. If someone decides to purchase a generic PSU, then they deserve whatever negative stuff happens to their box.

          • FireGryphon
          • 15 years ago

          Well, that’s the thing. Everyone claims that PSU’s are the cause of some problems, but can anyone prove it?

          About how long would you have to be using a “bad” PSU (either generic or name-brand lemon) before you run into problems?

            • muntjac
            • 15 years ago

            im curious about that too. the review says the generic has bad ripple but can point to no ill effects caused by it. i guess if some piece of hardware is especially susceptible and needs really clean signal it might cause problems. who knows.

            • Koly
            • 15 years ago

            I had to learn this the hard way. I was using generic PSUs as I knew nothing better. After I saw failing them one by one (as a rule of thumb in the end of summer) I switched to Enermax and Fortron. I haven’t seen a failed PSU since then.

            • Yahoolian
            • 15 years ago
    • droopy1592
    • 15 years ago

    Soon, digital tracking PS will make things more efficient and less power will be lost to heat. PS won’t even have fans.

    • JavaDog
    • 15 years ago

    We’ll, I’m also happy to see that the x-connect performed so well (even led in some areas) since I just recommended one to a friend!

    Good review Diss!

    • Jackson
    • 15 years ago

    My comment on this review could sound like a personal attack which is NOT my intent. You need technical help to do this kind of review. Your faulted definition of power factor vs efficiency proves that you have NO basic knowledge of simple beginning electronics. Thus the rest of your review is completely suspect!

    You should ask someone who is knowledgeable in electronics to review your stuff before you publish it..

    Or maybe you could do a Google search for “power factor” to learn the basics of what that term means. Then do the same for “efficiency”.

      • Dissonance
      • 15 years ago

      I actually did just that. Even searched for PFC and efficiency together. Imagine that.

      Granted, Google doesn’t necessarily tell you which information is right and which is wrong, but based on my research, it seems that active PFC can result in a higher PF than passive PFC, which has a higher PF than no PFC at all. The consensus, including test results thanks to Google, seems to be that a higher PF can improve overall efficiency.

      A couple of links, all thanks to Google:

      §[<http://www.seasonic.com/support/a01.jsp<]§ §[<http://www.overclockers.com/articles613/index02.asp<]§ §[<http://www.silentpcreview.com/article81-page2.html<]§ If you'd like to suggest further reading, I'm all ears.

        • Klyith
        • 15 years ago

        Dan’s Data did a pretty thorough debunking of PFC a while ago. Active PFC is LESS effiecint (by a small amount) than none at all, in that it actaully consumes more watts. A household meter is very good at measuring the real watts consumed, so the apparent load of a supply without PFC is not adding to your bills. Not directly, at least.
        §[<http://www.dansdata.com/gz028.htm<]§

          • muyuubyou
          • 15 years ago

          I hope Dan tested it under real conditions, as it’s with an increased number of appliances that distortion of the waveform happens. Whether that is true or not, tests should tell (unless he filtered out PSUs over that).

            • Klyith
            • 15 years ago

            It isn’t a test at all. It is a article explaining why PFC is not saving anything on your power bills.

            It isn’t like PFC doesn’t work! Supplies with PFC have a much higher PF, close to the 1 of a perfect resistor. That may be good for a number of things, but it isn’t more effecient (i.e. a PFC supply draws no less watts than a standard one).

          • Chrispy_
          • 15 years ago

          Damn, you beat me to it 🙂
          §[<http://www.dansdata.com/gz028.htm<]§ Anyway, PFC is marketing bullsh1t, Having a clean power factor of 1 saves the ELECTRICITY COMPANY money, because it reduces feedback into the power grid. Feedback into the system is not how you get charged for your electricity, it's measured on what power you draw from it, of course. Active PFC circuitry /[

          • Dissonance
          • 15 years ago

          That’s why I said that active PFC /[

        • Koly
        • 15 years ago

        Sorry Diss, but Jackson is right. You are messing up Power Factor vs. efficiency. Efficiency of the PSU, as you use it, is the ratio of the power supplied by the PSU to the system to that drawn by the PSU. The difference is wasted by the PSU as heat. It is the cost of transforming the AC 220/110V 50/60Hz to DC. However, in some of the links you provided they used the term ‘efficiency’ in a different sense, as a ratio between real and apparent power, i.e. PF.

        Power Factor is something completely different, it is almost independent on the PSU efficiency. Put simply, it is a number measuring how the electrical device is reacting to the supplied voltage. A device with PF equal to one reacts ideally, it does not distort the signal in any way. In contrast, a device with low PF is out of phase with the supplied voltage and it distorts the signal by taking or sending back voltage. That does not mean it consumes more power, it only consumes it out of phase, from the point of view of the electricity company in wrong times. So, when you have a PSU with low PF, it does not mean it is less efficient or consumes more power, it only means that you make your AC supplier unhappy. In fact, a PSU using active PFC, that means additional circuit for correcting the distortions, is likely to be less efficient due to the fact that this circuit consumes some power too. This is quite clearly the case with the Antec PSU you reviewed.

        BTW, the link provided by Klyith is excelent, you should read it.

        • Chrispy_
        • 15 years ago

        Seasonic aren’t lying, they’re simply not talking to consumers in that article.

        Businesses that draw large amounts of power from the grid are measured differently to consumers. Whilst we are measured on what we draw from the grid, A company running 5000 PC’s or using megawatt industrial machinery gets charged on the amount of power disruption they cause to the grid.

        So yes, for big businesses (which are primarily who Seasonic target) the PFC makes a *[

      • muyuubyou
      • 15 years ago

      I probably skipped the part you’re referring to (had to skip some parts – will read it all later at home).

      Are you talking about this part?:

      y[

        • Koly
        • 15 years ago

        You are missing that when Dissonance is talking about PSU efficiency he means how much power is wasted by the PSU (i.e.transformed to heat), not the PF (i.e.the ratio between real an apparent power). The difference is crucial, the latter does not effect your power bill.

      • kwarner
      • 15 years ago

      I just saw this review from a link from the Via page. Not bad. But I agree that you should get the help of a practicing analog EE (unfortunately they are rare) to get some of the technical stuff down.

      – On the presentation of voltages, the voltage regulation from light load to full load is what matters. Not the absolute voltage as long as it it within spec. A lot of times the absolute voltage is tweakable anyway.

      – I’m curious about transient response of the power supplies. That is, if the load jumped up because a high powered game started, how well does the power supply regulate its voltage? What you measure from the power supply is something called a “step response”. You will need a storage scope to measure it. You could see swings in the voltage that include overshoot that could take a power supply out of spec for a mS.

      – It would be good if you could convert the ripple graphs you made into a number for easy comparison between the power supplies. Most manufacturers use pk-pk voltage. You could again use a storage scope in accumulate mode to collect some numbers. Turn the timebase off and just record the voltages.

      Overall, a good effort that will hopefully get better.

    • Wintermane
    • 15 years ago

    Id imagine the reason the generic looks the way it does is its ment for a server where its likely needed to power 8 hdds some fans a gibit ethernet and loads of ram all of wich drain the 3.3 and 5 volt lines. Its prolly not made to run a high wattage cpu or gpu or specialy BOTH.

    Thing to look out for some of those supposedly good performners in this test had rather weak 3.3 and 5 volt rails so if your into lots of hdds and fans and lighthing and blah blah blah… foom!

      • Dissonance
      • 15 years ago

      The test results with our dual Opteron system, which had five hard drives (1 7,200 RPM and 4 10K RPM) didn’t show much in the way of 3.3 or 5V weakness.

    • madgun
    • 15 years ago

    so what sort of damage can the massive variation in ripples cause to the system…if any?

      • BooTs
      • 15 years ago

      system instability mostly. not sure about any physical damage other than data loss. ripple sux!!!@1.

    • Felix
    • 15 years ago

    Well its good to see the Ultra X-Connect didnt do to bad, im hoping to get one of those pretty psu’s soon, Xoxide has window modded ones for like $119, and im taken by the modular idea, although, i might make my own atx connector….

    • Convert
    • 15 years ago

    A great review as always. Excellent work on the graphs!

    Shame there wasn’t a thermaltake in there though. Ohwell.

      • Chrispy_
      • 15 years ago

      Last few times I saw a TT powersupply reviewed, they were overpriced, garishly-styled, underperforming failures. I’m sure some of their product reviews go down okay, but the fact that they have been heavily criticised in the past and that they did not send a unit for review here, of all places, is more than just coincidence, I think…..

      Seriously, there’s too much hype around that company. I have yet to see anything better than mediocre come from them, but in all fairness, I don’t buy from them anymore, I just feel glad that I bought something else when I see a TT product in the flesh.

      /[

        • Convert
        • 15 years ago

        Those are the ps’s that I am talking about (run of the mill looking ones).

        The reason why I wanted to see a review (here) was because they are darn cheap for starters and lots of people on the forums recommend them.

          • eckslax
          • 15 years ago

          Yep, I too am interesting in the TT Silent Purepower units. Specifically, the 120mm fan PSUs. I would like to see it matched up against the Antec unit from this review.

    • DeadlyFreeze
    • 15 years ago

    Good stuff, but I was really hoping to see a socket 775 system in there…

    • SuperSpy
    • 15 years ago

    May I be the first to say: Bravo!

    Articles like this are the reason I browse TR more than any other site.

    • muntjac
    • 15 years ago

    i have always thought sparkle (also known as fortron, spi, and a couple of other rebadges) offer the best performance for cheap. too bad one isnt compared to the more expensive units.

    • BooTs
    • 15 years ago

    Great stuff. I don’t know but I love seeing the voltage ripples graphed. Maybe its because thats one of the few things I learned at college… How to use an oscilliscope.

    • Spotpuff
    • 15 years ago

    On page 11, temperatures, there seems to be an error. CPU temps at idle are at ambient, which is impossible. Maybe the CPU and system temps got switched?

      • Dissonance
      • 15 years ago

      The temps are correct. The Silent Boost cooler does a heckuva job keeping the processor cool, and the processor’s sitting right next to the case’s 80mm exhaust. Not sure where the KV8-MAX3’s system temp sensor is, exactly, but I’m guessing it’s not as well-cooled.

Pin It on Pinterest

Share This