xentr_theme_editor

  • Please do not post any links until you have 3 posts as they will automatically be rejected to prevent SPAM. Many words are also blocked due to being used in SPAM Messages. Thanks!

9-Way 850W Power Supply Roundup

Status
Not open for further replies.
xentr_thread_starter
Xigmatek NRP 850W

Xigmatek NRP 850W


Most people who are reading this article probably know Xigmatek for their CPU and GPU coolers and to a lesser extent their enclosures. However, this is one company that isn’t a stranger to power supplies as they have been producing them for quite a few years now. Their NRP –No Rules Power- series represents one of Xigmatek’s first forays into the high-end market and are designed for enthusiasts but also carry with them more than fair prices. For example, the 850W model we are reviewing here goes for under $180 here in Canada while sporting a modular interface and an 80Plus certification.


The packaging we see with the NRP 850W is basic but wonderfully compact which should cut down on shipping costs for all of you on a budget. Within the box, everything is well protected with the modular cables off to one side and a box containing the accessories pushed off to the front while the power supply itself is wrapped in a form-fitting bubble wrap sleeve.

The accessories you get are basic with a sleeve for your modular cables, mounting screws, a warranty card and a quick install guide. Additionally there is a white fabric sleeve whose use we just couldn’t figure out. There is however one slightly worrying aspect here and that’s the power cord. Of all the power supplies in this roundup, the cord that attaches the Xigmatek 850W to your wall outlet is so thin it seems more suitable to a budget, off-brand 400W unit rather than a high-end product. While this shouldn’t have an impact upon performance, we would don’t like seeing costs cut in areas such as this.


The NRP is a compact looking 850W unit with a centrally mounted 135mm fan and a standard powder coated matte black finish. Interestingly enough, even though the accent color on the cable sleeving, on the box and on the side label is orange, the fan itself uses blue LEDs.

The modular interface is elegantly simple since there aren’t that many modular cables to begin with and all of the ports are clearly marked.


Even though the NRP isn’t fully modular, Xigmatek made some odd decisions when it came to choosing which cables should be fixed to the housing: a 4-pin CPU, two 6-pin PCI-E and the ATX connector are there. We wanted to see a 4+4 pin CPU connector along with 6+2 pin PCI-E cables considering the target market the NRP 850W is aimed at.

The modular cable selection is limited but you get what you need for literally any connection you could possibly want. The sleeving on these cables is very well done but we have to mention that the sleeving on the main ATX connector is atrocious. It is loose, poorly weaved and quite brittle.

PSU-ROUNDUP-46.jpg
PSU-ROUNDUP-47.jpg

We were greeted with a welcome surprise when we popped the top: a CWT PSH platform. This may not be the most advanced or efficient design on the market right now but it is perfectly suited to a budget 850W unit like the Xigmatek 850W. To put this into perspective, this is the same platform as the one used in the Thermalteke Toughpower XT but installed into a power supply that costs more than $80 CAD less. Xigmatek needs a pat on the back for this one.

PSU-ROUNDUP-48.jpg
PSU-ROUNDUP-49.jpg

As expected of most higher-end Channel Well platforms, this one is exceptionally well built with good soldering throughout. While the choice of orange throughout the innards is a bit odd, we can’t fault the capacitor selection which is a carbon copy of the Thermaltake unit: a Matsushita 85°C on the primary and Nippon Chemi-Con 105°C units on the secondary.
 
xentr_thread_starter
Cable Lengths and Connectors: ATX & CPU

Cable Lengths and Connectors




ATX Cables

PSU-ROUNDUP-211.jpg



CPU Cables

PSU-ROUNDUP-212.jpg
 
Last edited:
xentr_thread_starter
Cable Lengths and Connectors: PCI-E

Cable Lengths & Connectors



PCI-E

PSU-ROUNDUP-214.jpg
 
Last edited:
xentr_thread_starter
Cable Lengths and Connectors: Molex & SATA

Cable Lengths and Connectors



Molex Cables

PSU-ROUNDUP-213.jpg


SATA Cables

PSU-ROUNDUP-215.jpg
 
Last edited:
xentr_thread_starter
Testing Methodology

Testing Methodology



Test System Setup


Processor: Intel Core i7 920(ES) @ 4.0Ghz
Memory: Corsair 3x2GB Dominator DDR3 1600Mhz
Motherboard: Gigabyte EX58-UD5
Graphics Card(s): HD 5870 1GB Crossfire
Cooling: CoolIT Boreas mTEC + Scythe Fan Controller
Disk Drive: Pioneer DVD Writer
Hard Drive: Western Digital Caviar Black 640GB
Additional Fans: 2x Yate Loon 1200RPM
Monitor: Samsung 305T 30” widescreen LCD
OS: Windows Vista Ultimate x64 SP1
* Open test bench


General Notes

*Each test is run over the course of 30 minutes.

*Unless otherwise noted, the Boreas TEC runs at a constant 100%


Idle Conditions

Off: Is a load value where the system is turned off but a small amount of power is still required.

Idle: Idle values are determined by a stable Windows Vista x64 desktop.



Load Conditions

CPU Load: This test is run with 8 instances of a custom Prime95 test along with the Boreas running at max which we have found uses the most non-GPU power. The test is run for 30 minutes.

GPU Load: For this test we are aiming to show power consumption in a typical fast-paced gaming scenario. As such, we load the two HD 5870 cards with custom timedemo of Far Cry 2 at 2560 x 1600 with 4x AA enabled and set it to loop for approximately 30 minutes. The TEC unit is turned off.

Extreme Load Test: This is the big one which separates the boys from the men. Here we use a combination of the 3DMark06 Batch Size Test to load the two graphics cards along with the custom Prime 95 test and ramp up the TEC to max once again.


Voltage Regulation Testing Methodology

Multimeters Used:
Extech 430 DMM x3

*Note: All voltage readings indicated in the review are the minimum voltages seen over the period of our tests

We always take voltage readings from a loaded connector in order to more accurately see the voltage fluctuations our components are experiencing. Thus, this is how voltages are measured:

+12V:

In the CPU Load test the voltages are taken directly from the CPU connector and the used 8-pin PCI-E connector of the power supply. In the GPU Load test the voltages are taken from an 8-pin PCI-E connector which is plugged in to the topmost graphics card as well as the CPU connector. In the Extreme Load test, readings are taken from both the CPU and the PCI-E connectors and the lowest reading is recorded.


+12V Tolerances

Normal: 12.00V
Min: 11.40V
Max: 12.60V


+3.3V / +5V: From the main ATX connector.

Please note: Due to the lack of load on the +3.3V and +5V rails in a modern PC, we WILL NOT be including charts for their results unless there are noteworthy fluctuations.

+3.3V Rail Tolerances

Normal: 3.30V
Min: 3.14V
Max: 3.47V


+5V Rail Tolerances

Normal: 5.00V
Min: 4.75V
Max: 5.25V



AC Ripple Testing Methodology

Tolerances:

+12V : 120mV Max
+3.3V: 50mV Max
+5V: 50mV Max

The values were the highest peak ripple measurement across all of the +12V rails. So, if the +12V1 rail shows a ripple of 20mV and the +12V2 rail shows a ripple of 40mV, the highest value will be graphed.

Instruments Used:
USB Instruments Stingray Digital Oscilloscope
USB Instruments Differential Oscilloscope Probe

Since we do not have a load tester with a BNC connector for the standard o-scope probe, we needed a Differential probe in order to give us the proper capacitance to accurately determine ripple. In addition, the differential probe has a pair of connectors which are very much akin to a multimeter's probes which makes them ideal for use on SMPS designs. The locations of the probes for each test reflect the locations of the multimeter probes detailed in the Voltage Regulation Testing Methodology section.


Efficiency Testing Methodology


Instruments Used:
UPM Power Meter
Tripp Lite LC1800 Line Conditioner

The data points you see in our charts show the AVERAGE PEAK AC power consumption over all of the tests conducted.


Temperature Testing

Considering the amount of heat our open-air test system produces, it was found to be nigh-on impossible to properly regulate the temperature in the room even with a 10,000 BTU air conditioner. As such, we will be measuring the delta between room temperature and the exhaust temperature from the power supply.

To test temperature, we set up a pair of Type-K temperature probes. One is placed in the middle of the testing room at an elevation of 5 feet AFF to measure ambient temperatures. Meanwhile, the other temperature probe is placed 2” away from the exhaust grille of the power supply.


A few other tidbits

- AC Input Voltage: 120V constant
- Noise is subjectively tested
 
xentr_thread_starter
Efficiency Testing

Efficiency Testing



Unlike past reviews, we are going to break the efficiency section down into different sections ranging from the usual “Off” tests to the load tests.


Off

PSU-ROUNDUP-132.jpg

The nine PSUs we tested here seem to cover the full range of efficiency when the system is turned off. It is interesting to see how some power supplies consume nearly three times the wattage as others when plugged into the same system. Of the bunch, the most surprising thing here are the two similar CWT PSH platforms in the Toughpower and Xigmatek units with a gap of about 20 percent between them. Otherwise, the XFX with its Seasonic platform is the most efficient here while Antec and Silverstone bring up the rear.


Idle

PSU-ROUNDUP-133.jpg

Honestly, the “Off” chart we saw before really can’t tell the real efficiency of a power supply which is why we see the order of things being totally different here. Once again though, the XFX Black Edition as the most efficient unit with the Enermax and Seasonic not too far behind which is understandable considering their 80Plus Silver certifications. We can also see that on average the standard 80Plus and 80Plus Bronze units can consume 10 to 20 watts more than the most efficient ones.


GPU / Gaming Load

PSU-ROUNDUP-121.jpg

In higher load scenarios, the more efficient power supplies really start stretching their legs and pull out in front of the competition. Unfortunately, the high-end Antec Signature really suffers here and posts some of the worst results by requiring 37W more to run the same test as the Revolution 85+. What was surprising is that the Andyson platform the Ultra X4 is based off of seems to be coming into its own in this test and posts slightly better results than most of the 80Plus Bronze units. The Silent Pro M from Cooler Master also does quite well here.


CPU Load

PSU-ROUNDUP-122.jpg

Once again, the gap between the men and the boys widens significantly. We have to remember that as the load increases, a few percent increase or decrease in efficiency can make a world of difference for your electricity bill at the end of the month. The Silverstone Strider Plus seems to be getting more efficient as the load increases as has the Antec Signature but the Thermaltake and Ultra units have seriously taken a turn for the less efficient side of things. Considering the Thermaltake claims their unit hits 80Plus Bronze levels, it is disappointing to see it being less efficient than the 80Plus Standard certified Xigmatek.


Extreme Load

PSU-ROUNDUP-123.jpg

Nearly 100W. Let’s say it again for emphasis: 100W. That’s how much separates the most efficient power supply in this roundup from the least efficient. We were amazed by two things here: how the Enermax Revolution is able to absolutely dominate the field (we are sure it is flirting with 80Plus Gold status here) and how poorly the high-priced Thermaltake unit performs. It is also good to see how well the Cooler Master and Silverstone units do here as well.
 
xentr_thread_starter
Voltage Regulation

Voltage Regulation



Varying load scenarios have different impacts upon voltage regulation which is why we test the regulation on both the CPU and a PCI-E connectors. To do this, we use a “loaded” connector which is installed and insert the multimeter’s probe into the back.


+12V PCI-E Voltage

PSU-ROUNDUP-120.jpg

Considering the load we are putting these units under, it is great to see that so many of them did well in this test. However, there were two exceptions with the Ultra X4 and the Cooler Master unit returning what we consider slightly disappointing results but for entirely different reasons. The Ultra may be well within the ATX specified limits but in our opinion, it plays its game far too close to the maximum allowable 12.60V. We have to remember that the larger the delta between the highest and lowest voltage, the more the VRMs on your GPU or CPU have to work and with a delta of about 0.40V, things don’t look good for the Ultra unit. The Cooler Master on the other hand returns disappointing results by having its voltage tank but it did manage to stay above the ATX specified minimum.


+12V CPU Voltage

PSU-ROUNDUP-126.jpg

Since all of these units are basically based off of single rail designs and the separate “rails” are derived from current limiters, it isn’t surprising that these results closely reflect those of the last test. The Ultra unit does actually do quite a bit better here which could be due to anything from modular interface soldering joints causing the drops in the PCI-E test to the load on that specific connector. That isn’t to say that we were impressed with its results since like the Cooler Master, the voltage fluctuations were a bit too much for our liking.


A Note About +3.3V / +5V Regulation

As was stated in the methodology section, all of the +3.3V and +5V voltages were recorded but there was next to no variance so a chart here would be nothing but filler. Honestly, a modern PC system hardly uses either of these rails and to focus too much on them is simply a waste of time in our books.
 
xentr_thread_starter
+12V Ripple

+12V Ripple (PCI-E Connector)


PSU-ROUNDUP-127.jpg

While none of the power supplies we see here pushed ripple above the ATX specified maximum, this chart can be separated into three different categories: the good, the bad and the ugly. The XFX, Seasonic, Enermax and Antec units posted extremely good results with the Antec in particular showing next to no movement on its traces throughout the tests. The budget-friendly Silverstone Strider Plus also puts down some stunning numbers with a maximum ripple of barely 1/3 the amount the ATX spec dictates. The Ultra and Cooler Master units didn’t wow us with their performance here but they both kept things well under control by flirting with 50% of the ATX spec.

The disappointments of this particular test are the two CWT PSH-based power supplies with the Xigmatek and Thermaltake posting results that got quite close to the limits of acceptable. We can forgive the Xigmatek unit for this considering its asking price and the fact that it didn’t break the 80mV mark by polling similarly to the Cooler Master. Thermaltake’s unit on the other hand is an exceptional disaster here: it flirts dangerously close to 100mV without even being pressed to its limits AND it costs as much or more than the best of the best.


+12V Ripple (CPU Connector)


PSU-ROUNDUP-128.jpg

Here we see nearly the same results with a grouping of units that perform amazingly and another group of the slightly underperforming products. Once again, the Thermaltake Toughpower XT is the worst of the bunch while the Cooler Master jumps behind the Xigmatek and for some reason posts slightly poorer results here. Once again, this could be due to anything from wire resistance to a difference in soldering joints on the modular interface. Otherwise, all of the other products perform up to or exceed our expectations.
 
xentr_thread_starter
+3.3V / +5V Ripple

+3.3V Ripple+5V Ripple


PSU-ROUNDUP-129.jpg


+5V Ripple


PSU-ROUNDUP-130.jpg

Luckily for all of the contestants, there really isn’t anything to report here since a modern system just doesn’t put enough strain on either of these rails to cause them to jump out of sorts. So, all of the units do extremely well.
 
xentr_thread_starter
Temperatures & Acoustics

Temperatures


For this test, we placed an Extech Type-K thermocouple probe approximately 2” away from the back of the power supply in order to measure temperatures.

PSU-ROUNDUP-131.jpg

This charts tells an interesting story, doesn’t it? First of all we see that despite the Antec Signature not being the most efficient unit around, it also posts the best temperatures of the bunch. There is also a large number of power supplies clustered around the 19 to 25 degree mark which is historically where most of the better units usually end up after the testing regime reaches its peak.

Meanwhile, the PSH-based units are quite hot running and that shows here but the real star of this little show is the Cooler Master Silent Pro M. The main issue with this particular unit is the fact that its fan controller sacrifices cooling performance for excellent acoustics which is fine and dandy right up until load increases to the point of no return. We hit a blisteringly hot exhaust temperature of 59 degrees (37 degree temperature delta) until the fan realized that it was time to wake the hell up and start moving some air.

All in all, we can really see the impact of slightly different fan speed profiles in this test.


Acoustics


To be perfectly honest with you, the vast majority of the power supplies featured in this roundup were quiet to the point where there is no way you would hear them over the sound of your case, CPU or GPU fans. The only exception here is the unfortunate Cooler Master unit which actually lived up to its name right up until its internal temperatures hit stratospheric levels. It then let out a plaintive wail as the fan whirled into motion in a vain attempt to cool things down as quickly as possible. While we are sure you will never push your 850W power supply to these levels, you should still be aware that the Silent Pro M exhibits a Dr. Jekyll / Mr. Hyde mentality when running near its limits.
 
Status
Not open for further replies.
Back
Top