EVGA SuperNOVA 1000 G3 PSU Review

We already reviewed the SuperNOVA 850 G3, which left a good impression on us. Today, we're looking at the 1000 G3, this family's current flagship model. It's a physically diminutive PSU that offers high performance in every area.

By the time we finished testing EVGA's SuperNOVA 850 G3, we turned around and put the 1000W version on our bench. It's currently the family's highest-capacity model, at least until a 1.2kW or 1.6kW unit emerges.

EVGA adopted the new Leadex II platform quickly, replacing its aging (but still highly competitive) G2 series. Those are the power supplies that helped EVGA make a name for itself, and they were followed by even higher-efficiency P2 and T2 implementations.

Our test results from the SuperNOVA 850 G3 were really good, so we're expecting a lot from this more potent flagship. 

All G3 units share the same 80 PLUS Gold efficiency certification. Given that EVGA didn't go for a higher efficiency level, it is safe to assume we'll see Platinum-rated P3 and Titanium-rated T3 lines as well.

The SuperNOVA 1000 G3 is fully modular like the other members of this line, and it features a 130mm fan packed into a compact chassis. Seeing a 1kW PSU in a 15cm-deep enclosure is truly amazing! Up until now, the smallest 1kW PSU we tested was SilverStone's ST1000-P with its 16cm depth and 431W per liter power density. But the 1000 G3 changes this, even if we're not always enthusiastic about downsized PSUs. Invariably, they lead to more noise as PCBs are overloaded with hot components that can't be cooled quickly enough. The 850 G3 showed that it isn't a quiet power supply, especially under tough conditions. We can't help but assume the 1000 G3 will fare even worse in the same tests, given higher capacity. In fact, the 130mm fan will have to spin faster to push enough air. And if EVGA wants this PSU to survive its 10-year warranty period, the fan profile needs to be aggressive. If you really need a quiet PSU, small dimensions and high capacity don't go particularly well together (even if it features a semi-passive mode).

Specifications

The set of protection features that Super Flower builds into this platform is complete except for OCP at +12V, since this PSU only has a single +12V rail. An HDB fan is definitely an upgrade over the G2 series' double ball-bearing one. However, a smaller diameter means that it has to rotate faster in order to move the same amount of air.

Again, we marvel at EVGA's 10-year warranty. The company was first to offer this on its high-end models, forcing the competition to follow suit. Seasonic recently set the bar even higher, upping its coverage to 12 years on the Prime models. That's crazy to us. We think anything longer than five years is a gamble for the companies backing those guarantees, since a PSU's reliability can be hugely affected by the quality of the mains network feeding it with power. On top of that, environmental conditions can easily destroy a PSU.

Power Specifications

Rail3.3V5V12V5VSB-12V
Max. PowerAmps242483.330.5
Watts120999.6156
Total Max. Power (W)1000

The single +12V rail can provide the unit's full power on its own with more than 83A of maximum current output. The 5VSB rail has enough capacity for today's needs as well. 

Cables And Connectors

Modular Cables
DescriptionCable CountConnector Count (Total)Gauge
ATX connector 20+4 pin (600mm)1116AWG
4+4 pin EPS12V (700mm)2218AWG
6+2 pin PCIe (700mm)2218AWG
6+2 pin PCIe (700mm+150mm)3618AWG
SATA (550mm+100mm+100mm)41218AWG
Four-pin Molex (550mm+100mm+100mm+100mm)1418AWG
FDD Adapter (+100mm)1118AWG
AC Power (1500m)1--

All of the cables are long, and the number of PCIe and EPS connectors is sufficient. The distance between PCIe connectors (on the cables that host more than one) is adequate, too. But we can't say the same for the four-pin Molex cables where the connectors are installed only 10cm away from each other. This distance should be at least 13cm in order to avoid compatibility issues.

Finally, EVGA was kind enough to provide an FDD adapter in its bundle for anyone who still needs a Berg connector.

Power Distribution

Since this PSU features a single +12V rail, we do not have anything to say about its power distribution.

MORE: Best Power Supplies

MORE: Power Supplies 101

MORE: How We Test Power Supplies

MORE: Picking The Right Power Supply: What You Should Know

MORE: Is 80 PLUS Broken? How To Make It A More Trustworthy Certification

MORE: All Power Supply Content

Create a new thread in the US Reviews comments forum about this subject
This thread is closed for comments
13 comments
    Your comment
  • powernod
    140$. Exactly the same cost with Corsair's RMx1000 current pricing. Not sure which unit i would choose between those 2.
    0
  • BugariaM
    Nice.

    >>>The 1000 G3's single bulk cap is provided by Chemi-Con (400V, 820uF, 2000h @ 105°C, CE)

    Strangely, but I did not find "CE" series of capacitors in the Nippon Chemi-Con catalog
    0
  • 10tacle
    Wow. The price point of this and overall performance makes this a serious player even for those who were in the market for nothing more than 850W gold or even platinum PSUs. So many people out there say "you don't need more than a xxxW power supply for your build." I like to buy based on <60% of calculated max load use of my build for less fan noise, heat output, and higher efficiency wall draw.

    Buying a gold 600W PSU when my rig uses 500W for example does not make much sense to me. I'd prefer to spend a little more and buy a gold 850W ($110USD vs. $130USD in the example of these two 650 & 850 G3 series on NewEgg [US] right now). Not only for a higher efficiency operation as described above, but for future headroom growth. But I understand many do not think that and look to save a few dollars everywhere...one being on the PSU.
    0
  • 10tacle
    Anonymous said:
    140$. Exactly the same cost with Corsair's RMx1000 current pricing. Not sure which unit i would choose between those 2.


    Keep in mind when comparing the RM1000x that it was introduced at $180 in November 2015 and it took nearly a year to see the prices drop down below $150. The EVGA was originally released at $170 and has only been out for four months. Also keep in mind that the Corsair is marketed more for silent operation (drawback to that shown below).

    If none of the drawbacks described and shown here do not bother you, and if overall "hot" efficiency and exhaust temperature are among top priorities for you, then the EVGA is the better performer. Average efficiency result of 7 load line tests, and the average exhaust temps under each test:

    EVGA - 88.571%, 45.0C
    Corsair - 87.857%, 49.8C

    ^^While the <1% efficiency difference is nominal and will not be seen on your monthly power bill, that nearly 5C difference coming out can make a slight difference in the temperature of your room in the summer. So the Corsair being quieter has a drawback.

    http://www.jonnyguru.com/modules.php?name=NDReviews&op=Story4&reid=494 (EVGA)
    http://www.jonnyguru.com/modules.php?name=NDReviews&op=Story4&reid=449 (Corsair)
    1
  • Co BIY
    I would have liked to see the EVGA 1000 G2 in the performance tables as well. Both to see the generational improvement and to make a value comparison while both units are on the market.

    Is EVGA stopping production of the G2 series ?
    0
  • hst101rox
    10TACLE, if the efficiency of the EVGA and the Corsair are nearly identical, the ~5 degree C temp difference could be just from the amount of airflow being less in the Corsair. Power consumption is the only thing that equates to BTUs, heat.
    1
  • 10tacle
    Anonymous said:
    10TACLE, if the efficiency of the EVGA and the Corsair are nearly identical, the ~5 degree C temp difference could be just from the amount of airflow being less in the Corsair. Power consumption is the only thing that equates to BTUs, heat.


    That's why I stated the downside to the Corsair is the hotter air coming out the back. Same thing with a video card: you have to sacrifice more heat output for a quieter fan speed.
    0
  • Aris_Mp
    I don't have 115V test results (but only 230V) for the G2s, so I couldn't add them to the graphs.
    0
  • Co BIY
    "We didn't get any OCP results because our unit died during the OPP test, meaning either that this particular sample had a problem or OPP is set very high."

    Seems like this would be a big deal ?

    Maybe at least one bullet in the Con column.
    1
  • Aris_Mp
    I have another one here, which come late, to test. If it dies as well then I will update the con list.
    1
  • TJ Hooker
    Anonymous said:
    Anonymous said:
    10TACLE, if the efficiency of the EVGA and the Corsair are nearly identical, the ~5 degree C temp difference could be just from the amount of airflow being less in the Corsair. Power consumption is the only thing that equates to BTUs, heat.

    That's why I stated the downside to the Corsair is the hotter air coming out the back. Same thing with a video card: you have to sacrifice more heat output for a quieter fan speed.

    His point is that without considering air flow rate, exhaust temp does not tell you much heat it's putting out. If you want to know how much heat the PSU is dumping into the room, divide system power draw by efficiency. Even if you're running those PSUs at 1000 W (and we assume for the moment that the average efficiency you calculated applies with a 1000W load), you'd only be looking at 9 W difference in heat output, which is pretty trivial. That's not going to make a difference to the temperature of your room.

    Regarding your video card example: slower fan speed results in a hotter GPU, but it doesn't change the amount of heat output.
    0
  • rhysiam
    Anonymous said:
    Buying a gold 600W PSU when my rig uses 500W for example does not make much sense to me. I'd prefer to spend a little more and buy a gold 850W ($110USD vs. $130USD in the example of these two 650 & 850 G3 series on NewEgg [US] right now). Not only for a higher efficiency operation as described above, but for future headroom growth. But I understand many do not think that and look to save a few dollars everywhere...one being on the PSU.

    I see this logic around a bit and I have to respectfully disagree. The problem with making an efficiency argument based on max load of the system is that unless you have a relatively unusual workload (like bitcoin mining with 3 GPUs, for example) your system very rarely actually hits those peak levels of power draw. Very few rendering/productivity tasks fully load up a GPU, and gaming workloads don't fully light up the CPU - and when they do you're probably bottlenecking the GPU anyway.

    Let's say a hypothetical build hits 500W when you fire up Intel burn test and furmark simultaneously while running a low-level disk check on all your HDDs, It's likely that same build sits at a much lower 300-350W in most normal gaming workloads. Obviously you want to be sure you can handle those peak workloads so a 550W (or, for some extra headroom) a 600W PSU is mandatory. But you should IMHO make your efficiency decision based on the most common load scenario the system will be sitting at. And 300-350W is right in the efficiency sweet spot for a 600W PSU.

    Obviously you don't want to skimp on your PSU, there's no question about that, but I feel like I see many inexperienced builders on the forums here getting talked into spending big dollars on premium 650 & 750W units for their single GPU builds "just to be on the safe-side" when something like a CX450M for ~$50 would suit them perfectly.
    1
  • MisterLeeD
    "We didn't get any OCP results because our unit died during the OPP test, meaning either that this particular sample had a problem or OPP is set very high."
    In normal operations is this of major concern??
    0