Posted On: Mon, May-19-2014
Reviewer: OklahomaWolf
Product: EVGA Supernova G2 750W
Product Link: http://www.evga.com/Products/Product.aspx?pn=220-G2-0750-XR
Supplied By: EVGA
Price: $129.99 @ Newegg

Introduction

A little while ago, we looked at the latest awesome unit from EVGA - the G2 850 watt. But there was another unit that came in the shipping box that day, the little brother to that unit. Folks, today I'm looking at the G2 750 watt model. Let's see if the awesomeness from EVGA continues.

Page 1 - Marketing

Greetings, humanoids and non humanoids alike. Are we ready for a power supply review? Good. Today, I'm looking at another unit from EVGA, the G2 750W. Presumably, this unit uses the same platform as the 850W model we looked at mere weeks ago, so it should turn out to be an excellent unit indeed.

I see no reason for delay, so let's get right into it now.

First, the box and its marketing. Nothing new here - everything we see here is the same as what we saw on the box of the 850. Fully modular. Single 12V rail. Ball bearing fan. Switchable semi-fanless mode. 100% Japanese capacitors.

Once again, EVGA would like you to know that it is #1. I'm done with the box, let's open it up.

Yes, there does appear to be a power supply in this box along with a few other things. Let me get it all out on the table for you.

A power supply, user guide, cable bag, modular cables, a functionality tester, a bag of screws, and some velcro cable ties were all found within the cardboard cocoon.

Since the user guide is identical to the one for the 850W, I'm going to push the lazy button and just re-use the image from that review.

Like the 850W, this unit comes with a doohickey you can plug into the ATX cable to test and see if the power supply works before you use it. Remember, the fan won't spin if it's in semi-fanless mode.



Page 2 - First Look

Here's the power supply itself. I have to admit that it bears little resemblance to an exploding star.

Another angle for you. The finish on this thing is still as cool looking as the day I first saw it.

EVGA is still calling their semi-fanless mode "eco mode," I see. Lots of nice open grille space for hot air to get through, though it could never possibly match the amount of hot air coming out of the FCC this month.

Ah, a load table. Let's get that done now, then. I like how EVGA is openly bragging right on the label that this thing is good for fifty degrees at full gallop. If it's anything like big brother, I don't think I'll have any trouble with the hot box.

EVGA G2
750W
3.3V 5V 12V -12V 5VSB
24A 24A 62.4A 0.5A 3A
Max Power
@ 50C
120W 748.8W 6W 15W
750W

As is usually the case, the labeling on one side is inverted from the other so you can make it right side up no matter how you mount it: fan up or fan down. Semi fanless units like this I often run fan up so they can get some convection cooling, possibly keeping the fan off for longer.

Speaking of fans, here's the grille on this one. Warranty? Prepare to be voided. We're two pages away from looking at the guts.

Do not remove this cover. Well, Mr. EVGA, I'm just not going to listen to that. I want to see the guts of this thing later.

Here are the modular cables. They are nearly identical to the 850's cables, so I decided to spare myself from the extra cable shots. All are black sleeved cables, except for the 3.5" adapter which is ribbon cable.

The PCI-E cables have small capacitors integrated in them. Not sure those are really necessary, but they likely do help a little bit in getting this platform those insane looking low ripple shots we've been seeing.

Cabling - EVGA G2 750W
Type of Cable Length from PSU
Modular Cables
24 pin ATX connector 590mm
4+4 pin EPS/ATX12V 700mm
4+4 pin EPS/ATX12V 700mm
6+2 pin PCI-E 700mm
6+2 pin PCI-E 700mm
6+2 pin PCI-E 700mm
6+2 pin PCI-E 700mm
5.25" to 3.5" adapter 105mm
5.25"+5.25"+5.25"+5.25" 550+100+100+100mm
SATA+SATA+SATA 550+105+105mm
SATA+SATA+SATA 550+105+105mm
SATA+SATA+SATA 550+105+105mm
Unit Dimensions (L x W x H)
180mm x 150mm x 86mm

Two less PCI-E connectors and one less SATA cable - that's all the difference between this unit and the 850. Lengths are close enough to be virtually identical.

Let's start load testing now. It's 9am as I type this, and I'm expecting a big day of lightning shooting with Mr. Nikon later. I gots ta be ready fo' dat, peeps.



Page 3 - Cold Testing

And now we do some load testing. Fortunately, that gutless wonder I blew up last week has not, in fact, damaged the Rek RF9901 power meter, so it will be used to get my power readings again today. One SM-268 will be doing the loading of the unit, while the rest of the usual gear gets the rest of the measurements.

Results from EVGA G2 750W
STANDBY load tests
Test # +5VSB DC Watts/
AC Watts
P.F. Eff.
1 0.5A 2.55W/
3.36W
0.338 75.9%
5.10V
2 1.5A 7.59W/
9.83W
0.470 77.2%
5.06V
3 3.0A 14.9W/
19.7W
0.531 75.6%
4.95V

First, the standby load tests. And like I've seen before with this platform, efficiency is pretty well average to above average here. Nothing to complain about.

Voltage regulation is 2.9%, or average. It's not a big deal for the standby rail to have 1% regulation, but at the same time this can often be a sign of looseness at some of the ATX connector pins; either on the power supply end or the motherboard end. One hopes that this unit still gives amazing stability results in the main tests.

Results from EVGA G2 750W low load test
Load
Level
+3.3V +5V +12V -12V 5VSB DC Watts/
AC Watts
P.F. Eff.
10.1% 1.5A 1.5A 4.8A 0.2A 0.5A 75.8W/
91.1W
0.966 83.2%
3.31V 5.03V 12.15V -11.90V 5.09V

But first, one hopes that you like what you see here, because one certainly does. We have very healthy voltage readings, and a very not bad for a Gold unit efficiency reading of 83.2%. If this were a Titanium certified monster like that AX1500i, it would be forced to do 90% here to get a pass. But it's not.

Let's run the main tests and see if we hit Gold.

Results from EVGA G2 750W COLD load tests
Test #
+3.3V +5V +12V DC Watts/
AC Watts
AC
Input
Eff. P.F. Intake/
Exhaust
Progressive load tests
1 1.5A 1.5A 11A 150.9W/
170.5W
119.5V 88.5% 0.984 24C/
24C*
3.30V 5.03V 12.14V
2 3A 3A 22A 298.8W/
331.1W
118.5V 90.2% 0.993 26C/
25C*
3.30V 5.02V 12.11V
3 3.5A 3.5A 28A 377.5W/
418.3W
119.9V 90.2% 0.994 28C/
27C*
3.29V 5.02V 12.09V
4 5.5A 5.5A 45A 600.6W/
674.6W
117.2V 89.0% 0.996 26C/
39C
3.28V 5.00V 12.06V
5 7A 7A 56A 748.7W/
854.4W
117.4V 87.6% 0.996 26C/
43C
3.28V 5.00V 12.03V
CL1 14A 14A 0A 121.1W/
146.8W
119.5V 82.5% 0.980 33C/
27C*
3.29V 5.01V 12.15V
CL2 0A 0A 62.4A 753.7W/
858.9W
117.5V 87.8% 0.996 26C/
42C
3.29V 5.02V 12.00V

Ah yes... that's Gold, all right. This unit had no trouble hitting those efficiency numbers, and I reckon even if the unit were in fully fanned mode it wouldn't be a problem. The unit ran fanless any time you see an asterisk in the above table, temperature column.

I talked about possible connector contact issues earlier on this page. Was it a problem? Uh... no. Not at all. I did hit 4.9V on the 5VSB rail on test five, but that is not out of spec and really... who cares about the standby rail when the main power is online? Not me, says I to myself. You'd probably like me to run some stability calculations. Try this on for size: on the 3.3V rail, I got 0.6%. On the 5V rail, I saw 0.6%. Finally, on the 12V rail I see 0.9%. Better than excellent on all three, and the average is a very impressive 0.7%.

If EVGA keeps this up, and units like the AX1500i keep coming in here and raising the bar, I might have to alter my scoring requirements for excellence. Currently, I need 1% averaged in the hot box to take no points off the performance score. I may have to move that up to 0.5% soon. But how soon is now? I am human and I need to be loved, just like everybody else... does... ok, I think I'll turn off the Smiths now. No, wait. I'll leave them on. See, I've already waited too long.



Page 4 - Hot Testing

Overshoot Transient Testing - EVGA G2 750W
VSB On
VSB to Full, 12V
Off to Full, 12V

Before we do the old "heat it up and see if it croaks" bit, we'll look at the power on voltage spike tests. And marvel at their near flawlessness. There are no spikes out of spec... in fact, I'll be darned if I see any spikes at all. Rise time looks fantastic as well at less than half the spec. This platform is awesome.

But, we do still need to cook it if we can. I'm locking it into my hot box now and turning off all the extra fans... let's see what happens. I'm going to run it in fully fanned mode this time - not sure I've ever done that with this platform, as sometimes that can cause the unit to miss its 80 Plus target and I try to test them as 80 Plus would have.

Results from EVGA G2 750W HOT load tests
Test # +3.3V +5V +12V DC Watts/
AC Watts
AC
Input
Eff. P.F. Intake/
Exhaust
Progressive load tests
1 1.5A 1.5A 11A 151.0W/
171.7W
117.5V 87.9% 0.987 26C/
28C
3.31V 5.03V 12.14V
2 3A 3A 22A 298.8W/
332.5W
118.5V 89.9% 0.993 30C/
35C
3.30V 5.02V 12.11V
3 3.5A 3.5A 28A 377.8W/
419.4W
118.9V 90.1% 0.994 33C/
40C
3.29V 5.02V 12.10V
4 5.5A 5.5A 45A 600.1W/
676.5W
114.8V 88.7% 0.996 39C/
50C
3.28V 5.00V 12.05V
5 7A 7A 56A 748.1W/
857.1W
115.6V 87.3% 0.997 44C/
57C
3.27V 5.00V 12.02V
CL1 14A 14A 0A 121.1W/
148.4W
118.7V 81.6% 0.981 30C/
37C
3.29V 5.01V 12.16V
CL2 0A 0A 62.4A 753.7W/
860.0W
118.4V 87.6% 0.996 42C/
53C
3.29V 5.02V 12.00V

EVGA's got these tests handled once again. Forty-four degrees, and it's yawning along like I'm not even trying to hurt it. Awesome.

80 Plus Gold? No problem, even with the fan always running. I never really had much doubt it would do fine, there, as the cold tests were all far enough above the mark that no little fan was going to affect them.

Over in the voltage readings, we again see excellence. We're now at 1.2% for the 3.3V rail, 0.6% for the 5V rail, and 1% on the nose for the 12V rail. That's an average of 0.9%. What more do you want in a power supply?

Oh yeah... you want one with low ripple, too. Let's see what the scope has to say about that.

Oscilloscope Measurements - EVGA G2 750W
Test #
+3.3V
+5V
+12V
LLT
Test
1
Test
2
Test
3
Test
4
Test
5
Test
CL1
Test
CL2

Ha! Yeah, EVGA's coming out of this with another perfect performance score. You just can't hate this platform... it's amazing. Less than 15mV on all three rails.

Excuse me for a moment. I might need to print these scope shots out and frame them.



Page 5 - Disassembly

Time to take things apart. As you can see above, this is the same Super Flower Leadex platform as the 850W. In fact, since these are so similar to each other, I'm only going to do a few basic shots and call it a day. Full disassemblies take a lot of time, and often they aren't needed to identify most of the parts.

Here's our fan, and it is indeed a ball bearing model.

Once again, no line filtering parts are found at the receptacle.

Soldering and boardwork is excellent, though not as good as the likes of Delta.

All line filtering on the mainboard is the same as the 850W model - missing two Y caps again, but I won't score against it as there are still two more after the bridge.

Main switchers and PFC transistors are all 5R199Ps.

The modular board. True to their word, EVGA has gone with all Japanese capacitors from Nippon Chemi-Con.

The four 041N04N parts used for the 12V output can be found on these heatsinks.

Finally, the 5VSB output runs through that single S10C60C you see on the bottom left, next to the fan control PCB.



Page 6 - Scoring

Performance (40% of the final score) - 10. I mean, really... were you expecting any other score in this category today? Beyond excellent ripple suppression, beyond excellent voltage regulation, easy pass for Gold efficiency both times. I mean, come on... if anything deserves a perfect score here it's this unit.

Functionality (20% of the final score) - and it's getting another 10 here, as well. We have a fully modular unit with very nice black sleeved cables. The manual's decent. There are lots of connectors. There's a switchable fanless mode. EVGA gave us a little test adapter to see if it works before we use it. There are lots of other goodies in the box. Can you find something to complain about here? I certainly can't.

Value (20% of the final score) - Newegg's offering this unit for $129.99 right now, with a rebate on top of that. I can't score on rebates, but I can drool all over that price. It's directly competing with the Thermaltake Toughpower Grand, which may or may not still be the same one I reviewed. That said, the real competition is from Seasonic. Their KM3 750W is ten dollars more money. And you'd probably think EVGA has this all sewn up based on that alone. But... XFX has a Seasonic built unit at this same price point, so this category is not a lock like we saw with the 850W model. I'm going to go with a 9.5 here. When a Seasonic build is your only competition, you're doing it right. In fact, I'm almost ready to say that it is Seasonic who is competing with this one than the other way around.

Build Quality (20% of the final score) - as was the case with the 850W, I still want to see that fan control switch wiring cleaned up some. There's a little more glue on it this time holding it to the modular board insulation, but it still looks rather like an afterthought. There's room for that switch directly on the back panel... how about we do that instead, Super Flower? 9.5.

Performance

10

Functionality

10

Value

9.5

Build Quality

9.5

Total Score

9.8

Summary

The gravy train keeps on rolling for EVGA today with the G2 750W. Expectations were high, and EVGA rose to them without breaking a sweat. I have to imagine that over at EVGA headquarters, they have to regularly take breaks to go up on the roof and yell, "Top o' the world, Ma!" I certainly couldn't blame them. Now all we need from EVGA is their answer to the Corsair AX1500i, and we'll be all set.

The Good:

  • excellent voltage regulation
  • less than 15mV ripple on the outputs
  • switchable semi-fanless mode
  • excellent build quality
  • 10 year warranty
  • competitive price

The Bad:

  • nothing

The Mediocre:

  • 10 year warranty requires registration first




This review was provided by : JonnyGURU

The URL for this review is :
http://www.jonnyGURU.com/modules.php?name=NDReviews&op=Story&reid=380