BFGTech BFGR1000WPSU Review @ [H]

FrgMstr

FrgMstr

Just Plain Mean
Staff member
Joined
May 18, 1997
Messages
55,634
BFG Tech BFGR1000WPSU Power Supply - BFGTech did not disappoint when we asked them to participate in our inaugural PSU review program. Today we have their 1000 watt rated power supply that will be the first to run the [H] gauntlet.

If we were to ignore the failure to produce it’s full labeled output of 1000w either in the test chamber at 45c or outside of the test chamber at 20c the BFGR1000WPSU was still a mixed bag as seen by the units non existent ripple at all loads we were able to test, efficiency that was also pretty good for a unit that is not 80+ certified but with some less than desirable regulation on he 12v rails with their "bouncing".
Click to expand...

This is our first one guys, so let the slings and arrows fly. If you tell us we suck, please tell us why. That way hopefully we can fix things. There are still some shortcomings on format, but please chime in on these things too as we want our PSU reviews to be easy to read as well as informative.
 
Once I saw how HardOCP setup it's PSU reviews, I started to think that SilentPC Review may have a valid contender on it's hands for reliable PSU reviews.

But cumon Paul! Where's the PCP&C?
 
There's more than silentpcreview pumping out good reviews; many do a better job, like JonnyGURU.

As for PCP&C, only time will tell; I'm hoping for one too, but i wouldn't get my hopes up at the moment.
 
Karma said:
Once I saw how HardOCP setup it's PSU reviews, I started to think that SilentPC Review may have a valid contender on it's hands for reliable PSU reviews.

But cumon Paul! Where's the PCP&C?
Click to expand...

Two of them sitting here in my office. We will get to them.
 
Good first review. I knew that 45C testing methodolgy was going to be where units fail. Glad to see you pushing the envelope.
 
Karma said:
Once I saw how HardOCP setup it's PSU reviews, I started to think that SilentPC Review may have a valid contender on it's hands for reliable PSU reviews.

But cumon Paul! Where's the PCP&C?
Click to expand...

WoW!! You are kidding right???
SPCR doesn`t hold a candle to JonnyGURU`s reviews....
Valid contender what thats? SPCR is not the end all authority on PSU`s..that honor IMO is Jonny`s!!
HardOCP will be doing a real fine job!!
 
It's great to see another reliable and informative source of information about PSUs on the web.

My one request would be moving the data from pictures to HTML - the JPEG compression artifacts (on a .GIF image???) aren't pretty and seems to break up the article somewhat - you need to click on the images to view the data. Not only would it (I believe) improve the availability of the information you're presenting, it'd make it a lot easier to publish and/or correct the data (see this image for instance, the 1st Test's "Voltage" column says "31ow (input)").
 
Unfortunately once we tried to coax out a wattage close to its rated capacity, 958w in this case, the unit it died. Not once, but twice!

typo on page 4?


it would be nice if you could include a little blurb about the noise in the conclusion (i know you mentioned earlier in the review that you wanted to strap it down since the fans were going full tilt, it would be nice to have something about the noise in the conclusion)


The thing about SPCR is that they're anal about noise, which is great. They can tell you what PSU to put in your case so that you won't be able to distinguish it from background noise at 1 meter.
 
Very good first review, absolutely no punches pulled as to the units capabilities. As was summed up at the end, if it says 1000W, it does 1000W, plain and simple, i think some relabeling may be in order from BFG to clear this one up, but that's just a personal opinion.

Not doing it's rating at 20C is...well, i dont really need to say it do I. Looking forward to the next one up on the block, any guesstimates as to the date for the next review release, whatever unit it may be ? And can i assume that your begining at the top of the line testing 1kw units for the next couple of weeks or so before heading down the scale ?
 
wow great review! can't wait until the next one!

maybe this will tide us over until the R600 is out....:rolleyes:
 
tickle_me_emo said:
it would be nice if you could include a little blurb about the noise in the conclusion (i know you mentioned earlier in the review that you wanted to strap it down since the fans were going full tilt, it would be nice to have something about the noise in the conclusion)


The thing about SPCR is that they're anal about noise, which is great. They can tell you what PSU to put in your case so that you won't be able to distinguish it from background noise at 1 meter.
Click to expand...

Did you play the video?

That WHIRRING noise is the noise of the load tester's fan.

Noise levels can only be subjective when you use a real load tester to assess power supplies. And with the power supply sitting behind a door in an incubator, I'm sure not even "subjective" is a good word to use.
 
Oklahoma Wolf said:
Good job on the first review :)

The scope readings seem a bit strange somehow... as much bouncing around as the voltage readings did, it really should have showed up on the scope too, like in Jon's review of the 850W: http://www.jonnyguru.com/review_details.php?id=21
Click to expand...


Yeah I was pretty skeptical myself while watching the traces....but they checked accross two units. Unfortunately before I was able to really try and force the issue the units died.
 
Everything read well, however I was bothered by the last comment regarding the line regulation.
"These changes don’t show on the oscilloscope output because they occur at a periodicity that is much longer than the sample period of the oscilloscope. So while the oscilloscope showed us very little variation, the line regulation showed us a slightly different story"

The Easyscope software claims it can have a 50ms/div timebase, this should yield 2.5seconds of data on the screen. Although annoying, 2.5s update is slow, you should be able to screen-cap the variations shown by the load regulator.
 
oDii said:
It's great to see another reliable and informative source of information about PSUs on the web.

My one request would be moving the data from pictures to HTML - the JPEG compression artifacts (on a .GIF image???) aren't pretty and seems to break up the article somewhat - you need to click on the images to view the data.
Click to expand...

Agreed. The "easier to read" version wasn't :p Having the results image clickable and not readable in-line really breaks up the readability.

Also, your testing seems a little wide-brush - it works at 787W, and fails at 958? There are 171W between working and not! At least try halfway between here and there...
 
Anyone else not get the UL link on the first page to take you to any information about the unit? It seems to lack a search tag or number for me to do it myself.

Great review. It should be fun to see how many of these PSU fail such tough standards. It will be quite spectacular when one passes.
 
unhappy_mage said:
Also, your testing seems a little wide-brush - it works at 787W, and fails at 958? There are 171W between working and not! At least try halfway between here and there...
Click to expand...

Actually that isn't that big of a deal because if it is a 171w or 17w difference a failed run in its output range is a failed run. We could do the testing for weeks on end and simply increase 5w at a time but when it fails at say 901w instead of 958w that gives us no real difference. The unit failed.
 
Good first review. Id be real curious to see if the other "so called" 1kW PSUs are anywhere near what they say theyll do. Looking forward to more reviews in the future.
 
Paul_Johnson said:
Actually that isn't that big of a deal because if it is a 171w or 17w difference a failed run in its output range is a failed run. We could do the testing for weeks on end and simply increase 5w at a time but when it fails at say 901w instead of 958w that gives us no real difference. The unit failed.
Click to expand...

But given the high standards you're setting for the power supply, it's no great surprise to see units fail the test. I suppose more tests will be needed to see if anything passes. But if all the supplies you test fail at 75% load at 45C, what does that prove in terms of their relative goodness? I think as the loads get higher, the tests should be closer together in terms of output.

And a binary search algorithm (think of playing hotter-colder) will help you find and isolate the failure points faster. Test the easiest case first, then the hardest. If it passes that, then go through the rest of the pre-determined cases to see how ripple is and so forth. If it fails, go to the halfway point and see what it does there. Lather, rinse, repeat.
 
Well, the one thing I would say is that not only are the jpegs of your test results not happy with jpeg compression, but the tables need more info on them.

For starters, the name of the PSU being tested, ambient temps, what % of load, things like that. Seriously, your video card review charts have all the info on them to know exactly what they are showing you in absence of a supporting article. These tables I could open up two from the same article and not have a clue which represents which test.

The testing itself is nice though.

I think one thing you may wind up running into long term is that you are not going to be passing many power supplies. It might be nice to have a quick summary of the [H] rating of the PSU that breaks out the conclusion of what you are really buying. Which in this case is more realisticly a decent 800w power supply.
 
unhappy_mage said:
And a binary search algorithm (think of playing hotter-colder) will help you find and isolate the failure points faster. Test the easiest case first, then the hardest. If it passes that, then go through the rest of the pre-determined cases to see how ripple is and so forth. If it fails, go to the halfway point and see what it does there. Lather, rinse, repeat.
Click to expand...

There area numebr of problems with doing that. Time is one big point. To do one review that completes all the tests without having to reset because of a failure takes ~14hours for data collection alone. Another problem would be getting enough units to do that kind of search, quite simply we would probably need to buy/acquire 6-8 units. The third issue would be variation between those 6-8 units.
 
Very nice first review, the results were about what I expected for a topower unit rated at 1KW. Please in the future just make the charts in HTML for legibility.
 
unhappy_mage said:
But given the high standards you're setting for the power supply, it's no great surprise to see units fail the test.
Click to expand...

We'll see after Paul runs some more power supplies through testing.

For example, the two BFG's blew fuses. The fuse is on the primary side of a PSU. So it's not as if regulation on the DC (secondary) side blew up because of the load. It's more likely that because of the heat, the windings of the transformer shorted.

I will say that I would rather have that happen than having my rectifiers blow up because a failure on the AC side of the PSU isn't necessarily take out any of your hardware. :D
 
Good review! Cant wait to read more, keep up the good work..

I am glad to see BFG step up to the plate, disappointed in the product though. A 1000Watt PSU being marketed for enthusiast should be a 1000Watt PSU at any reasonable temp…

I respect that [H] was honest and truthful about the product even if that said product happens to be a sponsor, I am very impressed. This is the main reason I have always trusted this site for my and others computer hardware purchases. THANKS [H]ardocp!!!
 
Very nice job! You guys are off to a great start with this PSU evaluation thing. I'm seriously impressed!

One thing that would have made reading a little easier for me would be a better description of what those ripple graphs are actually saying. Like what does one bar on the y-scale equal? You said it was below the 125mv cutoff, but it isn't clear from the graphs how close to that point the ripple is getting.

Another thing, I got confused by your use of the term "OCP"... I thought "Overclock Comparison Page ... huh?" I realize now that you're talking about overcurrent protection, but I was scratching my head for awhile. Like a fool.

The only issues I took with the review stem from my own ignorance, but since I doubt I'm much more ignorant than a typical reader, it might be something to look into :)

This site just keeps getting better and better. Having a great source of power supply reviews from my favorite tech site makes life a little less depressing!
 
seems like a much tougher and more thorough evaluation process than I have seen at 90% of other sites. great start guys. some good suggestions in this thread for readability as well :)
 
Excellent review guys. It's nice to see [H] branching out a little bit. I can't wait to see how my HX620W, aka the Beast, or PCP*C, aka King of PSUs, fares in this insane environment :)
 
Nice review, I enjoyed reading it.

Only thing I would request is that in the future on the load testing pictures add lines between the tests or something. It would make it easy to read.
 
I just want to echo what's already been said by 3 or 4 others about the charts. Nice idea, I think, having them as seperate pictures in seperate windows, but I too would prefer the chart be located right on the review page. Also, I don't know if the compression blurred it out but it was a little tricky following the lines across from the left side to the right side to see what the combined rating of, say, the 12v rails were. A couple extra lines across the chart would make it easier for the eyes to follow.

I love the start of these PSU reviews though. If you're not destroying at least one unit per review then perhaps it's not a proper review. :p
 
unhappy_mage said:
But given the high standards you're setting for the power supply, it's no great surprise to see units fail the test. I suppose more tests will be needed to see if anything passes. But if all the supplies you test fail at 75% load at 45C, what does that prove in terms of their relative goodness? I think as the loads get higher, the tests should be closer together in terms of output.

And a binary search algorithm (think of playing hotter-colder) will help you find and isolate the failure points faster. Test the easiest case first, then the hardest. If it passes that, then go through the rest of the pre-determined cases to see how ripple is and so forth. If it fails, go to the halfway point and see what it does there. Lather, rinse, repeat.
Click to expand...


/agreed Exactly what I was thinking. Id like to see failure points, or at least closer range.

As much as I envy the purity of your power, I'd also like to see real-world regulation, like when your wife flips on the disposal or runs the vacuum cleaner under your feet and the plug is at the same outlet as your unit.

Also, if you could, on first use of an acronym, use full wording and possibly a short explanation or link.

Last, although not a PSU, id like to see power conditioning and related items such as UPS, surge protect, line conditioning. It would be nice to know that multi-thousand dollar rig is really protected by that $50 surge protector.

Very informative. :cool:
 
Awesome review, thank you! I look forward to the next!

A couple things i would like to note that probably aren't very big.

You posted a rear shot of the closed unit, but i did not see a front shot of the unit.

Also i don't remember reading any mention of the length of the cables, this would be a nice thing to point out with a lot of modders with full tower cases, and some reversing the HDD for cable management. I understand the main intent is measuring performance.

Also it would be cool if the thumbnails for the DC output quality were zooms of the original photo so at a glance i can tell if I'm clicking on a photo with any noticeable difference or not.

Before
1171920100dW8getjZ8u_6_12.gif




After:
testrr1.gif
<- suppost to be a zoom icon
 
is an enermax galaxy 1000w gonna be on the test bench soon? i'm anxious to see how mine compares. changed from a 510w pcp&c when i had to rma cause one of the pcie connectors was bad.
 
Nice review but I would have liked to see some pics/descriptions of the cables themselves. Length and plugs per line can be huge for some folks.

I can't wait to see the Corsair 620W review.
 
I can see lots of power-supply OEMs/partners/branders/whateverthey'recalleds saying "Aw, crud!". The power ratings on PSUs have been inaccurate, exaggerated, or just plain BS for many years, and it's about time someone called them on the carpet. The fact that very few people's rigs actually consume more than 250W has given PSU manufacturers a lot of leeway, simply because few people will actually try to draw 450W from a "450W" PSU. I predict that 95+% of the power supplies you review will not meet their sticker specifications.

<pedantic>On page two, you use the phrase "begged the question" improperly. It should be something like "raised the question"</pedantic>
 
Some very solid and relevant feedback guys! Keep it coming and thanks for taking the time to let us know how to make these better.
 
You must log in or register to reply here.
Back
Top