The Power Supply Unit tier list Discussion thread

Page 28 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Toggle sidebar Toggle sidebar
Status
Not open for further replies.
I too think a partnership in fine tuning and expanding the list is a great idea, but I also understand on the other side of the coin if Dottorent was reluctant to want to travel that path or relinquish any control over the list after having put so much time and work into it. I believe it would result in a more reliable and accurate list, but I also think he's done admirably with it insofar as it's a better and more relevant list than the only other commonly referenced one which we all know about and has apparently not been regularly updated consistently.
 


Thanks! Any way to ID the good from the bad reliably?
 


Would you believe...... autocorrect? 😉
 


Wish I could say the same... people pronounce my name as Gurnell or Gudrell, heck even in work when someone had to write my name down, and had the correct spelling next to them, they spelt it with one L... no blaming autocorrect for that one.
 
A knock on the head to those sort of people does the trick to remember 😀 Provided its not your boss :)

I've had the same thing happen more than 10 times. I just go: "Ahem, Its Lutfij not Luftij(or Luftji"
 
Hmm...Anyone know why the Aerocool GT 'S' series are relegated to Tier 4 considering they haven't been reviewed? Sure the Bronze ones are pretty rubbish, but what about the Gold? Some of these are rated 80+ Gold The Aerocool GT 700SG was actually quite good under testing - http://www.techpowerup.com/reviews/Aerocool/GT-700SG/
 


The temps on that unit for crossload and above rated wattage is actually pretty impressive.
First impression was that high temps would cause some odd results and problems, but since it would almost never get anywhere near the 85C the lowest cap is rated for, it should be a Tier 2a unit IMO.

At least from a quick scan over of that review.
 


One thing to worry about: In the user comments on that review there is one user who had his blow up after a year. I think Tier 2b/3 sounds about right, 4 seems a tad low down
 


To quote TechPowerUp "Inrush current or switch-on surge refers to the maximum, instantaneous input-current drawn by an electrical device when first turned on. Because of the charging current of the APFC capacitor(s), PSUs produce large inrush-current right as they are turned on. Large inrush current can cause the tripping of circuit breakers and fuses and may also damage switches, relays, and bridge rectifiers; as a result, the lower the inrush current of a PSU right as they are turned on, the better."

Regarding the Antec HCG series - the Seasonic units are discontinued, and widely unavailable. That being said, in my opinion, Tier 2 is too high for the Seasonic units. The 620W failed ATX specifications on the 3.3v and 5v rails in TechPowerUp's Advanced Transient response tests, as well as failed ATX specifications on the 5v rail under crossload. The source of which I believe is, to quote their pretest analysis of the build: "In the secondary side we immediately noticed the two, only, toroidal coils in the filtering/rectification section of the DC outputs. This means that the PSU uses an outdated group regulated design, so don't expect top performance at crossload tests." The 520W failed ATX specifications in the Crossload tests at TechPowerUp in the 5v and the 12v rails, to source I believe to quote: "Also the presence of only two coils means that the outdated, but cheap to implement, group regulation method is used, so in cross load tests don't expect breath taking results. The bigger choke is for 12V&5V and the smaller one for 3.3V."

The Delta units, by dottorrents list, are more difficult to list. The HCG 900 failed at 100% load at HardOCP's test, both at 120v input and 100v. This made it impossible for them to test the ATX specifications, causing just a failed test in general. To quote them

"at the end of the day the unit simply was not an excellent 900W unit as billed, but rather a decent 700W-ish unit. In addition to the failure to meet its full advertised output in our test environment, the HCG-900 was also a bit hamstrung by poor Transient Load Tests results and perhaps some poor quality control on the part of its connectors as the voltage regulation on the motherboard connector was significantly worse than on any other unless the unit was assisted during the testing."

Your site did test this unit, supplied by Antec, and gave it a 9.5+Rec, reviewed by OklahomaWolf, iirc. By TechPowerUp's test, however, and this list's requirements, this unit might be considered a Tier 2b, or it might be Tier 4. It really depends on how much power the unit can provide. I'd be willing to say Tier 2b considering other sites, such as yours. Then again, Antec may have given a GS, or TechPowerUp found a lemon.

The Antec HCG M series has issues as well. These are all Seasonic units. The 620M failed ATX specifications under Crossload for TechPowerUp on the 5v rail. The 750M performed well for TechPowerUp, and even in a hot test at 230v, the unit was extremely close to Silver efficiency (tested 85.74/88.13/86.62, requires min 85/89/85), and then it failed ATX specification for hold-up time (min 16ms, tested at 14ms). The 850m was even closer to Silver at 230v while hot (87.28/88.75/86.77), and while they don't want to fail it, it does fail the ATX specs for holdup time at 15.8ms, and it has high inrush current at 45.14A.

Your site reviewed the 750M (and a case), and gave it a 9.2+Rec, again by OklahomaWolf.

By this list's standards, failing ATX specifications earns them Tier 4.
_________________________________________

The be Quiet Dark Power Pro 10 1000W on TechPowerUp's 230v hot test missed Gold at 50% (min 88/92/88, tested 88.2/91.17/90.65), but it was under the hot test.

The Dark Power Pro 10 850W on TechPowerUp's 230v hot test tested (90.14/92.3/91.06), good for Gold, but the unit is rated Platinum, which carries (90/94/91), failing at 50% again. But you already know that, because you tested this unit yourself and on your 115v you cold tested (88.7/90.3/87.6) and (88.9/90.1/87.6) where it requires (90/92/89), failing on all 3 efficiencies, twice. But, again these are hot tests. Kitguru found it missed Platinum at full load, at 35C. HardwareSecrets found they missed Platinum at full load. They don't run 20/50/100 to test the other spots though. Sweclockers found they missed Platinum at 50% and 100%. With this many testers in agreement, I believe the list should reflect that the 850 is not Platinum.

The 550W missed Gold at 20% and 50% in TechPowerUp's 230v hot test. But, you already know this, you tested the unit yourself at 115v, and saw in your first test it missed under 50% load by 0.9%, and when you retested using 80Plus' loads, (min 87/90/87) you got (85.6/88.8/87.4), under cold tests.

In regards to your FSP quip, "The Dark Power Pro is roughly based off of the FSP Xilencer, but there are so many modifications made to the platform by the engineers at beQuiet, that one could almost say the similarity stops at the PCB, component placement and topology, which uses a half-bridge primary, resonant mode secondary and DC to DC for the non-primary rails." Even you stated it was -really- an FSP per say. But, I digress. None of that was really to say about their quality.


The 650W failed Voltage Regulation on the 5vSB in HardwareSecrets test at 80% and 100%. Figured I'd point that one out.

 
After your eyes are done bleeding from that wall of text, I do wonder if it's possible for JG to do 230 tests as well.


Interestingly, Ecova's documentation for the Platinum rating does not include a Serial Number.
 
Quoted directly from HardOCP's review of the Antec HGC 900: "HardOCP’s testing methodology is intended to very much push power supplies to their advertised wattage rating in temperatures that will represent some of the hottest computer enthusiast cases. So if a unit passes all our testing it is definitely not something to take lightly. In fact we expect more power supplies to fail our testing than make it through unscathed. "

You saying that it is a tier 4 unit because it BARELY dropped below ATX specs at 2x the normal operating temp, is ludicrous and completely pointless especially when they CLEARLY state that they expect most PSUs to fail this test. And in fact, that would make almost 50% of the tier 1-2b units tier 4 according to you.

These units are not NEARLY as bad as you make them seem. They are actually quiet good.
 


It's why didn't hammer it in stone. I do realize that HardOCP runs their tests at 45C. After looking back wondering, I saw on JG's site that tidbit that Antec does not list out their operating temperature for the unit. He had to look it up, and found out that it's only rated to operate up to 35C, supposedly raised up to 40. JG had 2 units to test it out, and the 2nd unit shutdown at 46C. It is good to note for the future that HardOCP did not check the operating temperature when they baked the unit. 35C/40C is a summer day's temperature. 45C isn't exactly Death Valley. HardOCP did have it running at 80% load with exhausts at 57C, and in their torture test at 80% load with exhausts at 65C.
 
45c is 113f. NO computer gets that hot unless it is outside in the sun in the desert.

also 35-40c is NOT a normal "summer day". It is 95-104f.

I would be very surprised if any of these units are anywhere in use at close to those temps.

I maintain the opinion that these are actually pretty good units.
 
also 35-40c is NOT a normal "summer day". It is 95-104f.
Click to expand...

I wish you would come tell a typical Georgia summer that! Factor in 70-90% humidity and we easily have heat indexes over 110.

15qyv89.jpg
 


No, for blessed is the air conditioning that makes living in the deep south bearable. For without it, well screw that noise....I'd move to Alaska! 😀
 
Status
Not open for further replies.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom