Today's Hard|Forum Post
Today's Hard|Forum Post

Why 80 PLUSآ® is Irrelevant to You When Buying a PSU

A HardOCP editorial about the 80 Plus® program and how it is changing the computer power supplies you are buying. Is this good, bad, or ugly, and should you care? Did you know that you paid for that 80Plus rating on your shiny new PSU? Certainly PSU efficiency is a great thing, but what about the rating system?

Introduction

By this point in time, most people have at least heard of 80Plus and/or they have seen the 80Plus logo’s on their power supplies over the last few years. Perhaps many have even come to see the 80Plus logo as a key point in your decision making process when buying a power supply. But for those of you who don’t know what the 80Plus program is, the concept behind 80Plus is fairly simple and straight forward (with this being a simplification of that). Briefly, a manufacturer submits a product to ECOS (the company that started the 80Plus program with "industry partners" before being purchased by AIQ) for testing in compliance with the 80Plus guidelines; ECOS tests the unit, gives the unit an 80Plus certification of some level and manufacturers can then parade around this "certification" as a key marketing point to get users to buy their product because it is more efficient/environmentally friendly/going to save them boatloads of money. This sounds like a win/win for everybody, right? But what if it isn’t really a win for the people that should care the most (the end user paying for the power supply)? What if the 80Plus label on your power supply is all but irrelevant?

No Matter What the Advertising Tells You, You Aren’t the Customer ECOS is Targeting

Raise your hand out there if you have ever bought a product from ECOS, or a service. Anyone have their hand up (all of my product reps please put your hands down)? I didn’t think so. And herein we find one of the problems. You aren’t the customer. As an organization ECOS (now owned by AIQ) does not serve you the end user, they serve manufacturers and the manufactures are their customers. A manufacturer sends ECOS a product that the MANUFACTURER selects for testing (these products are not randomly sampled units from production runs and this is key as we will see later). They also send ECOS a $2500 check. If the product passes whatever level of certification the manufacturer thinks it should, ECOS has a happy customer who next month will send them another product that they cherry pick and, most importantly, the customer will send ECOS another check. If ECOS does not pass a power supply they now have an unhappy customer. Unhappy customers are not good for business and ECOS is in the business of making money (a quick calculation using the number of units certified by ECOS, 3,303, multiplied by the $2500 fee tells us that so far ECOS has had revenue of at least $8,257,500 from the 80Plus program). Alternatively, in the case of an organization like consumer reports, as much as many people like to bag on Consumer Reports, the good news is you are Consumer Reports customer, and not the manufacturer, so if Consumer Reports screws up they answer to you and your interests. That is what you, the consumer, should want the arrangement to be (One quick aside, ECOS is solely responsible for making users thinking that 80Plus certifications are all about them as the manufacturers/vendors play this angle even harder in order to get you to buy their product and not someone elses). But going back to the real arraignment, where the manufacturer is the customer and not you as is the case with ECOS customers, what do the customers do then in this arrangement? They cheat.

It’s Only Cheating if You Get Caught

This topic is not exactly new as a colleague, Gabriel Torres at Hardware Secrets, ran an article some time ago about manufacturers who were labeling units as 80Plus certified that were not indeed 80Plus certified (you can read it here along with some updates). Through that article was able to flip a light on that made a few roaches scatter, but it did not completely address the issue of cheating at 80Plus. After seeing numerous units fail the 80Plus certifications they were given in our testing, and other reviewers testing, it seems that the reason is ECOS does not verify that the product their label is being used on and shipped to YOU is what they "certified". In the case of deliberate failings, this takes on two aspects that we can document.

The first, and perhaps more minor that is related to what Gabe wrote about is that companies not only have units that they know won’t pass and have used the 80Plus logo but they also use the 80Plus logo when they have not submitted to ECOS for testing (which is against the contract for usage) but intend to send in for testing. This case of being ahead of the curve is perhaps not a huge concern if the product does ultimately make the level of certification claimed. However, what if it doesn’t? Users bought that product expecting a certain level of efficiency and in the end get something different. Do they get a refund? Shouldn’t ECOS be policing the usage of their certifications? I am not necessarily talking about a few days or weeks in these cases either. Recently, we had a very well known vendor with a unit that was billed on My 31st as being 80Plus Silver certified but it was not until September 15th, a full 4 آ½ months later, that the unit was available on the 80Plus webpage. Sure this is the lesser of the two evils we are going to look at, but still if you don’t enforce the rules here where else are you not going to? Perhaps they won’t enforce their usage rules when vendor’s swap out the actual unit but keep the model name/number the same?

The Raidmax RX-1000AE was certified by 80Plus as an 80Plus Gold power supply based on Andyson’s K-series power supply. When we actually opened a retail unit it was not an Andyson K-series, but rather an Andyson E-Series unit. Andyson’s E-Series, by the way, is rated for 80Plus Bronze. One last facet with all of this is that the products sent to ECOS only have to pass 80Plus’s standards not FCC, UL, or ATX/EPS standards. So what can a vendor do? Well, beyond just cherry picking a unit they can send products that are not what will ship at retail (for instance alter cabling length, gauge, etc) or send a product that will not pass these other requirements (FCC, UL, ATX/EPS). All of this is kosher with ECOS because all ECOS’s testing requirements involve is efficiency, not quality or compliance. Well, ECOS is supposed to care about compliance but they can’t be oblivious to the fact that manufacturers are cheating them can they? If they are then there are other issues and we have the interesting question of which is worse? Don’t take this to mean that ECOS is the only that gets cheated and can’t (or won’t) do anything about it, Energy Star is just as bad. http://www.gao.gov/new.items/d10470.pdf

One of the important aspects to note about ECOS and 80Plus testing that was mentioned earlier is that vendors supply the power supplies that are being tested. ECOS does not go to a retail location and pickup power supplies; instead they rely on vendors to send them products that are supposed to be honest representations of retail products. This is an area that is ripe for abuse, and is abused, as vendors can easily cherry pick the units they send to 80Plus or even build a unit specifically to send to ECOS. Unfortunately, since ECOS does not verify anything after they test a unit, vendors have almost no chance of getting caught doing this and who can blame them when the watchman is asleep on the job?

80Plus Gold By Any Other Name Might be 80Plus Silver

Assuming a vendor does not cheat ECOS’ testing, tolerances involved with the production of power supplies make it such that your brand new shiny 80Plus Gold/Silver/Bronze (whatever) power supply may not actually be an 80Plus Gold/Silver/Bronze unit but rather one step down (or one step up but that is by far less common) if the unit was a marginal 80Plus whatever level unit to begin with. Now, in this case do most vendors care that 80Plus generates too generous of a rating for their product? Likely no. And why would they? In these cases they can simply point back to the report from ECOS to wash their hands and users are, in all reality, not going to be able to check up on this assertion anyway. So who cares that the user is not quite getting what they paid for?

This is an issue that vendors are aware of but on this aspect at least some companies do try to do right by customers. For instance, we have been working on a review where this came up and when asked the vendors indicated that they designed the product for 80Plus Bronze even though ECOS certified it for 80Plus Silver and they had no intention of changing their advertising from 80Plus Bronze to Silver. If a user gets one of this product that is 80Plus Silver they got lucky, otherwise users should expect only 80Plus Bronze. Also, previously, Corsair had this issue with their HX850 which they were billing as 80Plus Silver but ECOS certified it for 80Plus Gold. Corsair however did not change their advertising on this product as they felt they could not guarantee it would meet this higher 80Plus certification on retail units. (http://forum.corsair.com/forums/showthread.php?t=81637) However, don’t take this as a solid Corsair and SilverStone can do no wrong by 80Plus so buy them as that is simply not the case. However, do take it to mean that 80Plus test reports should be taken with a generous pinch of salt and the vendors even know this.