anandtech starting psu reviews

[kwak]

Anand posted an article about the testing methods for psu's they are going to use from now on:
-input voltage both 110V/60Hz and 220V/50Hz
-monitoring pf/voltage/ripple/efficiency/temps with 25C room temp/fan speed
-programmable ac load up to 1500W
-noise testing by placing the psu in an diy anechoic box with the load tester outside of that chamber(if i understand it correctly)

http://www.anandtech.com/casecoolingpsus/showdoc.aspx?i=3021

[Super Nade]

It would be great if they could add EMI compliance and transient testing. Good to see more big-boys jump in the fray. I wonder why they picked Chroma though? Any specific advantages? IMO they are wasting money by trying to take acoustic noise measurements. Who the hell cares about acoustic noise? :D

[jonnyGURU]

Looks like they went with the Chroma. Cool. I love the little knobs for cranking up the load. :D

[Spectre]

It would be great if they could add EMI compliance and transient testing. Good to see more big-boys jump in the fray. I wonder why they picked Chroma though? Any specific advantages?

Usual reasons ;)

[galapogos]

Kinda hard to test for EMI compliance without a real anechoic chamber, spectrum analyzer, antenna and the works...what FCC part/class do PSUs fall under anyway?

[Super Nade]

Class B I presume.

[SKYMTL]

Woah, their first one is up.......

LINK (http://www.anandtech.com/casecoolingpsus/showdoc.aspx?i=3037)

All that and once again, no ripple....and some darn odd readings from the "rails". 4 rails on a single rail PSU???? 11.65V from the +12V rail????

[jonnyGURU]

It's a bit messed up.

[Spectre]

Is there a whole page for the fan :confused:

[SKYMTL]

16 pages....my eyes hurt....

And I could have sworn that all the +12V leads terminate into a single +12V rail on the PCB....

That was an OCP knob on the secondary PCB?

[Spectre]

16 pages....my eyes hurt....



People think I am wordy..at least my 17 page one had 5 PSU's!

[SKYMTL]

And yours has ripple. And no "OCP Knob". And no whole page about a fan.

Does this Christoph Katzer post here?

On way or another, the review is definately extensive.

[signmeuptoo]

There have been times long in the past that I found myself slightly annoyed with anand shimpi. While it is interesting, and even nice that they have decided to wake up and start doing power supply reviews, why weren't they doing it before? IMHO, they just want page views back from Jonny and team, maybe I am wrong, but there have been times I didn't like anand nor his attitude. It is a personal thing with me, sorry.

I hope people continue to support this site and I wish I had known about this sight much earlier on. When someone has enough vision to take a chance and take the lead doing something the big guys should have been doing all along, my loyalty stays with the guy that took the risks to do the right thing. It's like Mcdonalds, they have been responsible for plenty of heart attacks for long and it took them decades to wake up and decide to do what smaller, more consciensious companies were already doing: Selling more wholesome foods such as full meal salads. They should have been doing that all along, so my loyalty stays with the little guys because it is thanks to them that food is getting more wholesome.

I don't want to sound like sour grapes, I am just trying to make a point. Now what we need is a small guy to make a site the tests, reviews, and reports monitors and provides a forum on the topic: Jonny, what cha think?

[Super Nade]

I think the Blue ceramic caps are the Y caps unlike what they show.

The 5V nose diving to 4.65..hmmm...?

Is the Etasis connection so readily apparent?

[SKYMTL]

I am sure their Silverstone rep will have them do some corrections to that review pronto.

[Oklahoma Wolf]

On way or another, the review is definately extensive.

I would say "expansive." ;)

My own impression is that it's almost as drawn out as my 185,000 word first novel, and only one and a half times as interesting.

[dhanson865]

I read http://www.jonnyguru.com/forums/showthread.php?t=359 and I hope my first post here isn't out of line.

Is it ok for me to say that review was a steaming pile?

[Super Nade]

Why? Is it because it is wordy? I would not hold that against it. ;)

[John]

The article appears lengthy since each page is a separate category. It could have easily been 4 pages w/ 4 categories per page. Most of the pages were filled with images anyhow.

[Hutch]

Jonny, how do you feel about your PSU review methodology being completely ripped off? Honestly, you deserve all the credit in the world for your PSU reviews. F*** anandtech and hardocp for completely ripping you off. You came up with an original idea to help people and maybe make a little bit of money on the side to help your family financially and you get completely ripped off.

If it's not you doing the anandtech PSU reviews, or they didn't ask you if they could use your methodology, then f*** them, over and over again, f*** THEM.

[Spectre]

Jonny, how do you feel about your PSU review methodology being completely ripped off? Honestly, you deserve all the credit in the world for your PSU reviews. F*** anandtech and hardocp for completely ripping you off.

You know some of us are sitting right here.................something tells me that some people doing reviews didn't really happen without his blessing.

[Super Nade]

Calm down man. Having many people do proper reviews helps spread awareness. Let me be frank and tell you that there is no place for such hostility about this. It helps no one.

As for the "ripping off" part, it just takes some basic common sense and reading through various IEEE standards/tech docs to figure out what needs to be done/how a test bench ought to be set up. Jon, did the hard work but nothing tells me that he didn't help out everybody else following the same track...a real guru (teacher) so to speak.

[Das Capitolin]

16 pages....my eyes hurt....

You're not kidding. I thought my reviews were a little much at 5-8 pages, but I suppose there are more ways to bore readers than I thought. At some point the author has to decide which is more important: a technically sound review or a readable review about technology. I choose to blend the two as best as I can, but Christoph Katzer went all out with the labcoat review.

At any rate, nobody stole anything from anyone. That's like claiming that someone stole my technique for reviewing a CPU cooler; there are only so many ways to skin a cat, and none have a patent. By the way, you stole my methodology for breathing, so please stop. Hope you can read through the sarcasm, but our point here is that nobody can really take something that is considered a standard practice.

On the flipside, I personally appreciate Paul (Hard OCP) and Jonny's assistance with getting my own ripple and regulation reviews under control. There aren't too many sites doing them (I think Benchmark Reviews (http://benchmarkreviews.com/) makes #7) so it's always good to see someone else change the course of a wayward ship. There have been too many sites doing it the wrong way for too long, and even a little effort, and sometimes 16-pages worth of effort, are always a good start in the right direction.

[JEDIYoda]

Whats interesting and quite frankly I am baffled at all the hostility towards Anand`s site doing PSU reviews.....

What do those of you who are doing reviews actually think your heads and tails above anybody else doing reviews?

We have several people posting who do reviews for other sites yet while not all of you are being so hostile it would seem to me that those who are being hostile are being very immature and childish with such comments as "ripping off" and "F*** anandtech and hardocp for completely ripping you off" and "Is it ok for me to say that review was a steaming pile?" ...yet there others who are all for more getitng involved in the testing process.....


Peace!!

[signmeuptoo]

I apologize for my sentiments. Years back I had some communication with anand that left me a little steamed, but I don't even remember what it was over, so no big deal. All I remember was that he came off arrogant to me and uninterested in hearing other people's input. But you know how it is with humans: Our memories get skewed, so...

My only gut reaction is that I'd hate to see a big McDonald's burger site steal the fire from a smaller local chain who have been making for all along the way it should be. The analogy is like what Blockbuster is doing now: They should have all along made videos available online/through the mail, but they didn't care about their customers... ...But now, a smart and nicer minded netflix comes along with a better way and what does blockbuster do? They wouldn't have also done mail videos if it hadn't been for the new guys, so my loyalty, no matter price, goes to the guys that cared enough in the beggining to "make life better". I value Netflix because I live very far away from any video rental place, and they did what no one was willing to do. To heck with Blockbuster. If Netflix went out of business, Blockbuster would just stop doing mail videos again, they don't care.

I imagine it might be silly of me to see this as a good guy vs. bad guy thing, but I believe in loyalty to people who do the right stuff from the start.

[JEDIYoda]

I don`t think it is about being loyal to JohnnyGURU...although I do understand your thoughts..
The whole world looks to Jonny for PSU guidance.
What`s interesting IMO is that when there people over at HardOCP joined in this they were welcomed a party was thrown and a ticker tape parade there was....as if they were being welcomed into the fold so to speak...yet there is not room for others to join the "big boys"???

Even Anand, if he were to post here would have to admit it will be quite a while if ever that their PSU testing is not directly compared to JohnnyGURU`s expertise!!


This isn`t a club where everyone needs to seek the approval of a group of people to test PSU`s....

The more the merrier I would think...the ones who don`t do a good job in time will fall by the way and the sites such as Jonny`s will keep right on testing and being the guage by which all other sites are compared too!!

Peace!

[Super Nade]

I like the write-up, but it could have been a bit more technical i.e more nuts and bolts and less/smaller pictures. ;)

[Spectre]

Whats interesting and quite frankly I am baffled at all the hostility towards Anand`s site doing PSU reviews.....

What do those of you who are doing reviews actually think your heads and tails above anybody else doing reviews?

We have several people posting who do reviews for other sites yet while not all of you are being so hostile it would seem to me that those who are being hostile are being very immature and childish with such comments as "ripping off" and "F*** anandtech and hardocp for completely ripping you off" and "Is it ok for me to say that review was a steaming pile?" ...yet there others who are all for more getitng involved in the testing process.....


Peace!!

Ummm......Tonto....none of the people who do reviews are being hostile.

[signmeuptoo]

Well, it has the potential of being very useful to have another authority testing and verifying products, but I just don't want the thunder being ripped from this site too hard. Sure, it probably will increase, not reduce, the awareness of the importance of knowing the supplies we use, and in as much, if anand isn't arrogant and he refers to jonny, mat, and all here sometimes, if he works complimentarily, then great. I will continue to tell people about all fo you here and "promote", and I guess I will start reading anand's site again after how many years, but anand's attitude will remain in my memory a little bit. Yeah, I don't even remember what it was about, but I remember I felt done wrong somehow.

We need a LOT more technical/laboratory standards like testing of products. Hard drives, Monitors, Sound Cards, NICs, Drive Controllers, and Motherboards need more testing and reviewing. Hopefully a trend towards more professional minded reviewing will take place. For instance, there are devices that can be put in front of a monitor (I believe) that will accurately measure parameters such as color quality. Monitors should be reviewed that way, not subjectively.

Lot's of details are important, and enough of it, but at the same time, not a darned dissertation and graphics the size of watermelons of a fan. Sure, it might be good to know if a good fan has been used, and details about CFM, static pressure, and noise levels, and even types of bearings, but I don't need a whole page on it necessarily. I got the impression they were trying to be "better" too hard. Just my impression.

I did like that they tested at both international major voltages though, but the write up could be done in a shorter, more compact style, without losing any important details.

[Tazz]

To start off with. Spectre is correct, none of the big dogs here are doing any bashing of any kind.

I know there are several sites that are working with Jonny in anyway they can (mine being one). Jon has good relations with many a review site and does a damn fine job at not attacking individuals when they attack him (much better than I would). As stated any one can easily figure out the testing process that needs to be done to properly test a PSU. The entire problem for the majority of sites is cutting loose of the thousands of dollars to do so, but are doing so. The sites have to have support in any of several ways to be able to justify doing so unless the owner is going to flip the bill straight out of his/her pocket.

More in depth testing on all products would be great. It all comes at the same price though. The more you want to test the more you have to spend to be able to test.

[SKYMTL]

Personally I am not trying to be hostile but I find it amusing how many people take reviews from certain sites as the gospel truth. Hopefully, Silverstone will have a nice long conversation with the reviewer and the review itself will be modified. Other than that, I think we can all admit that it is a good first try....a big "E" for effort like Jonny said...

I just don't like it when people are not given correct information by a site they think they can trust. And I am not necessarily refering to Anandtech here....

[GalvanizedYankee]

Give it 6 months to shake itself out...then cast stones. :p
I am quite sure the presentation of the PSU reviews will change for the better over time.
Look how things have changed with jonnyGURU.com over the last two years. All for the better I'd say. :) jon has helped many others do what he does so well, so I doubt he is troubled by another review site dealing with PSUs.

[Super Nade]

Hopefully, Silverstone will have a nice long conversation with the reviewer and the review itself will be modified. .

Why?

[SKYMTL]

Why?

We tested the power supply with the little OCP knob off as we mentioned before (the default configuration). That means that we should have been working with only one 12v rail. As we connected the cables to our machines we discovered that there are in fact four 12v rails at all times.

[Super Nade]

I assumed as much. :) But, maybe it is Silverstone who should be doing the clarifying. AT don't tell us how they distinguished between a 4 rail and 1 rail config. I'd assume that the easiest way would be to overload one rail so that the OCP shuts off the unit.

[SKYMTL]

The "OCP switch" they are talking about is actually the potentiometer that controls the +12V voltage. I myself originally thought is was a fan speed pot but after Jonny published his review of the DA850 I went back to my OP650, ran a few more tests and realized he was right.

Both the OP650 and DA850 have that same vertical PCB.

[Zap]

For instance, there are devices that can be put in front of a monitor (I believe) that will accurately measure parameters such as color quality. Monitors should be reviewed that way, not subjectively.

Some things need to be measured subjectively. For instance, how do you scientifically measure two ice cream flavors to determine which one tastes better to all people?

IMO more "decent" reviews are always a welcome thing. At least Anandtech doesn't do the "it ran my uber overclocked Celeron without smoking, so I'm giving it a Gold award" type of review.

I just wish that Gary Key would finally publish some mATX motherboard reviews that he's been promising for... MONTHS!!! :wall:

[signmeuptoo]

You know what might be cool? An alliance of computer geek websites, and the members of the alliance work in harmony vis a vis a pact. I don't mean a simple webring, something more formal and useful. Then again, it probably wouldn't work and what would there be to gain from it. I don't know...

I do like how this site is supportive of others and refers to them, sending business their way. The big guy sites should do this too. The more the sense of community there is, the better. So long as no one tries to steal someone else's thunder.

[JEDIYoda]

Ummm......Tonto....none of the people who do reviews are being hostile.

Ummm Lone Ranger...hostile...immature...ok...whatever....
you obviously didn`t read this whole thread...
you can`t tell me the few comments that I highlighted..here let me remind you...

such comments as "ripping off" and "F*** anandtech and hardocp for completely ripping you off" and "Is it ok for me to say that review was a steaming pile?"

need I say more....at the least innapropriate.....immature......ok....

Peace!!

[Hutch]

Ummm Lone Ranger...hostile...immature...ok...whatever....
you obviously didn`t read this whole thread...
you can`t tell me the few comments that I highlighted..here let me remind you...

such comments as "ripping off" and "F*** anandtech and hardocp for completely ripping you off" and "Is it ok for me to say that review was a steaming pile?"

need I say more....at the least innapropriate.....immature......ok....

Peace!!

He's saying that the mods and higher respected people aren't the ones expressing hostile remarks towards the big guns stealing Jonny's ideas. If they wanted to give the least bit of credit to where it belongs, they'd try to hire the man himself. If it weren't for Jonny, they'd still be trying to review PSU's by using a multimeter and measuring the 12V voltage by the dongles under idle and load for their testing. They wouldn't have a clue of what they were talking about without reading Jonny's reviews. I didn't even bother reading most of the review, and quite frankly, I'd prefer not to.

I just re-read my previous post... I knew I was already in a bad mood yesterday, but I didn't know I was THAT pissed off. I didn't mean to sound so harsh, and I do apologize. Damn discriminating pharmacist had my blood pressure going through the roof yesterday morning. Went to fill my prescription and she refused to fill it because it was 2 days early. Then she started completely lieing to my face when I questioned her reasoning for not filling it. I told her that I felt she was discriminating against me based on my age and labeling me as an another college kid who sells his Adderall prescriptions. Tries telling me that it's a federal law stating that pharmacists can't fill a class II controlled substance more than 1 day early and she'll lose her job if she fills it. Right, if something happened to me because I got my prescription 1 day early, it'd be her liable huh... WRONG AGAIN! It'd be the Doctors butt, she's completely covered. THE DOCTOR, THE SAME ONE WHO WRITES OUT ALL MY PRESCRIPTIONS IS THE ONE WHO WROTE THE PRESCRIPTION OUT THAT SAME DAY KNOWING IT WAS EARLY. She wouldn't even call the doctor to make sure she knew it was a day early! Then I looked down at the prescription and she wrote on it "Do not fill until 7/15" and that's when I really let her have it. This isn't the first time this same pharmacist has pulled this crap either. I wouldn't go there if it wasn't the only 24 hour pharmacy in the area or even the only pharmacy in the area that has the brand name Adderall in stock. She then calls me a junkie and an addict and I nearly lost it. Somehow I restrained myself from jumping over the counter and looked at her with a puzzled face and said, "So that's why you won't fill it, because you assume all 20 year olds who take Adderall are addicts?" "I don't know where you're from, but here we call that stereotyping." She then proceeded to call the cops on me, for who knows what reason. I wasn't asked to leave until after she called, which I was going to do anyways, as the cops would've assumed that it was all my fault.

So later yesterday I had to have my mom go back to the same pharmacy to fill it with the same pharmacist still there. Yea, they filled it, no questions asked.

[jonnyGURU]

Wow guys.... :D

It's like this..

I don't mind "competition." Nobody is taking away "my customers." It's cool.

I just want to make sure correct information is out there.

Paul @ Hard didn't "rip me off." Kyle came to me first, but this is my hobby. I could easily go two weeks without wanting to touch my computer or load tester or anything else because I'd rather go to Disney World with my family. I couldn't live up to Kyle's regiment.... even for payola. So I sold Paul the load tester and taught him how to use it. No ripping off their.

Is Anand "ripping anyone off?" I talked to Anand about PSU reviewing before Kyle even approached me, but "ripped off?" No, not really. I mean, there's really only one way to test a PSU like a factory and that's with a factory floor PSU tester. The variables come into play when people start getting creative. Like I have my "hot box." Paul has his incubator.....

But ultimately, it's how you test a PSU. These other sites that take their little trips to the PSU companies and do their little round ups have been doing those way before me. Am I not ripping them off? There's only so much data to collect... it's how you display that information.

But, like I said, I do want to make sure things are done right.

Some things I've already commented on that I would like to see addressed (being constructive and not just critical.....) have already been posted in the review's comment thread:

1. How much of the voltage drop shown would you contribute to the resistance created by the interface board and wires located between the end of the PSU's connector and the actual load. This isn't my favorite power supply, but I know the voltage regulation isn't nearly as bad as you have it graphed. And naturallly, since load also creates resistance, your voltage is going to decrease exponentially if you're not loading and measuring at the end of the connector. Since you're using a Chroma, I know you're not loading and measuring at the connector. Perhaps you could rig a DMM to measure voltages at the end of the connector instead of reporting what the Chroma is telling you since the Chroma is going to be incorrect since it doesn't take into consideration this added resistance.

2. Do you think it's possible to define "10%, 20%, etc." as it pertains to your loads. Naturally, you can't load all rails by 80%, 90%, 100%" because you'd easily exceed the power supplies capabilities. And certainly you're not loading the +12V with 80%, 90% 100% and then filling up the remainder with a 3.3V and 5V load because then you would crossload the PSU with an unusually high +12V load. The efficiency curve is nice, but there's no input/output wattage numbers. The graph under the curve looks pretty, but again we don't know the numbers. All we know is "thinner is better."

3. And, at risk of sounding like I'm beating a dead horse, ripple and noise results. We all know spec is 1%, and ripple and noise does typically increase with load. It would be nice to see if the power supplies being tested stay within that spec. through out it's total advertised capability.

4. Not so damn long! I know you try to get as much ad exposure as possible, but it totally effects the legibility... and I'm not talking about potential ADD or short attention spans! Navigation is difficult enough, but now we have graphed voltage drops that take up entire pages per rail. And then we do it twice between the 115V and 230V input. How about plotting the 115V on top of the 230V so we can actually see a direct corelation between the two input voltages and their effects on the output voltage? Not only do we knock out four pages, but we have something pretty cool to look at.

I added number four based on something Kyle said over at Hard about how he liked the additional graphing. I didn't agree. Graphing is nice, but it needs to be an informative and legible graph.

All and all, this is their first effort. Let's see what happens if and when they read and respond to my comment post in that last reviews. ;)

[Oklahoma Wolf]

you obviously didn`t read this whole thread...

Since this is the fourth (or fifth) time I've seen this post, and you apparently do not recognize that these types of posts constitute a violation of the forum rules, I'm giving you a three day vacation to read them.

[SKYMTL]

Looks like a response from the reviewer was given:

Thank you all for the feedback again. I think I can answer most of you with saying that we already work on the additonal tests. Some additional tests do need more time but will follow. PCP&C Silencer will follow next and it's well worth waiting for it.

[signmeuptoo]

Wow, Jonny, I was actually thinking most of those things, especially the graphs. At tech college they taught us some rules and ideals about graphs and how meaningless they can be when done incorrectly. Essentially, you can make a graph look like what you want people to think, without numbers and graphing done right, it's kind of a waste of time and effort.

Also: What good is any review without measuring noise and ripple, they are certainly as important as anything else on knowing how a supply is going to treat your build.

In tech college I got ticks taken off just a couple of my lab reports for using more words than necessary, and on others for not saying enough, and my professors were cruel, heh. And thank goodness they were, they made me appreciate the stuff and got me ready for how company CEOs and CFOs and others that aren't technical will react with reports one writes: You have to use the right language and the right amount of it. People zone out when being hit with wordiness. It's like two of my Number one web turn offs: Poor English from English speakers and people who create wordy posts and posts with monolithic paragraphs, and those who don't use the proper spaces between words, sentences, and all. If it is unpleasant to read and I am not profiting from it, it doesn't get my time.

I don't take people seriously who do this,and this.And so on. There is a reason why sentences are made a certain way, and a reason why schools teach you rules: So people won't get annoyed when listening to your arguments!

While I can see the value in a 50 page article on skin responesivity to various LASER output wavelengths in my old trade, I don't see it being useful having to read, say, 10 pages when trying to find the right product, as it can quickly add up to a lot of lost personal hours just trying to get to the point...

But, I am getting too wordy myself here. The important thing is that they are trying and if they work with Jonny and others, if there is a sense of teamwork among the computer tech sites, we all benefit immensely.

It sure is nice to see people feeling a loyalty though. :beer:

EDIT: Speaking of which: Jonny, one of my favorite forums has a spell check tool, any chance of getting that for here? I get annoyed by poor grammar but to tell the truth, I am not a great speller myself...

[Spectre]

Essentially, you can make a graph look like what you want people to think, without numbers and graphing done right, it's kind of a waste of time and effort.


Its like statistics reanalyze the data until you get the result you want.

[larrymoencurly]

Did AnandTech ever explain why their memory interference test gave the results that it did?

http://anandtech.com/casecooling/showdoc.aspx?i=1841&p=25

I've never seen anyone else do a similar test for PSUs.

[Kvar]

Wow. I think their equipment is overkill.

[Super Nade]

Did AnandTech ever explain why their memory interference test gave the results that it did?

http://anandtech.com/casecooling/showdoc.aspx?i=1841&p=25

I've never seen anyone else do a similar test for PSUs.

It was done to presumably check the 3.3V rail. However, there is no clear knowledge of how noisy the 3.3V rail can be (while still within the ATX spec). Also, they may be looking at EMI from the power source. Not sure how they made the connection though.