PSU review on Tom's hardware

PSUs: The source of DC power for all components in the PC & often a big noise source.

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee, Devonavar

Post Reply
geordie
Friend of SPCR
Posts: 175
Joined: Mon Jun 09, 2003 4:00 am
Location: Reading, UK

PSU review on Tom's hardware

Post by geordie » Mon Jun 09, 2003 4:07 am

Tom's hardware has a good psu review including some quiet models.
http://www6.tomshardware.com/howto/20030609/index.html

Harry Azol
Posts: 228
Joined: Fri May 02, 2003 9:21 am

Post by Harry Azol » Mon Jun 09, 2003 8:00 am

I don't trust Tom's.. their last psu comparison featured fake pictures of psu's on fire.. They are a bunch of drama queens

kamina
Posts: 147
Joined: Mon Jun 02, 2003 4:14 am
Location: Finland
Contact:

Post by kamina » Mon Jun 09, 2003 8:17 am

I'm not a very big fan of toms either. I think either his personal taste, or wallet plays too much role in his benchmarks. ( I hope it's his taste though)

MikeC
Site Admin
Posts: 12285
Joined: Sun Aug 11, 2002 3:26 pm
Location: Vancouver, BC, Canada
Contact:

Post by MikeC » Mon Jun 09, 2003 8:43 am

It's not a great review and it has several obvious flaws:

1) No distance cited for the noise measurements. If it was measured with the mic fairly close (maybe 1/2 meter or less, which I suspect), then the differences between units will be exaggerated.
EDIT -- I stand corrected; the distances was 1/2 meter -- see next post.

2) High emphasis on max power. I have never even heard of a PSU that shut down because a PC drew too much power from it. Cause instability, yes, for sure, but shut down? That whole part of the review seems irrelevant & useless for most users.

3) Rather than all the high power testing, it would have been much more useful to show variances on the various voltage lines at different power levels. This would show up potential actual-use instability issues. In fact, voltage variance was not reported at all.

4) Efficiency measured only at full power. It varies considerably depending on power load, some more than others; it can be as much as 65% vs 75%. What about at the much more typical 80-120W? Also was this on 120VAC or 240VAC? Many PSUs show 2-3% improvement in efficiency when measureed on 240 VAC vs 120 VAC operation.

5) The 78% efficiency cited for the Fotron 400 is simply wrong. They actually say it drew 571W from AC to deliver 397W DC. Do the math; the correct efficiency is 69.5%, which jibes with measurements I have made on the same PSU.
Last edited by MikeC on Fri Oct 31, 2003 2:18 pm, edited 2 times in total.

dukla2000
*Lifetime Patron*
Posts: 1465
Joined: Sun Mar 09, 2003 12:27 pm
Location: Reading.England.EU

Post by dukla2000 » Mon Jun 09, 2003 9:24 am

Yeah - I was just looking through it and it seems they have taken a step backwards since the previous review, rather than forwards. At least they do load the psus, unlike the folk who give some MBM readings, but I for one missed the rawdata on how much voltages sagged under load (only some anecdotal stuff) as well as even a definition of what the minimum load was they used for noise measurements etc.

So as before I took a browse past the German version (cause the reviewers are native Germans) and again quite a lot got lost in the translation. Not least the table explaining how an AMD XP2100 needs 355W of power - oh dear :shock:

Mike - the distance was 50cm
The Mediator 2238 measuring device from Brüel & Kjaer measures the noise generated by the fans at a distance of 50 centimeters.
I regularly drop past Tom's, some of their stuff is good, this is not. It's a shame, for example, they didn't make a better point about the psus that failed on overload and then would NOT restart - the last version of the ATX spec I read said
If the power supply latches into a shutdown state because of a fault condition on its outputs, the power supply shall return to normal operation only after the fault has been removed and the PS_ON# (or AC input) has been cycled OFF/ON with a minimum OFF time of 1 second.
Then again - at least they try. I know it is a lot easier to slag off someone else's work than to actually do it yourself. Just like much of the www, you have to use some savvy reading things and not fall for every word!

grnarrow
Friend of SPCR
Posts: 60
Joined: Fri May 30, 2003 3:48 pm
Location: FL, US

Post by grnarrow » Mon Jun 09, 2003 11:15 am

MikeC wrote: 4) Efficiency measured only at full power. It varies considerably depending on power load, some more than others; it can be as much as 65% vs 75%. What about at the much more typical 80-120W?
I thought this was the most disappointing thing about the review. I'd really like to get an efficient psu, but I'll probably never use anywhere near the max load of whatever model I end up getting, so it would be much more useful to know the efficiency when powering a typical low-end and high-end system.

SometimesWarrior
Patron of SPCR
Posts: 700
Joined: Thu Mar 13, 2003 2:38 pm
Location: California, US
Contact:

Post by SometimesWarrior » Mon Jun 09, 2003 12:09 pm

Wow, Tom must be getting the silent bug; half the article is spent talking about power supply noise!

Unfortunately, even with all his fancy testing equipment, he still hasn't learned how to measure the actual load generated by a fully-decked-out machine. I bet he still thinks most "p0w3r u$3r$" burn up 355 Watts in their "typical high-end computer."

Mirar
Posts: 86
Joined: Sun Aug 11, 2002 3:26 pm
Location: Falun, Sweden
Contact:

Post by Mirar » Tue Jun 10, 2003 12:18 am

It's too bad they didn't measure noise level 1) at normal load (120W?), 2) at normal load when all the generated effect as heat has to be evacuated by the PSU, 3) at max load when all the heat has to be evacuated by the PSU.

ATX specs say that the PSU fan should be enough to get rid of the heat generated. Obviously (fanless) all PSUs doesnt' do that, but it would be interesting to see if they can and how much noise they generate in that, absolutely worst, but very common case.

Also, I miss an oscilloscope test on the voltage lines to see the voltage ripple... That would really show how stable they are.

larrymoencurly
Posts: 170
Joined: Tue Dec 03, 2002 11:29 pm

Post by larrymoencurly » Tue Jun 10, 2003 10:27 pm

I don't trust Tom's.. their last psu comparison featured fake pictures of psu's on fire..

But there was no attempt to fool anybody that way. The picture was clearly labelled as a fake.

Much worse are all those websites that "test" PSUs by measuring nothing but voltages -- no numbers for amps, decibels, efficiency, or temperature.

MikeC
Site Admin
Posts: 12285
Joined: Sun Aug 11, 2002 3:26 pm
Location: Vancouver, BC, Canada
Contact:

Post by MikeC » Fri Jun 13, 2003 9:13 pm

Mmm... maybe there IS deliberate funny biz going on at THG... See this thread: 78% efficiency for Fortron is WRONG!

dukla2000
*Lifetime Patron*
Posts: 1465
Joined: Sun Mar 09, 2003 12:27 pm
Location: Reading.England.EU

Post by dukla2000 » Sat Jun 14, 2003 12:54 am

Hi Mike - dont take the lack of (official) response to your posts on Tom's personally: again just checked the German language forums which are similarly moribund and the thread about the psu article also has no contribution from the authors. (Hasn't picked up the efficiency issue, does have 2 posts about the 355W requirement table which was included in the German review.) Perhaps we should mail the authors/Tom direct to try wake them up?

MikeC
Site Admin
Posts: 12285
Joined: Sun Aug 11, 2002 3:26 pm
Location: Vancouver, BC, Canada
Contact:

Post by MikeC » Sat Jun 14, 2003 7:13 am

Naw, no offense taken.

Years ago, when Tom seemed to have a ruckus with Intel "hawks" threatening to shut his site down via web attacks over a negative article he wrote about the then new slot P-2, I actually wrote an article supporting him and had it published in a local Vancouver paper. I was one of many Tom supporters all around the world.

Now I look at the utter commericialism of THG and back at that whole so-called incident and wonder if it wasn't entirely manufactured by Tom himself, because Intel's actions were never seen directly, only reported by Tom. Whether he took some trivial difference of opinion and blew it out of proportion... in the hopes of getting more traffic to his site -- which did happen big time as a result. It put his site on the map for lots of folks.

Yikes, am I getting cynical or what! :?

Zhentar
Patron of SPCR
Posts: 918
Joined: Thu Mar 27, 2003 4:00 pm
Location: Madison, WI, USA

Post by Zhentar » Sat Jun 14, 2003 11:09 am

well; I'll just say this: those power requirements are ridiculous. The hard drive power draw is utter crap; his power numbers are approximately double that listed on my maxtor drive, and how many people really run a 4 drive RAID? so lets call that 15 watts, not 112, okay? and that CPU number seems high too, unless he means a pally core and not tbred.

MikeC
Site Admin
Posts: 12285
Joined: Sun Aug 11, 2002 3:26 pm
Location: Vancouver, BC, Canada
Contact:

Post by MikeC » Thu Jun 19, 2003 6:46 pm

Well, I am really disappointed. Thom's Hardware seems to have ZERO editorial responsibility. Here is the latest post I made at THG --
You guys still don't seem to get it: This "error" might not be an error, it could well be a case of dubious marketing by RHG to fulfill some kind of marketing contract obligation to a client.

Client says: "Here's $20,000, Get me tons of positive PR and traffic to my website."

THG deliberately says this PSU's efficiency is 78% when it really is only 69.5%, and bases their high level coverage and recommendation based on this false info. In the span of a week or two, millions of people scan the headlines and say "oh cool, that's must be good" and never come back to the article again. They will never know that THG has effectively lied to them.

If caught, THG say "Oh so sorry for the mistake!" and changes the text. If THG issues an error statement it is hardly likely to reach more than a few. In the meanwhile, millions of people have the impression that the Fortron has 78% efficiency and because of that it is a great product THG recommends. The promotion is done, the deal is complete.

Don't you get it guys? It looks like a SCAM! The fact that they have not responded to this thread, not changed the "error", nor respond to my emails to the entire editorial team that did the review suggest even more strongly that it is just that -- a SCAM.
So what do you think? Am I way off base here or should this thread be entitled "Who is Tom?!"

dukla2000
*Lifetime Patron*
Posts: 1465
Joined: Sun Mar 09, 2003 12:27 pm
Location: Reading.England.EU

Post by dukla2000 » Fri Jun 20, 2003 12:49 am

MikeC wrote:So what do you think?
Not to follow the thread on their discussion board is sloppy.

Not to respond to your emails, and to fail to correct (or defend - not sure how) the 78% efficiency shows a disregard for any of the principles of journalism. So yup, we have to assume their motives are unrelated to journalism, and looking at the prominence they get at Fortron-Source and MSI homepages it seems there may well be money involved.

I guess we are the police force, judge & jury - I am not sure how best to prosecute things further. At least we can make a loud statement here (SPCR) about the integrity (or otherwise) of other reviews out there.

aphonos
*Lifetime Patron*
Posts: 954
Joined: Thu Jan 30, 2003 1:28 pm
Location: Tennessee, USA

AMDMB not happy with THG either

Post by aphonos » Mon Jun 30, 2003 8:57 am

AMDMB is apparently not too happy with THG either. Editorial: Online Media Morals

And this...
[H]ardOCP wrote:[H]ardNews 3rd Edition

Monday June 30, 2003
Posted by Kyle 3:33 AM (CDT)

More Hardware Hubbub:
For those needing a daily dose of drama, read on. I returned home on Sunday to find that Steve has posted a link to this editorial at AMDMB.com. Ryan Shrout, owner of AMDMB, goes into some detail about his recent experiences at a LAN event in which he was denied access to some publicly displayed technology from AMD.

I was told by a representative of AMD that I was not going to be allowed to even enter the building because Tom’s Hardware (THG) had forced a media blackout for all other online press. That means that as a member of the press, I was not allowed to enter the building, take photos or write about the hardware events that would occur inside.

I had made my mind up to keep my yapper shut on this topic as some of the issues were worked out after Ryan went public with the situation, then I read this mail from Omid Rahmat of THG at the InqWell.

There really does not seem much sense having to deal with this sort of thing, although it has to be said that I am sick and tired of fanboy sites that do nothing but slam everyone for their own edification. Soapbox amateurs who just want to get a few rants up on a page, and paint themselves as gurus. What has it got to do with hardware and technology? Nothing. They're Tom wannabes, or Anand wannabes, or Kyle wannabes, or even Mike wannabes. I would hope they aspire to something greater, but the nature of society today dictates that we all have a grasp that exceeds our reach.

This would of course seem to be obviously aimed squarely at Ryan Shrout of AMDMB.com. I do know Ryan Shrout and have had many conversations with him and he simply does not deserve the above statements. I have shared dinner tables with him as well a cab ride or two. He is a very intelligent young man with a good head on his shoulders. He knows right from wrong, and is one of the few people in the industry that I will vouch for and have done so. I think that if Ryan posted his experiences, that is basically the way it happened, no matter how much mud gets slung his way. AMDMB.com is an asset to our community and I am proud to call them a peer.

"Soapbox amateurs" are very much in part what makes our hardware community such a great place. Keep fighting the good fight Ryan. I will put my own amateur soapbox up now and let you get back to a bit of tech news.
Sorry to post the whole text from [H] rather than a link, but their news tends to change pretty quick and their archives are difficult to search.

dukla2000
*Lifetime Patron*
Posts: 1465
Joined: Sun Mar 09, 2003 12:27 pm
Location: Reading.England.EU

Re: AMDMB not happy with THG either

Post by dukla2000 » Mon Jun 30, 2003 4:57 pm

aphonos wrote:AMDMB is apparently not too happy with THG either. Editorial: Online Media Morals
OK - I voted with my feet and deleted THG from my favourites.

aphonos
*Lifetime Patron*
Posts: 954
Joined: Thu Jan 30, 2003 1:28 pm
Location: Tennessee, USA

Post by aphonos » Mon Jun 30, 2003 6:00 pm

The plot thickens....
[H]ardOCP wrote:H]ardNews 7th Edition

Monday June 30, 2003
Posted by Kyle 12:37 PM (CDT)


THG to Sue AMDMB.com?
As mentioned earlier, this editorial posted at AMDMB, chronicles Ryan Shrout's experience at a rather large LAN party this weekend. Here in the USA, we understand it to be our right to openly speak about our experiences in life, but quite possibly others do not feel that way.

Omid Rahmat of Tom's Guides Publishing, a.k.a TomsHardware.com, verified in an email to me that he made the following statements in reply to a reader's email explaining that he would not visit THG any more due to their recent actions.

-----Original Message-----
From: Omid Rahmat [mailto:[email protected]]
Sent: Monday, June 30, 2003 8:41 AM
To: Name Removed

Subject: RE: MML2

Sorry to hear that, but Amdmb.com fabricated a story, and we gave as good as we got. We'll let the lawyers handle it.

If you choose to believe the amdmb.com version, fine, but we're not going to just sit there and have someone lie about us.

Omid Rahmat
Tom's Guides Publishing
www.tomshardware.com


Then also in another email:

It's bull, and no rebuttal required. We'll take them to court instead. They might get the message that way.

Omid Rahmat
Tom's Guides Publishing
www.tomshardware.com


It seems as though that is exactly what they intend to do as Omid called Ryan Shrout this morning around 11:45AM CST and verbalized that Tom's Guide Publishing would be having their lawyers send Ryan a letter explaining that they would be suing them for libel. Ryan, being a "poor college student" working on his EE degree does not have the funds needed to defend an action of this sort, it is our opinion that this action is to force Ryan's hand and make him take down his page that chronicles the events.

I have committed to Ryan that I will personally help him with funding his legal council so he will not have to fold with the almighty dollar being the deciding factor. We have also put him in direct contact with our attorney we keep on retainer that specializes in Intellectual Property matters. I think Ryan is telling the truth and should not be stifled.

The fact of the matter is that Omid was not present at the event and does not know what happened first hand. I would guess that Omid might be getting his information from parties that are now concerned with keeping their jobs. Also worth mentioning is that Ryan has some recorded audio of THG and AMD representatives saying some very interesting things to him that would be terribly embarrassing, if not downright damaging, should they be released publicly.

We have tried to contact AMD about this, this morning, but have not been successful.

As of posting this, I have received another mail from Omid making the following statement:

We will deal with Mr. Shrout through the appropriate legal channels, and leave it at that
And thickens....
[HardOPC wrote:[H]ardNews 9th Edition

Monday June 30, 2003
Posted by Kyle 5:01 PM (CDT)


THG Backpedaling on Lawsuit?
Omid just posted a very interesting "Hard News" blurb at the THG site that makes the statement below:

14:50 EST [Omid]
One more time - last word on MML2

Now that HardOCP has decided to get in on the action, it's time to end this farce with regards to a media blackout at MML2.

Below is the actual email I sent the Webmaster at HardOCP, a fact that he neglected to mention, choosing to splice it for his own nefarious purposes.

It's exasperating to deal with this sort of thing, and perhaps the lawyers will make it clear that it is a waste of our time to deal with this sort of baseless rubbish.


I am still a bit confused as how an email to me ends anything when the issue is AMDMB.com, but I do have to agree with his assessment of the situation as a “farce”. THG needs to be making nice with AMDMB to “end” this all instead of threatening lawsuits and trying to dilute the real issue.

Clearly I think the last paragraph above shows that THG is weighing their options. I would have to think that suing AMDMB for telling their story would be a PR nightmare for THG, not to mention a blow to the hardware community overall. This ugly situation is obviously not good for anyone involved. Throw into the equation that THG now knows Ryan at AMDMB is not going to roll over a play dead as legal counsel now represents him and that certainly might be a bump in the road that was not expected by THG.

Sadly THG still does not see the light. This is about AMDMB’s right to post the information they want to post and that is what needs to be focused on. It is not about [H], or even THG or media blackouts or MML. It is about the right that AMDMB has in the United States of America to post their version of the events that happened. Do not let the spin-meisters make you think it is about something else as Omid does above. I believe Ryan’s post to be a fair account of what Ryan experienced this weekend. It is time THG extended some of that “clear cut professional manner” they speak of to AMDMB.com and retract the legal threats. Will THG swallow their pride and arrogance and move back from what is now an obviously disadvantaged political and legal position, or will THG dive head first into a pool of muck so deep that there is no way to emerge from it untainted?
Again, sorry to post full text. Please keep up with the rest of it via hardocp.com, amdmb.com, and THG. :roll: And, no, I'm in no way associated with any of the above-mentioned sites.

Roger Beckett
Posts: 37
Joined: Thu May 08, 2003 2:32 pm
Location: Maine

Go back to school

Post by Roger Beckett » Mon Jul 07, 2003 7:10 am

Efficiency isn't quite as simple as dividing a couple of numbers, especially when you're going from AC to DC. Simple division works fairly well in High School Physics class but anything beyond that, like the real world, things get a bit trickier.

I especially like the testing/reviews done on an un-named site wherein a 100 watt lightbulb is used for heat generation to "simulate" heat generated by the intra-case components of a computer.

How many BTU's of heat energy does a "typical" 100 watt bulb produce? Depends on efficiency obviously and shouldn't be all that hard to calculate.

How many BTU's does any give cpu or chipset produce?

I have a really hard time giving credibility to any testing that ignores, or doesn't even bother finding, such answers, especially as they aren't all that hard to find.

I am becoming increasingly frustrated at the house of cards various "review" sites build upon the foundation of arbitrary assumptions. Why doesn't anyone do the basics?

Granted thermodynamics isn't the easiest underclass engineering course one can take, but it ain't fourier analysis either.

For Anyone whom just had their toes stamped on, I'll apologise up front and assume that this lack of attention to basics is due to time constraints rather than lack of knowledge or education. :)

Tom P
Posts: 104
Joined: Wed Nov 27, 2002 2:35 pm
Location: San Diego County

Post by Tom P » Mon Jul 07, 2003 1:54 pm

Looks like somebody just volunteered to help out on power supply reviews!

Roger Beckett
Posts: 37
Joined: Thu May 08, 2003 2:32 pm
Location: Maine

Not enough money in it

Post by Roger Beckett » Mon Jul 07, 2003 3:25 pm

I quit doing reliability engineering some time ago, it's got to be the most boring job known to man. So boring that even in the early 90's I was getting paid $30 an hour to make great huge spreadsheets of electronic components on the engines of apache helicopters and F-18 aircraft.

If there is a more mind numbing job I do not know what it might be, paid well though.

But, I wouldn't want to do something so boring for money, let alone as a volunteer. I suspect that's why most reviews are full of simplistic tests and short on theory and calculations. Soldering and wiring are easy, math, even with spreadsheets, can be hard.

cliche
Posts: 150
Joined: Sun Apr 20, 2003 9:30 pm
Location: uk

Post by cliche » Mon Jul 07, 2003 3:27 pm

Tom P wrote:Looks like somebody just volunteered to help out on power supply reviews!

lol :lol:

IceFire
Friend of SPCR
Posts: 109
Joined: Thu Jan 09, 2003 12:00 pm
Location: USA

Post by IceFire » Tue Jul 08, 2003 7:21 am

I really think Tom (or who ever is running the show) at THG doesn't read anything and just site behind his desk. A good example is the responds (did a search on THG and couldn't find the responds anymore. It was display clearly on their main page last week and now I couldn't even find anything relating to it.) he gave to amdmb.com's comment about the lan party. In his response to Steve's editorial, Tom said that Steve claim not to be able to get into the show/lan part and that they (THG) had pictures/clips (forgot what it was) proving that Steve was there. Will Steve stated clearly that he got in after calling around and spoke to various people.

THG said they were not contacted or any of the sponsors. Hello! Steve again stated clearly that he called AMD heads and they said they have no idea that this was going on. Steve also contacted other sponsors and MML2.

What's even more funny is that Steve stated that "Even more interesting, was the fact that the THG rep that was responsible for the event did not get permission from his boss at THG to do so." So he wasn't blaming THG but the THG rep there.

THG didn't have to make such a big deal out of it. I think MikeC is right that they just want more traffic - good or bad.

Now I know why I don't like visiting THG anymore.

MikeC
Site Admin
Posts: 12285
Joined: Sun Aug 11, 2002 3:26 pm
Location: Vancouver, BC, Canada
Contact:

Post by MikeC » Tue Jul 08, 2003 7:43 am

yeah, commercialism.... :roll:

THG is such a visual mess these days, there are so many distractions it's hard to stay focused on what you're trying to read.

I received a news submission from shiva about a QDI board that stops the CPU fan when it's not needed -- in a massive P4 875/865 motherboard review at THG. I am reluctant to post the news link because I don't know that any of that info is really reliable. I suppose if you really dig through it, you might be able to find some info relevant to quieting / cool operation... THG calls itmobo tsunami.

Liquidated
Posts: 110
Joined: Fri Apr 11, 2003 12:53 am

*shudder*

Post by Liquidated » Wed Jul 09, 2003 1:15 am

Wow I just read that thread and the main 3 guys posting are just not there.

Seemed all they were responding for was to up their post count.

That's a scay thread.

seems everyone ignored you mike, even the people posting that thread.

Cheers!
-Liq

Post Reply