Better Power Supply Roundup

What's the deal on the Antec 330W and P4? by Anonymous Coward · 2003-08-11 00:47 · Score: 3, Insightful

The dude writes that the 204W at 12V isn't enough for a P4 for he writes that ONLY for the Antec 330 and its 204W is more than most of the other PSes put out on that rail. It's crap like that that make the WHOLE thing suspect. .

Re:What's the deal on the Antec 330W and P4? by drwtsn32 · 2003-08-11 01:14 · Score: 4, Informative

Agreed. IMO Antec makes some of the best power supplies. I use several Antec 330W ones at work in P4 systems.

PC Power and Cooling is very good, but their supplies are more expensive than Antec and are usually noisier.
Re:What's the deal on the Antec 330W and P4? by cowbutt · 2003-08-11 02:32 · Score: 1

...especially seeing as I've been successfully running a P4 2.40B/i845PE/Radeon 7500/RAID/DVD/CDRW/512M machine for nearly a year off of a 'P4 Compatible' PSU that only offers 180W on the +12V rail.
Nothing overclocked, I grant you, but...
--
Re:What's the deal on the Antec 330W and P4? by dietpepsiman · 2003-08-11 07:34 · Score: 1

Dell uses a 220W PS on their Poweredge 600SC which has a P4, and 4 drive SCSI RAID. That's up to 40W on the 12v just for drives during load when striping.

Shuttle SFF boxes use a 170W PS for their P4 model, which various people have run with a Geforce 4 and 7.2k drive.

He seems to think that the 216W rail on the Thermaltake is "Plenty of Muscle for an AMD or Intel system", does 12 more watts make the man?
Re:What's the deal on the Antec 330W and P4? by Anonymous Coward · 2003-08-13 00:11 · Score: 0

OK, this is gonna sound stupid, but I've been trying for ages to find a replacement 850 W (or more) ATX PSU. Anyone any ideas where to find such a goodie? Need to be able to get at least a continuous 26 A out of the 12 V line...
Suggestions welcome.

Forgotten Power Supplies... by MosesJones · 2003-08-11 00:49 · Score: 5, Funny

DOH! NOW I know why my home build PC isn't working. I also made the "often made" mistake of forgetting the Power Supply.

I'm glad to know this is a common mistake and not just me being stupid.

--
An Eye for an Eye will make the whole world blind - Gandhi

Re:Forgotten Power Supplies... by Anonymous Coward · 2003-08-11 01:02 · Score: 0

I'm glad to know this is a common mistake and not just me being stupid.

No reason it can't be both.
Re:Forgotten Power Supplies... by horcy · 2003-08-11 02:20 · Score: 1

MUAHAHAHAHAHAHA

ROFLMAO

--
Check my site: http://pixel.pagina.nl
Re:Forgotten Power Supplies... by wukie · 2003-08-11 02:55 · Score: 3, Interesting

When I built my PC, the power supply was the first thing I considered.

I was looking at building a Dual Athlon system on a Tyan motherboard with a few hard drives.

Essentially to cut a long story short, this combination requires a VERY HIGH power rating on the 5 volt rail. Some 550W power supply don't cut it, Enermax units below 400W were uncertain, while Antec's 350W or better are fine.

I went for an Antec 430W and their server case. 8 fans all attached to the "special" fan rail which adjusts their speed according to conditions keeps the computer humming.

Corsair ECC memory, HD's with 3 year warranty help with stability aswell.

Bottom line: Win2k - never had a BSOD (Blue Screen).Linux is obviously rock steady aswell.

I would have considered an Asus or Gigabyte motherboard. Kingston, Micron or Infineon RAM. Chieftec or Thermaltake case. BUT the power supply had to a very good one ... i.e. ANTEC!
Re:Forgotten Power Supplies... by Jucius+Maximus · 2003-08-11 04:57 · Score: 1

"DOH! NOW I know why my home build PC isn't working. I also made the "often made" mistake of forgetting the Power Supply."
I actually have 'solved' problems on IRC help channels where the person said that their computer would not power up no matter what they did. The solution of course was to flip the power switch on the back of the PSU ;-)
Re:Forgotten Power Supplies... by dago · 2003-08-11 06:10 · Score: 1

shit, I have installed a server with 2 AMD, 4 HDD and an enermax 350W...

you mean it won't work ?

(uptime is now around 150 days)

--
#include "coucou.h"
Re:Forgotten Power Supplies... by Anonymous Coward · 2003-08-11 06:14 · Score: 0

How were they able to get to an IRC channel if their PC wouldn't power up??
Re:Forgotten Power Supplies... by RandomCoil · 2003-08-11 07:20 · Score: 1

Looks like we're both screwed -- I have a workstation with a Tyan MB and dual 1800MPs running off some generic 350W PS. I only have two HDs, though. My uptime's only 12 days due to a kernel update. The last crash was, oh, who knows...

RC
Re:Forgotten Power Supplies... by wukie · 2003-08-11 16:20 · Score: 1

Enermax units below 400W were uncertain

NO I mean they are UNCERTAIN!

Did you read past the words Enermax units below 400W

Even 250W Enermax powersupplies MIGHT work without a hiccup. The reason that brand was mentioned is because it's on par with Antec in quality and performance. If you don't make it to 2 years, check the 5V rail. All the best!

How much power is drawn from the wall? by Anonymous Coward · 2003-08-11 00:53 · Score: 2, Interesting

OK I'm no power supply geek, or electrician type either, but how much power is typically drawn from the wall in a pretty basic machine with a 400w power supply? A friend of mine is constantly being taken offline by his parents because they only let him use the computer for 2 hours a day due to power usage, and they're going on the wattage of the PSU in the machine he built. Personally. I think his machine (I think it's an 1800+ with one 7200RPM drive and a cd burner) would barely be using a third of that power.

Re:How much power is drawn from the wall? by Anonymous Coward · 2003-08-11 01:06 · Score: 0

400 watts is 400 watts no matter which way you look at it. The only way to save any power is to get a lower wattage power supply, perhaos a 250 would fulfill your friends needs.
Re:How much power is drawn from the wall? by wirelessbuzzers · 2003-08-11 01:10 · Score: 5, Informative

No, that's a maximum rating. It doesn't draw nearly 400W if you aren't going to use all that, otherwise it would have to dissipate 200W on its own, which it certainly cannot.

--
I hereby place the above post in the public domain.
Re:How much power is drawn from the wall? by henele · 2003-08-11 01:25 · Score: 1

I'd of thought between and third and a half would be a reasonable guesstimate, but you've also got to remember they aren't 100% efficient, so you have to add on some wattage to acccount for that...

There should be some device he could pick up at a local electornics store that could measure the amount of power the machine is using right?
Re:How much power is drawn from the wall? by Detritus · 2003-08-11 01:28 · Score: 4, Informative

Your friend's parents are mistaken. Just because the computer has a 400 Watt power supply doesn't mean that it is actually supplying that much power. It is probably considerably less. Think of it like the engine in a car. It may be capable of 150 horsepower but the average power output is much less.

--
Mea navis aericumbens anguillis abundat
Re:How much power is drawn from the wall? by steelerguy · 2003-08-11 01:42 · Score: 1

Have your friends parents read this...then they can start taking his graphics card away after 2 hours instead.

http://computer.howstuffworks.com/power-supply.h tm
Re:How much power is drawn from the wall? by HidingMyName · 2003-08-11 01:44 · Score: 3, Informative

The best way to measure is probably experimentally. You could use a "Kill a Watt" device, that has a combination of a male and female plug with an output for indicating how much power is being drawn. You can get one for $40 U.S. or less.
Re:How much power is drawn from the wall? by Tragedy4u · 2003-08-11 02:51 · Score: 2, Informative

I remember my boss having power concern issues with our equipment at our co-lo facility. We had a bunch of low end RS\6000 kit that had 800W power supplies in them, thats almost 7 amps which is half the usage of most vaccum cleaners! It didn't seem right, these were older machines and not very powerful they couldn't possibly use that much power, so we grabbed a multi-meter to check the usage. On boot up (the most expected), the usage spiked up to just above 1amp (filling the capacitors I imagine) then dropped quickly down to about .5 amps thats the equivalent usage of an average light bulb 60Watts.

Unless your friend is running a server with lots of SCSI hardware and doing very CPU and disk intensive activity all the time, I'd imagine that his power usage would sit around that ammount.
Re:How much power is drawn from the wall? by Skater · 2003-08-11 03:02 · Score: 1

That's silly. In my condo, the power bill shows exactly how much we spend on heating/cooling and how much we spend on everything else.

In my 1-bedroom condo, my non-heating/cooling bill is about $17/month. That includes TWO computers running full time (one with a 400W supply: an Athlon XP 1900+; the other is a Pentium 133 used as an MP3 player), with a third computer that I frequently leave run on weekends. It also includes my stove, refrigerator, lights, television, DVR, fish tank with heater and filter, dishwasher, i.e. anything that's not heat or a/c. (I don't have a washer or dryer.)

My conclusion: computers are cheap to run. If you're going to use it on anything approaching a regular basis, let 'em run, because the expenses incurred when you aren't using it is minimal. (Now, for the computer I only use on the weekends, that makes less sense to run continuously.)

--RJ
Re:How much power is drawn from the wall? by evilviper · 2003-08-11 03:07 · Score: 3, Informative

There should be some device he could pick up at a local electornics store that could measure the amount of power the machine is using right?

Search froogle for "Kill A Watt". I've got one and it works surprisingly well. Good luck finding it at any stores nearby. It's even a bit of a rarety online. (Check solar or radio shops)

As I posted in a reply to his comment, my XP 2000+ is using up almost exactly 100watts, obviously that may vary a bit, but not much. Even an old 166MHz system of mine is using up about 60 watts. Interesting not though... Notebooks are FAR, FAR more power effecient. My 1.2GHz PIII Notebook with 10+" screen is only using up 19watts normally, and up to 32watts at moments with intensive processing or harddrive reads.

--
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
Re:How much power is drawn from the wall? by evilviper · 2003-08-11 03:12 · Score: 1

That's silly.

I agree.

In my 1-bedroom condo, my non-heating/cooling bill is about $17/month

I'm not sure what you mean by a "not-heating/cooling bill", but you obviously don't live in California, where rates are still rather high. Not to say that it's not silly...

--
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
Re:How much power is drawn from the wall? by evilviper · 2003-08-11 03:19 · Score: 2, Insightful

how much power is typically drawn from the wall in a pretty basic machine with a 400w power supply?

I have no idea, it could be from 1-400watts depending on the system.

A friend of mine is constantly being taken offline by his parents because they only let him use the computer for 2 hours a day due to power usage, and they're going on the wattage of the PSU in the machine he built.

Your friend's parents are idiots. Even if it was drawing 400watts, that could be recouped by just shutting a handful of lights off. 400 watts mean 4x100watt lightbulbs, or 7x60watt lightbulbs.

Better yet, they should try unplugging any devices that they aren't using, since most devices that are turned-off are still drawing 2-4 watts a piece 24 hours a day. Unplug 4 devices would recoup your friend's 2 hours of 'computer time'.

I think his machine (I think it's an 1800+ with one 7200RPM drive and a cd burner) would barely be using a third of that power.

Good guess. I happen to have tested my XP 2000+ (with a "Kill a Watt"), and it's drawing almost exactly 100watts.

Although, in addition to the computer, my 19" CRT monitor uses just under 60watts, so things begin to add-up. Maybe your friend can use power usage as an excuse to get a LCD, or even better, a notebook, which would use up about 25-35watts.

--
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
Re:How much power is drawn from the wall? by i8urtaco · 2003-08-11 03:23 · Score: 1

Actually, the computer only takes from the wattage what it needs. The average computer only draws about 90 watts at a given time, unless of course if they have dual processors, 2+ HDDs, etc. Plus, if he doesn't use an LCD, this would be a good time to try to get his parents to buy him one; these things take up a LOT less wattage than a CRT.
Re:How much power is drawn from the wall? by Skater · 2003-08-11 03:46 · Score: 1

Right, I live in Virginia. By "non-heating/cooling", I meant the part of my bill that shows the power I use ignoring the heaters/air conditioners.

--RJ
Re:How much power is drawn from the wall? by Alan+Partridge · 2003-08-11 04:05 · Score: 1

Funnily enough, they DON'T.

They should, but they don't. it's not unusual to see a TFT display with a 60W rating these days.

--
That was classic intercourse!
Re:How much power is drawn from the wall? by mt2mb4me · 2003-08-11 05:16 · Score: 1

there is , it is a little bid more difficult to use and you have to do some math, but get an ammeter they are these nifty looking things that have a big loop, and it measures amperage, the only thing is you will have to scrap a power cord to use it, so you can pull the bloack wire out of the cord sheatings and put it though the loop, it will give you an amp reading, multiply by 120v (if it is standard USA house voltage) and that will be your watts. Difficult, but effective
Re:How much power is drawn from the wall? by zakezuke · 2003-08-11 09:01 · Score: 1

"...constantly being taken offline by his parents ..... due to power useage"

Power supply is rated for up to 400watts of consumption. If your PC was actually consuming 400watts of power, you'd feel the heat. I have some old sun equipment that actually does consume that much power, and it did a nice job of heating the computer room in the winter.

If the parents are concerned about power consumption, then turn off a freaking light! Chances are your PC power consumption is roughly equal to 1 or 2 100watt light bulbs at most.

If they can see their power meter... it would be rather easy to note the kw/hr consumption with the computer, and without the computer.

--
There is no sanctuary. There is no sanctuary. SHUT UP! There is no shut up. There is no shut up.

memory faults by brejc8 · 2003-08-11 00:54 · Score: 5, Interesting

The results of the memory tests are amasing. The MTBF is about an hour on some of the power supplys. I'm not sure If I understand the setup but that is appauling. I expect a MTBF of about 100+ years not an hour.

--
Mouse powered Chips, Open source Processors and Lego

Re:memory faults by shoppa · 2003-08-11 00:58 · Score: 1

I agree completely. I run my machines for weeks with memtest86 before making them operational. And while I have found memory and motherboard problems, even with the most el-cheapo power supply it's either success or failure. This sort of reliably intermittent memory test failure that Anandtech regards as "normal" puzzles me to no end. Maybe it's just part of overclocker attitude?
Re:memory faults by dnoyeb · 2003-08-11 01:08 · Score: 2, Informative

Weeks? really? Usually electronics fails in the first 24 hours, or its good to go. Especially where there are no moving parts.

I dont think the added time is a proper measure of increased reliability.

anyway, PSUs are overrated. I have used elchepos to the top quality units, and found differences in voltages, but not performance.

I buy for sound now. But of course I dont run any high performance / demand servers either.
Re:memory faults by simong_oz · 2003-08-11 02:21 · Score: 1

The results of the memory tests are amasing.

Really??? 3 replicates of every test on a single sample. I haven't done the maths, but looking at the numbers I reckon some proper statistics on those memory test numbers would show up very little, if any at all, statistically significant difference.

--
"Because it's there." - George Mallory, when asked why he wanted to climb Mt Everest, March 18, 1923 (New York Times)
Re:memory faults by brejc8 · 2003-08-11 03:11 · Score: 1

Whats amasing is the fact that there errors are appearsing at all reather than one power supply being better than another.
These errors sould not appear in a system no matter how poor your supply is

--
Mouse powered Chips, Open source Processors and Lego
Re:memory faults by Jeff+DeMaagd · 2003-08-11 04:54 · Score: 1

I don't think that a memory flip in the RAM means that the PSU failed.

I dunno, the MTBF in RAM for any particular bit may be 100+ years, but once you apply such statistics on a system with multigigabite RAM, I wouldn't be too surprised to find a bit flip in a days's time.

I would like more information on the test system, RAM brand, etc.
Re:memory faults by brejc8 · 2003-08-11 06:21 · Score: 1

I did mean 100years+ for the whole memory.
There is no way you would accept your computer crashing every hour. A bit change in the program memory will cause wrong behaviour and most of the time crash the application. If its in the data than it wont crash but it will corrupt a lot of important info.
No way a computer would be useful with that sort of failure happening every hour. This would need heavy error correction.

--
Mouse powered Chips, Open source Processors and Lego
Re:memory faults by Anonymous Coward · 2003-08-11 21:20 · Score: 0

The results of the memory tests are amasing. The MTBF is about an hour on some of the power supplys. I'm not sure If I understand the setup but that is appauling.

As appauling as you'r speling?

Re:Power Supplys by jpmkm · 2003-08-11 00:56 · Score: 4, Informative

I'm thinking that's a UPS. Servers usually have redundant power supplies, not one for multiple machines.

Re:Power Supplys by Anonymous Coward · 2003-08-11 00:56 · Score: 0

you have absolutely no idea what you are talking about. I know this is a novel concept, but read the article before you open your flaps and lets out all that stinky hot air.....

Other types of power supplies? by Anonymous Coward · 2003-08-11 00:57 · Score: 2, Interesting

I would like to see a review of all of the tiny power supplies out there. Espically all of the P4 compliant ones. I think alot of PCs will soon not look like PCs at all and they will still need power.

Re:Other types of power supplies? by name773 · 2003-08-11 03:54 · Score: 1

these don't look like computers

Re:Power Supplys by Anonymous Coward · 2003-08-11 00:57 · Score: 1, Informative

That's called a UPS. Not the same thing.

Re:Dupe by twoslice · 2003-08-11 00:58 · Score: 1

Your were duped dude, have some patience, the dupe story will eventually be posted...

--

From excellent karma to terible karma with a single +5 funny post...

Forgotten? by j4ck50n · 2003-08-11 00:58 · Score: 1

"Power supplies tend to be one of those forgotten, but ever so important parts of machine construction..."

I don't think forgotten, unless you completely have missed the DIY/MOD craze of the past 3 years or so. Nothing has escaped the scrutinizing eye of the geek...and with all those AMD's and GPU's?

Re:Forgotten? by wukie · 2003-08-11 03:11 · Score: 1

I disagree. Most modders care about how the PC looks. They would rather spent $100 extra on an aluminium case with a window and save $50 on the power supply which is not seen.

Let's not even start on the memory ...
Re:Forgotten? by Jucius+Maximus · 2003-08-11 07:36 · Score: 1

" I disagree. Most modders care about how the PC looks. They would rather spent $100 extra on an aluminium case with a window and save $50 on the power supply which is not seen."
When your cheap-ass PSU dies and shorts on the DC side, you'll have a lovely front row view of the fireworks when your HDs go up in smoke ;-)
But seriously, I despise case windows. They leak light out of the case and who knows what kind of EM radiation is escaping. Aluminium is good, yes. I just bought a Lian Li Aluminium case recently, but I will leave the case windows for other people.

Not quite as amusing as dansdata.com, but... by Junior+J.+Junior+III · 2003-08-11 00:59 · Score: 5, Interesting

I really liked Anandtech's article. It actually had a lot of information and was fairly thorough -- measuring performance in a variety of ways, including not just output wattages but also noise levels, heat buildup, and cost.

A bit more explaining the basics of what each different voltage rail is for and why x-level of performance is important would have been helpful. Along with some more basic stats, such as how long the power cables actually are -- surely people still build full-tower PCs, don't they?

--
You see? You see? Your stupid minds! Stupid! Stupid!

Re:Power Supplys by Anonymous Coward · 2003-08-11 01:00 · Score: 1, Funny

I'm thinking that's a UPS.

not necessarily - the guy went on to say:
"If you do run a computer that is somewhat important run a power supply you wont be
disapointed!".

I consider my desktop computer at home 'somewhat important', so I put a power supply in that (of the 300W ATX variety) and I must say, I'm not disappointed that I did. If I hadn't put a power supply in it probably wouldn't run nearly as well as it does now.

External supplies by Anonymous Coward · 2003-08-11 01:02 · Score: 2, Interesting

All power supplies that I know of (not that much by the way) and all supplies tested are internal supplies. Since the powersupply is a major contributer to the internal heat of the box, and noise generation, wouldn't it make sence to make the power-supply external. It will decrease the temperature of the box, and by having it external you can place it at such a location where its noise is of least annoyance. You can even make them bigger, having larger (= less noisy) fans, and you could share them between boxes. What do you think?

Re:External supplies by ZorinLynx · 2003-08-11 01:32 · Score: 4, Informative

No, this would be a bad idea. The cables from the power supply to the system would be carrying low voltages, and to provide power to the system at those voltages would require high current. This would cause voltage drop (and thus loss) in the cables.

When wiring up electrical systems, it is important to remember that your long runs should always be done with high voltage cable. For instance, from the wall to the power supply is 120V AC. Once the voltage is reduced, the runs need to be kept as short as possible, since every foot of cable loses substantially more power at the increased current required.

50 watts at 120 volts is less than half an amp. 50 watts at 12V is about 4A. Big difference. }:)
Re:External supplies by plcurechax · 2003-08-11 02:42 · Score: 2, Insightful

When wiring up electrical systems, it is important to remember that your long runs should always be done with high voltage cable. For instance, from the wall to the power supply is 120V AC. Once the voltage is reduced, the runs need to be kept as short as possible, since every foot of cable loses substantially more power at the increased current required.

Well, you have to also factor in that AC travels much better than DC. That debate goes back to Edison and Telsa, and Telsa got it right. That's why your mains power (that comes out of your wall socket) is high voltage AC and not DC.

A minor reason is that I suspect that switching power supplies with long DC leads would act as unintential radiators (i.e. accidential antennas) at LF, 20-40kHz typically I believe in most computer power supplies. Thus they would be unable to get FCC approval due to the RF noise.
Re:External supplies by ZorinLynx · 2003-08-11 02:53 · Score: 1

I hate to correct you, but AC does not "travel" better than DC. In fact, it's the other way around, though the difference is insignificant.

AC is subject to the "skin effect", whereby more of the current flows toward the outside surface of the conductor. However, the skin effect is pretty insignificant at 60Hz.

The reason why AC is used for mains power is because its voltage can be adjusted very efficiently using a very inexpensive and reliable device: The transformer. If you put DC into a transformer, you just get a lot of smoke. Changing the voltage of DC requires either a voltage multiplier, a regulator, or a switching power supply, and these are a lot less efficient than a transformer handling AC.
Re:External supplies by GigsVT · 2003-08-11 03:13 · Score: 1

If you use large enough conductors, it wouldn't matter.

Alternatively, a 24 or 48 VDC power system might make the complexity and heat of the internal power supply much less. I hear the telco uses something like that.

--
I've had enough abrasive sigs. Kittens are cute and fuzzy.
Re:External supplies by name773 · 2003-08-11 04:01 · Score: 1

i've seen wiring for the 12v car supply going the length of my brother's wagon and his amp works nicely. just use an rf choke / capacitor to ground on the output of the supply. as in decoupling.... that should be done anyway
Re:External supplies by pjrc · 2003-08-11 05:18 · Score: 1

wouldn't it make sence to make the power-supply external.
Not if you must pass FCC (or CE in the rest of the world) radio frequency emissions tests.
Those wires are going to radiate both the switching noise from the power supply and noise from the motherboard. They must remain inside the steel case to have any chance of ever getting the radiated EMI under the FCC class B limit, which is required for sale to consumers for use in homes. Even the higher class A limit (allows more noise, for business use where people aren't as concerned about interference with their TV and radio) probably will be impossible with those wires exposed.

--
PJRC: Electronic Projects, 8051 Microcontroller Tools
Re:External supplies by Noehre · 2003-08-11 05:19 · Score: 1

Yeh, people always get that confused.

I've had EEs tell me that normal high-voltage power lines are all DC. 0_o

The interesting thing is that now that there are better methods for stepping up DC voltages to high levels, you're seeing more DC power lines.

Generally, those lines are used to connect together various segments of the electrical grid that aren't properly wired together via normal high-voltage lines. Its a way to more easily transfer power to where it is needed and thus increase the overall efficiency of the grid. I'm often amazed the grid system works at all.
Re:External supplies by JesseL · 2003-08-11 05:50 · Score: 1

Ever heard of shielded cables?

--
"Prefiero morir de pie que vivir siempre arrodillado!"
Re:External supplies by klui · 2003-08-11 08:34 · Score: 1

Minor correction: I think you meant Nikola Tesla, not Telsa.

Re:slashdot by YetAnotherName · 2003-08-11 01:05 · Score: 1

get a fucking grip on yourselves eds and start posting articles that promote DISCUSSION and DEBATE

All together now: News for Nerds.

(But does it matter? I had no idea how much until I RTFA. Please don't tell anyone I RTFA.)

Re:Power Supplys by Matt_Fisher · 2003-08-11 01:11 · Score: 1

Oh my bad.. Reading mistake.. But yes on the topic of power supplies.. I got mine of ThinkGeek and it is completly silent. It's great beucase the server in is my bedroom. I reccomend this one Best investment I have made in a while!

--
--Matt Fisher

28 pages? Give me a break. by Anonymous Coward · 2003-08-11 01:13 · Score: 5, Informative

Single-page link.

Re:28 pages? Give me a break. by Jeff+DeMaagd · 2003-08-11 05:04 · Score: 1

Thanks!

It would seen that they are wasting their bandwidth just to get 28 ad exposures, with each "page" having only two paragraphs worth of comments and reloading their huge menu bars every time. Why do hardware sites have to be so ugly and absurd?

Not wanting to put a downer on things... by WIAKywbfatw · 2003-08-11 01:15 · Score: 5, Insightful

...but it strikes me that these reviews of PSUs aren't as accurate as they should be. I'm not wanting to run the guys at AnandTech or elsewhere down (because, most of the time, they do a great job) but it strikes me that, when you look at PSUs (as opposed to CPUs, graphics cards or HDDs) then perhaps testing just one sample of each product is flawed.

After all, some of the measurements taken to distinguish good from bad were to the fifth significant figure. It strikes me that if you have to be that precise to differentiate between the winners and the also-rans then you've got to test more than one of each PSU - three would be a minimum, five or more would be better - and average out the test results to give you figures that are more representative of the quality of these products.

After all, not every Zalman ZM400A-APF is going to have a 12V min/max fluctuation of only 0.005V, and not every Enermax EG651P-VE FMA 550W is going to have a fluctuaction of 0.65V. Who knows, perhaps this was just a particularly good Zalman and a particularly bad Enermax? Testing more units means accurate results, which is a good thing.

I appreciate that testing three (or five, or however many) of each PSU means more work - you have to get x many more of each unit, test x many more times, process x much more data before averaging out your results - but, sometimes, I think it's warranted. Without wanting to get down on anyone, I'd like to suggest that, where called for, they try to source more units and test more thoroughly.

And, before people start flaming me for not knowing what I'm talking about, how much work is involved, etc, let me just say that I've run a review lab and I do know what I'm talking about, how much work is involved, etc. It's not a trivial amount but, sometimes, it is worth it.

(No doubt that's just a cue for half a dozen people to tell me where I'm wrong. I welcome objective criticisms but you can keep any childish flames.)

--

"Accept that some days you are the pigeon, and some days you are the statue." - David Brent, Wernham Hogg

Re:Not wanting to put a downer on things... by plcurechax · 2003-08-11 02:58 · Score: 1

After all, not every Zalman ZM400A-APF is going to have a 12V min/max fluctuation of only 0.005V, and not every Enermax EG651P-VE FMA 550W is going to have a fluctuaction of 0.65V.

If the fluctuations are due to the switching power supply and regulator design, then yes these measurements are going to be similar on other samples of these models (assuming no revision changes).

Modern electronics components are amazing reliable and consistent as long you don't use surplus parts (or capacitors made from incorrect stolen information) and operate them in their specifications. Most components are either DOA, and thus found before they leave the factory, or die from common but well understood semiconductor failures (thermal breakdown, static discharage, etc.)

Though I agree, for the most accurate testing, multiple samples should be used. That's just standard scientific method.
Re:Not wanting to put a downer on things... by Natalie's+Hot+Grits · 2003-08-11 03:06 · Score: 1

He didn't even test the power output of these PSU's, he just got the specifications off the sticker and compared them. The only real testing he did was the memtest86 results.

Overall, a very poor article. The reviewer is known for poor, and unscientific reviews on anandtech. I don't know why they let him print his articles on that website. Usually Anandtech knows not to print crap that isn't sound.

If you look up that reviewer's past articles (mostly on aluminum cases) he goes through the same crap. Testing 2 cases of the same design, 1 alu, one steel. but generic fans in one vs good fans in another, and then concludes "aluminum isn't so good at cooling or noise reduction afterall!"

Gime a break. I appreciate the amount of time he spends on his article, but he needs to take some science and statistics classes before trying to use the scientific method.

--
Two infinite things: your stupidity and mine. But I'm not sure about the latter. If my sig offends you, I'm sorry.

Where's the EE beef? by AragornCG · 2003-08-11 01:15 · Score: 5, Interesting

Great, there are more power supplies, but where are the oscilloscopes? Where is the detailed methodology for testing the cleanliness of the resultant power? They used some "stuff we found in our local university basement" to satisfy "you EE types". Oh, great. LIKE WHAT?

I'll take a smaller review with decent scientific methodologies, thank you very much. There's more to stability of a power supply than "interference testing" which reports an integer number of errors that could be caused by anything... What about the thermal testing... where was it measured? by what, the onboard mainboard thermistor, which is notably unreliable? Which "industry standard Chenming case" was it?

I really, really, really wish that these hardware sites would hire people with a decent understanding of the discipline of science - let alone engineering! - to write reviews. I could make MUCH more informed decisions.

-Ben

Altogether now... by Zog+The+Undeniable · 2003-08-11 01:16 · Score: 3, Insightful

Who, among home users, just buys the cheapest PSU they can find with enough power? Or is it just me? Some good reasons:

On my machine the CPU fan drowns out the noise of any PSU
With the way things are going, I'll need to buy a different PSU next time (ATX may die out, or everything will need 450W)
You can't see them, so they're not sexy
They make no difference to performance
Most of them last a reasonable time. I've been running different PCs for an aggregate of 11 years, and I've had one dead PSU in that time, costing 30UKP to swap out.

You could argue that people running servers should care even less, because they normally have redundant PSUs.

/me dons asbestos pants

--
When I am king, you will be first against the wall.

Re:Altogether now... by _|()|\| · 2003-08-11 01:50 · Score: 2, Interesting

On my machine the CPU fan drowns out the noise of any PSU
The power supply fan has always been louder than the CPU fan in my computers. I had a generic 300 W PSU that I thought was too loud. I replaced it with a PC Power & Cooling "Silencer" that was just as loud. I later got an "ultra-quiet" QTechnology PSU from Quiet PC. Little, if any, improvement. Noise is important to some people, and it appears to be a difficult problem to solve.
Not too long ago, 300 W PSUs were considered extravagant, but it's easy to overtax one with a few disks, especially if the PSU isn't up to spec. One of my computers rebooted sporadically, until I replaced the PSU.
Re:Altogether now... by Zathrus · 2003-08-11 02:08 · Score: 3, Interesting

On my machine the CPU fan drowns out the noise of any PSU

I'm sorry... maybe you should look into quieter CPU fans in the future? Personally, the Zalman flower fans look very attractive to me... not in a visual sense, but an auditory one.

They make no difference to performance

They certainly do. Inadequate power supplies can cause system lockups and shorten lifespan of components due to improper voltage regulation... sure, that card can run with the 3.3V line at 3.6V, but you think that's not going to have some effects down the line?

Put enough components in a system and you'll find out fast how much PSUs matter. Stick in 4 or 5 HDs and your system may not even power up -- even with a 400W PSU. Why? Because that "400W" PSU only has a 10A 12V line, which is utterly inadequate to spin up more than one drive along with a modern P4 or Athlon. Realistically it shouldn't even be used then, because you're well over 80% draw -- at which point voltage irregularities and noise concerns become a bigger problem. That system lockup? Yeah... it was caused by your CPU going wacky. Which was caused by the power supply introducing so much noise that the motherboard voltage regulators couldn't filter it out and fed the CPU bad power.

You're talking about another $30 to get a decent PSU... it's not going to break the bank.
Re:Altogether now... by ipjohnson · 2003-08-11 02:12 · Score: 1

Your right but if you ever have a PSU go on you and takes out the motherboard and graphics card. You'll pay more attention to which PSU you put in.

I don't know about you but I'd rather spend the time and a little extra cash in the beginning than risk alot more in the end.
Re:Altogether now... by 0123456 · 2003-08-11 02:12 · Score: 1

Why would you want to hang $2000 worth off hardware off a $10 power supply? Particularly as when you overload that cheap power supply by accident (putting in too many hard drives, upgrading to a far more power-hungry CPU or whatever) it may burn out and trash your CPU, motherboard and RAM in the process.

Personally if I'm spending that much money on components for my system I like to have a decent power supply to run them and the knowledge that if I do screw up and overload it it will just shut down nicely: but if you're happy with a cheapy, good for you.
Re:Altogether now... by evilviper · 2003-08-11 02:45 · Score: 1

Who, among home users, just buys the cheapest PSU they can find with enough power? Or is it just me?

That's a big ME TOO here. Although I usually go for the second-cheapest after checking the full specs.

On my machine the CPU fan drowns out the noise of any PSU

Same here currently, that's why I just invested in a Volcano 9. Extremely quiet at low speeds, and less noisy than my current CPU fan when it's at full speed, and twice the (CFM) airflow. Besides, it's the cheapest good CPU fan I could find.

Also, I have been looking into ducting the power supply fan output outside, which would redirect almost all power supply noise where I wouldn't hear it anyhow.

With the way things are going, I'll need to buy a different PSU next time (ATX may die out, or everything will need 450W)

Power usage isn't likely to go up much further... If it does, systems will be operating beyond the temperature capacity of their hardware, not to mention the incredible extra heat.

Although you are kinda right about not being able to use your current power supplies... Some newer systems are comming with smaller ATX power supplies, probably to save money as much as space.

You could argue that people running servers should care even less, because they normally have redundant PSUs.

Not really... If you've got thousands of dollars invested into your hardware, you don't want it all wiped away because the power supply (you saved $5 on) happen to short out and send 110V (or 220V) straight through to your mobo, hard drives, etc.

--
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
Re:Altogether now... by Teach · 2003-08-11 02:56 · Score: 2, Funny

Personally, the Zalman flower fans look very attractive to me... not in a visual sense, but an auditory one.

They look very attractive in an auditory sense? I think I see what you're saying....

--
Graham "Teach" Mitchell, computer science teacher, Leander HS
Re:Altogether now... by Pompatus · 2003-08-11 03:17 · Score: 1

In the past year in my household the following parts were fried due to inadequate power/protection

2 motherboards
1 CPU
2 hard drives
5(!) network cards
2 CD-Rom drives

these were out of three computers. One of these computers has a top of the line PS in it, as well as being hooked up to a UPS. That particular computer had 1 network card blow (some kinda surge hit our cable modem, took out the modem, router, and 2 network cards).

These experiences have taught me that power supplies do indeed matter. Surge strips seem to do nothing, but the UPS is a godsend.

--

----
Squirrel ... It's not just for breakfast anymore
Re:Altogether now... by Anonymous Coward · 2003-08-11 05:05 · Score: 0

maybe you should look into quieter CPU fans in the future? Personally, the Zalman flower fans look very attractive to me

As with anything concerning the noise computers produce, you should check out SPCR - it is totally hardcore :D Zalman's fans don't rate very highly on their scale. Popular opinion leans towards a Thermalright heatsink and an undervolted Panaflo fan.

Alternative power supplies. by Anonymous Coward · 2003-08-11 01:17 · Score: 1, Insightful

Are there any power supplies running with alternative energy sources that might be used by someone looking for something unique? I often thought about making a PC or perhaps just a server run completely on solar energy, even if it required a large array of panels. Perhaps a fuel cell type setup as well. These would obviously have to run a lot external to a machine, but getting the technology out and about among hobbyists is surely a good way to help debug it and perhaps bring its usefulness all the closer. Are there any other types of small alternative power supplies anyone can suggest?

( here is a possibility :) )

Re:Alternative power supplies. by Anonymous Coward · 2003-08-11 02:43 · Score: 0

wow, put something amusing in a serious question and get marked down a troll. gotta love mods without a clue.

I've investigated solar PSUs, but decided that getting a clean source of power from them was costlier than I was willing to spend. Something greater than $600 for a complete system. If you were without power for lengths of time and a maximum uptime was absolutely necessary then that would be a small amount to pay. It would be geekcool that's for sure.

I have no idea what other options there are. too much thinking for this early in the morning.
Re:Alternative power supplies. by fataugie · 2003-08-11 09:01 · Score: 1

I tried the "Squirrel on a wheel" method, but I guess the power requirements of my machine were too much for it. When I started the machine, it backfed through my system and turned the happy wheel into a Whirrling Dirvish of Death & Destruction...and it pureed the squirrel.
Until I catch a new squirrel to try v 2.0, I'm out of luck.

--

WTF? Over?
Re:Alternative power supplies. by jejones · 2003-08-15 15:08 · Score: 1

OK...the insolation constant is 1394 watts/m**2. Fortunately you're not talking about a laptop; the lightweight photocells are 15-20% efficient. Heavier ones are around 35% efficient, so let's ignore the monitor for now and say you run with a 350W power supply. That comes to around a square meter of photocells...unless you want your system to be up when it's overcast, or at night...and we didn't worry about your latitude or seasonal changes in insolation.

No dB ratings? by Anonymous Coward · 2003-08-11 01:19 · Score: 4, Insightful

I'd expect anyone reviewing power supplies to have provided some analysis of the noise generated by them. Since that isn't the case here, I'll stick with the original source material (read "manufacturer's promotional literature") instead of Anandtech's promotional review.

Less clicking, at any rate.

As a requisite disclaimer for the "It's Just Fine For Me" folks: While it may be true that certain electronic components are being engineered to be quieter, it's still the case that manufacturers rarely advertise/publish noise levels. The fact that there is a booming cottage industry devoted to quieter replacement components is evidence enough that manufacturers haven't yet fully addressed the needs of computer owners who are starting to discover that their nagging headaches and frequent innability to concentrate is due in large part to that box under their desks.

Re:No dB ratings? by richie2000 · 2003-08-11 02:12 · Score: 2, Insightful

I'd expect anyone reviewing power supplies to have provided some analysis of the noise generated by them.
Yes?

--
Money for nothing, pix for free
Re:No dB ratings? by Anonymous Coward · 2003-08-11 02:52 · Score: 0

I stand corrected. Click 23 times and there it is.
Re:No dB ratings? by Anonymous Coward · 2003-08-11 04:35 · Score: 0

I stand corrected. Click 23 times and there it is.
Alternatively, scroll down an at the "Review Index" combo box available on all pages and select "Page 23 - Noise Testing ...". Perhaps if you spent a tenth of the time it took to bitch on slashdot instead calmly understanding the available site features, you might have a better experience.
Re:No dB ratings? by ereuter · 2003-08-11 04:56 · Score: 1

The really sad thing is that the parent rated +5, and it is WRONG that there is no mention of noise. Ah, slashdot moderators.
Re:No dB ratings? by richie2000 · 2003-08-11 18:48 · Score: 1

"This comment has been moderated by a pack of clueless Slashdot moderators."

--
Money for nothing, pix for free

As we said last time... by Rogerborg · 2003-08-11 01:31 · Score: 1, Insightful

Who cares? More to the point, who cares about articles that decide reliability based on a sample size of one?

Are we just killing time while we wait for the next SCO rant?

--
If you were blocking sigs, you wouldn't have to read this.

For anyone who thinks PS don't matter by plcurechax · 2003-08-11 01:32 · Score: 4, Informative

I just built a new Athlon XP (2600+) based system, and the power supply was the most frustrated part to get working correctly. I bought a generic case with a 300W Pro-Power ATX power supply. It had useless regulations, the 12V rail measured at 12.8V with the system on and idle, and jumpted to 13.2V when running some math calculations that I use for stress testing (www.mersenne.org).

So I spent a Saturday trying to find a local computer dealer open that had any power supplies in stock. Most were sold out because of recent lightning storms (note: most didn't fail right away), anyhow about 2 hours of looking and driving I found a Codegen 300W ATX supply. It didn't setup off the SilentPC power alarms, but it failed the stress testing, with errors in about 1 hour of testing. The CPU temperature was fine (40-42 C) but I suspect the power wasn't clean which introduced memory or logic glitches.

Finally after a week, I got a ThermalTake 360W power supply and my system works fine.

So anyone who has an unstable system, it may not be all the fault of the OS, but poor or underpowered power supply.

Re:For anyone who thinks PS don't matter by evilviper · 2003-08-11 02:28 · Score: 1

the power supply was the most frustrated part to get working correctly

Although it's possible that you were having a power supply problem, believe it or not, it sounds more like your mobo is poor, and very picky about power.

I do buy very cheap power supplies, and after hundreds of them, not a single one has had a running problem (a few didn't power-on though).

So anyone who has an unstable system, it may not be all the fault of the OS, but poor or underpowered power supply.

It's not likely at all that the problem is the power supply. People are far better off checking to make sure they havea good mobo, memory, and primarily that their processor isn't running too hot.

--
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
Re:For anyone who thinks PS don't matter by tie_guy_matt · 2003-08-11 03:14 · Score: 1

Your system ran anyway even though the 12V line was running at 13.2V? I wonder how sensitive mobos really are to voltage fluctuations. They often do have lots of caps and voltage regulators that clean the power for any circuit that is sensitive. Also switching power supplies can have trouble regulating the power if there is no load. Maybe the 12V line wasn't regulated well because nothing was using the 12V line?
Re:For anyone who thinks PS don't matter by toddestan · 2003-08-11 11:13 · Score: 1

According to my motherboard monitor, the +12V line is +12.480V, and the +5V line is +4.590V. The computer is completely stable with it's Athlon 2000+, 3 HDDs, CD burner, etc. The voltages don't fluctuate more than about .05V either. The power supply is an Enermax 350W, and it also worked great on my older setup, a K6-III 450Mhz with 3 HDDs and other goodies.

So I have concluded that either the builtin monitoring is wacky, or the components just don't care.
Re:For anyone who thinks PS don't matter by lepton+noodle · 2003-08-11 13:25 · Score: 1

For the most part the 12V line is probably running things like fans, drive motors, and other non-logic related equipment. In other words, it isn't supplying anything that is especially sensitive to ripple or slightly out of spec voltage. The power supply is probably regulating on the 5 or 3.3V power output, leaving the others to wander according to their applied loads (they're simply tapped off different windings of the output transformer, they have no regulation of their own).

When we remember about power supplies... by xyvimur · 2003-08-11 01:37 · Score: 3, Interesting

"Power supplies tend to be one of those forgotten, but ever so important parts of machine construction..."

Here you can see why you should pay more attention when buying supply...

Fan by gr8_phk · 2003-08-11 01:38 · Score: 1

How about one with no fan. We build >600W 12V motor drives in small sealed boxes for under-hood (125C ambient) applications. We don't have a $&^%# fan. Why then must my PC be so loud at room temperature?

Re:Fan by really? · 2003-08-11 02:02 · Score: 1

Those motors don't run 24/7, do they?
Also, "under-hood" applications don't require so much silencing.

--

"Consistency is contrary to nature, contrary to life. The only completely consistent people are the dead." A. Huxley
Re:Fan by battjt · 2003-08-11 02:17 · Score: 1

I have an old 486 running as a router, that I ripped all the fans out of (cpu and power supply). It runs in the attic of a garage. No problem. I did open all the slots across the back. I'm probably getting some stack effect ventilation. Joe

--
Joe Batt Solid Design
Re:Fan by GigsVT · 2003-08-11 03:27 · Score: 1

If you build power supplies, you should know that standard components with smaller operating temperature ranges are much cheaper than mil-spec temperature range components.

I'm sure power supplies would cost a hell of a lot more with parts rated to operate at 125C.

--
I've had enough abrasive sigs. Kittens are cute and fuzzy.
Re:Fan by Anonymous Coward · 2003-08-11 04:12 · Score: 0

something else must be considered... PC PS's have three main power outputs, at 3.3V, 5V, and 12V (there are a couple others, but they are small) Converting 120VAC to 5VDC and 3.3VDC is a less efficient process than the conversion from 120VAC to 12VDC.
Also, proper regulation or ripple on the output is not of concern for driving motors, the motor really doesn't give a crap about the purity of the DC. On the other hand, computer equipment does, and surely the ATX spec has much tighter requirements for those items than a motor supply would.
Re:Fan by JonMartin · 2003-08-11 12:00 · Score: 1

silentmaxx and Engelking both make fanless (or mostly fanless) power supplies. It make take some digging to find them on the German websites though.

--
Serve Gonk.
Re:Fan by gr8_phk · 2003-08-12 04:52 · Score: 1

" Those motors don't run 24/7, do they? Also, "under-hood" applications don't require so much silencing."
Brushless electric power steering motors generally only run while you have torque on the steering wheel, so no, it's not continuous. The PWM FET switching is constant, and disipates several watts inside the box on it's own. The environment requires a sealed box, so no fan (besides, they're expensive). The motor is coupled to the steering column, so a silent motor is imperative.

Yeah right by Anonymous Coward · 2003-08-11 01:42 · Score: 0

Power supplies is one of the main sources of problems after overheating for P4/Athlons.

If your P4 is working with a 200W power supply, good luck to you, but don't put a floppy in it's drive!

Card power requirements? Warning lights? by dpbsmith · 2003-08-11 01:49 · Score: 3, Interesting

For as long as there have been micros, we have played the game of "by guess and by gosh."

Why doesn't every card and component in a system have a clearly marked indication of its power requirement?

And why don't power supplies issue unmistakable warnings when the system draws more power than the supply can reliably provide?

Every fuse and circuit breaker in a house has the amperage clearly marked on it, and so does every appliance.

Why can't the insides of our computers come up to the same standards as our toasters and washing machines? This isn't rocket science, this is simple arithmetic.

--

"How to Do Nothing," kids activities, back in print!

Re:Card power requirements? Warning lights? by Zeddicus_Z · 2003-08-11 02:18 · Score: 1

You know what - I'd settle for every RAM manufacturer to clearly label the amount/speed/type on their fscing sticks of RAM! I mean seriously, how hard is it to add the sticker?

--
Janie took my gun...
Re:Card power requirements? Warning lights? by chiph · 2003-08-11 10:42 · Score: 1

Why doesn't every card and component in a system have a clearly marked indication of its power requirement?

I strongly agree. I have a PC Power & Cooling 300w PSU, and I'm wanting to build another system using it.

But I'm playing the "Gee, I wonder if it's got enough mojo in it to run a P4 with 3 7,500rpm drives?" game. Without hard numbers from the motherboard mfr and Western Digital (I'm willing to put a barely-better-than-VGA video card in it, so no worries about AGP 8x or anything), I just don't know unless I'm willing to plug it in and hope it doesn't spit flames out the back when I hit the switch.

Chip H.
Re:Card power requirements? Warning lights? by toddestan · 2003-08-11 11:17 · Score: 1

Just so you have some numbers, the Maxtor 120GB 7200RPM drive I have right here states that it draws 670mA on the +5V line and 960mA on the +12V line. The Samsung 16X DVD-Rom draws 2.0A(!) on the +5V line and 1.0A on the 12V line. An ancient Seagate 2GB I have does not have any ratings.
Re:Card power requirements? Warning lights? by Anonymous Coward · 2003-08-18 06:19 · Score: 0

He also needs the drives spin up consumption or else it wont go.

Terminator by E-Tigger · 2003-08-11 01:49 · Score: 1, Funny

Am I the only one who when reading the blurb thought of the line from the Terminator where Arnie is asking to buy a: "Plasma pulse laser, 40 watt range" Except it's the Power Supply Terminator looking for a power supply in the 300+ watt range.

Just get a Mac by Anonymous Coward · 2003-08-11 01:51 · Score: 2, Funny

So you have decent memory which has been thoroughly tested, a hard disk drive which matches it's controller, no hot spots [espcially around the cpu,3D graphics card and northbridge], well sorted BIOS, no interupt conflicts, a decent OS and compatible drivers for each device, then yeah, it's probably the power supply.

double story. (?!) by leuk_he · 2003-08-11 01:52 · Score: 1

I is a shame the editors dont read the comments on their own site. But then, the previous story was posted by taco.

If something gets really important you wont miss it on /. . It will be posted regulary agian.

Re:double story. (?!) by Anonymous Coward · 2003-08-11 13:07 · Score: 0

whore. SUck it

Power Supplies Affect Stability by Anonymous Coward · 2003-08-11 01:54 · Score: 2, Informative

For you do-it-yourselfers:

Don't underestimate the importance of power supplies, nor assume because the box says "400W" that it really is, nor assume that any particular rail - +12V, +5V, +3.3V has the current you need. Sagging voltages on any rail can have mysterious effects that you will otherwise blame on your motherboard, the BIOS, memory, cards or the OS.

I was the victim of an underpowered power supply when I upgraded my motherboard, and of course, I blamed the motherboard. Of course, my suspicions were bolstered by the presence of a known defect in some of the boards having to do with a flaky overheating protection circuit. When I replaced the board with the same kind, directly from the manufacturer and it kept happening, I realized it must be something else.

I did a power resource calculation using the guide on the AMD site and viola, I was running my supply at 90% of capacity. A no-no. I upped the supply from 300W to 450W (a true 450W), haven't had any problems since.

Don't forget the Monitor by Anonymous Coward · 2003-08-11 01:56 · Score: 0

A laser printer can easily hit 400 watt, but a computer would use about 180 watt (power supply consumes about 20 watt) and the monitor about 70 watt.

Total about 250 watts.

ThinkGeek? You are an idiot. by rkz · 2003-08-11 01:56 · Score: 2, Informative

If you buy from thinkgeek you are being severely ripped off, newegg has it 20$ cheaper.click here

--
There is no god

Re:ThinkGeek? You are an idiot. by Shivaji+Maharaj · 2003-08-11 03:13 · Score: 1

Nope you got some other PSU.
Here is the exact same PSU from directron with the price you listed.

--
We do not have a history of profitable operations. Our future SCOsource licensing revenue is uncertain.

Re:slashdot by rkz · 2003-08-11 02:01 · Score: 1

he slashdot community is a bunch of idiots, trolls and karma whores, in contrast to k5 where people can actually spell and WRITE ARTICLES WITH CONTENT INSTEAD OF JUST PASTING LINKS AROUND AND EXPECTING READERS TO GENERATE DISCUSSION ON COMA-INDUCING SUBJECTS

That is exactly why I read slashdot and not K5.

--
There is no god

Some "quiet" PCs have external power supplies by Anonymous Coward · 2003-08-11 02:02 · Score: 0

The cpu needs to keep cool with a very large heat sink or a fan, therefore only notebook type processors can be used with an external power supply.

Internal power supplies DO NOT ADD heat, they DRAW it out. Power supplies are at the back of cases drawing [sucking] cool air into the case. Any heat produced by the power supply is blown out the back, NOT into the case.

But, it is the first where stabilty is shown. by leuk_he · 2003-08-11 02:13 · Score: 1

We all know good power is important. But here i have seen the first test that shows that the power supply is related to flipping random bits in the memory.

That is a mucht better test than tom did: Just put a lot of load on a PSU and see if it dies. If it dies in flames then take a picture of it and put it on the front page. (Since it did not flame like the picture it is a fake)

That is also better than the test of ars-technica: put a scope on the powerline and show the ripple. This looks horrible, but if the power stays in spec there is nothing to worry about.

hire people with decent understanding of the discipline of science - let alone engineering!
Well you cannot go to a computer shop and ask for a scientificly proven PSU, and you cannot engineer one by yourself (well you can, but it is expensive). And these sites want to sell pageviews, not PSUs.

Now i suggest you make a better power comparison in the basement of your high school. So doit and post it here.

Re:But, it is the first where stabilty is shown. by AragornCG · 2003-08-11 02:39 · Score: 1

This test is definitely better than the tom's hardware one. But the other review mentioned in the original post, despite only covering five power supplies, was an excellent model for reviews of this type: http://techreport.com/reviews/2003q3/psus/index.x? pg=1 It's not perfect - I agree with one of the other posters in this discussion that multiple samples should be tested, for sure - but it does show you more of what is occurring. Measurements were taken with an oscilloscope - which I assume was in current calibration, although it would have been nice to be assured of that - and most importantly, it shows the ripple voltage as a measure of how clean the power is.

You mention that you'd like to see stability testing, but that assumes the motherboard is perfect. How are we to know if the mainboard's circuitry is properly designed? The best solution is to find a power supply that nails 3.3V exactly, nails 12V exactly, etc. as close as possible, then use that power supply to stability test mainboards. One variable at a time, folks.

As for the stability testing performed in Anandtech's review, I would submit that bits are more likely to be twiddled by noisy input power than by "interference" that can be protected from by "more shielding and heavy construction" (although errors can be caused by any number of factors). A bit of memory is just a latch constructed of tiny switches. This latch stays latched as long as power is flowing through it, in a best case scenario. I would expect that the more stable the input power is, the more likely system components are to function properly. I am not against testing with tools like MemTest, but I'd like to see noise analysis like the TechReport review alongisde it to help evaluate exactly what is going on. (This ideal review would follow sound statistical procedures, though, and not create three significant digits of results from three integer trials that vary significantly (ranging from 2 to 6 on the same power supply? Come on, people...). It would also tell us which addresses in memory were flipped so we could see if it was a certain location on the DIMM or completely random.)

You say "That is also better than the test of ars-technica: put a scope on the powerline and show the ripple. This looks horrible, but if the power stays in spec there is nothing to worry about." Exactly! If the power stays in spec, there is nothing to worry about. So let's focus on making sure the power stays in spec in a review.

We're in agreement.
Re:But, it is the first where stabilty is shown. by fmaxwell · 2003-08-11 05:31 · Score: 1

You say "That is also better than the test of ars-technica: put a scope on the powerline and show the ripple. This looks horrible, but if the power stays in spec there is nothing to worry about." Exactly! If the power stays in spec, there is nothing to worry about.

What? A power supply that is noisy and has a lot of ripple will usually make for a less stable system, regardless of whether the voltage remains within upper and lower limits. Power supplies that inject noise into a system will degrade system stability. With differing filtering caps all over the motherboard and PCI cards, the instantaneous logic "1" voltage seen by different chips can vary and can cause "glitches." While that's somewhat of an oversimplification, this is not the forum to give the equivalent of an EE course. Such noisy supplies can also lead to noisy analog I/O (audio signals in and out immediately come to mind -- perhaps some types of video, also).

Examining the ripple and noise with a scope is a very good way to evaluate a power supply. It costs money and takes design talent to create a clean output and it's a good way to tell when a power supply is substandard. I would much rather have a power supply with a clean output that was slightly out of voltage spec than one with ripple and noise which caused it to swing wildly between the upper and lower limits.

Quality power can not be stressed enough... by Awptimus+Prime · 2003-08-11 02:17 · Score: 4, Interesting

I used to buy whatever 450 watt CompUsa model I could get for $19.00. That was a big mistake.

After about 6 months use, I had one make a popping sound, then erupt in smoke. All I had in the box was dual 700 cpus, 6 scsi drives, 2 ide drives, cdrw, GF3, and a bunch of neon lights. During that time, the system would freeze up a lot, usually while compiling.

Basically 450 watts doesn't mean as much as the quality of components and how cloesly the manufacturer was willing to run them to the point of breakage.

I swapped out my burnt 450 watt for a Antec cool blue running at around 400 watts and my system is quieter and never locks up anymore.

When reading wattages of cheap PSU's remind yourself of audio equipment. Like when a brand like Sparkomatic or Jensen promises 1000 watts out of an amp only for it to sound like shiat and blow speakers. Meanwhile, a Bose amp of 75 watts would be generating more power overall and not be tearing your speakers up. The same game is played with PSU promises. They, the generic cheapies, never planned on anyone using all 450 watts of their equipment. These big number, low performance parts are meant for a consumer with a very basic system seeking comfort big specs.

I guess it reinforces 'You get what you pay for'.

Re:Quality power can not be stressed enough... by jbrandv · 2003-08-11 02:36 · Score: 1

I just had one "pop" and go up in smoke. The PC was OFF! It also tripped the breaker on my UPS and the PSU was only two days old. Quality isn't what it used to be. The PSU was made by Chieftec. Tigerdirect is sending me a new one. I've got my fingers crossed.
Re:Quality power can not be stressed enough... by GigsVT · 2003-08-11 03:18 · Score: 1

Tigerdirect is sending me a new one. I've got my fingers crossed.

Good, you'll need luck to deal with Tigerdirect. They are considered one of the most scammy retailers out there.

--
I've had enough abrasive sigs. Kittens are cute and fuzzy.

My PSU by Renraku · 2003-08-11 02:19 · Score: 1

I'm running a Duron 850 with 256MB of 133MHz SDRAM. GeForce2 GTS video card, SB Live, an ethernet card, and a generic 56k modem. Also, two HDDs, a CD-ROM drive, and a floppy drive.

When my computer is off, the PSU switch in the back is also off. I turn on the PSU switch, and then press the case button. Drives spin up, fans come on, but no screen. No USB. No beeps.

However, if I 'rock' the PSU switch, causing the system to 'brownout', the system will boot just fine, but I have to 'rock' the switch just right, or else its back to step one. Been doing this for about a year now, almost two.

Anyone have any idea what's wrong with 'er?

--
Job? I don't have time to get a job! Who will sit around and bitch about being broke and unemployed then?

Re:My PSU by Durrik · 2003-08-11 04:02 · Score: 3, Informative

Sounds like your power supply can't handle the inrush current for the start up of your hard drives and fans.

When motors start up they have to get over the standing still friction of the unit before they get over the hump of startup and then into the lower friction of a moving unit. I would put in a nice graph but slashdot won't let me.

What the graph looks like is there's a rise up to a peak. Doing an integration under the line to the peak gives you the force needed to get to the peak. This is the force needed to get to the object moving. The graph then falls down a fair ways, (maybe almost to zero it depends on the two materials being being rubbed together), and then the line stays steady, horizontally. This represents the force needed to keep the object moving.

This is about all I remember from College physics on friction. There's also some other things that require a high inrush current during startup. But motors are the biggest thing.

So when you rock your power switch just right you leave the motors running enough not to require the inrush current. Shut off power to the motherboard to force it to restart, but you don't leave the power off long enough to cause the power supply to go into suspend mode and turn off all the power to the system.

Probably the best thing to do is get a power supply that handles a larger inrush current then the current one. But there could be something else wrong with the system too, and I haven't built a system for close to 8 years and my knowledge might be a bit dated.

--
Software Engineer & Writer of Military Science Fiction and Fantasy Blog: petermwright.com Twitter: WrightPeterM
Re:My PSU by multipartmixed · 2003-08-11 04:51 · Score: 1

> I haven't built a system for close to 8 years and my knowledge
> might be a bit dated

A little over 6 for me, I know how you feel.

As for his in-rush problem (I'm pretty sure that's exactly what it is, BTW), depending on his SCSI drives and controllers, he may be able to get the controller to spin them up one at a time after the controller's BIOS has initialized (and hence all the fans, etc are running).

Back in the day, the Adaptec-1542 supported this if you hit Ctrl-A during boot-up and walked the menus appropriately. And most SCSI disks had a jumper enabling or disabling auto-spin up.

Alternatively, an extra STDP switch hooked to the +5 and +12 lines of one of those disks might do the trick. Power on the machine, count to three, throw the switch to power on the last disk... Probably safer than what he's doing now (but still a lousy solution, especially if his OS is on that second disk ;)

--

Do daemons dream of electric sleep()?
Re:My PSU by Renraku · 2003-08-11 06:02 · Score: 1

The system does the same thing with or without HDDs. Using IDE drives, by the way. ATX power supply. I've disconnected everything but the processor fan, RAM, processor, and video card. Doesn't even beep unless I 'rock'.

--
Job? I don't have time to get a job! Who will sit around and bitch about being broke and unemployed then?

Comment removed by account_deleted · 2003-08-11 02:50 · Score: 1

Comment removed based on user account deletion

Yo, moderators: Bring this up!!! by swordgeek · 2003-08-11 02:57 · Score: 1

THANK YOU!

I have been looking for one of these things for ages. I keep thinking I'll have to make one eventually, because I couldn't find a commercial one.

I searched for watt meters. I searched for power consumption devices. I searched for everything under the sun that might come up with these things, and came up blank every time. Now I plug in the brand name, and about fifty hits come up.

Thanks. Going to see if anyone locally carries them, and get one ASAP.

--

"People who do stupid things with hazardous materials often die." -- Jim Davidson on alt.folklore.urban

Re:Yo, moderators: Bring this up!!! by HidingMyName · 2003-08-11 06:56 · Score: 1

Thanks. Going to see if anyone locally carries them, and get one ASAP.
I've seen them at radio shack (but I couldn't see it in the on-line catalog. I seem to recall the price being around $30 or so, but the on-line prices appeared a bit higher for other vendors so I was conservative in my original post, because I didn't trust my memory on the price).
Re:Yo, moderators: Bring this up!!! by NadMutter · 2003-08-11 08:52 · Score: 1

I bought mine from eta engineering (see http://www.etaengineering.com/killawatt.html) for $35.

Computer Refrigerator by jcsehak · 2003-08-11 02:59 · Score: 1

According to this, a computer draws less than half the power of a refrigerator.

I think their numbers are a little high, though. By their calculations, I should be paying $30/month just for the computer (which is on 24/7), yet I'm paying about $30/month for everything.

--

c-hack.com |

Wattage hype? by zapp · 2003-08-11 02:59 · Score: 1, Insightful

It seems to me that there is a trend going about for high wattage PSUs. Don't most systems (even high end Athlon XP's) idle at something like 150w with a hard drive and cdrom, and video card in?

Don't these systems peak at something like 200-250w (with 3d card suckin juice, and 100% cpu usage, etc).

Why the hell would you pay $150 or so for a 450 or 500w (or more!) system? Unless you're running multiple CPUs or some sort of file server with a whole bunch of drives... I just don't see the point, it's just gonna run up your electric bill and add more heat (and therefore fan noise) to the system.

--
no comment

Re:Wattage hype? by rreyelts · 2003-08-11 04:09 · Score: 1

Well, it really doesn't make any difference what "idle" usage is, because you don't want your power supply to go wacky any time you "max" out your system. You also don't want to run your power supply at peak usage for any extended period of time. Having a higher-rated power supply will actually give you more efficiency than a cheaper one, which translates to a lower electric bill in reality.

Personally speaking, I have a 400W Antec power supply, but I also use it to power a dual Athlon XP system. Not cheap, but not excessive either.

Design Engineer, Power Supplies, Computers by lcsjk · 2003-08-11 03:03 · Score: 5, Informative

As one who has spent some 16 years designing power supplies for computers and other devices, I found the article to be very weak in factual information.
Power supplies have an actual efficiency of about 75%, and no better than 90%, no matter what brand, Efficiency varies with loading.
Quick spin-up on hard drives takes about 5 times as much 12V power as is required for running, but little 5V change.
The actual voltage needed by components is not nearly so critical as the amount of noise (ripple and "switching" noise. (These are switching power supplies.) The nearest thing to a noise measurement test was the memory test.

Output regulation is also dependent on the loading of the 5 volts or 3.3 volt outputs. Expect the 12 volts to follow the 5 volts as the processor and MB requires more or less computing power during operation. Some are independent, but most are not.

Things like temperature testing can be very meaningless. Most components are designed for operation at an ambient of more than 70 degrees Celcius. The switching transistors will produce the most heat, but a small capacitor setting downstream in the airflow may be the first long term failure. Probably input to output is the best indication.

Note also that the power supply fan sucks hot air from the PC and sends it over power supply components before exhausting to the outside air.

The one test I found most interesting was the memory test, and although I am not sure of the methods used, the results do indicate overall performance. (Yes, I have also designed memorys and been involved it testing of memory chips with walking ones and zeros and checkerboard patterns etc. Without knowing which row of memory transistors is adjacent to the one being used, pattern testing is basically meaningless except for one word at a time.

A few years ago I took a Sparkle (SPI) power supply and a Fortran power supply apart and the insides were identical - completely. Fortran was about $3 more in hundreds at the time. My guess is that they were made on the same production line. Are they the same or had one just been copied.

The power supply continuous output rating should be the one used when talking about a 300 watt power supply, not the power line rating.

Finally, audible noise. Bigger fans, lower speed, less noise. Fans mounted on the inside instead of near the back panel produce less noise.
Fans blowing onto holes instead of through a circular guard produce more noise.

I agree totally except... by wukie · 2003-08-11 03:06 · Score: 1

Why publish only an average from the 3 [or more] supplies.

Publish each supplies results, averages and deviations.

My computer (Dual Athlon on Tyan mobo) hammers the 5 V rail, so test on a few motherbords with quirks would help aswell.

People usually get a 550W supply to power multiple hard disks, usually in a raid configuration. They should have done this aswell.

Lastly, there is a variety in case designs. They should have tried a few cases and test for temperature and noise levels.

Forgotten...again by swordgeek · 2003-08-11 03:08 · Score: 2

"Power supplies tend to be one of those forgotten, but ever so important parts of machine construction."

That's why they make an appearance on /. every ten days or so.

--

"People who do stupid things with hazardous materials often die." -- Jim Davidson on alt.folklore.urban

Lite-On 300W PS OEM for Compaq (Server-Class) by MsGeek · 2003-08-11 03:08 · Score: 1

Sometimes digging around on sites like Directron is a big help when you need to jam Econo but you want quality parts in your machine. I got a power supply made by Lite-On (same company as the Gold Standard of optical drives) for Compaq for use in their servers. It is conservatively rated at 300W but has been stress-tested at 380W and didn't even break a sweat. Here's the link, sorry they don't have any more of them. $20 to purchase. I paid $10 extra to test the thing as good before they pushed it out the door. There are 6 harness wires coming out of the PS...I think this was designed for a box with a RAID.

--
Knowledge is power. Knowledge shared is power multiplied.

Be careful with "cheap" power supplies by m33p · 2003-08-11 03:18 · Score: 2, Insightful

While you may never see any problems running PCs with cheap power supplies on a daily basis, they often suffer catostrophic failure after six months or more of use. I had one machine that ran flawlessly as my home server for over a year but one day when I was re-running power cables (exterior to the case) the power supply literally exploded when I powered the machien back up, destroying ever single component inside! Since I've started talking about this, I have found that something similar has happened to quite a number of people. And ever since, I have used nothing but higher quality power supplies.

BTW, the worst case was a guy who had build a large IDE raid box. He figured he didn't need backups because he had RAID and the likelyhood of losing two drives at the same time was pretty low -- or so he thought. When his PS blew, it took *all* of the drives (and all of his data) with it. Can you say suck-o-rama? I knew you could...

-p.

Re:Be careful with "cheap" power supplies by Anonymous Coward · 2003-08-11 04:40 · Score: 0

There is little difference in components between cheap power supplies and more expensive ones. Price is no guarantee that you are getting quality components.
Assuming the power supply is well cooled and designed by competent engineers a cheap supply will preform similarly to a more expensive one of similar rating. One problem is that the rating printed on the outside of the box (e.g. "300 Watts") may have no true relation to what the supply has to offer. Look at the 3.3 Volt, 5 Volt, and 12 Volt current rating printed on the side of the supply case itself when comparing supplies.
Probably the biggest cause of failure of a power supply is fan failure which leads to catastrophic failure of the active components (i.e. switching transistors).
The other critical component is the capacitors. Here is where the quality is needed. Cheap capacitors with high ESR (effective series resistance) and low ripple current rating are the problem. If a capacitor has high ESR and can not handle very high ripple current, then it has no place in a PC switching power supply. The capacitors on the input side of the switching transistors are particularly important since they deliver huge non-linear current spikes when the transistors turn on thousands of times per second.
There may be nothing wrong with the quaility of high ESR/low ripple current capacitors per se, however using this type of capacitor in a switcher is a misapplication. They will work themselves to death, while providing a noisy output. They will overheat in use and die prematurely. Unfortonately there is no way of knowing what kind of capacitors were used in your power supply. If there is a price difference, between supplies of similar rating, it is to be hoped that the higher priced unit offers properly speced capacitors. There really is no way of knowing unless you can find their spec sheet (very unlikely).
What can you do to ensure the odds are in your favor? First and formost is cooling. Forget about those "silent supplies" with the thermostat controlled fans. Bypass the control circuit and let her rip! Let the fan hum like a mad woman. Blow as much cooling air through the unit as you can. Keep intakes and outvents clear of dust and debris. Even if you have wrongly speced capacitors, the extra cooling will keep them in line longer and may prvent premature failure. Like the CPU, cooler is better with power supplies.

Power upgrade by gsparrow · 2003-08-11 03:19 · Score: 1

When upgrading from a p3 to a p4 should you also upgrade the power supply?

Re:Power upgrade by Zed2K · 2003-08-11 03:24 · Score: 1

You have to. The power requirements and even the shape of the mb power connecters are different for a P4.
Re:Power upgrade by gsparrow · 2003-08-11 03:29 · Score: 1

Thanks

Re:Computer Refrigerator by evilviper · 2003-08-11 03:22 · Score: 1

Their numbers are incredibly high... They are doing the same thing this guy's parents are doing, and assuming the rating on the power supply is the actual power drawn.

Your computer is probably using up about 100watts, not 400, so divide all their stupid figures by 4 (and you should also use your own kilowatt/hour rates, not theirs).

--
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant

Re:slashdot by Anonymous Coward · 2003-08-11 04:14 · Score: 0

28 pages on power supplies? James Michener would be proud

Plus three hundred watt? Darn... by aziraphale · 2003-08-11 04:27 · Score: 1

I'm looking for a power supply in the minus three hundred watt range.

Bad power supply tests by Animats · 2003-08-11 04:33 · Score: 4, Informative

That article was very lightweight. They didn't do tests that any competent QA shop would do.

Tests they didn't do, and should have:

Look up the UL certification for the unit, and check that the unit actually passed UL testing. Many power supplies, especially from China, have forged UL certifications. (UL had about 80 shipments seized at U.S. Customs last year for this.) UL's test include surviving a dead short and not catching fire for any single component failure. Uncertified power supplies are almost always the ones that catch fire.
Checking the UL database is easy. Check any power supplies you have, and if the E-number isn't in the database or doesn't match the unit, report it to UL and have a serious talk with the seller.
Test the power supply with input voltages at the bottom and top of the rated range, and measure the regulated outputs. This is basic stuff.
Test the power supply in a hot box, with temperature maintained at the high end of the specified temperature range. This should be done with the unit loaded up to its rated load, and should continue for at least 72 hours. This, again, is a basic QA test.

Those are all basic. The "memory test" was interesting. That's an indication that these power supplies differ in their ability to sink RF noise coming in on the power rails. That, too, needs to be tested.

Nothing Anandtech did seemed to use a 'scope, let alone a spectrum analyzer.

Design Engineer, Power Supplies, Computers-Outside by Anonymous Coward · 2003-08-11 05:05 · Score: 0

"Note also that the power supply fan sucks hot air from the PC and sends it over power supply components before exhausting to the outside air."

I've often wondered why they don't suck in outside air? It's easy regardless if it's a desktop, or a tower (especially a tower).

BTW The power supply really should be at the bottom of the case not the top, since heat rises.

Clarification and additional thoughts. by fmaxwell · 2003-08-11 05:06 · Score: 1

I have no idea, it could be from 1-400watts depending on the system.

The power supply itself, with nothing connected, will still draw more than one watt.

Your friend's parents are idiots.

More importantly, his friend is an idiot. His friend "built" the computer and then doesn't know that a 400W power supply doesn't draw 400W all of the time that it is on. If he doesn't even have basic knowledge like that, he has no business building PCs.

Better yet, they should try unplugging any devices that they aren't using, since most devices that are turned-off are still drawing 2-4 watts a piece 24 hours a day. Unplug 4 devices would recoup your friend's 2 hours of 'computer time'.

Something is seriously screwed up with this whole scenario. The kid has an Athlon XP1800+ system with a 7,200rpm drive and the parents are worried about him using a few cents worth of electricity per day? Even if their flawed supposition about the power draw was right, their limit is .8kwh per day (.4kw * 2 hours). That's about six cents worth of electricity per day or about $2 per month. If they can't afford more than $2 per month, the kid should be using something like a 200mhz Pentium MMX that his parents found at the thrift store for $40.

Re:Clarification and additional thoughts. by Junior+J.+Junior+III · 2003-08-11 06:19 · Score: 1

The parents don't really care about the electricity bill. They just don't want their kid being exposed to all the pr0n and other evils on the internet for more than 2 hrs daily. He needs to get out and exercise, too.

--
You see? You see? Your stupid minds! Stupid! Stupid!
Re:Clarification and additional thoughts. by fmaxwell · 2003-08-11 06:43 · Score: 1

The parents don't really care about the electricity bill. They just don't want their kid being exposed to all the pr0n and other evils on the internet for more than 2 hrs daily. He needs to get out and exercise, too.

If those are the reasons, then the parents should be honest with their son instead of claiming poverty. Lying is no way to raise a kid.
Re:Clarification and additional thoughts. by Anonymous Coward · 2003-08-11 06:56 · Score: 0

More importantly, his friend is an idiot. His friend "built" the computer and then doesn't know that a 400W power supply doesn't draw 400W all of the time that it is on. If he doesn't even have basic knowledge like that, he has no business building PCs.

Building a computer does not require knowledge of exactly how many watts a piece of equipment uses. And power supplies aren't exactly "basic knowledge". How to put ram in? Yeah, thats basic knowledge. The fact that you need a certain wattage power supply? Yeah, thats pretty basic as well. Knowing that of those 400watts, only 100 or 125 are going to get pulled? Uh... not basic. Knowing that adding 7 more SCSI drives uses another 150 watts? uh.. not basic either.

Perhaps in your day you "built" your own computers. However, today you can buy fully assembled parts, and put your own together with just a screw driver. No sodering iron required.
Re:Clarification and additional thoughts. by fmaxwell · 2003-08-11 08:31 · Score: 1

And power supplies aren't exactly "basic knowledge".

Yes, that's basic knowledge. Watts, volts, amps, ripple, etc. is about as basic as you can get.

Perhaps in your day you "built" your own computers.

I did, and it's still "my day."

However, today you can buy fully assembled parts, and put your own together with just a screw driver. No sodering iron required.

You can bolt engines together by mail-ordering parts -- all with no knowledge of the effects of camshaft profiles, compression, bore to stroke ratio, spark advance, the advantages of forged (vs. cast) pistons, or what cross-bolted main bearings are for. What you bolt together will probably run. But it may be short-lived, unreliable, inefficient, and/or underperforming.

Just because you can do something does not mean that you should.

I've seen countless computers built by "screwdriver engineers" who had no real knowledge of the basics. Because they didn't understand the basics, they often bought components based on cost, advertising, or appearance rather than quality. Those computers often had cooling problems from inadequate heatsinks, thermal pads (rather than thermal grease), cheap, sleeve-bearing fans, and poor airflow (often with drives stacked one atop the other with no air-space in-between).

Other computers had low-quality power supplies that were either grossly under, or oversized. One person brought me a machine that only crashed during video game play. He was sure that the problem was the "crappy video card" -- actually the only good component in the whole system. I put a scope on the POS power supply and had him start gaming. The voltage sagged and got grossly noisy and, sure enough, the system crashed. I put in a decent supply and the system stopped crashing.

The PCs would often have poor-quality RAM (rather than Crucial/Micron, Mushkin, Corsair or some reputable DIMM manufacturer) that led to system instability. Many of the owners did not understand why the DIMM carrier is important. They would buy something with "Micron" RAM chips and then not think twice about the cheap carrier PC board. One builder proudly said that they insisted on "Micron" RAM. When I opened the machine, the DIMMS were no-name and the RAM chips on them were cheap, CAS-3 junk from some manufacturer other than Micron. You could have sold him the innards of a cordless phone as "Micron RAM" and he'd have believed it.

I've seen crappy motherboards with inadequate power supply filtration caps. One owner was astounded that I could look at a motherboard and see that the caps were undersized and too few in number. His several-times-per-day crashes ended when I replaced the motherboard with a reputable one.

If Dell built PCs the way that the average screwdriver engineer does, they would have gone bankrupt due to reliability and performane problems with their PCs.
Re:Clarification and additional thoughts. by evilviper · 2003-08-11 16:44 · Score: 1

Lying is no way to raise a kid.

"Look! Santa brought you a new computer for Christmas. Remember, no more than 2 hours a day because of the electric bill."

--
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
Re:Clarification and additional thoughts. by evilviper · 2003-08-11 17:09 · Score: 1

The power supply itself, with nothing connected, will still draw more than one watt.

Just to be a smart ass, I just tested one... A power supply (with nothing connected) is drawing exactly 01watts. Of course, this varies from power supply to power supply, and isn't really important since the power usage of the system will dwarf the power usage by the supply itself.

In any case, I didn't mean to literally imply that his system could be drawing 01watt, I was just making a point that there is no way to even guess how much power the computer may be using, just by the wattage rating on the power supply.

His friend "built" the computer and then doesn't know that a 400W power supply doesn't draw 400W all of the time that it is on.

I'm not so sure... Maybe he does know that, but his parents won't listen to him. That's not an unusual situation, and I try not to assume anything, so I went with what was written.

That's about six cents worth of electricity per day

Well, the problem with this, is that everyone is using the price of electricity in their area. You obviously don't live in CA, otherwise the price you quoted would be 50-100% higher. Yes, these people are incredibly electricly thrifty for some reason... Again, I'm not going to assume why.

If they can't afford more than $2 per month, the kid should be using something like a 200mhz Pentium MMX that his parents found at the thrift store for $40.

Well, that's the interesting thing... Power drain isn't really that much lower with older computers. Let's say they got a 200MHz system, that might require 10x as long to do the same ammount of work as the faster system, which would (in the long run) be more expensive.

--
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
Re:Clarification and additional thoughts. by fmaxwell · 2003-08-12 04:30 · Score: 1

Just to be a smart ass, I just tested one... A power supply (with nothing connected) is drawing exactly 01watts.

I guess I should have said that a power supply that was "On" (vs. standby) mode. I'd guess that you hooked up an ATX power supply and didn't short the pins to turn it on.

Well, the problem with this, is that everyone is using the price of electricity in their area. You obviously don't live in CA, otherwise the price you quoted would be 50-100% higher. Yes, these people are incredibly electricly thrifty for some reason... Again, I'm not going to assume why.

Even if I was 100% off and the supply was drawing a full 400watts whenever on, we're still only talking 12 cents per day.

Well, that's the interesting thing... Power drain isn't really that much lower with older computers. Let's say they got a 200MHz system, that might require 10x as long to do the same ammount of work as the faster system, which would (in the long run) be more expensive.

I was referring to the initial cost of the computer, not the computer's power consumption. The parents are concerned about a few cents per day in electricity and the kid's got a computer that costs, at the least, hundreds of dollars. Something's wrong.
Re:Clarification and additional thoughts. by fmaxwell · 2003-08-12 10:52 · Score: 1

"Look! Santa brought you a new computer for Christmas."

"What are you talking about? I built the computer just like it said on Slashdot. And there's no such thing as Santa."

Remember, no more than 2 hours a day because of the electric bill."

"So why didn't Santa bring you enough money to pay the electric bill?"

---

Yes, lying is no way to bring up a kid and you gave two darned good examples.

I suggest something more like:

"Son, we're proud of you for building your own computer, but no more than two hours per day. You need to spend time with your friends, keep up with your homework, and get exercise."

If parents were a little more open and honest about the need for outdoor activities, socializing, education, and exercise, the U.S. might not be a nation full of maladjusted, semi-literate, obese kids.
Re:Clarification and additional thoughts. by evilviper · 2003-08-12 12:19 · Score: 1

I was referring to the initial cost of the computer, not the computer's power consumption.

As was I. Since an XP 1800 is much faster than a thrift store computer, it is concievable that paying $200 instead of $50 would be a good idea, since you would save more money than the cost difference in the power bill in about a year. That, of course, assumes the 1800 is maxing out it's processor for 2 hours every day, so the slower computer would have to be left on that much longer, which is probably a bit of an extreme asumption, but I'm just making a point.

--
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
Re:Clarification and additional thoughts. by fmaxwell · 2003-08-12 15:30 · Score: 1

Since an XP 1800 is much faster than a thrift store computer, it is concievable that paying $200 instead of $50 would be a good idea, since you would save more money than the cost difference in the power bill in about a year. That, of course, assumes the 1800 is maxing out it's processor for 2 hours every day, so the slower computer would have to be left on that much longer, which is probably a bit of an extreme asumption, but I'm just making a point.

But we both know that the average gigahertz-plus PC spends 99+% of its time waiting for user input, disk I/O, or network traffic -- especially when the computer is in the hands of a kid. It's not like a computer that's five times as fast will let him produce homework five times faster, play games five times faster, or download five times as much porn per hour.

Not news worthy by amightywind · 2003-08-11 05:16 · Score: 1

This story is not news worthy at all. I like Anandtech, but I think /. readers would be much more interested in their motherboard roundups.

--
an ill wind that blows no good

From humble intentions...The PowerHog! by ackthpt · 2003-08-11 05:39 · Score: 1

This review is certainly the kind I've been looking for. There was a review another site had some time back where they loaded each individual PSU until it cracked or shutdown, most cracked into a smoking ruin, which would take who knows what with them. (shudder)

I sprung for a PCP&C silencer 425 last Dec., intending to keep my PC simple, but fretting over what size PSU I "really" needed. Most USENET groups were full of useless information which boiled down to "just throw the biggest PSU you can get" at the problem. Well, informative to see Enermax 550w only pulling 200 actual. With all the voltage conversions going on, and loss therefore to inefficiency, I tried to get a grip on what I really needed, with a suitable safety margin. Since then I've added an ATI AIW 9700 (on a whim, I originally had no intention of adding such a card) and everything is still humming along, nice and cool. Considering it has 2600+ CPU, 768 Meg DDR, 2 80G HD, modem, GPU, 3 fans besides PSU fan, DVD, CDR and a floppy, it seems up to the task.

It really is exasperating how little true knowledge there is regarding piecing together a PC and what power requirements, and what you actually get.

--

A feeling of having made the same mistake before: Deja Foobar

Re:Design Engineer, Power Supplies, Computers-Outs by pjrc · 2003-08-11 06:00 · Score: 1

I've often wondered why they don't suck in outside air?

Then the power supply would be blowing heated air INTO the computer, where cooling is needed the most. Not a smart idea.

BTW The power supply really should be at the bottom of the case not the top, since heat rises.

Hot air rises, in the presence of relatively still cooler air. This is a relatively small effect in the presence of fans forcing airflow.

Nonetheless, the slight buoyant force causing heated air to rise also helps it reach the power supply at the TOP of the case, where it is exhausted to the outside by the power supply's fan.

.

Of course, if you really want to keep your power supply cool at the expense of the CPU, video, disk and other devices, then by all means reverse the fan direction and put it at the bottom.

--
PJRC: Electronic Projects, 8051 Microcontroller Tools

Think carefully about paying more. by Futurepower(R) · 2003-08-11 06:10 · Score: 2, Informative

More and more companies are basing their business plans on taking advantage of the overclocker, gamer, case modder, or PC enthusiast who doesn't have much technical knowledge. These have slick sales literature and fancy graphics. They also may have no one at the company who knows or cares about the technical details of the product. I feel qualified to say this because, back when I was an electrical engineer, I have designed power supplies and industrial power control devices.

"Monster cables" won't give you better audio than cheap wire of the same gauge. Similarly, paying a lot of money for a power supply is very unlikely to get you any difference in computer performance, provided you buy a quality low-cost supply.

We pay $26 wholesale for our 350 Watt KingWin Kwi-350W power supplies, and they are fine. Froogle.google.com says they can be bought retail for $32. The AnandTech article discusses the Kingwin KWI-450, which they say costs $95. Why didn't they review the supply from Kingwin that is 1/3 the cost?

One guess is that the entire AnandTech article is a hidden Directron.com advertisement. Maybe Directron.com does not want you to know that you don't need to pay more, and that you will derive no benefit from giving Directron more profit.

The AnandTech article says, "Directron.com let us pick one of every new power supply off their shelves in order to put together a sample of 18 of the newer power supplies on the market right now."

The article says about the Kingwin KWI-450WABK, "We received our quad fan Kingwin supply well before the TTGI units had hit American soil. The unit was well received in the lab, but as we noticed when the TTGI units arrived, practically identical in construction. TTGI and Kingwin obtain their components from the same manufacturers in Taiwan."

"Practically identical" does NOT mean identical. Directron's 350 Watt TTGI TT-350SS is $28.99. It looks a lot like the KingWin 350 Watt supply, but is different; the TTGI supply does not have most of the labels that are on the KingWin supply, for example.

Look below at that link for what Directron says about their warranty: "Customer is responsible for returning defective products including DOA due to manufacturer unless testing is paid before shipping." DOA means "Dead on Arrival". Very clearly, Directron says that they do NOT stand behind their products.

Directron's warranty may be translated as, "If what we ship you doesn't work, tough luck, don't bother mentioning it to Directron, you are on your own, complain to the manufacturer, and pay the shipping costs to where the manufacturer is located, but we won't tell you in advance where that is, because you might buy directly from the manufacturer." Only if you pay more will Directron stand behind what they ship. So the true cost is not $28.99, unless you want to take your chances.

I suspect, and I have plenty of evidence, that when companies ask you to pay more for testing if you want a warranty from them, you get absolutely no testing. What you get is only the right to return a DOA unit to them.

Notice something else about that Directron page. The price is an image, not text, from a different server, yimg.com. Does anyone know why they do that? I don't.

There is plenty of evidence that the people at AnandTech have little technical understanding. Maybe the site makes money by doing things their advertisers want, which may be different from what is good for you.

The first page of the AnandTech article says, "We then ran several exhaustive tests on these pow

Re:Think carefully about paying more. by gotak · 2003-08-11 10:28 · Score: 1

I think you are right. Bit errors resulting from PSUs? What kind of PSUs are those? Not to mention who made the electronics? That's horrible designs if a spike can alter your signal.

One would think they would have filters and quick response voltage regulators on the mobo itself. Hell you can see those voltage regulators.

So what does that leave? Well i suppose if the PSU produce large amount of RF noise that might cause errors. But that'll be one crappy PSU. Of course they didn't measure RF noise. But like you said that would most likely be beyond their skills.

PLEASE NOTE Re:memory faults by Geek+In+Training · 2003-08-11 06:23 · Score: 1

The results of the memory tests are amasing. The MTBF is about an hour on some of the power supplys. I'm not sure If I understand the setup but that is appauling. I expect a MTBF of about 100+ years not an hour.

MTBF means Mean Time Between Failure, NOT Mean Time Between Errors. What you are seeing is not a failure every hour or so, it's a certain number of memory errors during the specified test period. It's not amazing nor appalling that any of these supplies have the system generating an occaisional memory error; they happen very frequenty, hence why you have ECC and parity in many levels. We're not talking about situations where one bad bit causes a blue screen or kernel panic here.

As a personal opinion, I think this was an early comment which jumped in to garner moderation without really understanding the information that was being commented on. (Nods to the next three posters who are about to reply, "This is slashdot, what did you expect?") I expect when someone posts incorrect or inproper analysis in comments, that they get moderated appropriately. :)

--
SlashSigTheorem: Humorous, Political, Critical, Constructive- If you have a .sig, someone WILL complai

Re:PLEASE NOTE Re:memory faults by brejc8 · 2003-08-11 07:55 · Score: 2, Interesting

Yes but most computers do not have ECC nor parity on their memory.
ECC might protect you and parity will simply allow you to crash out safely.
Unless it was dont care memory this error would cause a failure of some form.
I have run memory tests with quite dodgey memory, power supplies, fans etc. over periods of many hours and these errors do normaly not occur.
Unless you went out to buy eec memory for your computer because you are scared these sort of error might occur over the lifetime of your machine then you are not protected from them.
I think the case is you dont know what youre talking about. The and coming along and looking like a fool stating that the moderators are stupid for also being cauteous of the results presented doesnt help. You seem to be very angry that a public forum is composed of people expressing their "wrong" opinions but at the same time you are happy to troll away on them. You you dont agree then dont read it. Unless you spent several hours working out what "amasing" means then maybe you should lay of the coffee.

--
Mouse powered Chips, Open source Processors and Lego
Re:PLEASE NOTE Re:memory faults by mamba-mamba · 2003-08-11 12:20 · Score: 1

[memory errors] happen very frequenty, hence why you have ECC and parity in many levels. We're not talking about situations where one bad bit causes a blue screen or kernel panic here.
I used to design Intel-based Compact PCI computers for a big company. We would definitely not consider memory errors to be normal or acceptable. Also, by default the BIOS would configure the chipset so that if the ECC circuitry detected an error, it would assert the system error signal, which, in Windows 2000 does blue-screen.
If a board couldn't run through memory tests reliably, or if it generated memory parity errors during Windows Hardware Compatibility (HCT) testing, it was considered to be a broken design. We ran the HCT for 8 hours at a time while cycling the ambient temperature between zero and 50 C. We also varied clock speed and power supply voltages. None of these things caused memory errors.
So, what is your basis for saying that occasional memory errors are acceptable?
MM
--

--
By including this sig, the copyright holders of this work or collection unreservedly place it in the public domain.
Re:PLEASE NOTE Re:memory faults by Anonymous Coward · 2003-08-11 13:05 · Score: 0

No, you fucktard, I'm quite angry that you claim to speak english, and yet you can't spell worth a second grader.

It's spelled AMAZING, lay OFF the coffee, DOESN'T, CAUTIOUS, DODGY, NORMALLY, APPALLING... and there are probably several other errors I've missed.
Re:PLEASE NOTE Re:memory faults by Beliskner · 2003-08-13 10:38 · Score: 1

So, what is your basis for saying that occasional memory errors are acceptable?
Sir, do you disagree with Anadtech's lowest error rate of 1 bit-flip in an 8 hour run as documented here. Anandtech seems to imply that one bit-flip in an 8 hour period is an excellent result. Please comment.

--
A caveman dreams of being us, the incalculable power and riches. We dream of being Q, then what?
Re:PLEASE NOTE Re:memory faults by mamba-mamba · 2003-08-13 11:22 · Score: 1

So, what is your basis for saying that occasional memory errors are acceptable?
Sir, do you disagree with Anadtech's lowest error rate of 1 bit-flip in an 8 hour run as documented here. Anandtech seems to imply that one bit-flip in an 8 hour period is an excellent result. Please comment. [Href omitted]
I am not accusing anandtech of fraud. I am saying that the prevailing attitude in the industry, at least for high-end Compact PCI boards, is not one of acceptance toward memory errors.
I submit to you that it is possible that the board or the memory that Anandtech used for their test is of inferior design. They quite rightly used the same board and memory for all their tests, and it may just be a junker. On the other hand, it may be hard to get a quality board to fail, so maybe using a junker is the way to go.
It should be noted that what Anandtech did is check the long-term persistence of memory. This is not a test that I performed on my boards, and I can't say whether they would pass Anandtech's test or not. But any board that can't hold its memory for six hours is defective, in my opinion, unless the voltage passes outside specified margins.
MM
--

--
By including this sig, the copyright holders of this work or collection unreservedly place it in the public domain.

Not only are power supplies taken for granted... by vudufixit · 2003-08-11 06:41 · Score: 1

But so is the electrical grid itself. Imagine how difficult computing life will get if rolling blackouts/brownouts strike all of this country. It could happen sooner than later if our national energy policy doesn't start tilting away from nonrenewables.

Antec seems expensive and limited to me. by Futurepower(R) · 2003-08-11 06:54 · Score: 1

Antec does NOT make power supplies. The buy them from a Chinese manufacturer, I forget who. I have never liked Antec supplies, they seem very expensive and often limited in features.

Try KingWin supplies. Read about them, and other power supply issues, in my other post, Think carefully about paying more.

Antec supplies seem extremely expensive to me. by Futurepower(R) · 2003-08-11 07:01 · Score: 1

"BUT the power supply had to a very good one ... i.e. ANTEC!"

Antec supplies seem extremely expensive to me. I've never noted any extra value.

Hours repairing a system thanks to a cheap PSU by Goldenhawk · 2003-08-11 07:29 · Score: 1

> Who, among home users, just buys the cheapest PSU they can find with enough power? Or is it just me?

Sorry to burst your bubble, but I *HOPE* it's just you.

For the last YEAR I've been fighting all kinds of system instability, basically ever since I upgraded to a high-speed Athlon motherboard. I naturally thought it was mobo related. I was having crashes about every 10 minutes of heavy dual-drive activity (any time I was copying large files from CD-to-drive, or drive-to-drive, and it happened in both DOS and from Windows, so I was certain it wasn't software-related. I was also having boot problems - about half the time, on power-up, nothing would happen - requiring a power cycle. I did enough other tests to prove to myself it wasn't memory related, etc. Migrating to WinXP helped (Win98 was a nightmare) but it didn't stop. Backup was a breath-holding nightmare, since it was disk-drive related.

The worst part was that since it was disk-related, the crashes occurred a few times when a FAT-table write was occuring, and twice I had to rebuild my system. Thank God for recent backups. Finally I bought a new mobo, kept the same processor, and things improved instantly - at least, so I thought.

I should also mention that I'd lost two hard drives in the last couple years - to no-kidding hardware failure. Strange that both were with this same case... too bad it wasn't just chance.

After a few weeks of relative calm, with my new mobo, I thought I had the problem licked, except the startup problems (no big deal, since I leave the system running 24/7). Everything looked stable. Hurray! But then WinXP locked up again, in the middle of a drive write, and I lost everything AGAIN. Thank God for recent backups AGAIN. (Lesson learned... buy an external USB hard drive, and USE IT daily... even with a XCOPY32 c:*.* d: /e /d batch file if you're too cheap to buy backup software.)

Thanks to Slashdot, thus, for the article a few days ago about power supply ratings. It got me thinking, and as I started doing some research, discovered that many of the problems (unexplained crashes, hardware failures, startup problems, etc.) could be directly related to a cheap or underrated PSU. The core of my problem was the upgrade. The two-year old case with the 250W or 300W power supply may have been fine for my Pentium 2, but putting a hot, hungry Athlon into the same case was a real minefield. As I researched I ran into many comments about this very problem - the Athlon, and just as much the fastest Intel chips, are very power-hungry, and when combined with disk-head motors, the power draw can be just enough to spike the processor or memory at a critical time.

After much shopping research, I just today forked over $90 for an Antec TruePower 430 at my local shop, not a bad price for the service and no shipping charges. I selected this unit on its independent rail power for each voltage, plus its temp-controlled case-fan plugs, and although I haven't installed it yet to prove it's the fix I want, I'm completely confident it'll do the trick.

I'm hopeful that this $90 upgrade will prevent the hundreds of hours of hardware repairs, drive swaps, and software reinstallation I've suffered over the last year or two. It's a small price to pay, given my hourly wage.

Go and do thou likewise.

--
--Brandon / Split Infinity Music

Quiet PSUs by Anonymous Coward · 2003-08-11 07:41 · Score: 0

For anyone looking for quiet parts, the Silenx PSUs are rumored to perform very well both voltage- and noise-wise. Google found reviews here and here.

Reliable comments: by Futurepower(R) · 2003-08-11 08:13 · Score: 1

Be careful. The AnandTech article and the comments to the Slashdot article are full of mis-information. Here are reliable comments:

Not wanting to put a downer on things..., #6664769 (Poster says he runs a review lab.)

Design Engineer, Power Supplies, Computers, #6665630

Bad power supply tests, #6666617

and my own comment. Read why I think the memory tests are not reality, but a sales tool: Think carefully about paying more, #6667576

Re:Design Engineer, Power Supplies, Computers-Outs by fataugie · 2003-08-11 09:17 · Score: 1

IIRC, the original ATX spec was to have the power supply suck in the air and blow it on the CPU. It had the secondary affect of "pressurizing" the case to reduce the dust infiltration. The design sucked (no pun intended).

My first homebuilt was an Asus P5-A with an Enlight case and powersupply. The supply sucked the air from outside and blew it on the CPU and did nothing for cooling. A popular mod at the time was to open the power supply and reverse the fans.

--

WTF? Over?

Sick and tired by gotak · 2003-08-11 10:11 · Score: 1

I am pretty sick and tired of reviews like this being made such a fuss over.

I don't care if you review 5 mobos to see which ones gives you 2fps better in Quake. I just want a stable and long life mobo.

I don't really care which power supply can drive more juice as long as my computer works with the one I have and the PSU can handle being on 24/7 for 3 to 4 years. Where's the review about that?

I am complaining like this because recently tom's site had a column titled "Will Open Source Save Hardware?" (http://www.tomshardware.com/column/20030704/index .html) where the author talks about hardware enthusiasts and their trouble keeping up the drive for "bigger, better and faster".

Grow up why don't you! What's "bigger, better, and faster" used for anyhow? Games! I for one no longer have that urge to play games all the time. It's a waste of my time that I could better make ranting post on slashdot (LOL) or better yet actually do well in my Computer engineering degree so I can one day build things I go "WOW!" at.

I am sick and tired of the people who keep trying to rope people into the forever upgrading path. I just saw my friend sitting in front of his computer watching is Star Wars MMOPG charactor "camp" so he can gain a level in the camping skill. This is a guy who failed 3 courses last team and now has an extra year to do. This is the guy who has 2 retry exam is in 3 weeks. This is the guy who has 7 fans in his computer and a 500 watt PSU just so he can have a 2 degree drop in his CPU's temp.

So please spare me the stupidity and attempts at making commodity PC hardware exciting. And please spare me the cases people spent months modding (dude with that 200 dollars in lights and shit you could have gotten more ram).

But hey if running 20000 fps in quake 3 floats you boat ignore me. I'll just go back to programming this 4 Mhz MCU here. It's more fun for me.

Arghh! by Anonymous Coward · 2003-08-11 21:24 · Score: 0

That was meant to be posted anonymously!

196 comments