Will this psu limit oc'ing

[kdream3]

So this is my upcoming rig:

A64 3200+ 939 Rev E
2x512 Ocz vx pc4000
bba x800xl
dfi nf4 ultra d
80gb hd
cdrw/dvdrom

So im planning on getting a 450w ocz modstream. Will i have any limitations oc'ing my cpu with this psu. thanks.

 

[jonnyGURU]

Absolutely not.

You do realize that the Modstream 450W has 28A @ 12V and 45A @ 5V, right?

 

[kdream3]

I dont really know much about psu's except that watts dont determine the quality of a good psu. But I was just asking because I heard some people saying bad things about modular psu's and how to get the powerstream over the modstreams. But anyways thanks for the reply.

 

[Oklahoma Wolf]

Some modulars are better than others. This is one of those ones that's better than others. You'll be fine with it

 

[Vertigo Acid]

Yep, compared to Ultra, the Modstreams are great. I'd still prefer an Antec Neo, however

 

[jonnyGURU]

Yep, compared to Ultra, the Modstreams are great. I'd still prefer an Antec Neo, however

Actually... I use the Ultra, thank you very much.

ALL modulars are going to have the same issue: Voltage drop due to the resistance created by the connectors. But this resistance is minimal. Like .6 milli-Ohms versus .65. Take a look at this thread: http://hardforum.com/showthread.php?t=895505

As for the Antec... eh... I'm still not gungo ho on dual 12V rails, yet. And the Neo is so freaking ugly, the only advantage to the modular design is the air flow improvement. Wouldn't put it in a window case. At least the Modstream looks somewhat cool (that is.. If you're into bling. )

IF you decide to get a Powerstream over a Modstream because you don't "need" modular cables, don't go with the 520ADJ. The 420ADJ gives you just as much usable power.

 

[Vertigo Acid]

Actually... I use the Ultra, thank you very much.

ALL modulars are going to have the same issue: Voltage drop due to the resistance created by the connectors. But this resistance is minimal. Like .6 milli-Ohms versus .65. Take a look at this thread: http://hardforum.com/showthread.php?t=895505

As for the Antec... eh... I'm still not gungo ho on dual 12V rails, yet. And the Neo is so freaking ugly, the only advantage to the modular design is the air flow improvement. Wouldn't put it in a window case. At least the Modstream looks somewhat cool (that is.. If you're into bling. )

IF you decide to get a Powerstream over a Modstream because you don't "need" modular cables, don't go with the 520ADJ. The 420ADJ gives you just as much usable power.

Heh, what about a window case like mine where the PSU isn't visable behind the metal covering
Anyways, why in god's name would you willingly submit your components to an Ultra PSU?!
They're like chinese fireworks: Build cheap and end with a bang!
Honestly, I hope you're joking

 

[jonnyGURU]

Anyways, why in god's name would you willingly submit your components to an Ultra PSU?!
They're like chinese fireworks: Build cheap and end with a bang!
Honestly, I hope you're joking

I'm not joking. And you're basing your comments on personal experience?

All of this has been said before, but I guess I have to say it again... sigh.. I have two of the X-Connects and one X-Finity, oldest one coming up on a year and have had absolutely no issues. Furthermore, the last two VAR's I've worked for has sold the product and have had no higher RMA rate on their product than other brands we've sold (about 2%.)

And that's all I want to discuss on the matter. I knew before signing up for the second time here that [H]ard|OCP is an Ultra bash-fest. Something I feel is unsubstantiated, but also something I don't care to get involved in (guess it's too late for that, eh?)

 

[Vertigo Acid]

I'm not joking. And you're basing your comments on personal experience?

All of this has been said before, but I guess I have to say it again... sigh.. I have two of the X-Connects and one X-Finity, oldest one coming up on a year and have had absolutely no issues. Furthermore, the last two VAR's I've worked for has sold the product and have had no higher RMA rate on their product than other brands we've sold (about 2%.)

And that's all I want to discuss on the matter. I knew before signing up for the second time here that [H]ard|OCP is an Ultra bash-fest. Something I feel is unsubstantiated, but also something I don't care to get involved in (guess it's too late for that, eh?)

Personal experience is *not* god when it comes to reliability. I'm sure there are people out there that happily drive their pintos and yugos to work everyday and have never had a problem, just as I have a topless Deer that runs a dual P3 server on occasion. Just because I haven't set it on fire doesn't mean that it is built well. As for RMAs, RMAs != Failure Rate, especially when you only have a 1 year warrenty, and quite possibly the people who buy these supplies aren't going to push them like some of us do.
See http://silentpcreview.com/article193-page2.html
All i'm saying is that I'd rather have a PSU built like a tank or close to it *protecting* my investment. Maybe it's not a big deal to you that your PSU puts out more noise and has a nice .5v overvolt on the +12v rail by default. But that's not good enough for me.

 

[Irishllama]

I'm not joking. And you're basing your comments on personal experience?

All of this has been said before, but I guess I have to say it again... sigh.. I have two of the X-Connects and one X-Finity, oldest one coming up on a year and have had absolutely no issues. Furthermore, the last two VAR's I've worked for has sold the product and have had no higher RMA rate on their product than other brands we've sold (about 2%.)

And that's all I want to discuss on the matter. I knew before signing up for the second time here that [H]ard|OCP is an Ultra bash-fest. Something I feel is unsubstantiated, but also something I don't care to get involved in (guess it's too late for that, eh?)

Why Did you sign up for the second time,eh? Just wondering.

 

[jonnyGURU]

See http://silentpcreview.com/article193-page2.html
All i'm saying is that I'd rather have a PSU built like a tank or close to it *protecting* my investment. Maybe it's not a big deal to you that your PSU puts out more noise and has a nice .5v overvolt on the +12v rail by default. But that's not good enough for me.

I'm well aware of Silent PC Review... Thanks for the link. It doesn't tell me anything I don't already know and their findings are circumstantial at best.

Yes, it is a bit noisy. But that doesn't bother me as my PC's aren't any where they need to be exceptionally quiet. Furthermore, the "nice .5V overvolt" only happens when you load the 5V up more than the 12V as if you were running power supply in a non-ATX12V environment. Are you meaning to tell me that those of us [H]ard enough to pass the X-Connect by are also using Athlon XP 2000+ CPU's on Micro ATX motherboards?

The same "problems" actually occur with most power supplies where the 12V rail is more capable than the 5V rail. Sometimes, the effect is a bit more dramatic than a "nice .5V overvolt." But until a review site can break down what "400W" is on one power supply they review (in terms of how many amps on the 5V, 12V, etc.) versus another, I will take such reviews and their results with a grain of salt.

Good choice on the PC Power and Cooling Turbo-Cool, by the way. I certainly can't fault you for your power supply choice at all. Even with the steep price tag, the Turbo-Cool gets a 9.5 in my book.

Why Did you sign up for the second time,eh? Just wondering.

After the great membership flush of 2004, I had quit coming around. But then Ice Czar linked up my power supply reviews so I decided to come back around and see how things were doing.

 

[jonnyGURU]

I'm going to go ahead and give you some solid numbers so you can understand better where I think the "nice .5V overvolt on the 12V rail by default" is pretty lame....

http://www.slcentral.com/ultra-x-connect-500w-titanium-psu/page6.php

Test one has 20A on both the 5V and 12V rails. Not a very likely scenario, but it shows that IF the rails were evenly loaded, you would be within spec.

Test three (I'll get back to two) put the 12V well over the 5V. 30A and 20A to be exact. This could represent most CURRENT builds. A Prescott, an A64. Throw some 6800 Ultra cards in there if you wish. If you take a look at the Full Load test, you'll see the more the merrier. The more stress I put on the 12V rail, the better the PSU performed (that's 34A on the 12V rail if you care to know.)

Test two represents a Pentium III or non-ATX12V socket A board. Something that doesn't regulate CPU voltage off of the 12V rail. In that test, the 12V is only loaded up to 10A (fans, lights, drive motors, etc.) while there's a whopping 30A on the 5V. Notice how the 12V is well above spec. Fortunately, nothing that's going to get harmed by a 12V that far over spec is actually running off the 12V rail, so IMHO the result is dust in the wind. I'd rather have that than have my voltages bouncing around by as much as 2%, even if they're within spec.

Oleg over at X-Bit puts it so nicely in a color graphic:

http://www.xbitlabs.com/articles/other/display/powerful-psu_11.html

Note the 520ADJ graph when the 5V goes over the 12V.

Here's a review of an X-Connect that's so horribly done, it made beer shoot out of my nose:

http://techgage.com/review.php?id=801

Not that you would ever give this a grain-of-salt of validity, but look what he's using for a test rig:

"AMD 2600+ @ 2.2GHz, 1024MB DDR400, MSI KT4V, MSI Radeon 9800 Pro (420/380), 200GB * 2 WD, 160GB WD, Lite-On 8x DVD+RW, Phillips PSC724, D-Link NIC, 4 system Fans, excluding PSU and CPU."

He's probably got all of 8A load on the 12V rail on a power supply that is capable of delivering 34A and he's wondering why the 12V was at 13.13V?

In all fairness to you and your choice of PSU, I present the Turbo-Cool 510:

http://www.slcentral.com/pc-power-and-cooling-turbo-cool-510-express/page5.php

Truly an outstanding specimen and no doubt superior in every way. Note that regardless of the load, the 12V never waivers more than .08V! Awesome craftsmanship there. No doubt.

So clearly, it's not EVERY power supply that goes haywire when the 5V is over the 12V. Take another look at Oleg's graphs and you'll see the 470ADJ is quite stable all across the board. I find the same to be true with all of the < 520 OCZ power supplies. For whatever reason, OCZ switches to whatever MOSFET they use to get that extra 1A on the 12V rail and all hell breaks loose!

The methodology is better details is here, if you're interested: http://www.slcentral.com/raidmax-rx-520xp-psu/page3.php

 

[jonnyGURU]

And I'm not sure what this facination is with power supplies that blow up, either.

I've read two reviews where the power supply I use blew up and only one (MaximumPC) could tell me what they were doing with the power supply WHEN it blew up.

Take Thermaltake for example. Can we all agree that the PurePower is a lump of crap or is someone going to come in here and defend it. Anyone?

I just reviewed my second one today. The first one was the "Butterfly" with all of it's pretty lights and the second one is the "TWV" model.

The first one blew up when I juiced it up above spec. YES, I juiced it up above spec, but isn't overload protection supposed to kick in?

The second one didn't blow up, but it didn't do it's supposed "Max Load" either. And the 12V on THAT power supply was so out of whack when the 5V was juiced up, it made the Ultra look like it could operate at military spec! What's sad about that is the PurePower's are of "old school" architecture. A mere 18A available on the 12V rail and a whopping 30A on the 5V rail, but if I put 30A on the 5V rail, my 12V goes from 11.79 to 12.34?!?! THAT makes a whole hell of a lot of sense.

 

[Vertigo Acid]

But indeed you've just proven my point; a PSU should be able to regulate all voltage lines independently of eachother, regardless of the load presented, as long as there is some load.

As you note about SilentPC Review's findings being "circumstantial at best", I would respond that so are you personal experiences.

Not trying to attack you directly at all; just sick and tired of people recommending PSUs based not on hard facts and accurate numbers, but anecdotal evidence and hearsay. If the Ultra line of PSUs can not keep within spec during foreseeable load conditions, then it should not be recommended!

BTW, the links you provided were great. One thing that I immediatly thought when I read that was, what about times when you've got other components drawing +5v power, but your CPU is idle, and/or your GPU is idle? Wouldn't that present as a possible scenerio when +5v draw would outpace +12v draw?

 

[jonnyGURU]

But indeed you've just proven my point; a PSU should be able to regulate all voltage lines independently of eachother, regardless of the load presented, as long as there is some load.

Absolutely. That is why the PC Power and Cooling is such a great power supply... Alas, it is pretty much THE ONLY power supply that has stable rails on uneven loads. It's just the nature of switching power supplies for rails to be interactive. The alternative to making a tank of a PSU like the Turbo-Cool is to have seperate rectifiers as well as seperate transistors for each rail. Although this may cost less than a Turbo-Cool in the grand scheme of things, it's not very practical size wise.

As you note about SilentPC Review's findings being "circumstantial at best", I would respond that so are you personal experiences.

Touche. But on the flip side of that, and I think you'll agree, I would say that review sites that give GOOD reviews are also circumstantial at best as well. Right?

I would like to think that my experience is LESS THAN circumstantial than most sites where you have kids sampling one or two units because I actually work in a retail environment and am exposed to a large customer base and their feedback. Note my FSP review. Even though that power supply blew up, I still graded on a curve as if the power supply didn't blow up at all. Why? Because, despite my review sample, I send out about 20 FSP power supplies a week and never have an incident.

That's why I was willing to give the Thermaltake the benefit of the doubt the first time around, but less so now that I've had a second.

And the A.C. Ryan didn't even get that kind of respect from me simply because the construction of it was similar to something my daughter made when let her play with Lego's while eating a peanut butter sandwich.

Not trying to attack you directly at all; just sick and tired of people recommending PSUs based not on hard facts and accurate numbers, but anecdotal evidence and hearsay. If the Ultra line of PSUs can not keep within spec during foreseeable load conditions, then it should not be recommended!

You're cool. You didn't come off as attacking at all. I don't "recommend" any power supply, but I also don't recommend against any "reasonable" choice.

Unfortunately, power supply choices aren't as cut and dry as Pepsi versus Coke. There's a lot of PERSONAL PREFERENCE variables involved. That's why I laugh whenever someone posts "What's the best power supply for this build?" They end up with no less than a dozen different responses. Then they come back with "I have an under $100 budget." Now a different dozen are recommended.

Low noise, to me, is as important or unimportant as "bling." What do either do for your computer's functionality? Nada. So you have budget, noise, bling, a little of both, neither... In my case, I wanted a power supply that looked nice in a window case with a capable 12V rail. That was MY needs. I'm not even going to try to justify or impose that on anybody else.

BTW, the links you provided were great. One thing that I immediatly thought when I read that was, what about times when you've got other components drawing +5v power, but your CPU is idle, and/or your GPU is idle? Wouldn't that present as a possible scenerio when +5v draw would outpace +12v draw?

Not likely in an ATX12V environment, because the only thing that using the 5V in such a build is going to be PCI cards and the circuit boards of the drives. IF the CPU and GPU was "so idle" that the 12V draw would be "that low" the entire wattage requirements of the entire system would also be so low, you'd be below any kind of power curve that would cause an unusual response from the PSU or cause any harm to any of your components.