DO NOT skimp out on a power supply EVER

[hat]

Here's proof...
My old power supply really sucked. I'm sure you know, I bitched about it alot. Here's a good example. My RAM now does DDR600 4-4-4-11-15-1T at 2v, my old power supply couldn't do 4-4-4-12-16-2T at 2.3v.

[DanTheBanjoman]

Never buy expensive motherboard. Mine died.

Pretty harsh to conclude such things based on a single incident. I'm currently using a cheap 350W PSU and it works fine. Besides, you're basing your opinion on running stuff out of spec. That's like saying "don't buy a processor, it won't do 10GHz".

[tkpenalty]

Never buy expensive motherboard. Mine died.

Pretty harsh to conclude such things based on a single incident. I'm currently using a cheap 350W PSU and it works fine. Besides, you're basing your opinion on running stuff out of spec. That's like saying "don't buy a processor, it won't do 10GHz".

"Never buy gigabyte, my one had a broken heatsink"

Agree with you Dan... comments like these are plainly stupid. One buyer gets a bad experience, and that's all it takes to make them not get the same thing again.

[hat]

I'm just re-stating the fact that good power supplies make a big difference.

[DanTheBanjoman]

I'm just re-stating the fact that good power supplies make a big difference.

Big difference? You claim the difference in memory speed is actually something you notice in every day use?

Besides, the point remains, you base everything on a single PSU. Perhaps it was defective? Perhaps it was too weak? If I'd connect my normal rig to any 350W PSU it'll run like shit, things might even die. Does this mean the PSU is crap all of a sudden? You're not saying what your old power supply is, or what the rest of the system is. Nor the new PSU. So even if it isn't defective your "proof" doesn't really prove anything.

[btarunr]

I don't get it, how can a PSU affect memory timings anyway? :confused:
It's something that draws less than 10W / module. Maybe there's something else. I remember once I was on an OC spree and hit a "limit" the POST beep (memory error). All I had to do was to fiddle with it and it worked.

The point I'm making is that such things are but isolated cases that can't be generalised and communicated as axioms.

[Ripper3]

I've had a bad experience with Creative products from thsi oh... first ever Creative product I have laid my hands upon.
Doesn't mean I'll never buy a Creative product again, just means I'll be more choosy next time around.
I only listen to other people's problems and take away from them when they're not the only people having problems. If that problem is somehow fixed, or turns out to be caused by something unrelated, etc. I learn from it.

Oh, and just to add, I've had (and in fact still have) a Q-Tec 400W PSU, runs cool, quiet, and with stable voltages, and has been doing so since I bought it back in 2005. It's not a big brand name, it's not high powered, and it's nothing special, but it's served me well.
As others have said, it just have been that your old PSU was faulty.

Oh, and bta, maybe it wasn't actually supplying the full 2.3v to the RAM. vDroop or something.

[JC316]

Never buy expensive motherboard. Mine died.

Pretty harsh to conclude such things based on a single incident. I'm currently using a cheap 350W PSU and it works fine. Besides, you're basing your opinion on running stuff out of spec. That's like saying "don't buy a processor, it won't do 10GHz".

Agreed there, I have a 300W power supply that I have used since 02, it came with my case. Still powers 2 case fans, an overclocked Athlon XP 2200+, 1GB ram, 2 hard drives and a Geforce FX 5900 that recommends a 350W supply.

[Namslas90]

I don't get it, how can a PSU affect memory timings anyway? :confused:
It's something that draws less than 10W / module. Maybe there's something else. I remember once I was on an OC spree and hit a "limit" the POST beep (memory error). All I had to do was to fiddle with it and it worked.

The point I'm making is that such things are but isolated cases that can't be generalised and communicated as axioms.

The stability of the power supplied can effect everything!

:toast:

[Woah Mama!]

I've never gone cheap for a power supply and never will. Why risk basically every single component of your computer because of a dirty psu current that could spike at any given time and blow your motherboard or something else.

Building a kick arse rig and then buying a cheap power supply doesnt make sense to me, no sense at all.

[3991vhtes]

My one system has a Generic PowerMan 235watt PSU. It's powered an old PIII system since '01.

[zCexVe]

Well It might be a random incident.You need more situations of the same thing.Like this.
Me and two friends bought manhatten HDD coolers with two fans each costed 8 USD with the 10% discount coz bought 3.Its all noisy starting up then it settles down but bit of bouncy like.Vibrating.
I had my Hitachi 250GB SATA never moved away from my case sinse it have all my data,and it died after 10 months.one of my frnds had a Seagate 120GB SATA it died after 11 months later after fixing the cooler.The other one had his 160GB Maxtor SATA died 8 months after fixing the cooler.Luckily two of us got recovered our data coz the HDDs gave some errors and sounds.One lost his 120GB data.All hards died coz of platter errors.Now we are avoiding those coolers.And I sold my one to a friend for cheap(He loved the lifetime warranty)Its 3 months and I'm waiting till he says it died.If so Im damn sure thats the cooler.A little thing can cause a havok.