antec 500he not enough power?

Okasa

Member
Jan 22, 2005
168
0
0
could my antec 500HE not be supplying enough power to my bfg 7900gt? it freezes up a LOT, and not just under load. could my PSU be to blame? here are antecs specs on the amperage:

23A on 3.3v
17A on 5v and 12v
 

SpeedZealot369

Platinum Member
Feb 5, 2006
2,778
1
81
I don't know what that A on V stuff means, but if I would guess I would say it's a bad card. Those cards don't need much power to run, that is unless it's voltmodded or oc heavily. They take like 50w under load.

I think it's definitly a faulty card.

-SZ
 

anthrax

Senior member
Feb 8, 2000
695
3
81
I don't think its your PSU. The Neo HE is triple rail.. 12V 17A . with a combined load of 456W for all 3 12V rails.

Also, i can O/C my x1800xt on a HE500, so I doubt its the powersupply.
 

Bobthelost

Diamond Member
Dec 1, 2005
4,360
0
0
Originally posted by: PurdueRy
That is a VERY weak 5V and 12V rail...yes I would say its likely the problem


Weak 5V rail = Good. ;)

No, i doubt it's the PSU, 7900GTs draw very little power, the entire system could probably be run on a single rail.
 

Aluvus

Platinum Member
Apr 27, 2006
2,913
1
0
Is it freezing up while you're running something intensive like a game?

I would tend to blame heat, power, or video drivers.
 

Okasa

Member
Jan 22, 2005
168
0
0
no, thats the thing, it freezes up during normal non-intensive browsing. it has NOT happened during a 3dmark06 run, or a prime95 run. also, i have a zalman heatsink on the card which dropped the card consistantly under 40C. so not intensive, not too hot. not the ram, did 35memtest runs with zero failure.
 

1N0V471V

Senior member
Mar 13, 2006
410
0
0
Sounds ALOT like a CPU problem.

Memory will normally BSOD and video will either lock up temporarily or artifact like crazy.

Is your system overclocked?
 

Okasa

Member
Jan 22, 2005
168
0
0
well, the ram is at the moment, but it was occuring when the ram was not overclocked. and its not even 'overclocked', i just changed the settings to 4-4-3-12 1t and upped to 2.1V. why would a bad cpu cause this problem? lol i dont even know where to start to troubleshoot that, it sqrewing up could make the comp do anything...tho it is a new AM2 chipset, so its possible they still have bugs.
 

1N0V471V

Senior member
Mar 13, 2006
410
0
0
How long have these lock ups been occuring?
I would get into the BIOS and load the optimized defaults to see what that does.
If lockups still occur, run Prime95 blend test for at least 6 hours (preferrably 24 hours)
If that passes, download Memtest86+ and create a bootable disc. Run that for at least 5 hours (that should suffice)
If for some reason you STILL have lockups, try using a different power supply. I doubt this is your problem but you never know.
 

Okasa

Member
Jan 22, 2005
168
0
0
lock ups have actually been occuring almost since ive had this build. i cant say for certian, because im not 100% on this, but the lockups may have occured once i installed my new zalman heatsink on my graphics card. its their nicest one and it dropped the temps to 37C so i dont think that is the problem. I have tried loading bios defalts, ran prime95 for a few hours (possibly not enough), and ran memtest through 35 runs (which it passed)
 

1N0V471V

Senior member
Mar 13, 2006
410
0
0
Borrow (if possible) a power supply from a friend or go to a computer shop and ask to use on of theirs. See what happens (it sounds like you've got everything tested properly).
 

Okasa

Member
Jan 22, 2005
168
0
0
allright ill see if i can bum my friends' PSu for a few days - only one with enough juice lol. i just wish i knew what i could do to recreate it, i guess it will just be a waiting game. thanks, and ill post with results
 

Okasa

Member
Jan 22, 2005
168
0
0
no, you helped, its quite an odd problem, all my builds have their amazingly horrible quirks lol
 

Bobthelost

Diamond Member
Dec 1, 2005
4,360
0
0
The voltage on the RAM is quite high, try reducing that to stock levels. I was getting crashes on my gaming machine because it was overheating, rather an odd situation because the CPU fan was sitting ontop of the RAM itself, zero airflow :eek:
 

Okasa

Member
Jan 22, 2005
168
0
0
i checked theaccording to the newegg posting on my ram, should be a link a few posts up, my ram voltage is from 1.9v-2.1v, also, i ran it at 'auto' voltage and 5-5-5-12 2t for a long time and it had the same problem. memtest86 passed 35 times while under 2.1v and 4-4-3-12 1t. so i dont think that it could be that.

I just thought of something i forgot to mention actually. i was overclocking my graphics card when i was trying to get the core temp above 40C, and when i overclocked it not horribly, i think it was running 500/700 the screen looked like it was full of dead pixels even when i was not running a game or anything else graphics intensive. is that normal when oc'ing a gfx card? im kinda new to gfx ocing.
 

RebateMonger

Elite Member
Dec 24, 2005
11,586
0
0
Jeepers. There's no reason why an Antec 500 Watt power supply shouldn't be adequate for virtually any PC on earth.

Turn OFF overclocking on EVERYTHING. Then see what happens. If you still have problems, do a memory test, as suggested earlier.
 

Bobthelost

Diamond Member
Dec 1, 2005
4,360
0
0
Little boxes and dead pixels are what are known as "artifacts", just as polygons that don't get drawn in so they're just black. It's a sign that there's something going wrong with the graphics card. If you're overclocking then it's a sign you're overclocking too hard, if it's at stock you've got a broken card and you should get rid of it.
 

Okasa

Member
Jan 22, 2005
168
0
0
as i have said before i have done extensive memory tests, and the problem started occuring before any overclocking was started.

edit: yea its being overclocked, i simply didnt think that i was overclocking enough to make it artifact...does anyone else have a 7900gt? what are you running it at?
 

1N0V471V

Senior member
Mar 13, 2006
410
0
0
650 on the core and 950 on the memory (1.9Ghz effective) -- but mine is volt modded

What are your current clock speeds on the card?
 

Okasa

Member
Jan 22, 2005
168
0
0
pulling from memory (im at work atm) the max mem speed i could get was 710? i think thats right. did you volt mod urs with pencil line? if so, could i get a link to a good site that you know of that shows how to do it. sorry i know this is a side topic lol - if you volt mod with pencil and then have to rma it, if you erase the graphite is there any other way they can detect it?
 

Bobthelost

Diamond Member
Dec 1, 2005
4,360
0
0
If you've got a fault with the card at stock settings (which you seem to) then it's a very stupid idea to overvolt it and throw away your warranty. You might have already done it by overclocking, i don't know about BFG's warranty.

(I wonder if i've been diplomatic enough this time, last time i said that i got three days holliday ;))