good airflow will improve the temp a lot..thats my past experience..
btw..may u state what PSU u are using?

This post has been edited by serigala: Nov 27 2008, 06:49 PM

WRONG. you can buy the same wattage PSU with 80PLUS certification, meaning more than 80% efficiency ...meaning true power la.
need to know your cpu also..c2d enuf already.. 560w is enough already..

This post has been edited by alpha_company: Nov 27 2008, 07:13 PM

okay. My PSU is 560W SilverStone Strider, it says certified by ati.

80%+ efficiency, dual 12v rails,dual pcie connectors, support for ATX 12V 2.2 & EPS 12V, active PFC

if tak cukup juice i think mine tak cukup 1st...

thats a weak 12V rail u have there...mind state how many HDD u have and how many fans connected?
for HD4870,recomended u use SLI/Crossfire certified PSU..

This post has been edited by serigala: Nov 27 2008, 07:33 PM

all strider psu's are weak

get single rail op's non-modular to power ure comps..

what?! idle 34 celcius + in CM690 ?

y my 4850gs is 40 degree + ....
(unless it is out from casing then 30+ )

did u change the fan on the side panel or having non traparent side panel ?

This post has been edited by angelic88: Nov 27 2008, 08:25 PM

I'm using SilverStone Strider ST50F 500w, can juice my PC 24/7 eventho i OCed my proc, and gc, and my temps are also normal

you have a point. the CPU does play a part. LOL ...but u didnt say why it shutdown?

to be honest, to cause a reboot or shutdown in the middle of gaming, it's either the psu or the heat. (i believe so)

the board can stand up to at least 100C, so i would call it the psu problem..

A psu with high wattage but low 12+v rails ampere could cause this.

this was discussed in the Gainward V1 thread.

so b4 you decide anything bro, run a stability test with furmark. if you comp didnt shutdown in the after 20 mins.

then it shud be related to other things already. and take a snapshot of the test too.

may i ask how long should i stability test with furmark to justify it should be "ok"?

now that's your prob there.. store people will say the silverstone good, cheap efficient and all but it's on a DUAL rail. they just want to clear stock..that's all.. some people will say the psu nice and everything, those are the lucky ones... u don't need to take chances on these stuff..


I got myself a SINGLE 12V rail Corsair VX550W, after doing extensive reading on psu's. My corsair 5 years warranty very much worth it.

From what i know a DUAL 12v rail psu are either impossible or problematic to use with Crossfire and SLI setups. Only single, triple and quad can handle those setups easily.

I recommend getting a SINGLE 12v rail -or- TRIPLE OR QUAD 12V rail psu's. Read the specs first.
You can get a modular PSU, but it'll be more expensive and it has weaknesses in heat performance and lower lifetime compared to non-modular psu because the of the extra wiring internally.

---

Added on November 27, 2008, 9:13 pmyou're talking about me right?

i got the cm690, 4850GS is around 33-35 celcius.
i got the transparent side panel, and didnt change the fan. (imagine if i got the non transparent one, it'll be wayy cooler.)

two 140mm blue on top (exhaust)
one 120mm at the back (exhaust)
one 120mm blue (90cfm) in the bottom (exhaust)

one 80x15mm at the back of the mobo (intake I think)
one 120mm on the side panel (intake)
one 120mm blue in the front (intake)


This post has been edited by alpha_company: Nov 27 2008, 09:13 PM

Ok, guys, i just did the stability test, and i was NOT happy.
Initial temps of my gfx card and CPU right after startup, note the fan is already 70%



But, i was running the stability test, just as i was going to take the initial screenshots in the first 2 minutes, the computer shutdown. All 4 cores of my cpu were at 60 > c, whereas before i upgraded to the HD4870GS 1gb, the highest temps i whad were 55c.

anyway, the furmark thing recorded my gpu temp to be at 78c before it crashed. hmm... so is it the PSU problem or is the problem with the casing? My casing is the HP m8190d one, quite some stuff, and pretty compact. One fan on the case.

hope its not the PSU....

ps: im not running crossfire, just 1 GPU

This post has been edited by ruffdawg: Nov 27 2008, 09:19 PM

The gpu and cpu won't just shut down at those temps. It's still within acceptable limits and won't cause shutdowns. Yes you're not doing CF but doesn't it makes sense that if a psu can't handle CF & SLI, it's not really that good. I'll have a clear conscience knowing my psu can handle SLI and CF setups.

I'm not a genius but if the temps aren't the cause, something else is and i do suspect the psu.

Ram probs? nop, it'll just bsod. won't just get into windows and cause a shutdown

This post has been edited by alpha_company: Nov 27 2008, 09:48 PM

shutdown? i think psu is the problem

quote from guru3d

There are many good PSU's available, over the years we reviewed a lot of them and have loads of recommended PSU's for you to check out in there, have a look. Things that can happen if your PSU can't cope with the load?:

* bad 3D performance
* crashing games
* spontaneous reset or imminent shutdown of the PC
* freezes during gameplay
* PSU overload can cause it to break down

http://www.guru3d.com/article/palit-revolu...870-x2-review/5

This post has been edited by pikacu: Nov 27 2008, 09:34 PM

Hey, i don't know if to trust the reviews, it seems my PSU is not too horrible. Now im playing open case, however, temps are still above average, and running at 70% fan speed under load is 70c.

http://techgage.com/article/silverstone_st...t56f_560w_psu/5
http://www.hardwaresecrets.com/article/548/10

This post has been edited by ruffdawg: Nov 27 2008, 10:34 PM

err, i do mean this one : * spontaneous reset or imminent shutdown of the PC

and that is ur pc problem nao rite? so it may leads to ur psu..

Most probably is the PSU. Im facing this kinda problem when using vantec 450w. But after i changed to ST56f, everything smooth....

Right now, with the lowered temps i have not experienced a restart. I'm still trying to find out the main cause. ohh well, till the next crash.

@alpha company
my card got problem. the display driver keep on crashing and BSOD on 8x AA. using 22inch samsung 226BW 1680X1050.

@scanart
i heard from CEX that u sending my card to taiwan? when i getting it back... i tot u guys gonna flash the bios? so what is the problem with it. i need a card back as soon as possible ler... my desktop now only temporary borrowed 6600GT which i need to give back soon... my friend asking for it already...... pls reply as soon as possible..
and... what is best time i can contact carlson coz just now when i heard from CEX is already night so i wonder should i call or not since it's after office hour.

and for those who wanna use CM690 case for 4870GS.. remove the back plate for the PCI device at the back... remove the mesh of the side panel of the case and have the fan exaust out the air... without the mesh the airflow is faster. 4870 hot air stray out from the heatpipe area. i used to use CM690 and now i am on HAF932 full tower which now i have 4 12cm fan on the side panel but have one of them on exaust and partition them with cardboard. those with 4870gs should have the fan speed fixed to 65% on the expert tool or CCC. barely noticeable noise increase. will keep ur card cool. mine was at 49-52c Idle at 50%. at 65%(1500+ rpm) idle at around 44C. but even on 50% fan speed (500+ rpm) mine never exceed 70C on full load. 65% fan speed best compromise for heat and noise..
you have to open up the expertool fan control window everytime u reboot in order for then fan speed to pick up even after u fix at 65%. or else it will stay on 50%. I never try fix the setting on CCC yet but u guys could try.

ONLY on 8xAA crash? Any specific game that it crased on?

It seems that it may be a game driver problem.