furpitas

Honorable
Mar 20, 2012
7
0
10,510
Hey i have a 560 ti sli and it is overheating or getting close to it in bf3 i have seen it at 90 c it doesnt seem like the gpu fan is compensating for heat it stays at 30 % no matter what would it be safe to leave it running at say 70%... Im not sure but it could be the power supply i have a 300w power supply could that cause it to overheat..I know this is way to low for the card.
 
There is no way that a 300W PSU is going to have connections for 2 560ti cards, and there is no way your system would POST if you were really running that kind of config, so you must be mistaken on the ratings somewhere. Having a lack of power can lead to all kinds of wierd problems though.

1) check your power rating again to be sure that it is not 300W on the 12V rails instead of 300W global. If your current power supply is not up to snuff then get a quality 80+ bronze or better power supply that is rated for this type of load (600-700W should do the trick)

2) use MSI afterburner or a similar program to control your fans with a better, but still automatic, thermal profile that will use the fan properly, but if you really are having power control issues it may be running at 30% because that it all it can muster.

3) after making sure 1&2 are properly sorted out then check your system temps as well as the GPU, it may simply be that you are not letting the case breathe properly, or it is blowing hot air out the back, and then recycling it through through a side panel or other intake which would keep things a bit toasty.. Running duel GPUs puts out a fair amount of heat that normal cases are not meant to handle, and it is expected that you have a case designed for a little more than your average airflow for these types of rigs. Always remember, when you want to do something special, then you need special support in order to do it right, so always factor in a little extra money with your future builds to meet the requirements of what you want to do, otherwise you can damage things and spend even more money in replacements than what you attempted to save by doing things the 'cheap' way.
 
1. A 300 watt PSU is woefully inadequate and I am surprised you haven't seen smoke coming out of the system. According to nVidia, one 560 Ti requires a 500 watt PSU.

http://www.geforce.com/Hardware/GPUs/geforce-gtx-560ti/specifications

Maximum Graphics Card Power (W) 170 W
Minimum System Power Requirement (W) 500 W

For two, 500 + 170 = 670 watts.

At the minimum, I would never use anything but a quality 750 watter on twin 560 Ti's. If overclocked versions, of if overclocking yaself, along w/ an OC'd CPU, I wouldn't use anything less than an 850 watter.

2. What 560 Ti is it ? Is it factory OC'd ? Is a "reference" PCB (4 phase VRM) ? Answers to these will be important if OC'ing the cards.

3. In order to form a basis of comparison, use a standardized "torture test" program like OCCT. OCCT will test your CPU / GPU for an hour using a standardized routine, record and graph the results, monitor temps and voltages with out the need for additional utilities and include sound alarms for temperature thresholds so ya don't have to sit there the whole time during a 60 minute test.

http://www.ocbase.com/index.php/download

4. Use Afterburner 2.2.0 beta 15 to do your OC'ing and fan profile

http://www.guru3d.com/news/msi-afterburner-220-beta-15-download/

5. Test case cooling adequacy by running the test a 2nd time with the case side / window removed and a small desk / window fan blowing air across the components. I use these:

http://www.vornado.com/product.aspx?CategoryID=2c76a58c-b875-4db6-9273-9ac300bba659

If temps drop significantly, you have a case cooling problem.

6. By way of comparison, Son No. 3's build has a OC'd 2600k (4.7 Ghz on 24/7 boot profile) and twin 560 Ti's running at 980 Mhz) .... used to be at 1020 but dropped it down to keep temps < 82C in CCT GPU test thru summer months. He's using an Antec DF-85 (all fans at lowest possible rpm settings), Antec CP-850 PSU, Asus 560 Ti CU II TOP (900 Mhz factory OC), Silver Arrow CPU cooler.

The fact that yours are hitting 92C .... presumably at only the "reference" 820 Mhz and when only under gaming conditions, indicates a problem that should be indentified in steps 1 thru 5 above.


 

furpitas

Honorable
Mar 20, 2012
7
0
10,510
for starters i only have 1 560 ti and Yes yes I know, it was a horrible idea to run the 560 on 300 watts and i know for sure it is 300 watts it already had one 6 pin connector i used the adapter that came with the card to get the other. It runs i dont know how, I have a 750 watt powersupply on the way. I was just trying to find out if it was the lack in power causing the heat issue. Here is my specs



Computer
Model : Dell Inspiron 620s To be filled by O.E.M.
Chassis : Dell Desktop
Mainboard : Dell 0GDG8Y
Serial Number : ..CN701631B5017K.
BIOS : Dell A04 11/21/2011
Bus(es) : X-Bus PCI PCIe IMB USB i2c/SMBus
Total Memory : 6GB DIMM DDR3

Processors
Processor : Intel(R) Core(TM) i5-2320 CPU @ 3.00GHz (4C 3.2GHz/3.3GHz Turbo, 3.2GHz IMC, 2x 256kB L2, 6MB L3)

Chipset
Memory Controller : Dell Core (Sandy Bridge) DRAM Controller 100MHz FSB, 2x 3GB DIMM DDR3 2x 667MHz (1.33GHz)

Memory Module(s)
Memory Module : Hynix (Hyundai) HMT351U6BFR8C-H9 4GB DIMM DDR3 PC3-10700U DDR3-1334 (9-9-9-25 4-34-10-5)
Memory Module : Hynix (Hyundai) HMT325U6CFR8C-H9 2GB DIMM DDR3 PC3-10700U DDR3-1334 (9-9-9-25 4-34-10-5)

Video System
Video Adapter : NVIDIA GeForce GTX 560 Ti (384 SM5.0 1.8GHz/1.65GHz 9% OC, 2GB 4GHz, PCIe 2.00 x16)

Graphics Processor
(EVGA)
OpenCL GP Processor : NVIDIA GeForce GTX 560 Ti (384SP 8C 1.8GHz/1.65GHz 9% OC, 2GB 4GHz)
Compute Shader Processor : NVIDIA GeForce GTX 560 Ti (384SP 1.8GHz/1.65GHz 9% OC, 2GB 4GHz)
CUDA GP Processor : NVIDIA GeForce GTX 560 Ti (384SP 8C 1.8GHz/1.65GHz 9% OC, 2GB 4GHz)

Storage Devices
Seagate ST3500413AS (500.1GB, SATA600, 3.5", 7200rpm, 16MB Cache) : 466GB (C:)
TSSTcorp DVD+/-RW SH-216AB (SATA150, DVD+-RW, CD-RW, 2MB Cache) : N/A (D:)

Logical Storage Devices
RECOVERY : 15GB (NTFS)
OS (C:) : 451GB (NTFS) @ Seagate ST3500413AS (500.1GB, SATA600, 3.5", 7200rpm, 16MB Cache)
Optical Drive (D:) : N/A @ TSSTcorp DVD+/-RW SH-216AB (SATA150, DVD+-RW, CD-RW, 2MB Cache)

Peripherals
LPC Hub Controller 1 : Dell H61 LPC Controller
LPC Legacy Controller 1 : SMSC SCH5514D-NS
Audio Device : Dell Cougar Point High Definition Audio Controller
Audio Device : EVGA 0E0Ch
Audio Codec : nVidia 0016h
Audio Codec : nVidia 0016h
Audio Codec : nVidia 0016h
Disk Controller : Dell Cougar Point 4 port SATA IDE Controller
Disk Controller : Dell Cougar Point 2 port SATA IDE Controller
USB Controller 1 : Dell Cougar Point USB Enhanced Host Controller #2
USB Controller 2 : Dell Cougar Point USB Enhanced Host Controller #1
SMBus/i2c Controller 1 : Intel ICH SMBus

Printers and Faxes
Printer : Microsoft XPS Document Writer (600x600, Colour)
Fax : Microsoft Shared Fax Driver (200x200)

Network Services
Network Adapter : Realtek PCIe GBE Family Controller (Ethernet, 100Mbps)
Wireless Adapter : Dell Wireless 1502 802.11b/g/n

Operating System
Windows System : Microsoft Windows 7 Home 6.01.7601 (Service Pack 1)
Platform Compliance : x64

Windows Experience Index
Current System : 5.9
 

furpitas

Honorable
Mar 20, 2012
7
0
10,510
Yeah i dont know man it baffles me but thats why i was thinking. there is only 1 case fan pushing air out but its not very efficient, i took the side of the case off to see if it would get better air and it only decreased the temp by a couple degrees. so i dont know. I dont know to much about air flow and what it needs from what direction,
 

fil1p

Distinguished
Nov 29, 2010
944
0
19,360
Well this is simple:
1. First of get a new PSU, as others have said it is not enough, nvidia recommends more than that for a single GPU not to mention 2 in sli. A bad PSU can cause problems.
2. After that you get a better PSU you should try custom fan settings, so that when the GPU reaches a certain temp the it will increase the fan speed.
3. How is the airflow in your case? Once you complete the first few steps make sure thats good.

Hope this helps!
 

bucknutty

Distinguished
You did not mention any symptoms....Is the computer locking up or rebooting? Is the driver crashing or the screen glitching? Why do you think you are over heating?

90 is normal for the 400 and 500 based nvidia refrence cards. My GTX 470s purr at 90 during games, and the fans are at like 50-60%. If the manufacture thought that the fan profiles should have the cards at 90 with a fan at like 50% and they run ok what do i care. For the first 6 months or so i cared and ran a custom fan profile to try to keep the cards in the 70s, but they are 20 months old now and I just leave them to do what they like. The upper card (card 1 in my system) often tops out at 92 while the lower card (card 2) has a better position in the case stays around 88. Just finished a 10 hour lan party day and of MW3 BF3 and Space Marine. Not a single issue.
 

furpitas

Honorable
Mar 20, 2012
7
0
10,510
Well i try to keep an eye on temperatures of my hardware there was nothing going wrong that im aware of. from what i was reading 90 c is high.. i mean its hot enough to heat up a whole room.
 

bucknutty

Distinguished
I have a set of 8800gtx that stay in the low 80s, I also have an ATI HD4850 that stays in the high 80s.

Some one else mentioned using MSI precision. That is a great program you can use it make a spread sheet of your temps at all times and then when your done gaming you can use excel to get the highest temp or average the temps. It has onscreen overlays, recording features, screen shots, and it works with any video card.
I think the best part of precision is that you can easily make custom fan profiles to make the fan ramp up faster to help keep the card cooler.
 

heids24

Distinguished
Apr 16, 2009
134
0
18,710


You run your cards at 90C, but yet don't modulate your fans past 60%??? Why would you pass on the additional capacity you have available for cooling? That is quite hot! :eek:

I'm blasting my fans at 90% if I reach 70C on my GTX580s, but I typically have them hover around 65C while gaming.

 

heids24

Distinguished
Apr 16, 2009
134
0
18,710
To the OP - as others have said, get your power supply issue taken care of first. You're not going anywhere until that is remedied. Then download MSI Afterburner and set your fan profile to increase with the gpu temperature. I don't know that there is a magic temp to stay below, but cooler is always better. Heat accelerates the life of gpu's (a general rule of thumb is for every 10c increase, you effectively cut the life in half), so the cooler you can keep it, the more confident you can be that your card won't prematurely die.
 

bucknutty

Distinguished


I used to use a custom fan profile to keep the temps lower, or I would manually set the fans to 80ish% while gaming and that would keep the cards in the high 70s. I reinstallted windows and forgot to save my fan profile and i was to lazy to take the 2 minutes to make a new one, so i just sort of let it be. The cards run fine when i let them take care of themselves. Im sure if they get too hot they will ramp up the fans as needed. Besides if they melt then i will have a good reason to upgrade. My 470s will turn 2 years old in June.
 

TRENDING THREADS