Search results

Just wondered if anyone here could help me with an annoying problem. I purchased a 3810TG on eBay and tested it perfectly with the stock configuration and it was working fine. I then sold the 4GB RAM and stock hard drive off to swap for an SSD and 8GB RAM. On fitting the 8GB RAM it booted and...

I actually just got round to testing the replacement SOC card OCUK sent me with my new Corsair TX850 V2 and the results are exactly the same as with the OCZ Modxstream 600W. I get errors in OCCT at the 78-80 degrees mark at the default voltage of 1.088V. At this voltage it is impossible to keep...

Have you actually ticked the check for errors checkbox to see if it detects any GPU errors? Or are you just watching closely for artifacts?

I think they just got overzealous with the clocks and voltage tweaking and didn't put them through their paces enough. I like my SOC but it bugs me knowing I probably have the highest voltage example with loads of heat and noisy fans to compensate. Also feel it won't last as long as the ones...

You are lucky with that card. My first one had stupidly low voltage (1.025) and gave errors in OCCT after 1 second. This replacement card has 1.088v and gets too hot... Looks to me like these are not cherry picked GPUs but cards that have had each individual BIOS modified to suit the quality of...

What is your card's load voltage set at in the BIOS? Mine is 1.088... Is yours an SOC or an SE?

Mine will idle at around 40 degrees in my case when it is closed with the fan profile I posted in this thread (about 40% at idle). When playing current games like Dirt 2 at 1080p with max settings it reaches about 70 degrees with the case closed and a ghetto 120mm mod side fan next to it...

I have been doing further testing. To get my slightly more stable replacement card OCCT stable I had to do the following: 1. Make my room artificially cold by opening all the windows. 2. Remove side panel of case and hold a high speed 120mm fan next to the graphics card on full speed (really...

I wish I had just stuck with my overclocked 4870 TBH. The SOC is a nice card but too many problems for me, my old card was rock solid.

I tested my replacement card from OCUK and found much higher GPU voltage on this card than the last one. The card doesn't give errors immediately on OCCT like the last one but it does eventually when the card hits 80 degrees. There are two problems with these cards as I see it. For people that...

Test quickly with OCCT GPU error test and see what happens. If you get any errors then you know the card is not up to the job at stock overclocked speed. I have a feeling the side fan will make a big diff on temps for you though.

Buzzing is probably a capacitor on the card itself or PSU, it is a common thing to happen under load. Loads of the HD4890 cards had loud buzzing caps under load.

Yeah! Did you not read my recent posts? I was waiting to test my card with another PSU that never showed up so I ordered a Corsair TX 850W V2. Gonna test with that soon and might RMA the second card if still bad...

LOL you are optimistic, considering I am on my second duff card... :)

I messaged Gigabyte tech support and it took AGES for them to fob me off with "try another PC..."

I wouldn't let mine get that hot personally, I am trying to keep mine well under 80 degrees to be on the safe side. It is interesting to see that the card hit 73 degrees on the Gigabyte test bench, I am going to use that as my target temperature from now on. I had a really nice Silverstone...

No the card is definitely different. I thought I would update on my solution for the meantime until I decide what to do with the card. I found that keeping the card well below 80 degrees helps massively with stability, although this card is still not perfect as it will not work error free at...

They don't look adequate really, it is all about getting as much cool air over the cooler from the outside as possible. Seems to be working OK for me in games with 100% fan speed set manually, and I can just about bare the noise for now. Will try OCCT with new 850W corsair PSU when it...

I am using OCCT with GPU: OCCT test and 'Error Check' ticked. My BIOS is the same as my old card: 70.00.21.00.02. I have really stress tested the card with certain games that take the GPU to 100% usage, Cryostasis with PhysX for example, and the GPU temp goes over 80 degrees with the stock...

OK a quick update on my new card. The VDDC under load has been set much higher in the BIOS so more power is being used, more heat is generated and the fans spin louder. I am not sure if this modification has been done at the factory or after using something like NiBiTor, but it has not helped...