Did you flash the bios to the Asus bios? then you could use MSI Afterburner to overclock it further and voltage up to 1.36v. I have seen some insane overclocks on these HD5870. I personally have hit 1020/1298 on mine stable, but temps start to get up in the 80c range.
no I haven't. do you have links to save me some time in research? let's see what this thing can really do with a decent cooling L/C system ;-)Quote:
Originally Posted by Utnorris
N/M I found it.
[edit] wow, seeing guys freezing at 960 under furmark after 60 secs.. when I run this Baby for 2 1/2 hrs at 970 really shows what good cooling does!
Yeah, the HD5870 is a monster overclocker and if you can cool it I am thinking 1100 shouldn't be an issue.
As I figured, the DD 5870 block is not all that attractive. However, they are gonna be charging only $99.95 for it. DD full coverage block under 100 bones? Thats a new one on me!
http://home.comcast.net/~jefftrexler/dd5870.jpg
There's an EK block that's the same price and imo looks better. I wonder how their preformance compares for being the same price.
The DD one looks like it has the tiniest amount of surface area where water flows.
Wow, so many choices. All nice looking blocks. I may just go with Swiftech's separate heatsink and keep using my MCW60
Don't knock the DD design (well, okay, you can knock the lack of shiny :D) - they're still using bead-blasting, which does wonders for cooling effectiveness. My DD GTX280 block has hung with the best of the claims, even though its internal cooling fins are practically afterthoughts. And extremely low-restriction to boot.Quote:
Originally Posted by Bojamijams
Really wish they'd start countersinking their screws, though. That design leaves about 1mm between the bolts and any card below it, and my open bench lets the cards tilt a little.
im not sure if beadblasting is gonna make up for that lack of surface area, or turbulence for that matter, something the EK for example does have. besides that it's just ugly as sin and DD is usually more expensive.
Well 1100 was an issue, at least with this card. I locked up at 1070 in Vantage. Dropped it to 1060, and seem stable.Quote:
Originally Posted by Utnorris
http://www.swiftnets.com/ASSETS/IMAG...@1060-1300.png
Here is a screenshot of my Koolance 5870 full cover performance. Room temp is 74F. Dedicated loop, DD-CPX1, MCR220 w/ low speed yates, Koolance 5870 block.
http://imgur.com/gpILT.png
It scores 18270 on Vantage but I don't have a screenshot of that yet.
Wow Gabe! That's a sweet overclock. These HD5870's are awesome. Babalouj, you can go higher, but that still a nice overclock. Nice job!
Oh I know I can get it higher, it doesn't even need 1.25V, this is the first setting I put in to test with. Its Georgia Tech homecoming weekend so I won't have time to sit down with it until Sunday.
Aquacomputer blocks (as usual) look the best. Awesome work guys :up:
i have the ek fc block and i am also stuck on 1060/1300 at 1.310v.
GPU Temp ~37C havnt reach 38C yet
VDDC Phase ~45C havnt reach 50C yet.
Any pics?Quote:
Originally Posted by Pen3
wow, that's interesting. let's compare the parameters affecting coolingQuote:
Originally Posted by babalouj
Your voltage = 1,250, Mine = 1,350 => .1v higher (more heat)
Your GPU clock = 1,000, Mine = 1060 => 60mhz higher (more heat)
Your Ambient air = 23.3C, Mine = 26.8C => 3.5C higher
Your GPU temp = 42C, Mine = 39C => 3C Lower
Now for the cooling gear
Your loop = dedicated, Mine = same
Your W/B= Koolance, Mine = Swiftech
your Rad = MCR220, Mine = MCR220
Your pump = DDC, Mine = MCP655
Your fans = Low speed yates, Mine = Medium speed Ruilian
I have a higher voltage, which really increases the TDP. For reference, before the volt mod, @970 core, I was seeing 33C @ 23C ambient @ 1.165v.
Things that make you go "Hmmmm". Usually you do not see much difference between FC blocks. Gabe, you may have a winner here.
Might be the HD5870 as well, they seem to have different VIDQuote:
Originally Posted by Utnorris
let's push a little further then:Quote:
Originally Posted by Mech0z
He is pulling 43A, we are pulling 51A, 8A more
His avg VDDC is 55C, ours is 59.5C, 4.5C more
So we are definitely heating up our card much higher
Yet, we show 39C GPU temps vs. his 42C.
There is also a time issue: his Furmark test is 3 1/2mn, and we don't know if his loop had enough time to stabilize before that. Ours is taken after 6 1/2mn, immediately after we ran Vantage. So we can assume that our temps were fairly stable, whereas we don't know if his were. I'm sure he can clarify that though. However to be fair in a comparative, both tests should exhibit at least 20~30 minutes of constant load.
You draw your own conclusions anyways ;)
Sounds good. Please clarify:Quote:
Originally Posted by Pen3
Dedicated loop?
Rad?
Fans?
Pump?
Ambient temp (critical information) ?
Screenshot?
Personally I wouldn't dare draw any conclusions, there are far too many variables between two systems for a fair comparison. Loop and air flow rates for a start. I'd also want to see a couple of mount values for an average. With the same rig in a reasonably controlled environment and I'd accept it.Quote:
Originally Posted by gabe
slow down your fans gabe, to his levels, and then lets see whats left of thet 3C difference.
Starting to sound like a p:banana:ssing match to me.
I don't see any reason to continue this e-testing/comparing thing. Either someone gather some/all FC 5870 blocks and do some proper testing or just leave out comparisons...
My :2cents: