![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
ASUS 7970 DirectCU II available now
- Thread starter BM00
- Start date
I hope this is not true, Im going to run mine pretty hard to day and see if it crashes. D0 you have the VRM slot on the plate? Temps? and where did you purchase? sorry for all the questions
Hey guys, I noticed something earlier.
Crosspost I made from another forum:
Hum, I'd like your opinions about something...
The Asus 7970 DCII has basically been ripped to pieces so far because it seems to have a very high rate of return (out of 35 reviews on Newegg, 11 are 1 egg DOAs).
There are two Asus cards:
- ASUS HD7970-DC2T-3GD5 at 1000Mhz;
- ASUS HD7970-DC2-3GD5 at stock 925Mhz.
The "T" version is the one that seems to fail a lot. The interesting thing is that on Newegg.ca, the T version has been completely deactivated (http://www.newegg.ca/Product/Product...82E16814121499), which usually is reserved for end of life products. It seems Asus might be doing something about this? Also, the other version is now sold at $499 USD, with no reviews.
It's pretty weird, though, as if the only difference between the two is the clock, that means the one remaining is a custom PCB card with stock clocks?
Crosspost I made from another forum:
Hum, I'd like your opinions about something...
The Asus 7970 DCII has basically been ripped to pieces so far because it seems to have a very high rate of return (out of 35 reviews on Newegg, 11 are 1 egg DOAs).
There are two Asus cards:
- ASUS HD7970-DC2T-3GD5 at 1000Mhz;
- ASUS HD7970-DC2-3GD5 at stock 925Mhz.
The "T" version is the one that seems to fail a lot. The interesting thing is that on Newegg.ca, the T version has been completely deactivated (http://www.newegg.ca/Product/Product...82E16814121499), which usually is reserved for end of life products. It seems Asus might be doing something about this? Also, the other version is now sold at $499 USD, with no reviews.
It's pretty weird, though, as if the only difference between the two is the clock, that means the one remaining is a custom PCB card with stock clocks?
Oddly enough, I have the T(op), and mine's been fine for a few months now.
Sure, it's doesn't OC beyond 1175MHz no matter how much voltage I give it, but it's still a decent OC and cooling works fine. I ignore the VRM's since Asus stated that software reads the VRM temps wrong since it's a non reference design and it's actually much lower.
Sure, it's doesn't OC beyond 1175MHz no matter how much voltage I give it, but it's still a decent OC and cooling works fine. I ignore the VRM's since Asus stated that software reads the VRM temps wrong since it's a non reference design and it's actually much lower.
I've already RMAed it. Don't remember did it have a VRM slot or not.
Bought it from a local store here - Zagreb, Croatia, Europe, so I don't thing that concerns you.
It worked normal first 30 mins and boom crash in battlefield. Restarted, tried furmark - instacrash. Tried battlefield - same. So I was going to remove the card to see what's up and saw that cooler moves 0.5-1 cm from the PCB.
Ok here is a small update. so far i have played BF3 maxed out for a couple of hours and it ran fine. the max temps are
GPU 52c
VRM 80c
I understand no one is really paying attention to VRM. but the reading still gives you an idea of what its doing. During game play it did not crash or artifact at all. Im going to run it this way for a week. making sure i get enough play time in. then ill overclock it, to see if i can push a little. If the card continues to pass, ill change my rating on newegg.
GPU 52c
VRM 80c
I understand no one is really paying attention to VRM. but the reading still gives you an idea of what its doing. During game play it did not crash or artifact at all. Im going to run it this way for a week. making sure i get enough play time in. then ill overclock it, to see if i can push a little. If the card continues to pass, ill change my rating on newegg.
Im contemplating returning my 680 GTX and buying one of these since the price has come down. Is it still risky? Or have the recent batches been improved?
are you not happy with the GTX? as far as the batches are concerned, the only deffinate answer i can give you is. my first one was bad which was around the 15 of march. the one i have now seems to be good. i received it Friday. The last couple of reviews from newegg have been positive. it seems to be a hit or miss
Please DO post here when you OC the card, I'm totally into buying one of these for my 3xU2412M currently I'm still at my 6970DCII but really no one with an actually fullcover waterblock had managed to push this card to 1250+... DO try, DO post, please!
Are you returning your GTX 680 just to save a few bucks or is it giving problems ? If you are doing it because of both the reasons above its worth the effort. Otherwise if its working fine not worth the hassle.
ShamisOMally
[H]ard|Gawd
- Joined
- Apr 24, 2010
- Messages
- 1,480
The Asus Direct CU II 7970 TOP edition has been removed from newegg, cause morons don't know how to follow instructions.
On Asus forums, 99% of the BSOD or "Dead Card" issues I have seen is cause people incorrectly hook up the wrong ports to the wrong monitors so the cards mess up. Asus reps even said that almost -EVERY- RMA 7970 Direct CU II top edition they got back was working just fine, and it was most likely they hooked up the monitors incorrectly
On Asus forums, 99% of the BSOD or "Dead Card" issues I have seen is cause people incorrectly hook up the wrong ports to the wrong monitors so the cards mess up. Asus reps even said that almost -EVERY- RMA 7970 Direct CU II top edition they got back was working just fine, and it was most likely they hooked up the monitors incorrectly
EarlyAdopter
Weaksauce
- Joined
- Jan 16, 2012
- Messages
- 88
So you're suggesting that I was doing it wrong by hooking my 3 monitors up to 3 displayport ports on the ASUS DCUII I had that died after 4 hours? Yeah, right. LOL.
ShamisOMally
[H]ard|Gawd
- Joined
- Apr 24, 2010
- Messages
- 1,480
Ok then, I dunno who to believe then, Asus says almost every card they get back works, and it was most likely user/system error
Cause its funny cause the 925mhz version is selling like hotcakes, and people aren't having issues, when the 1000mhz version is the exact same, just with a mild overclock
Cause its funny cause the 925mhz version is selling like hotcakes, and people aren't having issues, when the 1000mhz version is the exact same, just with a mild overclock
Ive just got one (HD7970-DC2-3GD5) without cutouts, its does have the dual-link 2560x1600 noise problem.
The other problem, it could just me being a moron, it will not oc on the core at all, cant seem to change voltage (GPU tweak, or ABwith unofficial oc (also tried release drivers and AB, nothing)) or monitor VRM temps with AIDA or HWInfo, am i missing something... or is something else wrong with it?
The other problem, it could just me being a moron, it will not oc on the core at all, cant seem to change voltage (GPU tweak, or ABwith unofficial oc (also tried release drivers and AB, nothing)) or monitor VRM temps with AIDA or HWInfo, am i missing something... or is something else wrong with it?
Well for another update, my card is dead. It would not take an overclock so i left it at stock for a TOP anyway. Playing BF3 and getting my butt handed to me, it crashed and that was the end. Reset and all the normal tail tail signs started. I will call ASUS again, have a tech walk me through all of the steps they want me to try . then i will get a refund. Im not buying another one i know that. Im going to try nvidia for while i think. I will post back and let everyone know what the ASUS tech says. If your on the wire about this card, i dont think i would get an ASUS TOP version right now. saphire and msi and all them may have better luck . I dont see as many complain about the stock version of this card. To be honest, I think ASUS just screwed up for some reason with this particular TOP version. If you have a good one, sing to it and tell it you love it. maybe buy it some flowers. Ill let you know Monday what ASUS says. have a good weekend guys
Last edited:
I stayed on the phone for an hour trying everything the ASUS tech could think of. After we finished he said sorry, but your card is deffinetly bad. He admitted to a batch with bad heat sink mounting. But that was not the case with my second one. Just one of those things, you either get a good solid card or not. I was lucky with one day to spare for newegg return.
ASUS 7970 TOP no VRM slots
ASUS 7970 TOP no VRM slots
ShamisOMally
[H]ard|Gawd
- Joined
- Apr 24, 2010
- Messages
- 1,480
I don't think anybody has got that working properly
I don't even use it, its flat out useless given that everything it can do can be done CPU wise for almost the same resources cost
Are you sure your voltage is actually being increased? Reason I ask is that my DCII Top is voltage locked. Have posted a thread on it ( http://hardforum.com/showthread.php?t=1694176 ) and responses there are reporting the same.
Really want to overclock my card but without being able to increase the voltage (mine droops to 1.06v under load) I can't get beyond 1040mhz
spinejam
Gawd
- Joined
- Nov 29, 2011
- Messages
- 721
ShamisOMally
[H]ard|Gawd
- Joined
- Apr 24, 2010
- Messages
- 1,480
I don't know about voltage unlocking, as I just overclock without altering voltage but set clocks to 1125/1575 without fail
I think its stable, I've thrown 4 hours of metro 2033 at it and it worked fine. Highest videocard got was 70C, which is a record for it
I think its stable, I've thrown 4 hours of metro 2033 at it and it worked fine. Highest videocard got was 70C, which is a record for it
Llama[Style]
Limp Gawd
- Joined
- Mar 7, 2011
- Messages
- 482
got my 7970 top today... Zero issues here, ran unigine for a good hour and benched it. My scores seemed pretty on par to what others were getting with their 7970s. runs very cool, 29-35 deg c idle... and as high as 65-70 deg c 100% load. I'm very pleased with my purchase. played diablo iii/tf2/bf3 for a few hours with no hiccups whatsoever. (yeah, 2 of the games not that demanding--but hey, no crashes on 100% load = no crashes)
I had trouble getting 120hz to display--there was a switch on mine for single link and dual link... only found out after reading others having the same issue. only thing I ran in to...but of course that is a user error, nothing wrong with the card.
these temps are on stock clocks and stock fan speed--according to gpu-z the fan speed only got as high as 28% (recorded logs while i was playing games)...which, to me, is very impressive. I'm sure if I was to ramp that up some or make it a constant 28%~, I could get the temps far cooler. I'm content with my PC not being audible though, it is DEAD silent.
I had trouble getting 120hz to display--there was a switch on mine for single link and dual link... only found out after reading others having the same issue. only thing I ran in to...but of course that is a user error, nothing wrong with the card.
these temps are on stock clocks and stock fan speed--according to gpu-z the fan speed only got as high as 28% (recorded logs while i was playing games)...which, to me, is very impressive. I'm sure if I was to ramp that up some or make it a constant 28%~, I could get the temps far cooler. I'm content with my PC not being audible though, it is DEAD silent.
Zangmonkey
Supreme [H]ardness
- Joined
- Jul 6, 2005
- Messages
- 5,064
so then are all the HSF contact issued resolved on these now for new stock?
I think I missed it but how do you know whether or not you have a "new" revision with the correction?
I think I missed it but how do you know whether or not you have a "new" revision with the correction?
ShamisOMally
[H]ard|Gawd
- Joined
- Apr 24, 2010
- Messages
- 1,480
Only the 7950 had the heatsink connection issue, and it was fixed within a week of review samples being sent out. The 7970 came out months after, and its never had that issue
Llama[Style]
Limp Gawd
- Joined
- Mar 7, 2011
- Messages
- 482
Llama[Style];1038801023 said:
Maybe I spoke too early... I got a couple of black screens, couldn't even get in to games at one point. I started logging in gpu-z and noticed that the voltages were undervolting like crazy... obviously, it only uses as much as it needs, but the max was only getting up to 1050 (1.05v) and only when it was at 100% gpu usage...(yes i know 1050 is by design)-- I raised the max to 1100 (1.1v) and the black screens stopped... ran unigine for 30 mins and haven't had issues with it all day long..
kind of disappointed, because you shouldn't really have to jump through loopholes to get it to work the way it should.Does anyone think this will continue to happen at a higher voltage?
I know it's a common problem w/ the DC2T, but c'mon ASUS, if the card needs more power, damn it--give it more power!
Last edited:
I bought mine open box and it didn't want to seat in the PCI-E slot due to the shroud part where it mounts to the case, had to push and I guess straighten the shroud for it to seat. DId anyone have that issue or did the person who returned it the first time cause it?
Llama[Style]
Limp Gawd
- Joined
- Mar 7, 2011
- Messages
- 482
I had to do this as well...
--
My problems posted above haven't resolved--will be sending it back to NewEgg on Monday.
Sometimes I can game for hours without a hiccup...other times I will get a black screen 15-30 minutes of gaming. le sigh... I've tried everything suggesed on the ASUS ROG forums and tried just about every set of 12.x drivers... It was sent with the latest BIOS and I'm at my wits end. I give up. Definitely should have read up on this f*ing card before making my purchase.
not entirely true i have one of the first 7970 dc2t
the heatplate and its pads have a 1-5mm gap between them and some of the memory chips
but the memory still does 7500mhz so it doesnt seem to be a problem
at least the heatplate does make good contact with the vrm
I just packed up my voltage locked 7970 DCII TOP for return. While doing that I checked how well the heat plate was seated and noticed that some of the screws were not tight enough. Few memory chips which I could see didn't make contact to it but had ~½mm gap. Also card wasn't straight but bended quite a bit. For now I'd stay clear from this card.
I noticed the thermal tape on the memory chips too when i took the heatsink off. It covers about half of each chip. Ive had problem free card since 7970 dc launch so i guess im one of the lucky ones... Vrm's hit constant 107 under load at 1150/1575 but the readings are wrong thank god.
I have revision 1.0 which doesnt have backplate cutouts above the vrm's. Anyone tried to put thermal tape between the backplate and vrm's? I think that could help it a little, might test it today.
Please test and report back.
Also can you alter the voltage more then 1.175 or its locked?
Yes i can alter the voltage just fine, default is 1.112, max oc 1.139v 1150/1600. Ab voltage slider goes all the way up to 1.3v. Extra voltage above that 1.139 dont seem to do any good other than increasing instability due the heat i guess.
OK i did some testing. I used hwmonitor to monitor my vrm temperatures. 20 minutes of heaven 3.0 maxed out.
before (1150/1600, 1.139v)
vrm's max temp 108
after (1150/1600, 1.139v)
vrm's max temp 101
Looks like it lowered the vrm temperatures by 7 degrees. So i wouldnt say it was a waste of my time. It did not help to overclock my card further tho, shame on that. Brick wall at 1180/1600 as before(artifacting in 3dmark11). BUT i was able to loop heaven for an hour with higher voltage (1.151) and higher clocks (1170/1600). Before i applied the thremal tape, absolute maximum heaven oc was 1150/1600 1.139v. Anything above that caused artifacting within 20 minutes of heaven.
a little edit: this is an interesting card though, about 2 weeks ago i played bf3 at 1.2v 1200/1600 clocks shy of an hour wihout any problems what so ever. After that session tried to run 3dmark 11 and got artifacts. Heaven was also stable for few minutes before core temperature went to 74. During bf3 core temperature stayed under 70c so its definitely a temp issue. I didnt look vrms at the time.
Last edited: