ASUS 7970 DirectCU II available now

Hanakuso said:
Even though I have the 1.00x model, I'm not concerned about returning it anytime soon. If we find out the later models did fix some serious stuff I will just go through Asus for the RMA. I've gone through at least 6 RMAs with Asus and everytime it's been a good experience. I especially liked how they handled the Sandy Bridge chipset recall. They actually sent me a free set of headphones, bear in mind that fiasco was Intels fault, not Asus :p

On a side note, the EK waterblocks should be coming out any day now. An EK rep mentioned they will be out around Cebit and that is going on this week. Now I just need to figure out if I should buy direct or go through a US vendor..
Click to expand...

yes the only thing i'm worried about is a card revision after i have waterblocks so will wait a little for blocks
 
Alkihaul said:
How long were you with out your GPU when you RMA'd through Asus?

My big fear is RMA'ing it once a major title comes out like Guild Wars 2 and I want to do some serious gaming I could end up with out a card 6 weeks or something. (Being in Canada)

Running the card at 40% fan speed seems to make it usable. If it wasn't for the sound proofing in my case though i'd probably go nuts with noise lol.
Click to expand...
I RMA'd my GPUs with Asus 3 times and each took a little less then 2 weeks. This is because I used ground shipping and i'm pretty far away from there RMA department. They usually had my GPU for 2 days and then sent out a replacement.

bill_d said:
yes the only thing i'm worried about is a card revision after i have waterblocks so will wait a little for blocks
Click to expand...
If you end up getting the 1.00x you should be good to go. If you do get a revised card I can take any close up pictures/dimensions of my PCB if you would like. It would suck waiting all this time and then bricking your card because of compatibility :p
 
Even tho iam scared to death i bited the dust and took my usual retailer advise, buy it cause Asus RMA policy is very gool, plus we get 3 year warranty here. So weedness day ill be testing this one out.
 
Neon01 said:
How do you figure? Amazon has probably the best return policy in the business. It's zero effort - go online, print the form, put it back in the box and put it on your doorstep; done. Some items even qualify for immediate refunds (that they take back if they don't get your item within 15 days or something). In contrast, I've never made a return at newegg that I didn't have to call to get the restock fee refunded, even if it wasn't supposed to be taken in the first place (like video cards).

Also, If I order something overnighted from Amazon with prime for the $4 fee and it's defective, not only do they pay me for the return shipping, but they even refund my $4 overnight shipping fee. Most retailers will pay for return shipping, but they won't always refund your initial shipping cost.

I know this sounds like I work for them or something. I don't, but I really think they're the best retailer I've ever done business with.
Click to expand...

I second that...I had a defective tablet a while back and Amazon advanced shipped me a new one. I got unlucky and the new one had a different problem so Amazon advanced shipped me yet another one! I actually had 3 tablets in my possession no questions asked. Plus they paid the return shipped for both tablets. There is no return policy better than Amazon unless there is a company that will actually pay you for a return, because with Amazon it costs nothing to return an item. I don't buy anything from anyone except Amazon unless the price is significantly lower.
 
Just got my 2 from Newegg tonight. The great thing was they're sequential serial numbers, so hopefully that'll bode well for their ability to match one another on OCing for CFX. They came in an Asus box like they were straight from the manufacturer, but Newegg obviously opened them because they used the typical green peanuts and the invoice was inside instead of outside.

I only checked one card, but it definitely had the backplate that covered the entire back. :( With sequential serial numbers I didn't think I needed to check both....


Regarding the water block - I've been emailing the EK rep, and here was a bit of our exchange from today:

Niko: "The [Asus HD7970 DCII Top] is sold out in Europe as well as in most of United States. We have one coming from France by the end of the week to 100% verify the flawless operation on a RETAIL card. I cannot rely on the ES sample ASUS provides. Please bear with us"

Me: "Thanks for the reply, do you mind if I ping you again for status around middle of next week? I take it the block is already designed and built, but that you're just fit-testing it with the retail DCII at this time? I assume if no problems, the block can be produced and sold relatively quickly."

Niko: "You can nudge me anytime, no problem. Yes, the sample water block has been built for weeks unfortunately we have hard time finding a retail card around here. Even our buddies at FrozenCPU and PerformancePCs couldn’t score us one."

Unfortunately since they still don't actually have the card in their hands, this likely means we'll be waiting another 2-3 weeks at a minimum until the blocks are actually at retailers here in the states. That's just a complete swag, but it seems pretty reasonable. A shame...I thought we were looking at it going up for sale within the next few days/week.
 
Carbon_Le said:
You guys who received your cards today, what are your vram temps like? TIA.
Click to expand...

Yes I am curious about this as well. Please let us know how your experience is. I am curious to see what kind of OC you get as well if you don't mind. I am hoping that the people on here that have had issues was a fluke...I really want this card to work out!
 
Well NC IX Canada received 8 in stock this morning so I queued up and RMA for the one I have and ordered a new one on cross ship. I'll document the revisions numbers and temps after receiving the replacement in a few days.
 
Let me know because mine is from NCIX a while back, so if there are better temps on the VRMs, I will have to RMA mine as well.
 
Siryak said:
Yes I am curious about this as well. Please let us know how your experience is. I am curious to see what kind of OC you get as well if you don't mind. I am hoping that the people on here that have had issues was a fluke...I really want this card to work out!
Click to expand...

I got mine in and running, and let me say this - testing these things with any degree of accuracy or control is one tough chore. The long and the short of it is that the versions that were sent out this week (or late last) are no better than what's been circulating for some time.

For ref: this is with a clean install of 12.2 precerts. No AB, Trix, or AsusTweak or anything (for now).

My cards had ASIC of 84.8% and 75.5%, and with eyefinity and stock fan profile idled at 57 C and 29 C (pretty normal for eyefinity and cfx, if a tad on the high side). I ran through a couple loops of heaven and my VRM temps (stock clocks) got to 109 deg. They idled at around 78 C. Keep in mind this is in eyefinity res (3800x1600). My stock VDDCs are 1.02 for the primary card when in 3d apps, and the secondary card stays no higher than 0.9 V, so at least there's some room to move up there.

I tried bumping my core clock speeds on both cards to 1100 to see if it could handle it, and I got some strange results. I went back into heaven (this time doing single screen 1920x1200 to give a more comparable result) and started a benchmark test. Aside: holy shit are these things fast!! At 1920x1200 with 4xAA, 16xAF, and normal Tess (DX11), I was averaging around 110-120 fps....until it locked up about 60 secs in. What's interesting here is that it didn't lock up my PC, and I was able to alt-tab out and close the program. It gave me this error (unigine fatal error) "D3D11 AppWindow::swap_window():device removed". I had forgotten to bump the power boost up to +20%, so I went and did that, and got back into Heaven. Right away I could tell it was only using 1 GPU based on the frames, so whatever happened before to crash actually disabled crossfire. I had to go into CCC, and disable it (they were still supposedly enabled), then enable them again. Back into heaven and obviously both cards working again. But then...same error dumped me, only this time it was about 90 second in. Another note though is that both times, my VRM temps hadn't even gotten to 90 C yet, so I really doubt that was the problem. My GPUs were climbing toward 70 C when it crashed.

My next step was going to be to try AB to bump the voltage a bit and see if that makes things more stable, but with the problems I'm encountering, I really doubt it'll work. Not to mention I've read that you have to do all kinds of workarounds to get AB to work and give you voltage control with 12.2 precerts.

So now I think I'm going to try the new 12.4 betas and see what that gets me, if anything. If that doesn't work, I'll go back to the late Jan RC drivers.

So far my intuition tells me that these cards are going to overclock nicely, and that most of the problems I"m seeing so far are because of shitty drivers. Once I get some decent drivers and I get some water blocks on these things to cool the VRM temps down, I'm hopeful that I can get some serious performance out of them.
 
100-110 for me. Single and 2x.

My idle temps (core, mem, vrm) - 39, 27, 60 respectively.
 
109C at stock clocks...That seems way too high to mean. If it was overclocked to the max it would still be a little high IMO, but it would be far more reasonable. If it is already that high at stock, what is going to happen when you really push that thing?
 
Siryak said:
109C at stock clocks...That seems way too high to mean. If it was overclocked to the max it would still be a little high IMO, but it would be far more reasonable. If it is already that high at stock, what is going to happen when you really push that thing?
Click to expand...

Watercool.
 
I haven't seen this report mentioned anywhere - it is related to 7950's, but the cooler in general on both the 7950 and 7970 model.

http://www.behardware.com/news/12153/asus-hd-7950-directcu-ii-fault-report.html

This sounds much like my problem

If however we applied a slight pressure (such as even only a slightly rigid cable would exert) on the top of the cooler, it detaches itself even further from the GPU and the system crashes immediately.
Click to expand...

Where I would run the card for a few minutes and it would crash. I never got reported high temperatures.

After taking the card out (effectively squishing the heat sink back together) it would run again for 10 minutes or so before crashing.

This is a bit of skepticism at the moment but an interesting read.

I'll see what my replacement card brings.
 
If that is the case then my core temp will be ridiculously high. Mine never hits 75+ on FurMark.
 
That isn't relevant to the 7970. The GPU temp on the 7970 isn't the problem, it's only the VRM. The 7970 and 7950 DCII don't use the same cooler
 
Alkihaul said:
I haven't seen this report mentioned anywhere - it is related to 7950's, but the cooler in general on both the 7950 and 7970 model.

http://www.behardware.com/news/12153/asus-hd-7950-directcu-ii-fault-report.html

This sounds much like my problem



Where I would run the card for a few minutes and it would crash. I never got reported high temperatures.

After taking the card out (effectively squishing the heat sink back together) it would run again for 10 minutes or so before crashing.

This is a bit of skepticism at the moment but an interesting read.

I'll see what my replacement card brings.
Click to expand...

This was posted about 200 posts before. Not the same card though and seems like a different issue.

http://hardforum.com/showpost.php?p=1038423437&postcount=309
 
Ok, now I'm starting to get a little worried. I just played Rift for about 60 mins in eyefinity using the 12.2 precerts with the RC .dll file added back in. I installed AB beta 14 and it seems to be working just great. It's set both of my cards to 1112mV VDDC, which is definitely higher than my primary card was at stock, but the extra voltage made it really stable. Right now I'm testing a 1050mHz core clock with stock 1400mHz VRAM, and I've added in more aggressive custom fan profile.

Only problem is, I'm seeing INSANE temperatures. During that Rift play session I was monitoring temps and fan speeds via AB on my G15 keyboard and I maxed at 89C (primary - secondary was at 54-56C), and the never dipped below about 83C while I was playing. At those temps the primary card is at a fan speed of 70-75%. Further, when I exited Rift, I checked HWiNFO to find my VRMs had hit a peak value of 118C!!!

What's really strange is that despite all of this, there were absolutely ZERO hiccups, artifacts, stutters, anything. The performance was a bit slower than I was expecting, but I think that's probably just a lack of CFX optimization for Rift.

So it looks like I may have some decent performers on my hands, but it really scares me to think that I'm hitting temps around 15C higher than other people are hitting in Furmark, only in regular games. Insane. Do these sound like defective cards? I personally can't tell at this point because everything sort of seems abnormal to me with the DC2.

I'd like to think a pair of water blocks are going to fix this, but I feel like I'm going to need at least one more radiator to keep pace with all that heat. My office feels like a sauna!
 
Neon01 said:
Ok, now I'm starting to get a little worried. I just played Rift for about 60 mins in eyefinity using the 12.2 precerts with the RC .dll file added back in. I installed AB beta 14 and it seems to be working just great. It's set both of my cards to 1112mV VDDC, which is definitely higher than my primary card was at stock, but the extra voltage made it really stable. Right now I'm testing a 1050mHz core clock with stock 1400mHz VRAM, and I've added in more aggressive custom fan profile.

Only problem is, I'm seeing INSANE temperatures. During that Rift play session I was monitoring temps and fan speeds via AB on my G15 keyboard and I maxed at 89C (primary - secondary was at 54-56C), and the never dipped below about 83C while I was playing. At those temps the primary card is at a fan speed of 70-75%. Further, when I exited Rift, I checked HWiNFO to find my VRMs had hit a peak value of 118C!!!

What's really strange is that despite all of this, there were absolutely ZERO hiccups, artifacts, stutters, anything. The performance was a bit slower than I was expecting, but I think that's probably just a lack of CFX optimization for Rift.

So it looks like I may have some decent performers on my hands, but it really scares me to think that I'm hitting temps around 15C higher than other people are hitting in Furmark, only in regular games. Insane. Do these sound like defective cards? I personally can't tell at this point because everything sort of seems abnormal to me with the DC2.

I'd like to think a pair of water blocks are going to fix this, but I feel like I'm going to need at least one more radiator to keep pace with all that heat. My office feels like a sauna!
Click to expand...

hot air from second card blowing into primary card?
 
PeeWeePee said:
hot air from second card blowing into primary card?
Click to expand...

You're probably right - I was hoping that the 200mm side panel fan on my HAF X would actually help with that, but it seems it might only be making it worse.

I will say this though - the fans on the DC2 are not obnoxious at almost any level. Even at 75% I felt it was loud, but not overly so. With the reference cooler, 45% sounded loud, and 75% sounded ridiculous.
 
For me anything above 35% is loud. The only thing I'm hearing right now are my rad fans.
 
i'm holding out for a toxic sapphire card

or whichever card oc's the best
 
gigatexal said:
i'm holding out for a toxic sapphire card

or whichever card oc's the best
Click to expand...

I am thinking MSI Lightning for me. I am curious to see the pricing on the Sapphire and MSI cards. Both are looking to be the best cards on the market. Lets just hope they price them competitively.
 
Great, now the stupid things are crashing my PC on me. Weird too, because I switched to display output on (only) my HDTV, so my temps went way down (34/24C core, 59C VRM). Despite this, I'm getting crashing a lot now (probably once every 10 mins or so). Heard some bad stuff with 12.2 precerts doing this, so I tried RC11 drivers with a clean sweep in safe mode - still doing it.

Really wanted these cards to work since they're all that's out there for 6 display support, but I can't deal with this. I'm going to try taking one out of the case and playing around with them individually to see if that helps.

If anyone has any recommendations for things to try, they're appreciated.
 
Just got this email from Amazon. Seems they had a shipment arrive and backorders are about to get filled. We'll see. I'll report back here on temps, including VRMs, and stability. I just want my third screen tearing to go away. Why the stock cards only have 2 of the same connectors I'll never understand...

email:
We have good news! We're able to get this part of your order to you faster than we originally promised:

"ASUS HD7970 DirectCU II 1000MHz Overclocked GPU and GPU Tweak Graphics Cards HD7970-DC2T-3GD5"
Previous estimated arrival date: March 13, 2012 - March 23, 2012
New estimated arrival date: March 08, 2012 - March 09, 2012
 
Voltage hasn't seemed to affect my VRM temps at all. I've found my card to be stable @ 1020mV (looped Heaven for about 20 minutes) my VRM still hit a max of 100C. While it is lower than many people here have stated theirs to be, it's still the same temp i get at stock 1050mV.
 
well, never really had bothered to check my VRM temps in actual gaming, so not sure if it's better, but just played a 30min game of BF3 and my VRM temp maxed at 92C! That's nice :D The backplate screws were a little loose.
 
Siryak said:
I am thinking MSI Lightning for me. I am curious to see the pricing on the Sapphire and MSI cards. Both are looking to be the best cards on the market. Lets just hope they price them competitively.
Click to expand...

yeah i think i saw the lightnig do 1.3ghz core
 
Has anybody been able to actually get a descent overclock out of the DC2 yet though? I have seen people with reference cards getting to 1300. I was ready to pull the trigger on the DC2, but the MSI Lightning has made me hesitant. It is looking like it may end up the superior card. Especially since it has 4 DP and is only 2 slots.
 
Siryak said:
Has anybody been able to actually get a descent overclock out of the DC2 yet though? I have seen people with reference cards getting to 1300. I was ready to pull the trigger on the DC2, but the MSI Lightning has made me hesitant. It is looking like it may end up the superior card. Especially since it has 4 DP and is only 2 slots.
Click to expand...

not yet it needs a waterblock gpu looks to be cooled fine but vrm's are not cooled well enough
saw someone at 1250
http://www.xtremesystems.org/forums...-DirectCU-II&p=5061287&viewfull=1#post5061287


will know more soon my two have shipped from newegg now
 
bill_d said:
not yet it needs a waterblock gpu looks to be cooled fine but vrm's are not cooled well enough
saw someone at 1250
http://www.xtremesystems.org/forums...-DirectCU-II&p=5061287&viewfull=1#post5061287


will know more soon my two have shipped from newegg now
Click to expand...

That isn't really all that impressive when reference cards are hitting 1300. 1250 on water is underwhelming to say the least. Lets hope MSI keeps its track record of great overclockers going with the 7970 Lightning. Asus usually puts out some pretty good stuff, but so far from what I have seen they dropped the ball on this one.
 
Siryak said:
That isn't really all that impressive when reference cards are hitting 1300. 1250 on water is underwhelming to say the least. Lets hope MSI keeps its track record of great overclockers going with the 7970 Lightning. Asus usually puts out some pretty good stuff, but so far from what I have seen they dropped the ball on this one.
Click to expand...

he is on air
the waterbolcks for dc2 are not out yet
 
I gotta say having had a gtx 580 DCUII I was very impressed with this card they made for 7970 has been a huge let down. I think you guys are crazy keeping these with vrm temps and faulty mounting on coolers.
 
QwaarJet said:
Voltage hasn't seemed to affect my VRM temps at all. I've found my card to be stable @ 1020mV (looped Heaven for about 20 minutes) my VRM still hit a max of 100C. While it is lower than many people here have stated theirs to be, it's still the same temp i get at stock 1050mV.
Click to expand...

1020? Is that a typo? Or undervoltage? Ref volts for the ASUS 7970 DCUII are 1130-1150 so you're not really pushing them, you're giving them a break :)

Rattle said:
I gotta say having had a gtx 580 DCUII I was very impressed with this card they made for 7970 has been a huge let down. I think you guys are crazy keeping these with vrm temps and faulty mounting on coolers.
Click to expand...

Yeah it's a shame, but the mounting problem is only the 7950.
 
Mine keeps crashing in Furmark and OCCT stress tests at stock voltages using 11.x since that's the only driver that won't crash my desktop.
 
Try setting the fans manually to 60% or higher. The ASUS fan curve is not aggressive enough.
 
Shurakai said:
1020? Is that a typo? Or undervoltage? Ref volts for the ASUS 7970 DCUII are 1130-1150 so you're not really pushing them, you're giving them a break :)
Click to expand...

Yup, undervolted. I have a high asic score, so figured I could get away with some undervolting. My card stock was 1050, I kept knocking down by 10mV until Heaven failed to run, then crawled back up. Seems to be happy with 1125mhz core @ 1020mV.

Heaven brings my VRM to 100, if the room is hot, then 101 :p But as I stated before, BF3 only brings my VRM's to the low 90's
 
Shurakai said:
1020? Is that a typo? Or undervoltage? Ref volts for the ASUS 7970 DCUII are 1130-1150 so you're not really pushing them, you're giving them a break :)
Click to expand...

I don't know about his, but mine default to 1020mV (84.8% ASIC) and 1112mV (74.5%) without any adjustments at all. I'm not talking about Vdroop either.
 
Sounds like the high VRM temps are normal on these cards.
Pretty much matches what my DCUIIT does as well on the VRMs. My core though never gets really hot, so it does a great job there, it's just too bad the VRMs don't match what my Gigabyte Windforce can do (70C on the VRMs at higher clocks than the Asus).
 
You must log in or register to reply here.
Back
Top