Overclock.net › Forums › Graphics Cards › AMD/ATI › Problems of a Low ASIC quality card? [Sapphire 7970 - 60.4%]
New Posts  All Forums:Forum Nav:

Problems of a Low ASIC quality card? [Sapphire 7970 - 60.4%] - Page 9

post #81 of 106
@Luke, I've tried everything. Doesn't change anything. As i stated though, it seems to be locked but at the same time... it's not haha.

Well I've managed to Flash my Bios to another and it seems to have a higher voltage on load by .1-.2v So I'm just gonna stick with that one lol.

To be honest, I'm just saying all of this so that other people who are curious may know that they aren't the only ones who are having this issue. My card is stable where it's at right now @ 1200/1700. So it's not too big of an issue for me haha.

But I ditched Furmark because I'm fairly positive i will never run into a scenario like that haha. My VRAM shot up to 92c @ 1700/1.6v loool and my Temps went up to 75c @ 1200/(What Trixx/AB states 1.256v).

However through Heaven and Valley my GPU core only hits 58-61c and Vram is the same.
post #82 of 106
id be happy with 1200/1700 man, just see what its like playing crysis3 or BF3 and other games
post #83 of 106
Quote:
Originally Posted by Cyph3r View Post

Lower ASIC is better for extreme overclocking with volt modding, which is why all premium GPUs like the MSI R7970 Lightning, all have ASIC qualities under 70%. The ASIC on mine is like 59% and I can reach 1300Mhz @ 1.3v so...

This is a double edged sword i think, They charge a butt load more for the premium features on the PCB (seriously overhauled VRM's and power delivery) slap a great air cooler onto the card then put a crappy chip on it. True those cards shine under LN2 and other extreme overclocks but your average user is going to buy a lighting stick it in his box and call it a day. What sucks is paying alot more money then for reference and even other custom air GPU coolers then being limited to the same OCs they get because of a very leaky core. Like building a ferrari with a toyota engine in it.

I think MSi would do good to add very high ASIC quality chips to their lightning line so they get the absolute best Overclocks they can out of air and actually make use of the excellent PCBs and Coolers they design. I've seen a bunch or reviews for the lightning with mediocre overclocks on air even with extremely high voltages and it's a shame.

which brings me back to the thread topic,

Basically from my understanding ASIC determines how leaky your chip is.

Low Asic = Leaky chip, can take alot more voltage than a high Asic chip but unless you're using extreme cooling they are not very good overclockers on air. Requires more voltage because of bigger Vdroop.

High Asic = Lower leakage chip, overclocks better on air/ water but cannot go as high a a Low Asic chip on extreme cooling. Requires less voltage because of smaller Vdroop.

Which brings me to another theory, i'm having a hell of a time cooling my Cards VRMs, Two theories i have is that:
  • VRMs for both Low and High Asic cards operate at the same temperature but when a person with a high Asic card chooses a voltage (say 1200mv) their card stays closer to that voltage than a person with a low Asic which results in higher temps for the person with a high Asic .
  • For reasons unknown to me high Asic cards just heat up faster.
  • Something else tongue.gif

now i'm no GPU architect, so take all this with a whole bunch of salt lol.
Edited by th3illusiveman - 3/4/13 at 11:13pm
Bender
(18 items)
 
  
Reply
Bender
(18 items)
 
  
Reply
post #84 of 106
bios.zip 42k .zip file

Try this maybe? Good luck
post #85 of 106
Thread Starter 
Quote:
Originally Posted by Derpinheimer View Post

bios.zip 42k .zip file

Try this maybe? Good luck

I know this is a little late, but could I know a little more about that BIOS file you uploaded? Where did you get it from? redface.gif
post #86 of 106
I was reading your previous thread about problems with waking up from sleep with the hd 7xxx series cards and I have the exact same problem too. Did getting a new card solve the problem?
post #87 of 106
I have came to conclusion I give zero ****s to the ASIC score.
Fall of a Jedi
(14 items)
 
  
CPUMotherboardGraphicsRAM
AMD FX 8350 @ 4.4MHz ASUS Sabertooth 990FX Sapphire 7970 3GB OC CORSAIR Vengeance 8GB 
Hard DriveOSMonitorPower
Seagate Barracuda 1TB Window's Ultimate 64 bit Asus VE249H 24 inch 1920x1080 Coolermaster 800w Silent Pro 
CaseMouseAudio
Corsair C70 Logitech G9x Creative Sound Blaster Sigma 
  hide details  
Reply
Fall of a Jedi
(14 items)
 
  
CPUMotherboardGraphicsRAM
AMD FX 8350 @ 4.4MHz ASUS Sabertooth 990FX Sapphire 7970 3GB OC CORSAIR Vengeance 8GB 
Hard DriveOSMonitorPower
Seagate Barracuda 1TB Window's Ultimate 64 bit Asus VE249H 24 inch 1920x1080 Coolermaster 800w Silent Pro 
CaseMouseAudio
Corsair C70 Logitech G9x Creative Sound Blaster Sigma 
  hide details  
Reply
post #88 of 106
Thread Starter 
Quote:
Originally Posted by Ionospheric View Post

I was reading your previous thread about problems with waking up from sleep with the hd 7xxx series cards and I have the exact same problem too. Did getting a new card solve the problem?

This is a little late, but I'm not sure. My issue has been fixed about a month ago when I RMA'd my 7970 for the last time. But on the same day, I got a new monitor as well. Though I'd like to say that my issue was caused by a faulty GPU, right now I really think the root of the problem lies with my old monitor. It was a Samsung SyncMaster 2032MW which is a pretty cheap and old monitor. Maybe it was due to hardware incompatibilities. I really can't put a finger to it. But ever since I got my new monitor, which is a ASUS VX238H (fantastic monitor btw), things have been working as smooth as butter.

Well hope this helps smile.gif
post #89 of 106
Thanks Gabriel,
My monitor is not that old maybe a couple years old and has never had any issues with my previous cards so I doubt that in my case at least that the monitor is the issue. I still haven't ruled out a driver issue but I've uninstalled and reinstalled numerous times to the latest catalyst drivers. An odd problem I have is I am unable to downgrade to any previous driver versions.
I googled this issue and it seems many people have had the same problem and the only solution seems to be to not let the monitor go to sleep.
I have been meaning to get a bigger monitor so I will see if that helps the issue and if not I might just RMA this card and see if the problem is a defective card. If that is the case I will post the results here. Cheers!
post #90 of 106
Figured I'd chime in. I've owned 2 different Sapphire 7970 Vapor-X GHz Edition cards. Both low ASIC. First was 63.4%, second was 56.3%. They sucked at overclocking. I mean really sucked. Neither card could truly get past 1100Mhz (only a 50Mhz increase over this card's stock clock). I could run them at 1150-1175Mhz totally fine for about 1-2 hours, but artifacts/crashes would eventually show up inevitably. When stress testing with Unigine Heaven/Valley they never went past 66C even at an unstable 1200Mhz and 1.35v. Maybe it was the Vapor Chamber cooling, but they really didn't run hot like low-ASIC cards supposedly do. Still, 66C shouldn't be hot enough to start causing instability. They were just garbage cores that couldn't clock high.

But I've seen some chatter about TriXX (or others) changing only the non-boost voltage? I know on the Vapor-Xs I had, both BIOSes were identical (they were both boost BIOSes) and IIRC, the voltage under load was always the same regardless of what I set VDDC to in TriXX. I was using Trixx 4.4b (the modded version for MVDDC). It makes sense now that you mention it changing the non-boost voltage only. Has anyone tried flashing a GHz Edition card with a non-GHz (925Mhz) edition BIOS to see if that changes the way voltage behaves under load?
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: AMD/ATI
Overclock.net › Forums › Graphics Cards › AMD/ATI › Problems of a Low ASIC quality card? [Sapphire 7970 - 60.4%]