Overclock.net › Forums › Industry News › Rumors and Unconfirmed Articles › [VC] AMD Radeon HD 9000 Series to Feature New Naming, Meet the Radeon R9-XXXX
New Posts  All Forums:Forum Nav:

[VC] AMD Radeon HD 9000 Series to Feature New Naming, Meet the Radeon R9-XXXX - Page 13

post #121 of 127
Quote:
Originally Posted by TheBlademaster01 View Post

@maarten

G92 didn't have bad solder that were the G80 and GT200 dies. G92 was fine, can always die of course.

It also depends on what you do with the cards. I can't be bothered to have a hit and miss experience with drivers at this point though.
Overclock all the values in the control panel!
Running 100% for 2 months straight was no exception. (non watercooled)
Edited by maarten12100 - 7/25/13 at 3:36am
post #122 of 127
Quote:
Originally Posted by TheBlademaster01 View Post

They were almost all from different AIBs and reference

That doesn't discount it. Most main system motherboards are nasty. It's possible if one made a shuffle to get bad ones.

But if you knew you get something like an 8 real-phase equivalent on a graphics board package.. it'd be different.
PC
(9 items)
 
  
CPUMotherboardGraphicsRAM
4790K 4.6G 1.24v/1.74v, 5.1G 1.35v validation. GA-Z97X-Gaming 7 Tri-X R9 290 1100/1350 +0.012v G.Skill 2400 c10 
CoolingMonitorPowerCase
Noctua NH-D15 24EA53 IPS 76Hz OC EVGA 1000 G2 Phanteks Enthoo Pro 
Audio
SoundMagic E10 
  hide details  
Reply
PC
(9 items)
 
  
CPUMotherboardGraphicsRAM
4790K 4.6G 1.24v/1.74v, 5.1G 1.35v validation. GA-Z97X-Gaming 7 Tri-X R9 290 1100/1350 +0.012v G.Skill 2400 c10 
CoolingMonitorPowerCase
Noctua NH-D15 24EA53 IPS 76Hz OC EVGA 1000 G2 Phanteks Enthoo Pro 
Audio
SoundMagic E10 
  hide details  
Reply
post #123 of 127
Quote:
Originally Posted by fateswarm View Post

That doesn't discount it. Most main system motherboards are nasty. It's possible if one made a shuffle to get bad ones.

But if you knew you get something like an 8 real-phase equivalent on a graphics board package.. it'd be different.

Quality is an intrinsic property, considering I kept the board stable/constant there was a clear correlation between low quality products/drivers and better ones. Let alone that I wasn't the only one. When looking at problems in society for example you can't deduce much from the people that have no problems or the ones having specific problems atypical for the norm. You curve the ones with problems and look at which is more prominent. Also narrowing your scope far enough may yield a perfect result as well.
Quote:
Originally Posted by maarten12100 View Post

Overclock all the values in the control panel!
Running 100% for 2 months straight was no exception. (non watercooled)

Yeah 100% running the same tested/tailored code for 24/7 all my cards should have been perfect tongue.gif, I guessed something like that. Those are usually the people that only encounter hardware failure. Just let it run out until the circuitry fails.

In that world hardly any driver issues exist.
post #124 of 127
Quote:
Originally Posted by TheBlademaster01 View Post

Quality is an intrinsic property, considering I kept the board stable/constant there was a clear correlation between low quality products/drivers and better ones. Let alone that I wasn't the only one. When looking at problems in society for example you can't deduce much from the people that have no problems or the ones having specific problems atypical for the norm. You curve the ones with problems and look at which is more prominent. Also narrowing your scope far enough may yield a perfect result as well.

Using fancy words doesn't make an argument logical. You have absolutely no indication quality is of a certain level by just "keeping the board stable/constant" (whatever than meant). Also, I've no idea in reference to what argument you're beating around the bush afterwards but it seems you just want to say the sample must be big. Yes, the sample must be big enough to get a good statistical indication, but a couple of users in a forum thread is certainly not that sample.

I didn't provide any sample myself, but that's precisely the point: If I have not enough data I ought to say: "My sample is too small to rule out mere chance".
PC
(9 items)
 
  
CPUMotherboardGraphicsRAM
4790K 4.6G 1.24v/1.74v, 5.1G 1.35v validation. GA-Z97X-Gaming 7 Tri-X R9 290 1100/1350 +0.012v G.Skill 2400 c10 
CoolingMonitorPowerCase
Noctua NH-D15 24EA53 IPS 76Hz OC EVGA 1000 G2 Phanteks Enthoo Pro 
Audio
SoundMagic E10 
  hide details  
Reply
PC
(9 items)
 
  
CPUMotherboardGraphicsRAM
4790K 4.6G 1.24v/1.74v, 5.1G 1.35v validation. GA-Z97X-Gaming 7 Tri-X R9 290 1100/1350 +0.012v G.Skill 2400 c10 
CoolingMonitorPowerCase
Noctua NH-D15 24EA53 IPS 76Hz OC EVGA 1000 G2 Phanteks Enthoo Pro 
Audio
SoundMagic E10 
  hide details  
Reply
post #125 of 127
Quote:
Originally Posted by fateswarm View Post

Using fancy words doesn't make an argument logical. You have absolutely no indication quality is of a certain level by just "keeping the board stable/constant" (whatever than meant). Also, I've no idea in reference to what argument you're beating around the bush afterwards but it seems you just want to say the sample must be big. Yes, the sample must be big enough to get a good statistical indication, but a couple of users in a forum thread is certainly not that sample.

I didn't provide any sample myself, but that's precisely the point: If I have not enough data I ought to say: "My sample is too small to rule out mere chance".

Fallacies don't make an argument logical either. It's how I write for a lack of other words describing precisely what I mean, note that I'm not a native English speaker but learned it from reading. I fail to see how it's very different from how you write either.

To elaborate, my post was indeed vague at some points so sorry for that wink.gif

Keeping board stable/constant implies that I had a constant system/setup with the only variables being the GPUs. In here some GPUs failed and had driver issues and others didn't. IAW, the GPUs were the points of failure (again intrinsically). It also wasn't restricted to this site/forum but rather across most forums I frequent and reviews. It was no coincidence. At some point there wasn't even support for the GPUs (hardware acceleration, 4770 on release, hotfix for games needed)...
post #126 of 127
I don't see how that argument was a fallacy. I said that I don't see your sample being big enough to discount the probability of mere chance. It's a consistent stance of statistical analysts that only for very large samples you can make confident estimations.

Plus, people shouldn't be comparing GPUs without taking into account the quality of graphics boards and all their components, VRAM, VRs, capacitors and everything else.
PC
(9 items)
 
  
CPUMotherboardGraphicsRAM
4790K 4.6G 1.24v/1.74v, 5.1G 1.35v validation. GA-Z97X-Gaming 7 Tri-X R9 290 1100/1350 +0.012v G.Skill 2400 c10 
CoolingMonitorPowerCase
Noctua NH-D15 24EA53 IPS 76Hz OC EVGA 1000 G2 Phanteks Enthoo Pro 
Audio
SoundMagic E10 
  hide details  
Reply
PC
(9 items)
 
  
CPUMotherboardGraphicsRAM
4790K 4.6G 1.24v/1.74v, 5.1G 1.35v validation. GA-Z97X-Gaming 7 Tri-X R9 290 1100/1350 +0.012v G.Skill 2400 c10 
CoolingMonitorPowerCase
Noctua NH-D15 24EA53 IPS 76Hz OC EVGA 1000 G2 Phanteks Enthoo Pro 
Audio
SoundMagic E10 
  hide details  
Reply
post #127 of 127
Quote:
Originally Posted by Gexx View Post

Guess they didn't want to have a Radeon 9800 Pro again tongue.gif

I'd love another 9800 pro, or an x1900 that was a sweet hammer.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Rumors and Unconfirmed Articles
Overclock.net › Forums › Industry News › Rumors and Unconfirmed Articles › [VC] AMD Radeon HD 9000 Series to Feature New Naming, Meet the Radeon R9-XXXX