AMD acknowledges 390 series crashes, looking into it

[Kinda Bottlenecked]

Unfortunately, however, it is the direct replacement of the 270. It's not as powerful as a 270.

 

It isn't, the 270 was in the r9 line, the 370 is in the r7 line. 

[Dabombinable]

1442 is baaad for a 970 lol hahaha. if you want to compare gpus, at least use the same manufacturer  

 

all 970 gaming I've had seen can reach 1500+ mhz

Not really, the only way to get a semi stable 'overclock' with my GTX 970 G1 Gaming is to increase the voltage, which give me a 50Mhz core clock increase. Anything else and boom, driver crash. I honestly want to know what the person who binned my 970's GPU was smoking.

[asker509]

Not really, the only way to get a semi stable 'overclock' with my GTX 970 G1 Gaming is to increase the voltage, which give me a 50Mhz core clock increase. Anything else and boom, driver crash. I honestly want to know what the person who binned my 970's GPU was smoking.

my classified 980 is in the same boat I can't get a over 40+ mhz core

[Dabombinable]

my classified 980 is in the same boat I can't get a over 40+ mhz core

So much for GPU binning....its non existent if cards like ours exist.

[asker509]

So much for GPU binning....its non existent if cards like ours exist.

it makes me feel like this custom pcb, dual 8 pins, and triple bios is just gonna go to waste. I was gonna watercool but my card isn't stable on air with low over clocks I don't see much of a point

[patrickjp93]

Yes, the Nasdaq is a joke, look at the Dow Jones or the S&P500. Talking about crashes, wasnt it funny when the nasdaq crashed...twice. Besides a single day of trading on one stock market index is as important as a youtube comment, and a quarterly earning result can't be blamed on one single factor, so, let's keep stocks out of this? I mean, anything you have ever disliked about a product was decided by those fucking people, they are the hidden enemy that makes intel step back from a breakthrough, prevent AMD to coming up with a new arquitecture in a timely fashion and make Nvidia fucking mislead their customers on specs, which i would hope someone would be in jail for...

No one is stopping Intel on the investor side. Intel pays dividends regardless of the stock price. If Intel comes up with a breakthrough, no one in my court's stopping them. If the stock price goes up due to a breakthrough, we're doubly happy. If Intel makes a venture and has a small failure, we don't care. The revenues and profits have been increasing, and the dividends have too, regardless of what the markets do with the share price. I only own 100 shares myself, a $2800 value on paper (though I also sold those 100 shares in May when they were worth $3460, and I expect the price to begin ticking up shortly after Skylake's release anyway), but I also get paid $300 in dividends each year (amortized over the last 5 years, going upward). Intel's pursuit for phones I supported. The Xeon Phi project I support. Quark I disagree with unless they can manage to find some profit margins above 10%, but that's about all that I've raised any disagreement over. Kirzanich is not stupid, and neither are most of Intel's investors. Now, if Warren Buffett jumps, that'll raise some alarms, but even he pretty much doesn't disagree with any decisions.

[patrickjp93]

Not really, the only way to get a semi stable 'overclock' with my GTX 970 G1 Gaming is to increase the voltage, which give me a 50Mhz core clock increase. Anything else and boom, driver crash. I honestly want to know what the person who binned my 970's GPU was smoking.

 

it makes me feel like this custom pcb, dual 8 pins, and triple bios is just gonna go to waste. I was gonna watercool but my card isn't stable on air with low over clocks I don't see much of a point

You two need to realize binning is a form of market and product segmentation. In a bin, 100% of the chips must be able to perform at the slowest speeds guaranteed by that bin, much like Communism enforcing equal results. Some less than ideal chips will make it into upper bins to help puff up profit margins. No one was high. An anticipated demand for that level of bin was calculated, and that number of cards were made. After that you get the same bell curve that you see across the entire GPU spectrum. Most members of the bin will get within a certain overclock range. Some won't do so well, very few will not go past boost, and very few are LN2 competition wet dreams.

[Dabombinable]

You two need to realize binning is a form of market and product segmentation. In a bin, 100% of the chips must be able to perform at the slowest speeds guaranteed by that bin, much like Communism enforcing equal results. Some less than ideal chips will make it into upper bins to help puff up profit margins. No one was high. An anticipated demand for that level of bin was calculated, and that number of cards were made. After that you get the same bell curve that you see across the entire GPU spectrum. Most members of the bin will get within a certain overclock range. Some won't do so well, very few will not go past boost, and very few are LN2 competition wet dreams.

Gigabyte's GPU Gauntlet binning process-straight up false advertising since they don't actually do it.

[patrickjp93]

Gigabyte's GPU Gauntlet binning process-straight up false advertising since they don't actually do it.

Do you have proof of that?

[Dabombinable]

Do you have proof of that?

Let me see, the GPU Gauntlet is supposed to weed out GPU that don't overclock that well, and what do you know? I've got a GPU that can't manage a stable overclock at all.

[patrickjp93]

Let me see, the GPU Gauntlet is supposed to weed out GPU that don't overclock that well, and what do you know? I've got a GPU that can't manage a stable overclock at all.

Could be the motherboard if the PCIe controller is the temperamental part of it. Could also be the vBIOS. Could also be unstable power delivery through the PCIe connector or your cables above a certain voltage. If you haven't tried everything, you really can't reach that conclusion, as annoying as that may sound. Further, have you asked Gigabyte about this? Could damage have happened in shipping?

[Dabombinable]

Could be the motherboard if the PCIe controller is the temperamental part of it. Could also be the vBIOS. Could also be unstable power delivery through the PCIe connector or your cables above a certain voltage. If you haven't tried everything, you really can't reach that conclusion, as annoying as that may sound.

Asus P5Q Deluxe, problem appears. Asus H87M Pro, problem appears. Asus P5Q turbo, Problem appears. Various drivers, problem appears. Used 3 different PSU, problem appeared with each one. Update the vBios, problem still appears. Used a custom Bios, problem still appears. Windows Vista, 7, 8.1-problem still appears. Yes I can reach conclusions.

[patrickjp93]

Asus P5Q Deluxe, problem appears. Asus H87M Pro, problem appears. Asus P5Q turbo, Problem appears. Various drivers, problem appears. Used 3 different PSU, problem appeared with each one. Update the vBios, problem still appears. Used a custom Bios, problem still appears. Windows Vista, 7, 8.1-problem still appears. Yes I can reach conclusions.

I have to ask, or I wouldn't be doing anyone any favors. Which brings me to this: shipping damage? Have you asked Gigabyte about this or raised Hell on their support forums? Have you used that ASIC quality checker to bring up as partial counterproof?

[asker509]

You two need to realize binning is a form of market and product segmentation. In a bin, 100% of the chips must be able to perform at the slowest speeds guaranteed by that bin, much like Communism enforcing equal results. Some less than ideal chips will make it into upper bins to help puff up profit margins. No one was high. An anticipated demand for that level of bin was calculated, and that number of cards were made. After that you get the same bell curve that you see across the entire GPU spectrum. Most members of the bin will get within a certain overclock range. Some won't do so well, very few will not go past boost, and very few are LN2 competition wet dreams.

I think we both realized that seeing how our cards barley overclock

[Dabombinable]

I have to ask, or I wouldn't be doing anyone any favors. Which brings me to this: shipping damage? Have you asked Gigabyte about this or raised Hell on their support forums? Have you used that ASIC quality checker to bring up as partial counterproof?

Blocked from Gigabyte's forums, my topics got ignored (and deleted) and after several RMA I received my original card. The ASIC score is 62.1%, and I've seen other G1 Gaming with ASIC around the high 70, low 80's.

[patrickjp93]

Blocked from Gigabyte's forums, my topics got ignored (and deleted) and after several RMA I received my original card. The ASIC score is 62.1%, and I've seen other G1 Gaming with ASIC around the high 70, low 80's.

I mean, you're only guaranteed the speeds on the box, so Gigabyte's binning exists. You just got burnt in the lottery. It happens to all of us once in a while.

[Dabombinable]

I mean, you're only guaranteed the speeds on the box, so Gigabyte's binning exists. You just got burnt in the lottery. It happens to all of us once in a while.

You don't own a Gigabyte GTX 970 G1 Gaming so you really don't know what I'm talking about:

 

Read the box.

[Mattb6288]

Had this problem playing dragon age inquisition with my HD 7950 too on 15.7. It's probably just rushed beta drivers. I hope they can fix these issues soon since I plan on upgrading to a 390

[Notional]

Had this problem playing dragon age inquisition with my HD 7950 too on 15.7. It's probably just rushed beta drivers. I hope they can fix these issues soon since I plan on upgrading to a 390

 

Meh, all drivers have issues. Next WHQL driver should be out next week for the windows 10 launch.

[Sonefiler]

to put it simple, the system sends a request to the video card, if the video card takes more than 2 seconds to respond (I think it's the default response time), the OS will reset the video card driver

 

 https://msdn.microsoft.com/en-us/library/windows/hardware/ff569917%28v=vs.85%29.aspx?f=255&MSPPError=-2147217396

wtf? Ive had this in TF2 for years.

[zMeul]

wtf? Ive had this in TF2 for years.

then, I'm sorry to inform you that your video card could be defective 

[nunya bus]

for what its worth. I have a R9 380 and it doesnt crash. I wonder what it is.... That's causing it 
Infact this card is doing so damn well at 1440P i've cancelled my 980TI hybrid lol and will wait till Q1 next year to get HBM v2 news/ products reviews, and then maybe get a 980ti at a lower price, and there might be some in stock. been waiting 3 months almost for my TI hybrid. EVGA is so far behind!!

[zMeul]

the Dow Jones

yes, but AMD is not part of Dow Jones indices

who is? Intel, MS, Apple

[Sauron]

Who wants to bet that the memory of the 290/290X was overclocked too far.

 

it's straight up different memory

[asker509]

wtf? Ive had this in TF2 for years.

are you running a overclock or not running at stock voltage I know for me 2k15 doesn't not like any overclock I throw a at it and does this