AMD RX Vega Leaked Benchmark Shows It Ahead Of GTX 1080 – Specs Confirmed, 1630MHz Clock Speed, 8GB HBM2 & 484GB/s Of Bandwidth

[Delicieuxz]

Just now, Colonel_Gerdauf said:

Didn't a few people here say something to the effect of nVidia being frightened by Vega being the one and only reason the 1080TI exists at all?

What they are saying now?

 

<crickets>

 

Figured as much.

That could well have been true (though I think 4K gaming would also have been a major consideration). If Vega doesn't end up impressing, that doesn't mean Nvidia would have had magic goggles to see that would be the case when they chose to make the 1080 Ti.

[Colonel_Gerdauf]

23 minutes ago, Drak3 said:

To be fair, they might not exactly be wrong. Any 'fear' on Nvidia's part would now be proven a waste. In the end, we got the 1080Ti so it don't really matter.

Meanwhile what has AMD done in response to the 1080TI? So far, it looks to be absolutely nothing. They cannot afford to go "yo I am doing my own thing" when the results on that attitude are visibly lackluster, nor can they keep doing the price cuts needed to allow a more basic and higher-level form of competition against nVidia. That runs back to the original budget gamer problem with AMD products; very bad PR. While I can give some benefit of the doubt with Ryzen's promising but selective future, AMD very clearly cannot double down on both CPU and GPU space at the same time.

 

The ye olde sentiment holds true to this day; the buyout of ATI was a grave mistake.

[Colonel_Gerdauf]

11 minutes ago, Delicieuxz said:

That could well have been true (though I think 4K gaming would also have been a major consideration). If Vega doesn't end up impressing, that doesn't mean Nvidia would have had magic goggles to see that would be the case when they chose to make the 1080 Ti.

These kinds of things do not get worked out in one day - and also, NVidia and AMD tend to have very deep insights into each others road-maps and even paper designs. There is no legitimate reason nVidia could not have seen the issues coming; especially not when competitive bias comes into play. If NVidia is blind to the issues of Vega, then they are blind to Vega in general; marketing does not affect them the same way it affects us. In that sense, they had nothing to worry about and released the 1080TI either as a result of internal schedules or as a Viagra pill.

 

And thus we have run into a massive logical hole - a paradox, in other words.

[Jito463]

22 minutes ago, Colonel_Gerdauf said:

And thus we have run into a massive logical hole - a paradox, in other words.

Did someone mention a pair-a-docs?

 

Spoiler

 

[leadeater]

30 minutes ago, Colonel_Gerdauf said:

-snip-

Well Nvidia was always going to release a 1080Ti, just like 980Ti and 780Ti. Recent trend from Nvidia is to always hold back on launching new products with fully enabled dies so they can give a mid generation performance jump and assess other factors like yields and optimizations that might need to be done.

 

When Nvidia released the 1080 everyone at the time knew that the 1080Ti would be coming, the performance increase it gave from memory was higher than previous Ti's but I could be wrong on that.

[AlwaysFSX]

4 hours ago, Taf the Ghost said:

A lot of people are defaulting to the "Because Nvidia does X, AMD has to do X as well!" argument, when Raja specifically said wasn't the case. Once we get the optimized drivers, though, I do expect the FE's results to pick up.  There's also the point the FE is clocked lower than the RX Vega, but that's not up for discussion either.

 

It's in the ballpark, but it's not the same as Nvidia's approach, especially if there's anything interesting going on with that ability to switch Modes so easily.

How are you going to clock Vega up when it already has issues hitting its rated clocks out of the box?

[leadeater]

3 minutes ago, AlwaysFSX said:

How are you going to clock Vega up when it already has issues hitting its rated clocks out of the box?

Increase the default fan speed profile, wait for AIB boards with better cooling solutions, slap a water block on it, move to the arctic, buy a ton of dry ice and put a chunk on the backplate. Plenty of options, not all of them good .

[AlwaysFSX]

3 minutes ago, leadeater said:

Increase the default fan speed profile, wait for AIB boards with better cooling solutions, slap a water block on it, move to the arctic, buy a ton of dry ice and put a chunk on the backplate. Plenty of options, not all of them good .

(maybe make a card that isn't a slight incarnation of Fermi)

[MageTank]

Would be pretty sweet if the title of this thread was changed to "AMD RX Vega Leaked benchmark shows it ahead of a crippled GTX 1080". Am worried people are going to think that GTX 1080 3dmark 11 score is representative of a real GTX 1080. The same goes for that GTX 1070 score, which was clearly clocked at 1750mhz~. 

[ivryk]

Thats some serious gimp on the 1080 that they used, I just got 34k (33,963 ) on mine.

http://www.3dmark.com/3dm11/12258645

[Foxxer]

I'm still holding of 1080ti, until I see real results from both. 

and what the hell are those clock speeds on vidia lul.

[MageTank]

49 minutes ago, ivryk said:

Thats some serious gimp on the 1080 that they used, I just got 34k (33,963 ) on mine.

http://www.3dmark.com/3dm11/12258645

Yeah, they reduced that 1070's clock speed to 1750mhz in order to obtain it's score. I had to underclock mine from 2050mhz down to 1750mhz to match it. 200mhz lower than my stock turbo-boost clock speed (which is 1950mhz). 

 

May I ask what your 1080 was clocked at when you got 34k? My bet is around 1950-2050. 

[Red Hardware]

1 hour ago, MageTank said:

Would be pretty sweet if the title of this thread was changed to "AMD RX Vega Leaked benchmark shows it ahead of a crippled GTX 1080". Am worried people are going to think that GTX 1080 3dmark 11 score is representative of a real GTX 1080. The same goes for that GTX 1070 score, which was clearly clocked at 1750mhz~. 

Hi

The source is WCCFTECH!! Did you expect anything else ?

[Red Hardware]

2 hours ago, AlwaysFSX said:

(maybe make a card that isn't a slight incarnation of Fermi)

Hi

Nothing can be as bad as Fermi............. nothing

[MageTank]

Just now, Red Hardware said:

Hi

The source is WCCFTECH!! Did you expect anything else ?

Hi

Did you read ANY OF MY!!! previous posts in this thread? You'd already have your answer!  

[Red Hardware]

Just now, MageTank said:

Hi

Did you read ANY OF MY!!! previous posts in this thread? You'd already have your answer!  

Yes , i did 

I just wanted to make it clear for others that didn't realize that the source is wccftech

 

[MageTank]

Just now, Red Hardware said:

Yes , i did 

I just wanted to make it clear for others that didn't realize that the source is wccftech

 

By quoting a man that already knew? That seems... counterproductive. Such effort would be better spent on those that actually fell for it. You know, like the vast majority of others in the thread? lol

[ATFink]

9 hours ago, Terryv said:

I sold my fury x for 450$ (also what I paid for it when i got it new) and pick up a local 1060 for 180$ to hold me over.

Nice man! My card was a fury, not fury x and I paid nothing for a new GPU since I didn't buy a new GPU. I've had the gtx 760 since 3 years ago.

[Red Hardware]

3 hours ago, Colonel_Gerdauf said:

These kinds of things do not get worked out in one day - and also, NVidia and AMD tend to have very deep insights into each others road-maps and even paper designs. There is no legitimate reason nVidia could not have seen the issues coming; especially not when competitive bias comes into play. If NVidia is blind to the issues of Vega, then they are blind to Vega in general; marketing does not affect them the same way it affects us. In that sense, they had nothing to worry about and released the 1080TI either as a result of internal schedules or as a Viagra pill.

 

And thus we have run into a massive logical hole - a paradox, in other words.

Hi , 1080ti wasn't the result of anything other than their schedules , the ti thing was't anything new as we had 980ti , 780ti 

The thing the surprised most of the people and made them believe that nvidia did that because of vega was it's performance , it was faster than titan x pascal (old xp) but it was significantly cheaper. 

3 hours ago, Colonel_Gerdauf said:

the buyout of ATI was a grave mistake.

No , it wasn't !! I know it caused budget problems and damaged their CPU department but they could've used it really well (they didn't) . They focused on APUs and that caused a lot of problems . The old CEO made some mistakes with their strategies. 

Also some of their top guys left the company at the time ( jim keller , raja koduri , Bob Drebin ) so you shouldn't blame it all on ATI

For those who don't know : jim keller came back did zen and left amd again(unfortunately) ; raja is back so there is some hope for navi and bob drebin is still working at apple

[Red Hardware]

30 minutes ago, MageTank said:

By quoting a man that already knew? That seems... counterproductive. Such effort would be better spent on those that actually fell for it. You know, like the vast majority of others in the thread? lol

You are right , sorry 

[Foxxer]

31 minutes ago, MageTank said:

-snip-

You need to relax brother. 

[MageTank]

Just now, Foxxer said:

You need to relax brother. 

I am. Though, I find it odd that people get the misconception that I am worked up, when this is my normal way of typing, lol. Perhaps I need to alter the tone of my text slightly to avoid that in the future.

[Red Hardware]

1 minute ago, GamingMemeKing said:

I wonder what the price will be compared to the 1080.

I'm sure it's cheaper. Around 450 USD would be nice for it

[ivryk]

5 hours ago, MageTank said:

Yeah, they reduced that 1070's clock speed to 1750mhz in order to obtain it's score. I had to underclock mine from 2050mhz down to 1750mhz to match it. 200mhz lower than my stock turbo-boost clock speed (which is 1950mhz). 

 

May I ask what your 1080 was clocked at when you got 34k? My bet is around 1950-2050. 

Either 2050~2067 ( was jumping back and forth ) on the clock and 500 on the memory. Which seems pretty standard ( give or take some) for most of pascal-based gpus. That gimp thou.