Overclock.net › Forums › Industry News › Rumors and Unconfirmed Articles › [VC]GTX 1060 specifications leaked - faster than RX 480
New Posts  All Forums:Forum Nav:

[VC]GTX 1060 specifications leaked - faster than RX 480 - Page 40  

post #391 of 735
Quote:
Originally Posted by Orthello View Post

One of the german reviews their ref sample did 1420mhz . It was the highest i have seen. So their might be the odd Golden AIB card that does 1500mhz yet. Still i think its golden card at 1500 rather than 1600.

Plus I think it is a matter of stability testing too. Considering his results are exceptionally higher than everyone elses, I think it is more likely they were not as thorough with their stability testing.
Quote:
Originally Posted by GorillaSceptre View Post

If that leak is real, then the 1060 on a Nvidia favored bench is only 8% faster than a "atrocious" reference 480? And it will probably have most of it's OC headroom used up by boost 3.0, so it shouldn't have nearly the same potential as the AIB 480's.

But i guess this 3GB card will be hailed as a resounding success, even when it ends up costing $250+.. Remember guys, it needs to handily beat the 980 to be a success. thumb.gif

Most of the criticism launched at rx 480 isn't it's price to performance, it's the engineering let down vs Nvidia. It is just so much slower than Polaris even with the die size taken into account.

If the 1060 is 8 percent faster, while being 20% smaller and achieves this using 20-25% less power, it basically means nvidia can play around with AMD because they beat them in every important metric.

It means they control pricing, which means AMD future is in Nvidia's hands. Being so behind Nvidia in the engineering department, all AMD can do is value price their products to compete with Nvidia. However because of the performance per mm2 aspect, Nvidia is capable of producing a faster product than AMD that is cheaper to make. This means AMD is forced to deal with lower margins, and if Nvidia decides to be aggressive like the 970 again, it can force AMD to take losses on their GPU department.

People wanted AMD at the very least show they closed the engineering gap and had perhaps Fury levels of performance for the full polaris chip.

I.e The 1080 being 35% larger than polaris is 42-45% faster.

What we got is the 1080 for being 35% larger, shows a gulf of 75% faster 80+% gap at higher resolutions. That performance per mm2 gap means they are just as far behind as Tonga vs the gm104 generation. Meaning AMD hasn't closed the gap at all. And if you remembered, this was a bad time for AMD marketshare and profits.

If this gap isn't closed, it also means vega has a very good chance of needing to be near hawaii sized to compete with a chip much smaller than it. And that isn't good when you add HBM to the overall costs. The current performance of polaris lowered peoples expectations for vega.

AMD needs to compete on the engineering level because this is what they need to if they want to be profitable again and this is what everyone desires.
Edited by tajoh111 - 7/5/16 at 3:26pm
post #392 of 735
Quote:
Originally Posted by tajoh111 View Post

Plus I think it is a matter of stability testing to. Considering his results are exceptionally higher than everyone elses, I think it is more likely they were not as thorough with their stability testing.

Yes could be right.

What we are yet to see though is a non throttling overclocked result. Eg you mention 1375 boost at guru3d.. did it maintain that boost at Guru3d through the benchmarks ? If the AIBs can maintain their boosts that also will make a difference as my pic is those boosts are throttling down to lesser numbers.
post #393 of 735
Quote:
Originally Posted by tajoh111 View Post

Plus I think it is a matter of stability testing to. Considering his results are exceptionally higher than everyone elses, I think it is more likely they were not as thorough with their stability testing.

GN got theirs to 1390 when eliminating the reference heat issues, the AIB's will have vastly superior power delivery too. I think 1400 is low-balling the AIB's imo, but we'll see, hopefully not much longer to wait.

We've seen the pcie power draw crash systems with low quality boards, it's not hard to believe that some of the stability problems past 1400 might of been down to more than the chip itself, even though the higher quality boards didn't entirely crash.
post #394 of 735
Quote:
Originally Posted by GorillaSceptre View Post

GN got theirs to 1390 when eliminating the reference heat issues, the AIB's will have vastly superior power delivery too. I think 1400 is low-balling the AIB's imo, but we'll see, hopefully not much longer to wait.

We've seen the pcie power draw crash systems with low quality boards, it's not hard to believe that some of the stability problems past 1400 might of been down to more than the chip itself, even though the higher quality boards didn't entirely crash.

Hopefully some leaks of the Nitro will come out soon , i think we will see more % gain from an AIB overclock at same mhz matched to a ref card OC. I'm sure the ref cards are throttling any overclock from its full potential. PCPer alluded to this in their review.

I don't want to bring out that age old drivers will improve things arguement but .. they simply will going by AMDs past history. In another year add 10-15% to these numbers and it will be over GTX980 stock level and well beyond GTX970 level for sure and even beyond both in DX12. So yeah if you are looking now to buy in this price category then only cards to compare against in this price bracket would be GTX1060 (if its in the price bracket in all reality see 1070 vs retail pricing). Not liking the lack of SLI either .. or 6gb vs 8gb on the 1060.
post #395 of 735
all new cards get better drivers post release, 1060 is no different


ref 1060 will probably be better than ref 480, AIB vs AIB might be closer
The Green Beast
(16 items)
 
  
CPUMotherboardGraphicsRAM
Intel i7 3770K @ 4500 Mhz ASRock Z77 Pro3 Gigabyte GTX 1080 G1 Gaming Crucial Ballistix 2x8GB DDR3-1600 
Hard DriveHard DriveHard DriveCooling
SSD Crucial M550 500GB SSD Samsung 850 Evo 1TB HDD Seagate 7200rpm 3TB Cooler Master Hyper 212 EVO; Xilence X5 
OSMonitorKeyboardPower
Windows 10 Pro 64-bit Acer Predator XB271HU 27" IPS Gsync 1440p 165Hz CM Storm QuickFire XT Cherry Red 800W modular 
CaseMouseMouse PadAudio
Fractal Design Define R4 Black Logitech G900 Chaos Spectrum SteelSeries QcK+ 4mm SK Gaming Realtek On-board 
  hide details  
The Green Beast
(16 items)
 
  
CPUMotherboardGraphicsRAM
Intel i7 3770K @ 4500 Mhz ASRock Z77 Pro3 Gigabyte GTX 1080 G1 Gaming Crucial Ballistix 2x8GB DDR3-1600 
Hard DriveHard DriveHard DriveCooling
SSD Crucial M550 500GB SSD Samsung 850 Evo 1TB HDD Seagate 7200rpm 3TB Cooler Master Hyper 212 EVO; Xilence X5 
OSMonitorKeyboardPower
Windows 10 Pro 64-bit Acer Predator XB271HU 27" IPS Gsync 1440p 165Hz CM Storm QuickFire XT Cherry Red 800W modular 
CaseMouseMouse PadAudio
Fractal Design Define R4 Black Logitech G900 Chaos Spectrum SteelSeries QcK+ 4mm SK Gaming Realtek On-board 
  hide details  
post #396 of 735
Quote:
Originally Posted by Orthello View Post

Hopefully some leaks of the Nitro will come out soon , i think we will see more % gain from an AIB overclock at same mhz matched to a ref card OC. I'm sure the ref cards are throttling any overclock from its full potential. PCPer alluded to this in their review.

I don't want to bring out that age old drivers will improve things arguement but .. they simply will going by AMDs past history. In another year add 10-15% to these numbers and it will be over GTX980 stock level and well beyond GTX970 level for sure and even beyond both in DX12. So yeah if you are looking now to buy in this price category then only cards to compare against in this price bracket would be GTX1060 (if its in the price bracket in all reality see 1070 vs retail pricing). Not liking the lack of SLI either .. or 6gb vs 8gb on the 1060.

Well, how much did 380's performance improve 960 over a year? Dare to take a guess? Well, according to TPU

https://www.techpowerup.com/reviews/ASUS/R9_380X_Strix/23.html

https://www.techpowerup.com/reviews/Palit/GeForce_GTX_1080_GameRock/25.html

It (the lead) has dropped from ~17% to ~6% averaged over fairly large game sample.
post #397 of 735
Quote:
Originally Posted by lolfail9001 View Post

Well, how much did 380's performance improve 960 over a year? Dare to take a guess? Well, according to TPU

https://www.techpowerup.com/reviews/ASUS/R9_380X_Strix/23.html

https://www.techpowerup.com/reviews/Palit/GeForce_GTX_1080_GameRock/25.html

It has dropped from ~17% to ~6%

Because the 380 is not the same card as the 380x lachen.gif
Workstation
(4 items)
 
  
CPUMotherboardGraphicsMonitor
Xeon E5-2690 Supermicro 2011 Nvidia GP100/ Vega FE Dell ultrasharp 4k 
  hide details  
Workstation
(4 items)
 
  
CPUMotherboardGraphicsMonitor
Xeon E5-2690 Supermicro 2011 Nvidia GP100/ Vega FE Dell ultrasharp 4k 
  hide details  
post #398 of 735
Quote:
Originally Posted by Orthello View Post

Hopefully some leaks of the Nitro will come out soon , i think we will see more % gain from an AIB overclock at same mhz matched to a ref card OC. I'm sure the ref cards are throttling any overclock from its full potential. PCPer alluded to this in their review.

I don't want to bring out that age old drivers will improve things arguement but .. they simply will going by AMDs past history. In another year add 10-15% to these numbers and it will be over GTX980 stock level and well beyond GTX970 level for sure and even beyond both in DX12. So yeah if you are looking now to buy in this price category then only cards to compare against in this price bracket would be GTX1060 (if its in the price bracket in all reality see 1070 vs retail pricing). Not liking the lack of SLI either .. or 6gb vs 8gb on the 1060.

~50€/$ more doesn't make the 1060 another category, also the 8GB that nobody needs, won't make any difference over 6, especially on mid range cards which have limited life span
post #399 of 735
Quote:
Originally Posted by sugarhell View Post

Because the 380 is not the same card as the 380x lachen.gif

0/10, both linked summaries contain 380.
post #400 of 735
Quote:
Originally Posted by sugarhell View Post

Because the 380 is not the same card as the 380x lachen.gif
he is comparing 380 to 960 in both of those ....

nothing to do with 380X (its just a link to an older review/summary)
The Green Beast
(16 items)
 
  
CPUMotherboardGraphicsRAM
Intel i7 3770K @ 4500 Mhz ASRock Z77 Pro3 Gigabyte GTX 1080 G1 Gaming Crucial Ballistix 2x8GB DDR3-1600 
Hard DriveHard DriveHard DriveCooling
SSD Crucial M550 500GB SSD Samsung 850 Evo 1TB HDD Seagate 7200rpm 3TB Cooler Master Hyper 212 EVO; Xilence X5 
OSMonitorKeyboardPower
Windows 10 Pro 64-bit Acer Predator XB271HU 27" IPS Gsync 1440p 165Hz CM Storm QuickFire XT Cherry Red 800W modular 
CaseMouseMouse PadAudio
Fractal Design Define R4 Black Logitech G900 Chaos Spectrum SteelSeries QcK+ 4mm SK Gaming Realtek On-board 
  hide details  
The Green Beast
(16 items)
 
  
CPUMotherboardGraphicsRAM
Intel i7 3770K @ 4500 Mhz ASRock Z77 Pro3 Gigabyte GTX 1080 G1 Gaming Crucial Ballistix 2x8GB DDR3-1600 
Hard DriveHard DriveHard DriveCooling
SSD Crucial M550 500GB SSD Samsung 850 Evo 1TB HDD Seagate 7200rpm 3TB Cooler Master Hyper 212 EVO; Xilence X5 
OSMonitorKeyboardPower
Windows 10 Pro 64-bit Acer Predator XB271HU 27" IPS Gsync 1440p 165Hz CM Storm QuickFire XT Cherry Red 800W modular 
CaseMouseMouse PadAudio
Fractal Design Define R4 Black Logitech G900 Chaos Spectrum SteelSeries QcK+ 4mm SK Gaming Realtek On-board 
  hide details  
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Rumors and Unconfirmed Articles
This thread is locked  
Overclock.net › Forums › Industry News › Rumors and Unconfirmed Articles › [VC]GTX 1060 specifications leaked - faster than RX 480