Now We All See the Genius of AMD Going Lowend First

[Thala]

Default as fast as the GTX1060 FE in the five games. Even with an increased powerlimit it is slower than a MSI GTX1060. Power consumption is way higher (~200W vs. 120-140W for the MSI).

You forgot to mention, that the Strix RX480 is already faster than GTX1060FE without removing power target and even faster than the OCed MSI GTX1060 when overclocked....all in a tests containing of only DX11 games! Some of them Gameworks!!!
I guess i dont have to mention what happens if you give this beast DX12/Vulkan titles if it already beats MSI GTX1060 at DX11 gameworks.

 

[linkgoron]

Here is a preview of the Asus RX 480 Strix: https://www.computerbase.de/2016-07/asus-radeon-rx-480-strix-test/2/#abschnitt_benchmarks_in_full_hd

Default as fast as the GTX1060 FE in the five games. Even with an increased powerlimit it is slower than a MSI GTX1060. Power consumption is way higher (~200W vs. 120-140W for the MSI).

Like i said: A price cut is necessary that these AIBs can survive.

Looking at the increased power limit you mention, it's beating the 1060 gaming in COD by 20%, Witcher 3 by 7%. The AIB 1060 is beating it by 29% in Anno, by 8% in fallout 4 and 6% in Far Cry Primal.

According to their average, the 1060 gaming is just 3% faster in DX11, and 4% faster than the non-MT OC. Definitely a win for the AIB 1060. Looking at the games themselves, the 1060 wins in 3 out of 5 games. That's actually not a bad showing for the 480 consider that the 480 would beat the AIB 1060 in Quantum break, Hitman, Doom, Ashes and Total War DX12 (i.e. most dx12 and Vulkan games, the strong side of Polaris, which are mysteriously missing).

Also according to that review the difference is 50 watts, not 60-80 as you wrote in your post. While it is a difference, it's just 50 watts - not something major. Note that the non max target OC is using only 22 watts more than the 1060 gaming but the 1060 is just 4% faster in DX11, so we all know what'll happen if we add in the aforementioned DX12/Vulkan games.

I actually believe that you're exactly proving and driving home the point of this thread. DX12 is not quite here. For some reason a lot of sites are still not benching Doom Vulkan, or a significant amount of DX12, and therefore AMD are still being hindered by their poorer DX11 performance and not getting enough press about their superior DX12 performance. I believe that such a review, six months from now, will have at least 2 or 3 DX12 games.

On another note, does anyone have any insight on why Polaris has such a bad showing the FC Primal? Grenada is really performing well there.

 

[3DVagabond]

What's the difference between this and the reference card, outside of the fan and 8 pin? I just bought an accelero mono plus for my reference card.

Are you serious? Look at the card. Except for the GPU and RAM they are completely different. Nothing's the same.

 

[Bacon1]

Anno 2205 is terribly optimized on AMD hardware, I'd say its a worse test than Project Cars. Its also played as much as Ashes of the Singularity which sontin considers a benchmark only, so surprised he cares so much about it

Almost no core scaling for a CPU bound game, so makes sense it hurts AMD:

http://www.guru3d.com/articles_pages/anno_2205_pc_graphics_performance_benchmark_review,9.html

 

[poofyhairguy]

On another note, does anyone have any insight on why Polaris has such a bad showing the FC Primal? Grenada is really performing well there.

Yeah that is the big mystery. It's not like the consoles are expecting 512 busses.

 

[Sweepr]

Here is a preview of the Asus RX 480 Strix: https://www.computerbase.de/2016-07/asus-radeon-rx-480-strix-test/2/#abschnitt_benchmarks_in_full_hd

Default as fast as the GTX1060 FE in the five games. Even with an increased powerlimit it is slower than a MSI GTX1060. Power consumption is way higher (~200W vs. 120-140W for the MSI).

Like i said: A price cut is necessary that these AIBs can survive.

So according to ComputerBase still slower than the MSI 1060 with increased power limit, extra 50W on top as well. Also whoever believed the 1500 MHz guaranteed / some hitting above 1600 MHz OC rumors will be very disappointed:

The Polaris-10 GPU to the Asus Radeon RX 480 Strix can be around 40 MHz to 1,370 MHz overclock, before it comes to the first graphic errors. That is about 100 MHz more than the standard clock of the reference card, but only minor additional 40 MHz compared to Strix itself.

 

[USER8000]

Awesome Asus cooler,LOL!!

https://pics.computerbase.de/7/3/5/3/1/14-1080.2245261692.jpg

Edit to post!!

Supposedly,the overclock was on stock voltage.

 

[Jaskalas]

I actually believe that you're exactly proving and driving home the point of this thread. DX12 is not quite here. For some reason a lot of sites are still not benching Doom Vulkan, or a significant amount of DX12, and therefore AMD are still being hindered by their poorer DX11 performance and not getting enough press about their superior DX12 performance. I believe that such a review, six months from now, will have at least 2 or 3 DX12 games.

A six month time frame for NVIDIA to improve their drivers for DX12.
Surely it seems reasonable that they can close the gap?

 

[Snarf Snarf]

This isn't the first time ASUS has had a heat sink designed for another chip slapped onto the Radeon version, I think Hawaii had the same issues with DCUII coolers. This is why I said earlier using the Strix as a data point is a bit misleading, Sapphire and XFX will have MUCH better versions out.

I have Strix 1070 and 1080 installed in rigs here at our shop for show room builds and while they look great with RGB LED strips in cases the cooling was kind of meh for a triple fan set up. EVGA ACX or MSI Gaming coolers are at least in my opinion much better solutions.

 

[nurturedhate]

A six month time frame for NVIDIA to improve their drivers for DX12.
Surely it seems reasonable that they can close the gap?

Except for the fact that Nvidia has been telling us about their dx12 and async compute drivers for close to a year now with nothing to show for it. So it's not just 6 months.

 

[poofyhairguy]

Surely it seems reasonable that they can close the gap?

If the hardware isn't capable software can't close the gap.

Plus historically we see with Nvidia the performance for a card is at its best when its new, and over time the performance in relative terms drops. You have to give them credit- they know how to make a best first impression.

AMD could learn a thing or two about that.

 

[sontin]

And here is a full review of the Strix: https://www.techpowerup.com/reviews/ASUS/RX_480_STRIX_OC/23.html

Still slower than a GTX1060FE.

 

[amg377]

1060 still gets smoked in hitman =p

 

[sontin]

And that even without DX12.

 

[linkgoron]

And here is a full review of the Strix: https://www.techpowerup.com/reviews/ASUS/RX_480_STRIX_OC/23.html

Still slower than a GTX1060FE.

I like how, in the 1060 review, the 1060 was 7% faster than the reference 480 and now two days later in this review the 1060 is 14.4% faster than the ref 480 and 11% faster than the strix.

note that there is one game missing in this review, WoW which was previously in favor of the 1060.

1060 review (July 19th):

Strix review (July 21st):

 

[Madpacket]

1060 to me really only makes sense in a power and heat constrained HTPC where you can't fit a card longer than 6.8 inches. There are a few exotic ITX cases that fit this profile (HG Osmi for example) where an EVGA 1060 makes for the perfect card. However if you have room for the Rx480 and can exhaust an extra 20-30W of heat waste I believe it's the better overall card given the market is heading towards DX12/Vulcan.

 

[Bacon1]

I like how, in the 1060 review, the 1060 was 7% faster than the reference 480 and now two days later in this review the 1060 is 14.4% faster than the ref 480 and 11% faster than the strix.

note that there is one game missing in this review, WoW which was previously in favor of the 1060.

1060 review (July 19th):

Strix review (July 21st):

Uhh wow, that is pretty crazy. It went down a lot and now is next to a 780 ti.

970 got a whole lot closer as well.

What is going on over at TPU :O

 

[nurturedhate]

Uhh wow, that is pretty crazy. It went down a lot and now is next to a 780 ti.

970 got a whole lot closer as well.

What is going on over at TPU :O

small bag o cash

 

[Maverick177]

Lmao, wonder how Wizzard will spin this.

Edit: False alarm probably. Wizzard can't math.

 

[SolMiester]

1060 still gets smoked in hitman =p

Is that supposed to be a humorous post?....2 frames is smoked?...LOL

 

[SolMiester]

I like how, in the 1060 review, the 1060 was 7% faster than the reference 480 and now two days later in this review the 1060 is 14.4% faster than the ref 480 and 11% faster than the strix.

note that there is one game missing in this review, WoW which was previously in favor of the 1060.

1060 review (July 19th):

Strix review (July 21st):

You are using different charts with different ref points!...It shows nothing!
Ist chart is % of 1060, 2nd chart is % of ASUS 480!

 

[Rifter]

Strix isn't the best aftermarket cooler around, also that's all DX11 titles half of which tend to be NVidia skewed and the RX 480 looks pretty good in all of those FPS comparisons. 50 watts more for DX11 performance parity and better DX12 and Vulkan performance at very close to the same cost is the bottom line here.

You can spin it however you like but both the GTX 1060 and RX 480 present excellent values for a market segment that was in need of it. Both are compelling products and no amount of you bashing the RX 480 makes that any less true. It comes down to what titles someone plays, and how long they hold onto their cards that is going to determine what card gets recommended. You also can't discount the value proposition of cheaper FreeSync monitor with an RX 480 as a longer term solution than a GTX 1060 with a G-Sync alternative for $200 more.

well said

 

[linkgoron]

You are using different charts with different ref points!...It shows nothing!
Ist chart is % of 1060, 2nd chart is % of ASUS 480!

I'm not sure if you're joking or not.

Anyway,

First chart:
1060=100
480=93
ratio: 100/93=7.5%

Second chart:
1060=108
480=97
108/97 = 11.3%

Note that they've already admitted and fixed their problem in the second chart, they changed the image from what it was when I first posted. The numbers are a bit different now.

Apparently they've been doing wrong calculations for a while now, although as already stated in another thread- they've got other problems with their math.

 

[Rifter]

Awesome Asus cooler,LOL!!

https://pics.computerbase.de/7/3/5/3/1/14-1080.2245261692.jpg

Edit to post!!

Supposedly,the overclock was on stock voltage.

the 1070/80 asus strix i believe was even worse, the Strix is a marketing gimmick plain and simple, they didnt even bother designing a cooler that actually makes full contact with the GPU.

 

[Bacon1]

You are using different charts with different ref points!...It shows nothing!
Ist chart is % of 1060, 2nd chart is % of ASUS 480!

Considering Wizzard from TPU already admitted there was a mistake, apologized and fixed it, you look like a fanboy