RX 480 Pics and Benchmark Leaks Emerge from China

[Dabombinable]

My main questions. Does it sound like a hairdryer? And is its cooler quieter than the Gigabyte Windforce x3?

[Big-G]

2 hours ago, Dabombinable said:

My main questions. Does it sound like a hairdryer? And is its cooler quieter than the Gigabyte Windforce x3?

As I don't own a Windforce... wouldn't know the difference.

For me it's not loud at all... I'm still running stock fan profiles as they come from EVGA.

[Briggsy]

Here's another leak, directly from 3DMark results. Seems promising as this is stock with Graphic score of 2906, which is stock 390x territory.

 

http://www.3dmark.com/fs/8900159

 

[Bouzoo]

20 hours ago, Dabombinable said:

My main questions. Does it sound like a hairdryer? And is its cooler quieter than the Gigabyte Windforce x3?

Apparently, stock cooler version sits at 71.9 C, while 1070 stock sits on 71.1C for comparison, was published yesterday here, but the video was removed. Apparently that was while using 100W on load which is also interesting, and if that all is true that is horrible. And apparently it's quiet, which might hint that the fans could keep it colder etc etc. However, leaks leaks leaks, we'll know all in few days.

[Dabombinable]

1 hour ago, Bouzoo said:

Apparently, stock cooler version sits at 71.9 C, while 1070 stock sits on 71.1C for comparison, was published yesterday here, but the video was removed. Apparently that was while using 100W on load which is also interesting, and if that all is true that is horrible. And apparently it's quiet, which might hint that the fans could keep it colder etc etc. However, leaks leaks leaks, we'll know all in few days.

The main reason I want to know is that:

• Gigabyte went full retard with vRAM placement on the GTX 970 G1 Gaming-half the vRAM on the back with no cooling at all despite the GTX 980 G1 Gaming having it all on the front and having a pin compatible GPU and not having that flaw
• It gets distractingly noisy-far noisier than the case fans at 100%-one alone moving more air than 2x of the 80mm fans on the Windforce X3

[Briggsy]

6 minutes ago, Dabombinable said:

The main reason I want to know is that:

• Gigabyte went full retard with vRAM placement on the GTX 970 G1 Gaming-half the vRAM on the back with no cooling at all despite the GTX 980 G1 Gaming having it all on the front and having a pin compatible GPU and not having that flaw
• It gets distractingly noisy-far noisier than the case fans at 100%-one alone moving more air than 2x of the 80mm fans on the Windforce X3

I know you you mean about the windforce x3 fans, I have a 290 version of that and its gotta be the noisiest GPU cooler I've ever had under load.

[No]

just wait for drivers

[Dabombinable]

5 minutes ago, Briggsy said:

I know you you mean about the windforce x3 fans, I have a 290 version of that and its gotta be the noisiest GPU cooler I've ever had under load.

I still don't know why people say its quiet-its far from it and is as loud as the stock fan on my Seidon 120V (rev 1)-and the stock fan is notoriously loud. 

[Briggsy]

15 minutes ago, No said:

just wait for drivers

 

 

(sorry, it had to be done  )

[Briggsy]

Just found a 1080p crazy score for a single RX 480.

 

36 FPS average is a couple FPS higher than I was expecting, given that AMD_Robert said the crossfire test was scaling at 183% for ~63FPS at Computex, this would mean that AMD have improved 480 performance by roughly 5% since then. Either that, or this 480 is overclocked.

 

[Big-G]

4 hours ago, Briggsy said:

Here's another leak, directly from 3DMark results. Seems promising as this is stock with Graphic score of 2906, which is stock 390x territory.

 

http://www.3dmark.com/fs/8900159

LOL... mkay...

 

RX 480 vs my EVGA GTX 980 Classified as out of the box...

 

http://www.3dmark.com/compare/fs/8900159/fs/6661531

 

I'll keep my 980

[handymanshandle]

On 6/20/2016 at 11:43 AM, awesomeness10120 said:

How is this a bad thing? It's only 200 to 270 dollars. Devalues every previous generation card.

If the FS:E and FS:U scores are low because of drivers then okay sure.

If the FS:E and FS:U scores are closer to R9 380X because of hardware, $200 to $280 is the price range that the R9 380/380 are at. I'm hoping it's just drivers and we do see it actually being 5% to 10% slower than the GTX 1070. 

 

I don't really want to go with 1080p (would rather solid 60fps at 1440p with high to ultra preset.) but I guess maybe for the RX 480 I'd consider it.

[don_svetlio]

11 hours ago, Bouzoo said:

Apparently, stock cooler version sits at 71.9 C, while 1070 stock sits on 71.1C for comparison, was published yesterday here, but the video was removed. Apparently that was while using 100W on load which is also interesting, and if that all is true that is horrible. And apparently it's quiet, which might hint that the fans could keep it colder etc etc. However, leaks leaks leaks, we'll know all in few days.

Horrible? 100W power draw with barely any fan speed and reasonable temps is horrible?

[LAwLz]

I am still expecting it to be around the 390 in terms of performance.

That level of performance, plus the lower power, new features and low price would make it a really good card. I think that if you expect 980Ti performance you will probably end up disappointed. 

 

It's better to have low expectations and then have them exceeded,than to have high expectations that aren't met. 

[Bouzoo]

34 minutes ago, don_svetlio said:

Horrible? 100W power draw with barely any fan speed and reasonable temps is horrible?

Only 100W (~61% less than a maxed 1070) power draw while running at allegedly 300-400 MHz (~20-30%) lower clocks than a 1070 and hitting same temps is horrible. If it's while fans are not being maxed then it may be a small difference, but based on leaked info, cooling might be worse than what NV has.

[Mr.Kios]

Just watched this 

He says may or may not be 4k max settings. Now i don't own overwatch. Does anyone know how taxing is overwatch on max settings at 4k? I am no taking it as true, but i might be a good indicator if true. 
The monitor though is WQHD.

[don_svetlio]

3 minutes ago, Bouzoo said:

Only 100W (~61% less than a maxed 1070) power draw while running at allegedly 300-400 MHz (~20-30%) lower clocks than a 1070 and hitting same temps is horrible. 

Actually, no. You see, you are powering 400 more compute cores than the 1070. 1070 also draws around 170W under max OC (as evident by the 1080 going towards 200W with a proper OC) thus meaning you are drawing 70% less than a 1070, powering 20% more compute cores. We also don't know fan speeds.

As for clockspeed - are you really comparing clockspeed across vendors AND architectures? I expected more from you man

[YubinTheBunny]

9 minutes ago, Mr.Kios said:

Just watched this 

He says may or may not be 4k max settings. Now i don't own overwatch. Does anyone know how taxing is overwatch on max settings at 4k? I am no taking it as true, but i might be a good indicator if true. 
The monitor though is WQHD.

I get about 80-90fps on Overwatch @4k with all the settings on max (Minus the AA) on a single R9 390 so I'm assuming the RX 480 should perform similar based on these leaked benchmarks.

[Bouzoo]

1 minute ago, don_svetlio said:

Actually, no. You see, you are powering 400 more compute cores than the 1070. 1070 also draws around 170W under max OC (as evident by the 180 going towards 200W with a proper OC) thus meaning you are drawing 70% less than a 1070, powering 20% more compute cores. We also don't know fan speeds.

As for clockspeed - are you really comparing clockspeed across vendors AND architectures? I expected more from you man

Clockspeeds are the main factor for higher temps. Not the only one, but main. Voltages etc, not going in specifics. And that higher power draw even more so backs up what I'm saying. 200W vs 100W and same temps? I am absolutely comparing apples to oranges here because from info that we have, if NV can deliver higher clockspeeds, noticeably better performance and higher power draw at the same temps, then they did the better job. No arguing that I am jumping to conclusions (as in leak info conclusions, I still expect everything and nothing from Polaris at the same time), we know nothing about the fans speeds, if those leaks are even true, yada yada yada, but everything stands. Either way, I'll be getting an AMD card this year just to be clear, but doesn't change the fact that based on this info (which again, may or may not be true), NV did a better job in everything except price. 

[Lays]

14 minutes ago, Mr.Kios said:

Just watched this 

He says may or may not be 4k max settings. Now i don't own overwatch. Does anyone know how taxing is overwatch on max settings at 4k? I am no taking it as true, but i might be a good indicator if true. 
The monitor though is WQHD.

 

 

I can play at 1440p Ultra and get about~80-100 fps on a 780, so that tells you about how demanding it is.

 

With my 980 TI before, I could easily game at 120-144 most of the time at 1440p.

 

(Not demanding at all really)

[don_svetlio]

7 minutes ago, Bouzoo said:

Clockspeeds are the main factor for higher temps. Not the only one, but main. Voltages etc, not going in specifics. And that higher power draw even more so backs up what I'm saying. 200W vs 100W and same temps? I am absolutely comparing apples to oranges here because from info that we have, if NV can deliver higher clockspeeds, noticeably better performance and higher power draw at the same temps, then they did the better job. No arguing that I am jumping to conclusions (as in leak info conclusions, I still expect everything and nothing from Polaris at the same time), we know nothing about the fans speeds, if those leaks are even true, yada yada yada, but everything stands. Either way, I'll be getting an AMD card this year just to be clear, but doesn't change the fact that based on this info (which again, may or may not be true), NV did a better job in everything except price. 

You're comparing Intel to Samsung actually... - if the 1070 needs 50% fan speed for 70*C and the RX 480 needs 30% fan speed, what would that mean? Also, the COOLER is not the same - comparing temps is SO pointless as  - fan speed, cooler, GPU, voltage - EVERYTHING is different - all we know is the 480 is more efficient than the 1070 using 70W less to power more compute cores and that is to be expected. Pascal is a modded version of Maxwell on 16nm - it's performance stems from higher clockspeeds and thus more heat than Maxwell. And we already know GCN scales MUCH better with clockspeed as evident by a 1200MHz 390 beating the 1500MHz 970.

[Mr.Kios]

5 minutes ago, YubinTheBunny said:

I get about 80-90fps on Overwatch @4k with all the settings on max (Minus the AA) on a single R9 390 so I'm assuming the RX 480 should perform similar based on these leaked benchmarks.

 

3 minutes ago, Lays said:

 

 

I can play at 1440p Ultra and get about~80-100 fps on a 780, so that tells you about how demanding it is.

 

With my 980 TI before, I could easily game at 120-144 most of the time at 1440p.

 

(Not demanding at all really)

So IF (Big IF though) video is true (4k max settings, no FPS to help us) then it should be probably 390x level of performance, with a really small possibility of performance close to a 980 ti. Would you agree on that? or what i am supposing is BS

[AlwaysFSX]

1 hour ago, don_svetlio said:

Horrible? 100W power draw with barely any fan speed and reasonable temps is horrible?

Yup, it should be less power, less heat, and less fan speed.

 

AMD dun goofed this one.

[don_svetlio]

1 minute ago, AlwaysFSX said:

Yup, it should be less power, less heat, and less fan speed.

 

AMD dun goofed this one.

And if the cooler is running at 10% fan speed?

[AlwaysFSX]

3 minutes ago, don_svetlio said:

And if the cooler is running at 10% fan speed?

Not good enough.