Hi /g/. A channel here in Brazil released a video with some RX 480 tests.
Here the tests: https://www.youtube.com/watch?v=eUiaJXLoKnE
>>55304774
How can it even use above 150w?
I mean it has just one 6 pin.
>Poland
>now BR
Is Venezuela next?
>>55304940
The chinks did it first though.
HUEHUEHUEHUEHUEHEUHEUEHUEH
>>55304898
Good PSUs can deliver a bit more than 75W through a 6pin plug
>>55304971
It's highly inadvisable to do that for long periods of time though
>>55304774
How's it even possible ? Technically speaking, the card can draw only 150W at most
>>55305001
Oh yeah, its Furmark though, so you won't be seeing that kind of power draw or temps in real life usage
>>55304774
Top lel, next thing you'll do is post a well-frame, unmistakable plane from an airport loading station and call it a UFO.
>>55305001
>>55305020
295x2 was powered by two 8pins and slot, so 375watt max whilst the card could pull well over 400watts.
Most decent PSUs are over engineered as they should be.
>it's fucking nothing
oh well, I'll stick to the 280x
>>55305033
Furmark is a joke, both AMD and Nvidia throttle during it so they don't blow up the cards, Nvidia even put a hardware limiter starting with the 500 series just for these situations.
>>55304898
They're using a handful of cheap multimeters. The peak they're reading is likely a 1ms~ momentary peak that every GPU in existence sees.
If average pull in a Furmark test remains around 150w then it will be peaking higher than that for brief periods.
>>55304774
Brazil can into relevance!
>>55304774
Sorry, I dont speak spanish
>>55305478
Brazil speaks Mexican, not Spanish fool
>>55304774
It looks.. uhm
>witcher 3 4K: 20 fps
>witcher 3 1080p, no filters: 55 fps
>division 1080p: 57 fps
>mordor 1080: around 60-80 fps
>ashes cwaizy 1080p: heavy batches 25 fps
Is this like, shit?
those division framerates are quite close to 980 actually
https://www.youtube.com/watch?v=ENoQZcLvS20
>>55304971
75 watt from the PCI slot and the rest from the 6-pin. Not that hard is it?
>>55305055
>Cheap multimeters
>Fluke
>Cheap
KEK
I don't know what to make of this wetback huehue fuckery. Does it mean the rx480 is underwhelming?
>>55306099
A handheld meter doesn't have the accuracy or resolution of a professional oscilloscope. Toms uses two of them together for the sake of accuracy, and being the gold standard in reviewing.
The nig rigged BR testing equipment isn't even in the same realm.
>>55304774
jesus christ they actually filmed their monitor with a cam for this.
>>55304774
>55304774
78 degrees in fucking Witcher 3. This isn't even a benchmark. HOUSEFIRE
>>55306177
Its exactly where its supposed to be. Around 390X/980, both stock. The earlier aftermarket 970 OC misled people into thinking it was just like 970. Sure it is like 970 OC, but remember, thats base RX480. When you OC 970, you get close to 980. Then 980 can also OC. So the point is, RX480 is exactly where it should be as a reference card. With OC and aftermarket mods, this will compete with 980 OC/aftermarket easily.
>>55304774
it's thermal throttling too. notice the fluctuating clocks in witcher 3 when it goes above 75C.
Sorry i cant understand illegal immigrants language
>>55306495
No, the clock fluctuates at much lower temperatures too, and the bar for thermal throttling wouldn't be set as low as 75 anyway. It's power throttling.
Going by the division benchmark it's actually as fast as a 980. WHAT TO BELIEVE? THE RUSE CRUISE ENDS IN A FEW HOURS.
>>55306540
why did they aim for such high clocks if it can't reach it using the single 6pin? am i missing something here or did they fuck up?
>>55306540
NOPE.
The card doesn't pull anywhere near 150w in gaming workloads that it pulls in Furmark.
>>55305717
Nodrivers yet.
>>55306768
so it was thermal throttling like i said? >>55306495
>>55306555
Division likes more cores.
>>55306555
Every architecture has its nuances, and performs better in certain workloads.
>>55306878
Maintaining a target temp is a possibility, though if that were the case it could just ramp up the fan speed, unless fan speed was locked at something low for noise levels.
FUCK YOU BRAZIL, YOU RUINED TIBIA
>>55306772
Benchmarks tomorrow, curious if 2x480 can beat the 1070.
>>55305717
how is it shit you retard, its faster than the 970
>>55304774
damn, TW3 runs like ass, and I've been looking forward to playing it the most.
>>55307524
it was a question, not a statement.
you have downs?
>>55307905
Oh I see you're fucking retarded
How about you shoke on your dad's dick faggot?
>>55307993
You forgot to eat today or something?
>>55308027
I forgot to not reply to people with obvious mental handicap
>>55308063
i apologise.
when you add this in your "arguments won.txt" file please input my name as buttking mcburger.
>>55304774
The result in The Witcher isn't bad, cosidering how hairwork is such difficult to render on AMD card.
>>55304774
I'm waiting to see how the aftermarket brands will make it run. My experience with reference cards is to never buy them because they all suck dick compared to what you can get some few weeks after. Also I just hate blowers in general and am too lazy to install my own coolers over it.
>>55305803
And considering the drivers will be optimized, it will reach the 980 level in most situations.
>>55308202
The result in The Witcher isn't that good, considering that hairworks were disabled.
>>55308274
>>55308274
>>55305478
>>55305491
Muricas...
>>55308293
I didn't know this. Thanks for the information.
I have mixed feelings about the card.
I hope prajeet poolooloo delivers.
>>55308309
>>55304774
Why did AMD send their latest video card to literal monkeys who leak everything before NDA?
>>55304774
Why the fuck is 4k a relevant measurement on a $200 card?
>>55308550
>expecting Brazilians to do anything right
>>55305717
Could some kind anon tell me which of these are which DirectX version?
>>55308693
Ashes could be either; The rest are DX11
>>55306601
>>55306878
>>55306931
Most likely the clocks were fluttering because it was hitting its frame rate target and clocking back to save power, then clocking back up as frames dipped.
