ASUS ROG STRIX RX 480 O8G GAMING Video Card Review @ [H]

I agree with Kyle, that it appears to be the top 480 card people can buy....So the Gold award seems justified to me. The power consumption is common knowledge for 6 months now lol and +100 watts is no big deal to me
Historically ATI/AMD cards get better with driver optimizations. It will probably only get better versus the 1060.
 
Kyle and Brent
Why do newer games like The Division get removed from your game test suite while older and less graphically impressive ones like Fallout 4 still continue to remain. btw Division also saw a DX12 patch which shows a nice bump in performance for a few cards like GTX 1080, Rx 480 and Fury X. Division is showing that DX12 could improve the user experience. Removing newer and more graphically demanding games while retaining older less impressive titles does not help when trying to determine the capabilities of these latest GPUs and their architectures in latest apis like DX12. Whats more relevant is you are searching for titles which improve the user experience while running on latest APIs. So its really puzzling to see you leave out The Division while continuing with Fallout 4.

https://www.computerbase.de/2016-12/the-division-dx12-benchmark/

Because a new title that is also gameworks and by the same publisher replaced it.

That being watchdogs 2. Ubisoft published watchdogs 2 and the division.

The division is simply hardly played anymore even those it is a bit newer than fallout 4.

Games that will stay in charts longer are typically more popular games, which makes sense because people makes videocard purchases on games they are playing.
 
Internal benchmarks:
  • Only useful information I get from them is that I can directly compare what I get to what others gets - gives a data point which is useful and to use to see changes due to system changes such as OCing, faster memory etc.
  • They don't tell you about stutter, missing textures, corruption or any other issue which affects game play for a given piece of hardware
  • They don't always even remotely reflect real game play for the title
    • GTA V with my Nano, setting the options for a good benchmark was utterly unplayable in real game play, I am talking high 40's in the benchmark and less then 20fps in the game on one mission!
    • Dues X MD - bench mark like so many others is not even from a FPS prospective - it also does not reflect FPS in real game play - it seems to be much less. So setting game settings for that benchmark would lead to reduce IQ settings which you could use in the real game
  • Then it also becomes obvious that Nvidia and AMD then waste time to optimize for the benchmark and not always the game making real game play suffer
I value real game play from professionals as the elite in reviews which is by far the hardest to achieve. I would have a very hard time duplicating the consistency that HardOCP routinely does with their testing. That is OK since that is why you have professionals who know what they are doing getting the job.
 
Enjoyed this article. Went red this round upgrading from a GTX 770 to an MSI RX480. Was hesitant to pick the 480 over the 1060 (1060 is still amazing), but with generous rebates (including a discount for BF1) I haven't regretted buying the 480 this time.
 
Enjoyed this article. Went red this round upgrading from a GTX 770 to an MSI RX480. Was hesitant to pick the 480 over the 1060 (1060 is still amazing), but with generous rebates (including a discount for BF1) I haven't regretted buying the 480 this time.

Have you had any issues with crashes? I just bought this card and have had many crashes in BF1 and Witcher 3.

Update: I am getting a new card today. After changing to a 750 Watt power supply, and trying the newest beta drivers and the newest WQHL, underclocking the card, increasing the power limit at stock, etc. Hopefully the card was just faulty and the new card fixes my problems.
 
Last edited:
Fine..... I bought one..... Twistin' arms n stuff..... Now to upgrade the other box with some new green team flavors as well.
 
Good god.... I know they say this is a big card in the reviews, but man is it big damned card. It's like stuffing a size 13 shoe into the case..... Very pleased with this purchase though. I may go with the green versions of this for the other box. Although I think two Strix 1070's or 80's may not fit in a Haf 932 case. I'll put my shoe in there first.
 
Good god.... I know they say this is a big card in the reviews, but man is it big damned card. It's like stuffing a size 13 shoe into the case..... Very pleased with this purchase though. I may go with the green versions of this for the other box. Although I think two Strix 1070's or 80's may not fit in a Haf 932 case. I'll put my shoe in there first.
Yeah that was one advantage of the Sapphire (relative to other 480 cards), not sure how many other 480 are similar dimensions to the Sapphire 480 models.
Cheers
 
Have you had any issues with crashes? I just bought this card and have had many crashes in BF1 and Witcher 3.

Update: I am getting a new card today. After changing to a 750 Watt power supply, and trying the newest beta drivers and the newest WQHL, underclocking the card, increasing the power limit at stock, etc. Hopefully the card was just faulty and the new card fixes my problems.

Maybe an occasional crash or freeze, one thing I did see initially was this light bloom just on certain BF1 maps. Have not seen that for some time now. Still satisfied with this purchase.
 
Maybe an occasional crash or freeze, one thing I did see initially was this light bloom just on certain BF1 maps. Have not seen that for some time now. Still satisfied with this purchase.

The new card fixed all my problems. Quite weird that Asus would let that slip through the QC cracks, but I know it happens.
 
The new card fixed all my problems. Quite weird that Asus would let that slip through the QC cracks, but I know it happens.

That's good they took care of you and your issue. Overall Asus products have served me well.
 
Benching mine right now with Time Spy. Was able to run 1410/8800 150% with stock volts although went ahead and bumped them up to 1.3v for the test I'm on now. What I don't get is the detailed results not showing the gpu on 1410 instead it shows 350.

Ok some crazy stuff here.
My stock voltage run @ 1410/8800 150% shows core clock at 350MHz 1,067MHz 4303 score
My 1.3v run @ 140/8800 150% shows core clock at 2,065MHz mem at 3241MHz 4279

This has gotta be an issue with 3DMARK. I'm sure the card would go into nuclear meltdown at 2000MHz. I also don't get why it doesn't report the ram speed properly either.

Just did the unlock on Time Spy and now it's reporting 1410/2200 gpu/mem. Maybe it was a limitation of the demo version. Not able to get above 4300 though with forced 1.3v lol Funny how the stock voltage would get me there but adding the extra voltage is hurting the score.
 
Last edited:
I eventually bought the PowerColor 470, so far it is okay it just gets very very hot.
I have felt the metal pipe that you can see on top the the card, it actually burns your finger.
69-73 degrees celcius when i play starcraft 2.

Not sure what other games to test, i have Call of duty black ops 1 and battle field 4 on dvd,havent tried them yet.
Maybe because i think i am stuck in the world of Socket 939 with DDR1 ram and a X2 4400+ CPU....
 
I picked up one of these open box at Microcenter for around $180.

Nothing "wrong" with it, but memory had no margin for overclock.
Previous owner(s) may have returned it for that. There's no telling.
At original marked price, I would have been greatly dissapointed.
For what I paid, just plain working was acceptable.

Gave it to a friend (purpose of the purchase) and did not have time to
attempt much in the way of GPU overclock. I will say: did not throttle.
Timespy benched about 200 points better than XFX RS at the same
Wattman settings. I assume the difference was "not throttling"?

All three GPU fans were spinning very quiet under benchmark load.
I could not hear them over standard power supply and CPU fans.

The default core voltage was slightly higher than mentioned the article.
I'm using Crimson + Wattman, have not tried the ASUS tweak driver.

It is a long and heavy card. Maybe worth your time to support the free
end with a wire sling or something.
 
Last edited:
I picked up one of these open box at Microcenter for around $180.

Nothing "wrong" with it, but memory had no margin for overclock.
Previous owner(s) may have returned it for that. There's no telling.
At original marked price, I would have been greatly dissapointed.
For what I paid, just plain working was acceptable.

Gave it to a friend (purpose of the purchase) and did not have time to
attempt much in the way of GPU overclock. I will say: did not throttle.
Timespy benched about 200 points better than XFX RS at the same
Wattman settings. I assume the difference was "not throttling"?

All three GPU fans were spinning very quiet under benchmark load.
I could not hear them over standard power supply and CPU fans.

The default core voltage was slightly higher than mentioned the article.
I'm using Crimson + Wattman, have not tried the ASUS tweak driver.

It is a long and heavy card. Maybe worth your time to support the free
end with a wire sling or something.

I have one of these also. My first one was returned because of instability, even at stock clocks. I also have very little room for memory overclocking, only getting 8400 MHz effective. Does anyone else notice Wattman doesn't allow any voltages above 1175 mV and that it doesn't even seem to change it when checked with GPU-z?
 
Back
Top