My RX480’s are here. Time to give ’em a go!

IMG_0598.JPG

I have been waiting a couple months for my RX480’s of which I have wrote about already on this site. I ended up with some regular ASUS branded reference 8GB RX480 cards instead of the mildly overclocked and backplated XFX ones but that’s not what we are here to talk about.

Stock on these cards has been miserable here in Canada but I received my new RX480’s on Friday and have been diligently running benchmarks on my computer all weekend to get you guys some numbers before Monday. And it seems I have been successful.

IMG_0618.JPG

With a few minor hiccups with Crossfire I have had a pretty good look at the performance of my new cards versus the cards of the past. I included my faithful XFX Double Dissipation R9 280 overclocked to 1200MHz (it’s sweet spot), the ever relentless Sapphire R9 290 Tri-X OC factory overclocked to 1000MHz and my RX480’s that are clocked at the stock 1266MHz straight from AMD.

IMG_0606.JPG

Below are my PC specs listed for your information. You can also see these specs listed in the About section of the website:

CPU: AMD RX 8320E @ 4GHz

CPU COOLER: CORSAIR H80I V2

MOTHERBOARD: ASUS 970 PRO GAMING/AURA

MEMORY: 16GB G.SKILL RIPJAWS X @ 1866MHz

HDD: SAMSUNG 850 EVO 250GB SSD

POWER SUPPLY: EVGA SUPERNOVA G2 750W

CASE: IN WIN 303

First we will start with the synthetics to get them out of the way. I ran all of these benchmarks at default settings and only recorded their final graphics score.

3D Mark Sky Diver

skydiver

This benchmark is getting a little long in the tooth but I kept it in with the tests as I was including my oldie but goodie R9 280. From these you can see a pretty clear increase in graphics score with the RX480s in Crossfire clearly taking the lead. The R9 280 isn’t all that far behind the R9 290 in this test but that changes soon.

3D Mark Firestrike

firestrike

Here we really start to see which cards choke and which prevail. The R9 280 is clearly out of it’s element here and I believe it’s really due to it only having access to 3GB of memory. The R9 290 pulls ahead clearly being helped by having access to 4GB of memory. The RX480 on it’s own is just ahead of the R9 290 but we really see an increase in performance with the RX480 in crossfire doubling the score of the R9 290 and nearly doubling the score of the single RX480.

3D Mark Time Spy

time-spy

This is Futuremark’s answer to the DirectX 12 benchmark and it really pushes current hardware to it’s limits. We get relatively the same story here though the crossfire configuration really runs away with it. The R9 280 really has a hard time here and the single RX480 and R9 290 perform decently.

Valley Benchmark

valley

The results here are a little more mixed though generally, with the exception of the R9 290 pulling ahead of the RX480 in this case, the scores follow the same path as the other synthetics.

Now we move on to the non synthetic tests. These are all performed within game benchmarks. All are run at 1440P and each chart denotes their graphics setting.

Dirt Rally

dirtrally

While admittedly this game is optimized for AMD hardware it’s still worth it to run as it can be hard to get high frames per second still at high presets. Here you can see we see a similar result to the Valley Benchmark.

Metro Last Light

metroll

This is also a very demanding title still and we get a similar result to Dirt Rally and Valley. Here though we see a more significant increase in the crossfire setup over the rest.

Grand Theft Auto V

gtav

I am convinced here that crossfire was not functioning properly. The game scored lower in crossfire even though this game is supposed to scale relatively well. In this case it scored lower than the single RX480 and the R9 290. I worked on this for several hours and can’t seem to get the score any higher. If you have ideas let me know. Other than that issue you can see that the RX480 wins the cake in this case though all cards are within 7 frames per second.

Tomb Raider

tomb-raider

Most people have dropped Tomb Raider from their benchmark scores but I think it’s relevant. In this case it shows that you nearly double your frames per second with the RX480 in crossfire configuration and without the RX480 still takes the lead by a smooth 4 frames per second.

So there you have it performance-wise. It’s a pretty decent card.

Of course the big word out there today is VR compatibility. Well the only benchmark that you can run right now is Steams SteamVR Performance Benchmark. I ran this on all three cards and this is what I got.

steamvr

I don’t think the SteamVr Performance test is set to scale with crossfire yet but at least we don’t lose performance. All the card are READY according to the utility with the only exception of the R9 280 which is only CAPABLE receiving half the score of the other cards.

Also all this was done without the cards screaming loud. They stay relatively quiet considering they are blower style cards. They perform quite nicely in their crossfire configuration never going above 78 degrees in my IN WIN 303 case. I’m very impressed.

The question remains though, is it worth it to buy two of these over a GTX1080? For me personally the answer is yes. But I’m biased. I love AMD. That said the best advice for most people is to buy the best SINGLE card you can get your hands on. The GTA V benchmark at least clearly shows that while crossfire is relatively reliable it does not scale in everything.

Have you purchased the RX480? How have you liked the card and which card did you buy? Let me know below.

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s