Mantle Benchmark Thread

[uksoldierboy]

First load of results
BF4 Multiplayer
Full 64player servers used
Full Ultra settings-4x Msaa-90FOV-1080P

System used
I7 4770K @4.6
Asus Reference 290X @1150/1400

Paracel Storm
Direct X
Frames:- Avg: 93.791 - Min: 63 - Max: 192

Mantle API
Frames:- Avg: 108.924 - Min: 74 - Max: 228 (16% Faster)

Dawnbreaker
Direct X
Frames:- Avg: 92.786 - Min: 81 - Max: 102

Mantle API
Frames:- Avg: 119.644 - Min: 8 - Max: 228 (29% Faster)

Golmud Railway
Direct X
Frames:- Avg: 112.552 - Min: 50 - Max: 154

Mantle API
Frames:- Avg: 122.262 - Min: 12 - Max: 228 (9% Faster)

Locker room
Direct X
Frames:- Avg: 116.688 - Min: 82 - Max: 161

Mantle API
Frames:- Avg: 107.833 - Min: 79 - Max: 185 (-7% Slower)

As you can see Mantle for the most part is working brilliantly in my system, but it's not without it's problems.

:Cpu spikes/stutter in most maps resulting in very low minimum framerates.

:Random crashes when joining or leaving a game (Battlefield stopped responding)

:Known fog/colour loss bug

Even with the issues though i think mantle is off to a great start, the performance increases are there for those with the currently supported hardware.

The issues i'm sure will be fixed soon enough and Support and optimizations for all GCN added asap.

Things are looking good some far though and i'm eager to see what AMD can do with Mantle.

 

[LtMatt]

Good testing Reece.

 

[uksoldierboy]

Cheers matt, going take a little break and try some more maps later.

 

[Silent_Scone]

First loads of results

Nice work, some nice gains. Weird how Locker is slower of all maps.

Is anyone in the position to do some 1440P tests? I've a 290Pro to try out but it's all too much effort this week. Noticed the fogging / colour palette is a little off too but I think that might be something for DICE to amend.

 

[ashmanuk67]

Best bit of testing results I've seen so far great work

Very early days but looking promising, But i thought it would be more shall we say ready than it is, As i thought AMD was waiting on Dice??

It must be very exciting for you guys because every new bit of software will give you better gains

Also i have seen some with Xfire on here saying they have really good results? i thought this was not working yet

 

[Beerbaron]

I completely removed my AMD drivers and then reinstalled 14.1. I still get the massive CPU spikes. Looks like it's quite a common issue.

 

[Gregster]

First loads of results
BF4 Multiplayer
Full 64player servers used
Full Ultra settings-4x Msaa-90FOV-1080P

System used
I7 4770K @4.6
Asus Reference 290X @1150/1400

Dawnbreaker
Direct X
Frames:- Avg: 92.786 - Min: 81 - Max: 102

Mantle API
Frames:- Avg: 119.644 - Min: 8 - Max: 228 (29% Faster)

Something wrong there, as you previously scored this on DawnBreaker.

Just had a few rounds making sure my OC was stable, it is which is great.

Card now running at 1150/1450 until it's underwater, i've asked Santa for a water loop so we'll see what happens.

Fps have increased nicely.

Siege of Shanghai
2013-11-15 13:43:22 - bf4
Frames: 178424 - Time: 1365703ms - Avg: 116.069 - Min: 73 - Max: 202

Dawnbreaker
2013-11-15 14:05:03 - bf4
Frames: 198813 - Time: 1619469ms - Avg: 122.764 - Min: 81 - Max: 201

Have they crippled DX to make Mantle look better or just a driver bug?

 

[Silent_Scone]

Yeah that's really confusingly inconsistent TBH! Between that and Star Swarm, it's becoming really hard to get some solid comparisons so far. Which is annoying because it's making me want to plumb in the 290 Pro now lol.

 

[uksoldierboy]

Cheers guys, lol no Greg just different play styles i think, i spent all of my time in the action on these mantle runs as an engineer.

Frosty in regards to Locker, i really don't have a clue why that was slower, i played as support like i always do on locker, same choke points etc.

 

[Gregster]

Cheers guys, lol no Greg just different play styles i think, i spent all of my time in the action on these mantle runs as an engineer.

Frosty in regards to Locker, i really don't have a clue why that was slower, i played as support like i always do on locker, same choke points etc.

Sorry bud but it wouldn't make those differences. Something very wrong to me.

 

[uksoldierboy]

Would be so much more easier if it had a built in benchmark i know.

My results are MP after all and no two runs will be the same.

 

[uksoldierboy]

Sorry bud but it wouldn't make those differences. Something very wrong to me.

Yes it would, if i was to spend the whole map looking through a scope on the edge of the map, my averages would be huge, sorry pal but i can't agree with you there.

DX performance is still extremely good for me, as good as i remember it playing.

 

[Gregster]

Yes it would, if i was to spend the whole map looking through a scope on the edge of the map, my averages would be huge, sorry pal but i can't agree with you there.

DX performance is still extremely good for me, as good as i remember it playing.

Not worth arguing over and if you posted those first results in the BF4 thread looking up at the sky, that is your choice/style of play

 

[uksoldierboy]

I'm not arguing over it bud

Those results are from playing as an engineer, mixing it up with tanks, but hell, if thinking that helps you look at those titans of yours without feeling a tiny bit sad, i'm happy for you.

 

[Gregster]

I'm not arguing over it bud

Those results are from playing as an engineer, mixing it up with tanks, but hell, if thinking that helps you look at those titans of yours without feeling a tiny bit sad, i'm happy for you.

Yep, I am very sad with my Titans

 

[uksoldierboy]

Bite bite Greg.

 

[Gregster]

Bite bite Greg.

Ok.

I could sell my 2 Titans, buy 2 290X's and have plenty left to go on a weekend bender

But nothing worth me selling them for in truth.

But I am getting off topic. I wasn't picking fault and just saying it looks strange. Those are some long play times, so unless you are face down in the grass the whole time, I would have expected similar frames, especially as it was the same person doing the testing.

 

[uksoldierboy]

Lol sounds like a plan to me bud!, you get rid of that green crap, name a pub and i'll join you for said bender where the only discussion will be Liverpool

In all seriousness though, i won't defend Mantle, if i think something is not quite right, i'll say, i don't see the point in blind loyalty to company that don't even know i exist.

AMD have done me no favours, we aren't friends, the only reason i have an AMD card in my system is that for me, at this moment in time, it offers me everything that i want at a price i'm comfortable paying.

Greg if you like i could benchmark while playing alongside you? i really do wish they'd do a benchmarking tool for this though.

 

[SiDeards73]

guys keep this thread to purely benchmarks etc if possible, if its going to end up being a debate about mantle vs dx, or AMD vs Nvidia, or shoes vs trainers etc, or anything thats not really a mantle parse then go take it to the other threads.

Thanks

 

[uksoldierboy]

Rgr that Si, back on task.