Soldato
Makes a huge difference on the 8350 minimums.
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
**Official THIEF-Incorporating AMD™ Mantle™/True Audio™/Bench mark Thread**
- Thread starter tommybhoy
- Start date
- Status
Associate
Looks like it works according to benchmarks.
Interested in how well the TrueAudio works, and if theres any hit to performance with it? Haven't taken the plunge and bought the game yet...
Interested in how well the TrueAudio works, and if theres any hit to performance with it? Haven't taken the plunge and bought the game yet...
Yep that was it thanks. I used AMD CCC uninstaller before installing 14.3 for the 1st time. This time I used Display Driver Uninstaller and True Audio is now an option.
Unfortunately my hopes that it would somehow make the Thief Audio experience better were seriously misguided. People 2 floors away still sound like they are right beside you.
Here's hoping AMDs next big game to showcase Mantle and Trueaudio isn't a complete turd like Thief.
Soldato
Managed to fix my problem, I had disabled crossfire because it didn't work in this game. Enabling crossfire sorted it even though mantle doesn't work with crossfire
Glad it's fixed. Can you post your bench scores like the other guys please BG?
Caporegime
- Joined
- 18 Oct 2002
- Posts
- 33,188
So a 4770k minimum in DX is 37fps while it's 29 fps on the FX8350, but under Mantle the 4770k minimums don't move and FX8350 minimums move all the way up to 41fps? That is pretty damn impressive.
Though that points to, as Mantle is no slower for the i7 under Mantle it's not been "un-optimised" but is DX itself not optimised for AMD cpu's? It's removing much of the overhead, but it actually got faster. Is DX itself not optimised to take full advantage of an AMD cpu, wouldn't surprise me as so much software is compiled with Intel compilers which flag everything possible on Intel CPU's and not much beyond the basics for AMD cpu's.
Removing the CPU overhead really shouldn't boost the FX8350 ahead of the i7, just bring it much closer.
Though that points to, as Mantle is no slower for the i7 under Mantle it's not been "un-optimised" but is DX itself not optimised for AMD cpu's? It's removing much of the overhead, but it actually got faster. Is DX itself not optimised to take full advantage of an AMD cpu, wouldn't surprise me as so much software is compiled with Intel compilers which flag everything possible on Intel CPU's and not much beyond the basics for AMD cpu's.
Removing the CPU overhead really shouldn't boost the FX8350 ahead of the i7, just bring it much closer.
Permabanned
Cant post attachments but here's a quick run.
FX8320 @ 4100
8GB PC1600 CL9
Win 8.1
Sapphire ref R9 290 @ 1100 / 1325
Korean 1440p @120
everything maxed in settings, trueaudio enabled, 2560 x 1440 vsync off
DX11
Min: 13.4
Max: 57.1
Ave: 42.1
Mantle
Min: 42.3
Max: 69.0
Ave: 50.0
FX8320 @ 4100
8GB PC1600 CL9
Win 8.1
Sapphire ref R9 290 @ 1100 / 1325
Korean 1440p @120
everything maxed in settings, trueaudio enabled, 2560 x 1440 vsync off
DX11
Min: 13.4
Max: 57.1
Ave: 42.1
Mantle
Min: 42.3
Max: 69.0
Ave: 50.0
Caporegime
- Joined
- 30 Jul 2013
- Posts
- 28,909
Stupid question: Where do I actually enable Mantle on Thief? I couldn't see it on any of the options screens.
I have the option for Trueaudio now which I turned on, but nothing about Mantle.
I have the option for Trueaudio now which I turned on, but nothing about Mantle.
The possible answer is cores. AMD's only advantage over Intel equivalents is the extra cores. Mantle is making more use of the cores and as such playing to AMDs only advantage.
I have an Intel i7 4770K overclocked to 4.5GHz and Mantle actually has lower average FPS in this game for me.
2560x1600 maximum details apart from SSAA off
DirectX
Min 41
Max 118
Avg 77
Mantle
Min 51
Max 98
Avg 68
So Mantle is ~12% slower on average. The difference is that Mantle gives ~25% higher minimums and ~20% lower highs. This keeps the game looking and feeling much smoother since the frame times remain tighter.
The option to enable Mantle is in the options screen on the game launcher. Make sure the 32bit option is unticked.
Caporegime
- Joined
- 30 Jul 2013
- Posts
- 28,909
Steam didn't appear to have a game launcher though, It just boots straight in to the game.
Caporegime
- Joined
- 18 Oct 2002
- Posts
- 33,188
Problem with that being that the quad core shows similar performance. Under mantle ultimately the most that can happen is offloading the non game threads, just the rendering threads to the extra cores and only on a chip with more cores. It's significantly faster on the quad core AMD chip also and importantly higher minimums on the quad core AMD than Intel chip.
It is very odd, but the massively higher minimums is great, not too surprising as minimums will often occur in complex scene situations which is why the draw count goes through the roof. So it's when DX starts using a heck of a lot more cpu time. But the massive improvement on AMD chips in mins and maxes really does hint at something beyond just the dx overhead happening, and how competitive a quad core AMD chip is with both the AMD octo and Intel quad.
It could point to something more basic than just overhead with DX but simply code that is effectively not optimised for AMD chips at all, the gap is simply too big and disappears too quickly and the advantage for Intel in minimums is there in DX but not in Mantle.
Soldato
I know it's not many games that support mantle, but those that do show it's a capable performer (with an FX). Some will still not want to accept it though, although we have been saying for a long time about the intel compilers and optimisations.
What I am seeing is healthy minimums which completely counter the 'go intel' advice train, making the odd
Caporegime
Can the guys who have posted DX11 results, post their Mantle results please. It would be very interesting to see how Mantle stacks up against DX11 
Permabanned
You get a mini screen where you have play, options etc before you actually launch the game. It's in that one.
Last edited:
Caporegime
On Mantle I got mins of 55.5FPS on my R9 290 with 4770K, for whatever reason it was locked to 60 FPS, so max was 61.3 with average at like 59.9 or whatnot.
Mins in DirectX before that were 40 odd.
Mins in DirectX before that were 40 odd.
Last edited:
Caporegime
DX xFire 40 FPS.
Mantle Single GPU 68 FPS
He has an FX and Phenom II rig, wonder which one he did that on?
http://www.overclock.net/t/1429303/amd-mantle-discussion-thread/1990#post_21971155
Mantle Single GPU 68 FPS
He has an FX and Phenom II rig, wonder which one he did that on?
http://www.overclock.net/t/1429303/amd-mantle-discussion-thread/1990#post_21971155
Permabanned
Sounds like a beta driver issue.
One thing I will say is that under mantle it's a hell of a lot smoother when picking pockets etc.
Trueaudio is a bit of a joke though, there are more and clearer sounds but the sound staging is just terrible still.
Hopefully someone decides to code up and do it from scratch instead of just patching in mantle and TA - even if it's a short demo that's playable.
One thing I will say is that under mantle it's a hell of a lot smoother when picking pockets etc.
Trueaudio is a bit of a joke though, there are more and clearer sounds but the sound staging is just terrible still.
Hopefully someone decides to code up and do it from scratch instead of just patching in mantle and TA - even if it's a short demo that's playable.
Caporegime
- Status