AMD Mantle API [updating]

My Prediction: Lower Fps, lower power. ;)

Given it's early state, I think Mantle looks very promising, though I in AMDs stead would have it introduced with Kaveri, rather than with Hawaii.
 
I get a feeling some people here can't be pleased no matter what.

I on my side am exited to get free performance uplift no matter how big or small.
People saying the benefit is too small on top of the range Intel 6 core CPU paired with cheap Radeon R7 need to have a reality check. Maybe you can do better?

There is one aspect of Mantle I'm interested in. How big of a difference will it make to computer power efficiency. Imagine 2x290X in CF capped to 60FPS constant. Using DX11 requires quite a substancial CPU to feed GPUs which burns power. What effect will Mantle and it's more effiecient nature have on total system power? I'm going to find out later on today :)

Well, sorry if from a company as big and important as AMD for the CPU/GPU market, I expect a more professional showcase of their technologies. Right today, they should at least had reviewers ready with performance numbers right of the bat, instead of presenting half-baked numbers in completely random test configs.
 
My expectation for Mantle was basically this. It's something to significantly help AMD's own CPU's first and foremost, was never expecting big gains with high-end GPU's and fast CPU's.

However, what I also expected from Mantle (or at least titles where Mantle is taken advantage of it) is much better frame pacing overall - basically more consistency in frame times, especially with lower-end CPU's.

yleruLW.jpg


By looking only at the two frame rate counters, you could say "WTF? Mantle only gives a 4fps boost!" - but then look at the graph on frame times in the lower left.
 
Well, sorry if from a company as big and important as AMD for the CPU/GPU market, I expect a more professional showcase of their technologies. Right today, they should at least had reviewers ready with performance numbers right of the bat, instead of presenting half-baked numbers in completely random test configs.

why they have already done that at a big conference........ today is DICE releasing a patch.....

My expectation for Mantle was basically this. It's something to significantly help AMD's own CPU's first and foremost, was never expecting big gains with high-end GPU's and fast CPU's.
.
.

when then your ignoring data points that dont fit your position..........
 
My Prediction: Lower Fps, lower power. ;)

Given it's early state, I think Mantle looks very promising, though I in AMDs stead would have it introduced with Kaveri, rather than with Hawaii.

Ha! Reread my post as it specifically states constant FPS. If I need 3GHz i5 under Mantle and 4.5GHz under DX11 to achieve that shoul show some difference on my power meter :)
Even at the same CPU settings there should be measurable gap I think.
 
why they have already done that at a big conference........ today is DICE releasing a patch......

I'll just quote myself, ok?

Right today, they should at least had reviewers ready with performance numbers right of the bat, instead of presenting half-baked numbers in completely random test configs.

If I would be running AMD or DICE (they are both at fault here really), I would be seriously pissed at my Marketing employees. Further, if there is a partnership between AMD and DICE on Mantle they should have advanced together with numbers, instead of going each their own way about it.
 
I'll just quote myself, ok?

and i would quote myself but i CBF...... AMD already did that... remember.

why would DICE do this?

also how do you get predictable runs in high server load situations unless you "half bake" it , aka use bot/something predefined on unreleased code.......


the hardware configs them selves are perfectly reasonable....
 
and i would quote myself but i CBF...... AMD already did that... remember.

why would DICE do this?

also how do you get predictable runs in high server load situations unless you "half bake" it , aka use bot/something predefined.


the hardware configs them selves are perfectly reasonable....

What runs, nevermind predictable? Do you really think it is a coincidence that 5 out of 6 screens show exactly or very approximate (less than 2 fps difference) FPS to the ones used as "average"? And the fact that the 6th screen, which was the only one that showed very different FPS from average (116 vs 121.5), was taken out after someone pointed it was from Single Player?

Lots of coincidences here...
 
Last edited by a moderator:
when then your ignoring data points that dont fit your position..........
I'm ignoring "data points" that don't affect my position of what I "expected"? Suggest you read my post again, as your response makes no sense in context. I am not making any declarative statements on Mantle performance as an API now or in the future.
 
Thanks, that is why I included it. Wanted a good spread. But also don't have the resources or time to do super extensive testing and comparison. There will be plenty of hardware review sites that do their own extensive comparisons. Though you should ignore sites that test multiplayer on empty servers.

Good to know (about the empty servers).

If I recall correctly you once said that for Battlefield 4, you would use Mantle mostly to optimize the CPU's load, which certainly seems to be the case based on those early numbers (very nice work, by the way).

I would imagine that in the future, you'll turn your attention a bit more towards the GPU side of things and further improve Frostbite 3. [Edit: I've just read your blog post which mentions GPU optimizations, so this is apparently already underway] I have a couple of questions:

1) Is there really much to be gained on the GPU side?
2) Will Battlefield 4 (and other, upcoming FB3-based games) be retrofitted with future improvements to the engine, in particular Mantle-based improvements?
 
Last edited by a moderator:
What runs, nevermind predictable? Do you really think it is a coincidence that 5 out of 6 screens show exactly or very approximate (less than 2 fps difference) FPS to the ones used as "average"? And the fact that the 6th screen, which was the only one that showed very different FPS from average (116 vs 121.5), was taken out after someone pointed it was from Single Player?

Lots of coincidences here...




well you know if you stop being a whinny little bitch and ask nicely(twitter etc?) repi might clarify or confirm for you, seeing as he is the one who wrote the post ..........


i do see what so hard understand about the numbers, why must the picture match up exactly, you have like 3 seconds worth of data on the screen.
 
I'm ignoring "data points" that don't affect my position of what I "expected"? Suggest you read my post again, as your response makes no sense in context. I am not making any declarative statements on Mantle performance as an API now or in the future.

the multi gpu config with a 6core i7 showed significant performance improvements according to Repi. So its not just low end amd cpu's like you claimed.
 
the multi gpu config with a 6core i7 showed significant performance improvements according to Repi. So its not just low end amd cpu's like you claimed.
Again, we have a reading comprehension problem here. How could benchmarks released today affect my expectations of what advantages the API would bring for the majority?
 
Again, we have a reading comprehension problem here. How could benchmarks released today affect my expectations of what advantages the API would bring for the majority?

you never said that, you said

My expectation for Mantle was basically this. It's something to significantly help AMD's own CPU's first and foremost, was never expecting big gains with high-end GPU's and fast CPU's.

stop moving goal posts. this statement is wrong.

WTF is wrong with you? NeoGaf forums are thataway kiddo.

never been on NeoGaf, and im not a kiddo old timer...... i just hate pretensions BS, be honest about your agenda dont try to dress it up. if it looks like a duck, walks like a duck and quacks like a duck. then its a duck........

my agenda, i want to see mantle succeed, and ultimately what people are talking about in the grand scheme is so trivial and inconsequential its not funny.

question would people rather a rapid software development lifecycle approach for this or still be waiting?i'll take rapid everyday of the week.
 
i do see what so hard understand about the numbers, why must the picture match up exactly, you have like 3 seconds worth of data on the screen.

I did not get exactly you are trying to say here. But the point is right that. How much time did they bench? Did they even walk around? Because the screenshots look to be in exactly the same position in the map, just with slightly different content like smoke or outdoor advertisement. Its like they just drop the player in the exact same coordinates.

I do not use tweeter or whatever, and he already wrote on this forum, so why not answer here as well?
 
The lower numbers were taken with a Core i7 4960X, which is a $1000 CPU.

The purpose of Mantle is to have gamers taking full advantage of their GPUs without requiring a $1000 CPU.
What will be interesting is to know the difference in performance between a $1000 Intel CPU and a $180 AMD CPU.
What is not interesting is to show how a very expensive CPU won't benefit at all from Mantle, because that was a given from the start.


Whoever decided to leak those numbers that way clearly harmed AMD's whole point with Mantle.
 
Back
Top