• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

The AMD Mantle Thread

Page 72 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Status
Not open for further replies.
The closer it got the the presentation, the more I thought "I bet they dont release any numbers at all". They sure are milking it to the max without giving any hard proof of what the improvements will be. Pretty soon they are going to have to "show us the beef" as the old Wendy's commercials used to say, instead of just making rosy promises without any data backing them up.
 
https://twitter.com/ryanshrout
ryan is checking out a mantle base rendering engine!

Ryan Shrout - "Oxide claims that it only took 2 "man months" to support Mantle. "Not much work or expensive" to integrate it."

"API overhead reduced by at least 10x. In some cases StarSwarm is 3x faster with Mantle."

"Completely changes the landscape for multi-core. Claims an FX-8350 is now just as fast as a Core i7-4770K"

James Prior - "With Mantle, an @AMDFX 8350 underclocked to 2GHZ, the game is still GPU bound on a R9 290X - no loss in frame rate"

I would call that some staggering info.
 
Last edited:
Also one of things that I want to know more about was when they talked about Mantle being able to run crossfire without needing to share memory between gpus.

They claimed to have crossfried 8 cards in a workstation using this method for use in rendering, pushing 40 tflops.

Can anyone elaborate on this?

Will this be implemented in BF4 or is this just vaporware?

I want to use my Quadfire in BF4 without it being terribad.
 
Can anyone elaborate on this?

Will this be implemented in BF4 or is this just vaporware?

I want to use my Quadfire in BF4 without it being terribad.

They did not seem to be talking about it in relation to BF4 but more of something that was theoretically possible with Mantle.
 
They did not seem to be talking about it in relation to BF4 but more of something that was theoretically possible with Mantle.

Sadface.

Is there any higher quality video that will be made available later? The VOD AMD is using is pretty bad. I'm guessing they can't use Twitch.tv because they have a deal with Nvidia or something.
 
Sadface.

Is there any higher quality video that will be made available later? The VOD AMD is using is pretty bad. I'm guessing they can't use Twitch.tv because they have a deal with Nvidia or something.

Well, they said we will get a free demo later this quarter, whenever that is. That space demo based on what Ryan says sounds cool.
 
Ryan Shrout - "Oxide claims that it only took 2 "man months" to support Mantle. "Not much work or expensive" to integrate it."

"API overhead reduced by at least 10x. In some cases StarSwarm is 3x faster with Mantle."

"Completely changes the landscape for multi-core. Claims an FX-8350 is now just as fast as a Core i7-4770K"

James Prior - "With Mantle, an @AMDFX 8350 underclocked to 2GHZ, the game is still GPU bound on a R9 290X - no loss in frame rate"

I would call that some staggering info.

this does bode well for AMD and mantle!
 
this does bode well for AMD and mantle!

Should be wonderful for people trying to get 120-144 fps constant in BF4 with lowered settings.

Also should recover lost ground to Nvidia in BF4 since AMD's cards are slower than Nvidia's in CPU bottlenecked scenarios, as well as lower the amount of frame stuttering in single GPU caused by CPU bottlenecks.
 
Last edited:
Ryan Shrout - "Oxide claims that it only took 2 "man months" to support Mantle. "Not much work or expensive" to integrate it."

"API overhead reduced by at least 10x. In some cases StarSwarm is 3x faster with Mantle."

"Completely changes the landscape for multi-core. Claims an FX-8350 is now just as fast as a Core i7-4770K"

James Prior - "With Mantle, an @AMDFX 8350 underclocked to 2GHZ, the game is still GPU bound on a R9 290X - no loss in frame rate"

I would call that some staggering info.

Staggering, indeed. Looks like we might actually see gains well above 20% on average. If so, then I completely understand how Mantle would "ridicule" Titan.
 
Johan Andersson's keynote alludes that BF4's mantle implementation will be 4 threaded rendering.

This is apposed to what he shows as 2-3 threaded rendering from DX (probably alluding to driver command lists)
 
Guys? Bueller?

Come on. Let's talk about the only thing that's not vaporware about mantle, Johan's Keynote about the implementation in BF4.

It's the most substantial thing in this entire thread. Any technical geniuses that can give us any deep insights about Johan's Keynote to laymen?

He goes pretty detailed.

Alot of the specifics go over my head, and I can only get a general idea of what he is talking about.
 
Last edited:
Towards the end of Johan's keynote he mentioned that Dice has to implement it's own multi-gpu framework when using Mantle. This bodes well for people stuck on Tahiti or Pitcairn if they actually implement Mantle Multi-GPU and that it's not vaporware.

----------------------------------------------------------------------------------------------------------

I can't believe no-one posted about this yet.

 
Last edited:
As a game studio, you're a money making machine. You cannot/will not ignore nvidia's x86 market share, nor you cannot direct dev resources from dx to mantle(this will translate into worse dx perfomance, more bugs, etc..). This requires additional dev resources, which means additional dev cost. No company follows this logic. It's the other way around: decrease dev cost by any means possible, increase profits. DX's and opengl's strength is that they are portable between the two of the most dominant gpu manufacturers. This is how and why they were born in the first place. I can't see how people keep saying that mantle reduces dev cost.

I think people are taking things out of context. It costs less to do a Mantle port than it does a DX port and the quality will be far better. Devs have said that the additional costs are minimal, so they'll get enough of a return on investment for porting to Mantle to make the cost of doing so profitable.

Add to this that game devs have been wanting an API like Mantle for a long time. They feel that DX limits what they can do, creatively. Keep in mind that games are an art form like making movies. Everyone wants to make the next classic, the next blockbuster, and they are willing to invest the time and money to accomplish this. AAA games are more than a mass produced commodity. How much fame and fortune do you think the original Crysis has made for Crytek and the individuals who run it?

Contrary to what many believe the top developers aren't simply trying to make the cheapest piece of crap they can and laugh all the way to the bank.
 
IMG0043514_1.jpg


IMG0043515_1.jpg


IMG0043516_1.jpg


IMG0043521_1.jpg


I think that seals the deal on whether or not Mantle is worth it...
 
Last edited:
this sounds really amazing, can't wait to see it in action. It's a game changer in many ways, really interested to see laptops with it. While I don't care about what card sits in my desktop (AMD or Nvidia), I would still lean towards nvidia for laptops, but if mantle delivers - then AMD may end up being significantly preferred in laptops for gamers.
 
Didn't know if this warranted a new thread, but AMD has confirmed the PS4 supports TrueAudio
http://www.anandtech.com/show/7513/ps4-spec-update-audio-dsp-is-based-on-amds-trueaudio

I think it does TrueAudio is not Mantle.

So no Mantle performance figures untill late december? Wouldn't be a surprise if aftermarket 290(x) won't be released until late december as well.

What does one have to do with the other?


After gleaming what was presented, Mantle isn't going to be minimal gains. Very interesting.
 

It is said in this article that complex effects were limited
in usual APIs due to staggering induced costs , a limitation
removed by Mantle.

According to the slide for a given scene they went
to 33ms/frametime from 99ms in usual APIs.

All in all Mantle seems firstly designed to increase
image quality more than fps but i may be wrong,
there s quite a lot of infos to digest.
 
Status
Not open for further replies.
Back
Top