• 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.

An AMD Mantle users Testing BF4.

Page 5 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
What happens when a new version of direct x is released?

Only specific games support it, and all your old hardware is now useless. Direct X and mantle are the same. The problem here is that mantle hasn't had time to get off the ground yet. It's still in beta.

Generally that's not how things have worked on a given time period though. There is a rollover, and DirectX has been fairly decent about backwards compatibility until it gets to the point where games require a feature set beyond what an old DX uses.

They also have been pretty stable over time. We had garbage DX era (largely ignored), then DirectX 7, which was the first fairly well received DX. From there we had 8.x, which didn't last all that long. Then DX 9 family, which lasted a loooooooong time. In fact, tons of games still work fine with DX9 even to this day. Then DX10, which was a waste of time (similar to 8, but I don't think hardly any games were DX10 only), and finally DX11 which is what, 4 years on now?

I hate saying all that because I hate Microsoft overall, and wish that OpenGL was the standard for many reasons.

Adding Mantle to the mix is just an entirely new mess, but I could be wrong and it could turn out awesome. It's all going to depend on if it reminds us of the Glide days (bad), or of the more stable days. And I'm going to lose my head if Nvidia pulls the same crap. Imagine if you're a game developer, and you have to code for :

DirectX (most of the potential market)
Mantle (for GCN cards, what .. ~5% of steam users?)
Nvidia Api (probably limited to Kepler and newer? wouldn't doubt it)
OpenGL possibly (for PS4 version)

And oh well, let's throw in an Intel version too, potentially bringing the total 'extra' APIs needed to make a game in 2015-2016 to FIVE. Glorious.
 
It's even worse than that because you will also need the newest copy of windows while mantle happily works with Windows 7 or 8.

That's why I liked OpenGL, there's no reason even Windows 2000 or Ubuntu couldn't run an OpenGL title with much less effort than just having a DirectX version that doesn't port anywhere (other than perhaps XB1)
 
Really, when dx12 launches, mantle is going to have more marketshare than dx12. Both games and GPU's.

I honestly have serious buyers remorse right now with this gtx780. It's an overclocking dud, and I still have terrible minimums in the 70's in the only game I really play. May return it and try for a r9 290.
 
Really, when dx12 launches, mantle is going to have more marketshare than dx12. Both games and GPU's.

I honestly have serious buyers remorse right now with this gtx780. It's an overclocking dud, and I still have terrible minimums in the 70's in the only game I really play. May return it and try for a r9 290.

Mantle > DX12 for me. I plan on moving to SteamOS for my gaming needs. Will play any Windows-only games I just must have on Windows 7. Looking forward I'm more interested in OGL/Mantle than DX. Windows8 really ruined it for me and it's time to break the chains.. glad to have more than just OGL on SteamOS going forward.
 
Some large performance increases being reported in this thread, much higher than what the reviews stated..

This one in particular made me 😵



I would definitely need evidence to believe that. It's not like he was using a slow CPU either..
Maybe you should get a GCN card and see what you are missing? 😉

This isn't BF4 but it demonstrates the point where mantle shines, when you are CPU bound. He was on low settings which are CPU bound and there are massive gains to be had.

61930.png
 
Maybe you should get a GCN card and see what you are missing? 😉

This isn't BF4 but it demonstrates the point where mantle shines, when you are CPU bound. He was on low settings which are CPU bound and there are massive gains to be had.

61930.png

Nice. This is one of the few GPU innovations in a long time that actually matters. We all watched Physx, 3D Vision and a bunch of stuff float in and out..

I'm still in puppy love with my 5870 since 9/23/2009 (still going hard), but will definitely pickup a GCN2.0 APU or card when the time to retire this old workhorse comes.
First card that in my mind has trumped my original Diamond Monster 3D as 'best videocard of all time'.

Mantle means more to me than to others because I'm moving from Windows to SteamOS once it's ready, for my gaming needs. In that realm, it's going to be Mantle or OpenGL.

I'll take both rather than just 1. I'll be interested to see how SteamOS, Mantle, and how much difference those technologies make when we get a 2x4K Oculus Rift.. everything is going to count there and this is a BIG leap.
 
I honestly never thought I would see someone say "Get a $700 card and put the settings on low @ 1080p, and see what you are missing"
 
I honestly never thought I would see someone say "Get a $700 card and put the settings on low @ 1080p, and see what you are missing"

Sure, but what the low CPU dependency translates to for the rigs that are GPU limited is lower frame times. 20-45% less compared to DX in BF4 from what I've seen. That gives you higher minimum FPS and more consistent FPS. That means less noticeable hiccups or stutters. It delivers a better gaming experience. Ever since the frame time debate was brought we cannot solely look at average FPS. This is outside the fact that no current Mantle developer has leveraged the CPU being freed up for anything. It's purely just a performance advantage.
 
Except the only example of a game that Mantle helps in limited situations (64MP?) actually causes stutter, according to the BF4 forums and a thread on the front page of this sub-forum currently.

I have played with both, and have yet to see a "frame time" advantage mean anything to me personally.

I spend $$$ on decent hardware so that I have a visual advantage.
 
Except the only example of a game that Mantle helps in limited situations (64MP?) actually causes stutter, according to the BF4 forums and a thread on the front page of this sub-forum currently.

I have played with both, and have yet to see a "frame time" advantage mean anything to me personally.

I spend $$$ on decent hardware so that I have a visual advantage.

I dont feel any stutter on my kaveri apu...mantle seems to be working fine, IMO that is.
 
I know what op means with all settings maxed in bf4 with a 290x and mantel in mp modes with all sorts of weather effects going storms action ect i never see the game dip below my 120fps cap once i think i saw it hit 115 briefly.

same thing in dx11 and it was regularly dropping down to even 50fps at times which gets quite noticeable and in fact i wasnt even running the game at quite as high settings.

its not the maximum fps #s that seem to matter its the minimums and averages mantel just seems to keep things up higher during the hectic scenarios.

and ive never experianced stutter?? maybe because i capped the game at 119-120fps??
 
Last edited:
Except the only example of a game that Mantle helps in limited situations (64MP?) actually causes stutter, according to the BF4 forums and a thread on the front page of this sub-forum currently.

I have played with both, and have yet to see a "frame time" advantage mean anything to me personally.

I spend $$$ on decent hardware so that I have a visual advantage.

Works fine for me. Runs smoother than my friends GTX680.
 
:\
I dont feel any stutter on my kaveri apu...mantle seems to be working fine, IMO that is.
I think some strive for the best performance at the highest possible settings, but most are just looking for the best performance. Some of the former don't even play the game, or don't spend the amount of time as many serious gamers. I am willing to sacrifice some eye candy to achieve smooth game play.
 
Except the only example of a game that Mantle helps in limited situations (64MP?) actually causes stutter, according to the BF4 forums and a thread on the front page of this sub-forum currently.

I have played with both, and have yet to see a "frame time" advantage mean anything to me personally.

I spend $$$ on decent hardware so that I have a visual advantage.

As you see in the posts here many are finding great performance improvements and no stutter.

As for Frame Times, not everyone would notice the benefits. i'm one who does, certainly from a technical aspect lower frame time reduce input lag which is obviously very beneficial in first person shooter games.
 
They already had a Cryengine game who is actually going to implement Mantle. My guess is that once the initial time and $$ has been put into making Mantle work with one game, it is much easier to implement in other games within that engine, should the developer choose to.
 
They already had a Cryengine game who is actually going to implement Mantle. My guess is that once the initial time and $$ has been put into making Mantle work with one game, it is much easier to implement in other games within that engine, should the developer choose to.
According to a hothardware.com writer at the time of APU13:

"I was told that Crytek was extremely careful when it came to adopting new standards because those technologies had to fit within the company's workflow without making it too difficult to support multiple vendors or relying on flashy but limited tricks of creating one specific effect or capability. Crytek, in other words, was only interested in supporting standards that were flexible, scalable, and delivered a high enough performance improvement to be interesting. I was left with the impression that the company was still in "Wait and see" mode last November.
Evidently they've seen enough."
 
They already had a Cryengine game who is actually going to implement Mantle. My guess is that once the initial time and $$ has been put into making Mantle work with one game, it is much easier to implement in other games within that engine, should the developer choose to.

That's the case with the Frostbite 3 engine.
 
Back
Top