from what it looks like, Is Mantle a game changer to you?

is mantle a game changer for you?


  • Total voters
    202
Yeah but use of all of those things is quite ordinary in the PC world.
No, I wouldn't say their usage is "quite ordinary". In some domains, yes. In others, no.

Besides, usage is orthogonal to low-levelness/high-levelness.
 
I think Nvidia made a strategic mistake to give AMD 100% of the cur/next generation home entertainment market. That allows AMD to come up with a very disruptive product such as Mantle. Nvidia didn't think outside of the box and had their mind too stuck with the conventional x86 + Windows + DirectX PC gaming platform. Once AMD comes up with a gaming/home solution with ARM + Linux + Mantle alternative to achieve the equivalent performance with much lower cost, that could very well be close to the end for Wintel home PCs. Nvidia can catch up with their ARM solutions. However, it will loose its competitive advantage on drivers and designs if AMD can successfully execute its plan to make Mantle the industrial standard.
 
I voted no. I can 't pretend to speak for everyone who did, but since I have a decent processor (2500k), the results are uninspiring. Sure, a 10%-15% perf boost would be nice, but it won't singlehanded py make the decision for me. I'm never going to have a low-end proc for other reasons, so I'll never see the Hugh benefits that some people are touting.

This is all on a personal level, of course. I have have no idea how the industry will respond.
 
If Mantle was nVidia, there would be screams from the rooftops of 'proprietary, PhysX all over again, evil NVidia etc etc'

Because it's AMD, it's OK of course.

NVIDIA said PhysX was going to be open and that anyone could use it, look at it now.

AMD on the other hand has a good track record, nearly every CPU sold uses AMD64 tech.

So yeah, when AMD say that it is going to be open and free people are more likely to trust them over NVIDIA.
 
For me to get excited about Mantle, it will need to increase performance in a game that I actually play.

That was the thing that killed PhysX for me. At the very beginning I was super excited about the tech and even had an Ageia PPU, but as the years went by and almost no games came out that benefited from it, it became basically a useless tech.

Granted BF4 is a popular game and the fact that it has support must have a lot of people excited about it. Just having BF4 support makes it seem like it's on a better track to being commonly adopted than PhysX ever was. Only thing is, I don't play BF4 nor do I ever plan to.
 
Good point gotnorice. The game coverage is one of the keys things to make mantle successful. Mantle made console porting easier. Does that mean all console ports will use mantle? If mantle is open and popular, why not? If mantle is limited, then no and why throw extra money on unnecessary development. However, if game coverage of mantle is limited, then mantle won't be popular. Therefore, this becomes a chicken and egg problem. In such scenario, amd needs to persistently market and lobby many games, especially popular games, to adopt mantle. This continuous push would serve as a positive catalyst to kick the adoption of mantle in a positive cycle. I believe amd can achieve that with solid executions. From technology standpoint, mantle is a totally different beast from physx, as physx is more about extra eye candies, while mantle improves the overall experiences of games and making hardware more productive and efficient.
 
I liked the gains in BF4, but the only way this will be a game changer is if a majority of games use it. I buy mostly EA games, Dragon Age, Mass Effect, Battlefield, so if EA has all their studios use Mantle from now on then that's great but, I would still want to see other devs make use of it as well.
 
Definitely more interested after the BF4 results were posted. Still not sure how prolific Mantle will become.
 
I'm excited by it because prior to Mantle, upgrading to an R9-290X would be a bit pointless as I'm stuck with an overclocked i7-930 till I decide to revamp my system with a new motherboard, processor and depending on how long in the future maybe I'll require new ram as well. Thus upgrading to an R9-290X would be (500 for the gpu + another ~600 for ram/cpu/motherboard). Now though, it seems like I could pair an R9-290X with a i7-930 using Mantle and not have the i7-930 be the bottleneck.

Thus, I can potentially upgrade my video card and get fairly up to date performance in games (in terms of fps) at a little under half the cost. Is that a game changer to me? Yes. So I voted, Yes. That being said, I can't find a R9-290X anywhere near MSRP.... so I'm thinking I'll be stuck waiting for the 'R9-380' if I don't get lucky with a nowinstock website notification. I really would like to upgrade now though, I feel like it's been far too long and the latest games do not run as well as I'd like.
 
I'm excited by it because prior to Mantle, upgrading to an R9-290X would be a bit pointless as I'm stuck with an overclocked i7-930 till I decide to revamp my system with a new motherboard, processor and depending on how long in the future maybe I'll require new ram as well. Thus upgrading to an R9-290X would be (500 for the gpu + another ~600 for ram/cpu/motherboard). Now though, it seems like I could pair an R9-290X with a i7-930 using Mantle and not have the i7-930 be the bottleneck.

Thus, I can potentially upgrade my video card and get fairly up to date performance in games (in terms of fps) at a little under half the cost. Is that a game changer to me? Yes. So I voted, Yes. That being said, I can't find a R9-290X anywhere near MSRP.... so I'm thinking I'll be stuck waiting for the 'R9-380' if I don't get lucky with a nowinstock website notification. I really would like to upgrade now though, I feel like it's been far too long and the latest games do not run as well as I'd like.

...Must admit, past this few day since BF4 patch , reading this forums has been a blast
So much glorious butthurtness:D
...need[ed] a free performance increase and a kick in the stagnant DX arse...

I completely agree on all accounts. DX is a fairly tired horse and hasn't been getting as much attention as it deserved from Microsoft in favor of terrible development choices,time and money-spent elsewhere in Windows platform.
 
Good point gotnorice. The game coverage is one of the keys things to make mantle successful. Mantle made console porting easier. Does that mean all console ports will use mantle? If mantle is open and popular, why not? If mantle is limited, then no and why throw extra money on unnecessary development. However, if game coverage of mantle is limited, then mantle won't be popular. Therefore, this becomes a chicken and egg problem. In such scenario, amd needs to persistently market and lobby many games, especially popular games, to adopt mantle. This continuous push would serve as a positive catalyst to kick the adoption of mantle in a positive cycle. I believe amd can achieve that with solid executions. From technology standpoint, mantle is a totally different beast from physx, as physx is more about extra eye candies, while mantle improves the overall experiences of games and making hardware more productive and efficient.


If they are going to be porting anyways, why NOT use Mantle. It lowers development cost. In other scenarios, they can (in some games) decide if porting for a lower cost will give a net gain to the bottom line. Add to that a better experience on the PC from a ported game that otherwise may not have been ported and we have a win-win situation.
 
Mantle is not for me. For mid range it seems like a great solution but the FPS comparisons I have seen thus far in BF4 are not impressive at all. Since I will remain at one notch lower than fastest cards out Mantle may not entice me to go back to AMD given that I am quite comfortable with nVidia cards for past 1.75 years and really understand now what I was missing when I had 4870-7970 AMD exclusive run of cards for almost 4 years.
 
I'm excited by it because prior to Mantle, upgrading to an R9-290X would be a bit pointless as I'm stuck with an overclocked i7-930 till I decide to revamp my system with a new motherboard, processor and depending on how long in the future maybe I'll require new ram as well. Thus upgrading to an R9-290X would be (500 for the gpu + another ~600 for ram/cpu/motherboard). Now though, it seems like I could pair an R9-290X with a i7-930 using Mantle and not have the i7-930 be the bottleneck.

Thus, I can potentially upgrade my video card and get fairly up to date performance in games (in terms of fps) at a little under half the cost. Is that a game changer to me? Yes. So I voted, Yes. That being said, I can't find a R9-290X anywhere near MSRP.... so I'm thinking I'll be stuck waiting for the 'R9-380' if I don't get lucky with a nowinstock website notification. I really would like to upgrade now though, I feel like it's been far too long and the latest games do not run as well as I'd like.

http://www.amazon.com/gp/offer-listing/B00HSY1TBK/ref=dp_olp_new_mbc?ie=UTF8&condition=new

Buy that from Amazon and when it comes it they will charge your CC and ship immediately. Make sure you choose Amazon as the supplier.
 
As an Amazon Associate, HardForum may earn from qualifying purchases.
DO THAT!!!

I bought 2 of them a couple of weeks ago, 2 more weeks to go.(delivery between 2-19 to 2-25)
 
If they are going to be porting anyways, why NOT use Mantle. It lowers development cost. In other scenarios, they can (in some games) decide if porting for a lower cost will give a net gain to the bottom line. Add to that a better experience on the PC from a ported game that otherwise may not have been ported and we have a win-win situation.

Simple, because at this moment, DirectX still has 99.99% personal PC market, and AMD GCN gpu has < 25% video card market share. Plus nobody can even accurately predict a future beyond 3 months.
 
Last edited:
Mantle made console porting easier.
For Mantle-exclusive titles, potentially. There are no Mantle-exclusive titles, however.

Adding an entire renderer to existing multi-million LOC projects does not make cross-platform development easier.
 
For Mantle-exclusive titles, potentially. There are no Mantle-exclusive titles, however.

Adding an entire renderer to existing multi-million LOC projects does not make cross-platform development easier.

It is not a trivial job for sure.
BTW, are you implying that it is easier for console games to be ported using DirectX rather than using Mantle?
 
No, I'm only suggesting that it's more complex to port a game with two renderers than it is to port with one.
 
Yanmeng, I don't understand this correlation you're making between next gen consoles and Mantle. You are just wrong here. Both next gen consoles are using their own APIs (they cannot use Mantle), with the XB1 using a forked version of DirectX. The only things carried over in a port with either Mantle or DX are assets such as artwork, textures, and stuff along those lines - Mantle is not any less work than DX in terms of porting. I have no idea why you would try to suggest that Mantle makes porting easier, because it really doesn't. If DX takes two months for a port, Mantle takes another two months. It is not easier or harder to do any type of port from a console since the consoles are doing their own thing in terms of an API. If anything, since MS is using a modified version of DX for the Xbox 1, DX could potentially be easier in terms of porting for that very reason.

The fact of the matter is, porting a game to Mantle is additional work / manhours. The APIs used on the next gen consoles are not Mantle, nor are they even remotely close to Mantle. Same for directX. As mentioned, the only thing that can be carried over is artwork, textures, and assets along those lines. Mantle does not ease that transition whatsoever, period.
 
Last edited:
i wish i had waited a little longer before going nvidia esp since i ONLY play bf4
 
Yanmeng, I don't understand this correlation you're making between next gen consoles and Mantle. You are just wrong here. Both next gen consoles are using their own APIs (they cannot use Mantle), with the XB1 using a forked version of DirectX. The only things carried over in a port with either Mantle or DX are assets such as artwork, textures, and stuff along those lines - Mantle is not any less work than DX in terms of porting. I have no idea why you would try to suggest that Mantle makes porting easier, because it really doesn't. If DX takes two months for a port, Mantle takes another two months. It is not easier or harder to do any type of port from a console since the consoles are doing their own thing in terms of an API. If anything, since MS is using a modified version of DX for the Xbox 1, DX could potentially be easier in terms of porting for that very reason.

The fact of the matter is, porting a game to Mantle is additional work / manhours. The APIs used on the next gen consoles are not Mantle, nor are they even remotely close to Mantle. Same for directX. As mentioned, the only thing that can be carried over is artwork, textures, and assets along those lines. Mantle does not ease that transition whatsoever, period.

How do you know mantle isn't any less work than porting a game to DX? Your making pretty big assumptions too based on nothing.

Can I see the proof of everything you said? Specifically these comments:-

Mantle is not any less work than DX in terms of porting..

The APIs used on the next gen consoles are not Mantle, nor are they even remotely close to Mantle.

Can you provide some links to this? Sorry I should add that, yes I know the APIs in the consoles are not called mantle, but how do you know that they are not even close to mantle?

. Mantle does not ease that transition whatsoever, period.

And this as well.

The only sentences I agree with are that both consoles are using their own API and porting a game to mantle is going to need additional man hours. Well of course it is, no matter how easy/hard it turns out to be, it's still going to require some work and effort.

Since AMD are involved in Mantle and the development of both consoles, wouldn't it be a more logical assumption that this will make games easier to port to mantle? I don't know and you don't know.
 
i wish i had waited a little longer before going nvidia esp since i ONLY play bf4
If you overclock your GTX 780 to about 1100 on the core, you will get Mantle equivalent or slightly slower than Mantle performance. Why the long face?
 
Yanmeng, I don't understand this correlation you're making between next gen consoles and Mantle. You are just wrong here. Both next gen consoles are using their own APIs (they cannot use Mantle), with the XB1 using a forked version of DirectX. The only things carried over in a port with either Mantle or DX are assets such as artwork, textures, and stuff along those lines - Mantle is not any less work than DX in terms of porting. I have no idea why you would try to suggest that Mantle makes porting easier, because it really doesn't. If DX takes two months for a port, Mantle takes another two months. It is not easier or harder to do any type of port from a console since the consoles are doing their own thing in terms of an API. If anything, since MS is using a modified version of DX for the Xbox 1, DX could potentially be easier in terms of porting for that very reason.

The fact of the matter is, porting a game to Mantle is additional work / manhours. The APIs used on the next gen consoles are not Mantle, nor are they even remotely close to Mantle. Same for directX. As mentioned, the only thing that can be carried over is artwork, textures, and assets along those lines. Mantle does not ease that transition whatsoever, period.

There are few articles online describing the similarities between console APIs and mantle APIs. I don't know whether these sources are correct. If you have more detailed information, please shed some light.

"Feedback we&#8217;ve gotten from other sources continues to suggest that Microsoft&#8217;s low-level API for the Xbox One is extremely similar to Mantle, and the difference between the two is basically semantic. This doesn&#8217;t square very well with Microsoft&#8217;s own "

http://www.extremetech.com/gaming/1...mantle-and-ps4-is-also-unlikely-is-mantle-doa

"What Mantle creates for the PC is a development environment that's _similar_ to the consoles, which already offer low-level APIs, close-to-metal programming, easier development and more (vs. the complicated PC environment). By creating a more console-like developer environment, Mantle: improves time to market; reduces development costs; and allows for considerably more efficient rendering, improving performance for gamers. The console connection is made because next-gen uses Radeon, so much of the programming they're doing for the consoles are already well-suited to a modern Radeon architecture on the desktop; that continuum is what allows Mantle to exist."

http://www.techpowerup.com/192552/amd-explains-why-mantle-doesnt-work-on-xbox-one.html
 
As an avid BF4 player, I have to say that Mantle makes a pretty great difference in gameplay when it's working perfectly.

I would say maybe about 2-4 times per game, there like 5 second instances where the frames will drop very suddenly, and then pick back up. This is a widely reported issue and seeing as it's in beta, I'm willing to live with it until they can fix it.

I'm willing to live with it because the game overall just plays so much more smoothly with Mantle. While my frames aren't significantly improved, what is improved is just the entire feeling. It feels much much less choppy and more fluid. It's difficult to quantify, but it's immediately apparent. Apparent enough that I have no interest in turning DX back on. I was pretty surprised as, coming from AMD, I didnt expect to really see significant results.

All in all, I'm happy with the direction it's going and I'm very happy with how BF4 is performing with my 290x.
 
As an avid BF4 player, I have to say that Mantle makes a pretty great difference in gameplay when it's working perfectly.

While I'm not a huge BF4 player, I did try the Mantle version and DAMN it feels NICE! I don't know if it's a combination of substantially higher minimum FPS and reduced lag probably because the CPU it cut out of the loop, but I was REALLY surprised at the difference.

Smooooooth & Responsive !

One of those things which is a bit hard to quantify, you just have to try it... then you know.
 
AMD on the other hand has a good track record, nearly every CPU sold uses AMD64 tech.

So yeah, when AMD say that it is going to be open and free people are more likely to trust them over NVIDIA.

What???? You know why AMD64 tech is in nearly every CPU don't you?

AMD used MS to force Intel to obtain a license and use AMD64. MS refused to support more than one 64bit instruction set. Since X86-64 was built on x86 it was easier for MS to adopt. Remember Intel Itanium which used the IA-64 tech? It was out before AMD64. Intel hedged their bets on the radically new tech only to have MS force them to adopt AMD64 or lose the windows market.

Intel could either scramble and use AMD64 or do without Windows. That market was and still is a huge percentage of Intel's income.

So, AMD wasn't just a nice guy who took X86 and created X86-64 then shared with everyone for nothing. Intel and AMD now have mutual assured destruction licenses that neither can pull. AMD smartly, helped to secured their future by doing this.

AMD is no different than Nvidia at this point. Both historically have only cared for their own futures. If money can be made from a tech they will do it.

Back to the topic.

It is way too soon to tell whether Mantle is a game changer or just a cool new tech. More games need to be released/benched and more devs to adopt Mantle before it can be a game changer. Nvidia being allowed to use Mantle would greatly help but who knows if/when that will occur.

I would bet Mantle stays AMD only for quite some time.
 
Last edited:
What???? You know why AMD64 tech is in nearly every CPU don't you?

AMD used MS to force Intel to obtain a license and use AMD64. MS refused to support more than one 64bit instruction set. Since X86-64 was built on x86 it was easier for MS to adopt. Remember Intel Itanium which used the IA-64 tech? It was out before AMD64. Intel hedged their bets on the radically new tech only to have MS force them to adopt AMD64 or lose the windows market.

Intel could either scramble and use AMD64 or do without Windows. That market was and still is a huge percentage of Intel's income.

So, AMD wasn't just a nice guy who took X86 and created X86-64 then shared with everyone for nothing. Intel and AMD now have mutual assured destruction licenses that neither can pull. AMD smartly, helped to secured their future by doing this.

AMD is no different than Nvidia at this point. Both historically have only cared for their own futures. If money can be made from a tech they will do it.

Back to the topic.

It is way too soon to tell whether Mantle is a game changer or just a cool new tech. More games need to be released/benched and more devs to adopt Mantle before it can be a game changer. Nvidia being allowed to use Mantle would greatly help but who knows if/when that will occur.

I would bet Mantle stays AMD only for quite some time.


Sooo....
Lets say AMD did not do what business is suppose to do and protect the interest of their investment AND their investors and were no longer in business, only Nvidia. How much do you think you would be paying for a video card? How much would you be paying for a CPU? They just saved your butt in reality. Also monopolies are frownedupon by the US commerce department. That is why Bill Gates/Microsoft gave Steve Jobs/APPLE a HUGE loan to keep them in business. These practices you speak of are agreed to for other reasons than the spin you hear sometimes.
 
What???? You know why AMD64 tech is in nearly every CPU don't you?

AMD used MS to force Intel to obtain a license and use AMD64. MS refused to support more than one 64bit instruction set. Since X86-64 was built on x86 it was easier for MS to adopt. Remember Intel Itanium which used the IA-64 tech? It was out before AMD64. Intel hedged their bets on the radically new tech only to have MS force them to adopt AMD64 or lose the windows market.

Intel could either scramble and use AMD64 or do without Windows. That market was and still is a huge percentage of Intel's income.

So, AMD wasn't just a nice guy who took X86 and created X86-64 then shared with everyone for nothing. Intel and AMD now have mutual assured destruction licenses that neither can pull. AMD smartly, helped to secured their future by doing this.

AMD is no different than Nvidia at this point. Both historically have only cared for their own futures. If money can be made from a tech they will do it.

Back to the topic.

It is way too soon to tell whether Mantle is a game changer or just a cool new tech. More games need to be released/benched and more devs to adopt Mantle before it can be a game changer. Nvidia being allowed to use Mantle would greatly help but who knows if/when that will occur.

I would bet Mantle stays AMD only for quite some time.

Let's just gloss over the fact that Itanium ran 32 bit apps like crap and everything up until Itanium was introduced was *gasp!* 32 bit. MS was right to choose the x86-64 extension.
 
Last edited:
Let's just gloss over the fact that Itanium ran 32 bit apps like crap and everything up until Itanium was introduced was *gasp!* 32 bit. MS was right to choose the x86-64 extension.

And? I never said Intel's tech was better. In fact I said Intel hedged their bets on a radically new technology as opposed to one based on the X86 which was easier for MS to adopt. I never faulted MS or AMD.

However, the person I replied to seems to think AMD64 was some friendly tech AMD created for all to enjoy. It wasn't. AMD was competing with Intel on creating the first 64 bit chip and used MS to leverage their tech. Thus, as I said, creating a mutual destruction pact with Intel.

Sooo....
Lets say AMD did not do what business is suppose to do and protect the interest of their investment AND their investors and were no longer in business, only Nvidia. How much do you think you would be paying for a video card? How much would you be paying for a CPU? They just saved your butt in reality. Also monopolies are frownedupon by the US commerce department. That is why Bill Gates/Microsoft gave Steve Jobs/APPLE a HUGE loan to keep them in business. These practices you speak of are agreed to for other reasons than the spin you hear sometimes.

Your comment has nothing to do with my post. Read the members post I quoted first then my reply. Has nothing to do with monopolies.

BTW, MS didn't invest in Apple to prevent a monopoly. They did it to get Office and MS dev tools onto the Mac platform, 8M+ users. It created a win-win for both sides and cost MS little money ($150M).
 
And? I never said Intel's tech was better. In fact I said Intel hedged their bets on a radically new technology as opposed to one based on the X86 which was easier for MS to adopt. I never faulted MS or AMD.

However, the person I replied to seems to think AMD64 was some friendly tech AMD created for all to enjoy. It wasn't. AMD was competing with Intel on creating the first 64 bit chip and used MS to leverage their tech. Thus, as I said, creating a mutual destruction pact with Intel.
...

You forgot to mention two IMPORTANT facts:
1. AMD get the MS support for Server 2003 on x64 AFTER publicly stated FRAND licensing for the x64 architecture; yes, even VIA is using it.
2. Intel NEVER intended to license IA64 to any competitor. And I'm happy that monstrosity is dead, with no chance of revival.
 
BTW, MS didn't invest in Apple to prevent a monopoly.
They did it to get Office and MS dev tools onto the Mac platform, 8M+ users. It created a win-win for both sides and cost MS little money ($150M).

NOPE!!!, What really goes down and what is reported can be very different. DOJ was very satisfied with what happened.
__________________
 
Its only the beginning, but I think Mantle have proven some points already. Changing API gives performance improvements across the board, especially giving us higher minimum FPS and it also have a much faster frame delivery, which can lead to smoother games.

this is key. For even in situations where the framerate doesn't improve that much, Mantle consistantly delivers MUCH more consistant frame times across the board. Which is the source the "smoothness" benefit that many are talking about. Framerate is only part of the puzzle. As it is an average over time. But a lot can happen in 1000ms/1 second, that Framerate doesn't account for. Techreport.com has been a big source of awareness on this issue. I suggest everyone go read their "Inside The Second" article and the related articles which are linked within it.

Some of the reviews dropped the ball in their test of Mantle. Mantle benefits lowend and midrange systems the most (and ultra-highend), where CPU limited scenarios like BF4 multiplayer shows the advantage of Mantle more then any other. Most gamers doesn't have a $1000+ CPU and a $700 GPU. Most gamers have a midrange to lowend system and Mantle gives an extreme boost to them (which have a GCN based GPU) by just changing the API. Still, some reviewers didn't even test this.

We have here an API that gives a great boost to "most gamers" (of the group that can use Mantle) and some reviewers and even forum users are too stupid to see that or test for that. Instead, we are stuck with reading reviews of $2000+ systems by morons trying to figure out if Mantle have a benefit for a few percentage of the ones that can use Mantle in the first place, instead of testing if Mantle in general have a benefit for most gamers. :p

Yeah...I'm not sure what happened here. I suspect that maybe AMD told everyone to only publish articles with the 290x, since Mantle doesn't really officially support anything else yet. and if they did, I guess reviewers had to keep quiet about that.

Because otherwise, what the heck man??? I didn't see one review with anything besides a 290x. Very FEW reviews that tested more than one processor (and I had to laugh at sites that used the same processor, but downclocked or disabled cores). Techreport actually used three CPUs, but said the results between their "mid" choice processor and the lower performance processor were so similar, they skipped posting three sets of data!. Imagine if every review site had something similar. That would be good to know, yeah?

and even fewer that actually bothered to post some multiplayer numbers (thank you [H], for at least doing that). I know you can't get surgically identical testing out of MP. But so what? Gamers are interested in what they can expect in real world scenarios. Not a clean room.
My own tests suggest that simply testing on the same map in a similarly populated game is enough to expose dramatic improvement with Mantle.

Yeah, like I said, the term "beta" to AMD means "shit the bed" status whereas it doesn't mean that for other companies. As I said, i've used beta drivers from Intel, nvidia, Asus, and numerous others without issues - typically, for other companies the term "beta" just means that the release was submitted for WHQL status but had not been approved. Whereas, beta clearly has other meanings at AMD.
I'm no apologist. But as you pointed out here, the term "beta" has been very devalued, as a way to push out drivers sooner than the WHQL process allows. It's a scapegoat term to cover asses, just in case. Most of the time, beta drivers from Nvidia and AMD are fine, if not great. and that's because they are otherwise pretty normal driver updates.

When's the last time a GPU company had a true reason for a real beta? We haven't had a new API launched in YEARS. So yeah, I can give AMD a pass when their ACTUAL beta for a driver, supporting a brand new API, isn't a replacement for an everyday use driver. The fact that AMD was so late on delivery, really emphasizes that situation here.

Also so much for the massive gains console kids were talking about, now we got an example of what a low level API can bring you and its like lol 30% max.

All doom and gloom eh? Do you even know what you are saying? a 30% boost is like spending $100 extra dollars on a GPU.
There's also tests out there showing low potency CPU architectures like the A10 or old Core2quads getting 100% extra performance. Essentially making them perform about as well as i5/i7 do in D3D. That is absolutely awesome.
Did you know that only a small percentage of steam users actually have hardware that makes their PC significantly better than a PS4.
Only 2% have displays capable of higher than 1080p.
and Only about 25%f steam users have a PC roughly equivalent to a PS4.
valve publishes this data.

all these gains happened, at initial release, with some spotty beta drivers. Even if they just fix the VRAM issues and the lighting/vapor effects rendering issues in BF4 and never improve performance much more, I'll be very happy!

Here's a look at my own tests

Using these settings on a 7870 overclocked to 1150 core 1325 memory:
ScreenshotWin32-0002.png


I gained over 25fps on my average framerate, during 64 player siege of Shanghai.
I run a Phenom II X6 which I didn't think was holding me back all that much. It seems my 7870 still have plenty to give!

I suggest you look at my full post with all the data I collected. Screenshots showing image differences, raw data from test server laps, huge VRAM usage differences, etc.
http://hardforum.com/showpost.php?p=1040594668&postcount=29

Otherwise, here is the bit that shows my framerates:

I finally bothered to calculate my average frame rates


Earlier I posted raw perfoverlay frame data (CSV files) for 13.12whql, 14.1 DX11 and 14.1 Mantle.
There are two sets of data. One set is for a test server lap which was kept as similar as possible for each.
video of what a lap looks like:
https://www.dropbox.com/s/wgd3gr8fdlmn429/bf4testserverrunsamplea.mp4

The other set of data is for a roughly 6 minute play session, in 64 player Siege of Shanghai. (timed for no less than 6 minutes). I used a 24/7 Siege of Shanghai server so that I could ensure reliable testing of the same map on the same server. These multiplayer sessions also had AMD Performance Monitor dumping CPU core/s and GPU core utilization data. So the CPU load is slightly higher than what it normally would be. Which is fine, because many people are running overlays and/or monitoring software anyway.

Here are the average frame rates for the test server lap:

14.1 Mantle: 108
14.1 DX11: 103
13.12 DX11: 104

You can see here, the Test server probably doesn't stress CPU performance. As such, the average performance in framerate is basically the same. There are points at which Mantle performs better in the Test Server. But overall it is pretty much the same.

But it should be noted that Mantle still delivers much more consistant frametimes. Therefore the "feel" is smoother. (of course, ignoring the VRAM hitching).

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

Average frame rates for 64 player Siege of Shanghai, during 6 minutes of play:

14.1 Mantle: 85
14.1 DX11: 58
13.12 DX11: 54


Here is where we get HUGE gains. That speaks for itself. Even despite the fact that the Mantle run was hitching every few seconds, with occasional hard stops/several frame drops, it still turns in a much higher average number.

The hitching is likely related to the fact that Mantle uses a lot more VRAM, for the same settings, on my system. which I show in this post here:
http://hardforum.com/showpost.php?p=1040596739&postcount=76

I have a 7870, so it is probably a bug with how Mantle is working on those cards/my system.
-----------------------------------------------------------------------------------
As an avid BF4 player, I have to say that Mantle makes a pretty great difference in gameplay when it's working perfectly.

I would say maybe about 2-4 times per game, there like 5 second instances where the frames will drop very suddenly, and then pick back up. This is a widely reported issue and seeing as it's in beta, I'm willing to live with it until they can fix it.

I'm willing to live with it because the game overall just plays so much more smoothly with Mantle. While my frames aren't significantly improved, what is improved is just the entire feeling. It feels much much less choppy and more fluid. It's difficult to quantify, but it's immediately apparent. Apparent enough that I have no interest in turning DX back on. I was pretty surprised as, coming from AMD, I didnt expect to really see significant results.

All in all, I'm happy with the direction it's going and I'm very happy with how BF4 is performing with my 290x.

The difference is quantified when you look at frametimes. You can see them graphed in [H]'s review or on techreport.com's reviews (they post frametimes for all GPU and game perfomrance reviews. Not just Mantle).

Frametimes are how fast each individual frame is delivered. FPS is an average over time, which ignores the specifics on what is happening with each individual frame. The thing is, there is a lot of room in 1000ms/1 second, for variance. 5 frames could be delivered at about 7ms. But then you might have 1 or two that take 23ms. The more spikes like that you have, the less smooth things will feel. Mantle bring average frame times waaay down. There's still some work to do with spikes and VRAM, but overall you still feel the smoothness. I mentioned near the top of this post to read Techreport's articles, if you are interested in knowing the details on this stuff.
 
All doom and gloom eh? Do you even know what you are saying? a 30% boost is like spending $100 extra dollars on a GPU.
There's also tests out there showing low potency CPU architectures like the A10 or old Core2quads getting 100% extra performance. Essentially making them perform about as well as i5/i7 do in D3D. That is absolutely awesome.
Did you know that only a small percentage of steam users actually have hardware that makes their PC significantly better than a PS4.
Only 2% have displays capable of higher than 1080p.
and Only about 25%f steam users have a PC roughly equivalent to a PS4.
valve publishes this data.

for some set ups it does a lot more than to even buy a new rig.
its a good start, bit beta still and buggy IMO, but I gained 10fps with my resolution and set up. so its 125fps framelimit for me instead of 115.
and smoother except the hickups from mantle/driver
 
for some set ups it does a lot more than to even buy a new rig.
its a good start, bit beta still and buggy IMO, but I gained 10fps with my resolution and set up. so its 125fps framelimit for me instead of 115.
and smoother except the hickups from mantle/driver

Buying a new rig is what I did a few months ago in anticipation of R9 cards. In addition to the rig below which has been with me since 2009, I now have a 4670K platform. Will I get $500 more dollars of performance out of this new machine over my old machine? I highly doubt it, under Mantle the old machine shines once more. I could have immediately put that $500 towards another R9 card.

All the reviewers talk about the "game play experience". In this beta, a boost to the experience is clearly exhibited. If a developer can port a game to the PC and be assured of giving a better experience on a cheaper machine wouldn't this increase sales on the PC platform? Wouldn't the developers see that increased income stream and want to tap into it?

Chameleoneel wrote:
Did you know that only a small percentage of steam users actually have hardware that makes their PC significantly better than a PS4.
Only 2% have displays capable of higher than 1080p.
and Only about 25%f steam users have a PC roughly equivalent to a PS4.
valve publishes this data.

This is what makes it a game changer, at least to me. Some of us here with our Video cards that cost more than most people's whole systems including the display are saying Mantle will have no effect. Those of us with the higher level systems and cards have already purchased our preferred performance level. Those with lesser systems will be getting a major boost.
 
Gsync is more beneficial then Mantle, since a Mantle is still up in the air. With Gsync you get support for every game.
 
Back
Top