The AMD Mantle Thread

[Saylick]

lol, best post I've read in a while.

Haha, it's sig worthy material in my opinion.

Minkoff, what are you doing? You "know" performance gains are not in the single digits? Really? How do you know this when not ONE IOTA of performance information resulting from Mantle use has been released or posted anywhere in public? See, this is the kind of posting I have a tremendous issue with. You actually just said out loud, that Mantle was a brilliant move for AMD. Why? What have you seen that deems it brilliant?
See what I'm getting at?
And please, none of that "Any fool could see" nonsense.
Stop praising this thing until we get some concrete info for goodness sake.
When it comes out and it does well, then I'll raise the roof with you, but until then dude, stop promoting.

He doesn't know, and neither do any of us, but he's making an educated guess. $8 million investment for only single digit gains? Do you think that's a good investment? I mean, I know AMD's management isn't all that top-notch, but I'd give them the benefit of the doubt here and expect that they have something better than single digits gains up their sleeve, especially since Mantle is more-or-less the highlight of the convention for those who care about gaming. To champion Mantle that seriously at APU13 only to show that it gives a paltry 5% boost isn't exactly good PR...

 

[Minkoff]

Minkoff, what are you doing? You "know" performance gains are not in the single digits? Really? How do you know this when not ONE IOTA of performance information resulting from Mantle use has been released or posted anywhere in public? See, this is the kind of posting I have a tremendous issue with. You actually just said out loud, that Mantle was a brilliant move for AMD. Why? What have you seen that deems it brilliant?
See what I'm getting at?
And please, none of that "Any fool could see" nonsense.
Stop praising this thing until we get some concrete info for goodness sake.
When it comes out and it does well, then I'll raise the roof with you, but until then dude, stop promoting.

You are correct and I realized it as soon as I posted. Nonetheless I rephrased Skynner: “I'll put it this way. We're not going to develop an entirely new API just for a performance gain of 3 or 4 percent. It will be a significant increase.”.
Does significant mean 6% to you? Do you think that people do not messure their words, when something so critical is being answered?

The brilliancy is not always on the technology side... I
- do you have people talking about Mantle?
- does the competition has anything to counter at this stage?
- does the possibility of free gains do not influence the purchasing/holding of cards? and so on...
- when was the last time you saw such a movement, that got you interested in GPUs?

Praising? When did I do that? Go back to any one of my posts and show me! As I said earlier, I am no AMD employee...dude, nor promoting anything.

 

[frozentundra123456]

But the point is nobody "knows". One can make a reasonable guess, but it is simply false to say we "know".

 

[Saylick]

But the point is nobody "knows". One can make a reasonable guess, but it is simply false to say we "know".

From what I understand, he doesn't say he knows what the exact number is...

1. We know, that gains are not in the single digit. Let's take 10% gains.

Translation:

1. We know that the gains are not in the single digits according to THW's Q&A, so to provide some basis for my example, lets assume worst case scenario and say that the gains are 10% only.

 

[GaiaHunter]

I for one have no idea what the performance gains with Mantle will be.
AMD guy said they wouldn't chase 2-3% performance increase.

From all I know of PR dudes that means it could be 3.1%.

 

[Saylick]

I for one have no idea what the performance gains with Mantle will be.
AMD guy said they wouldn't chase 2-3% performance increase.

From all I know of PR dudes that means it could be 3.1%.

Potentially, yes. Realistically?

 

[Minkoff]

Saylick, thank you Yes, I should have measured my words more carefully, I admit!
The THWs AMA quote I was looking for was:
"Real numbers regarding Mantle will be published at the AMD Developer Summit running November 11-13 in San Jose. I will say that we are not undertaking such a mammoth effort to yield 3-4% performance--that would be a waste of time."

Let's hope, that no PR would risk a huge letdown with such statements.

 

[Paul98]

From what I understand, he doesn't say he knows what the exact number is...


Translation:

that is how I read it.

 

[AnandThenMan]

Minkoff, what are you doing? You "know" performance gains are not in the single digits? Really? How do you know this when not ONE IOTA of performance information resulting from Mantle use has been released or posted anywhere in public? See, this is the kind of posting I have a tremendous issue with. You actually just said out loud, that Mantle was a brilliant move for AMD. Why? What have you seen that deems it brilliant?
See what I'm getting at?
And please, none of that "Any fool could see" nonsense.
Stop praising this thing until we get some concrete info for goodness sake.
When it comes out and it does well, then I'll raise the roof with you, but until then dude, stop promoting.

If someone wants to wax poetic about the potential of an upcoming tech, they are free to do so. You don't have any numbers either, but that doesn't stop you from being extremely skeptical about Mantle, and you are free to voice that opinion.

Now if you have read anything about Mantle, you would know that it has tremendous potential, and applying small amounts of logic point to it being a leap forward. AMD64/GCN code on consoles, pushed over to the PC side running on a low level architecturally specific API, you don't see how this is something game changing?

 

[Keysplayr]

You are correct and I realized it as soon as I posted. Nonetheless I rephrased Skynner: “I'll put it this way. We're not going to develop an entirely new API just for a performance gain of 3 or 4 percent. It will be a significant increase.”.
Does significant mean 6% to you? Do you think that people do not messure their words, when something so critical is being answered?

The brilliancy is not always on the technology side... I
- do you have people talking about Mantle?
- does the competition has anything to counter at this stage?
- does the possibility of free gains do not influence the purchasing/holding of cards? and so on...
- when was the last time you saw such a movement, that got you interested in GPUs?

Praising? When did I do that? Go back to any one of my posts and show me! As I said earlier, I am no AMD employee...dude, nor promoting anything.

Sorry man, maybe came across to harsh. Just that this thread is full of such things and maybe praised was the wrong word. My basic point was to say do not jump the gun.
And about the percentages? There isn't any percent. None whatsoever, yet.
Nuff said on my part. :thumbsup:

 

[Keysplayr]

If someone wants to wax poetic about the potential of an upcoming tech, they are free to do so.

Now if you have read anything about Mantle (it appears you don't know what Mantle actually is), you would know that it has tremendous potential, and applying small amounts of logic point to it being a leap forward. AMD64/GCN code on consoles, pushed over to the PC side running on a low level architecturally specific API, you don't see how this is something game changing?

I think I actually have a better grip on reality than that my friend. And waxing poetic about something that has yet to be even demoed is a bit silly IMHO.

 

[AnandThenMan]

I edited my post for clarity, what you quoted came across as too harsh, my apologies.

IMO there are paradigm shifts, and Mantle is one of them. I vividly remember when I first read about Opteron and thought, this will transform the industry. I don't see Mantle quite on that level, but in the same ballpark. Just my opinion.

 

[Minkoff]

Sorry man, maybe came across to harsh. Just that this thread is full of such things and maybe praised was the wrong word. My basic point was to say do not jump the gun. Nuff said on my part. :thumbsup:

No offense taken! As I said, I should have edited my post, as I saw my mistake.

 

[AnandThenMan]

And waxing poetic about something that has yet to be even demoed is a bit silly IMHO.

In some cases I believe you're right. But IMO, not this time. You don't see devs talking so highly about something if it's all smoke and mirrors. Sure some will say, the devs are paid to use it. But even if they are being outright bribed, a bribe doesn't produce genuine enthusiasm.

 

[Minkoff]

In some cases I believe you're right. But IMO, not this time. You don't see devs talking so highly about something if it's all smoke and mirrors. Sure some will say, the devs are paid to use it. But even if they are being outright bribed, a bribe doesn't produce genuine enthusiasm.

Couldn't have said it better

 

[GaiaHunter]

Potentially, yes. Realistically?

No idea.

I hope not but I'm not betting either way.

 

[Noctifer616]

For several years now we have been working with AMD on exciting game titles like Deus Ex, Hitman, Tomb Raider, and we are currently working on the latest version of Thief, to bring the best possible experience to AMD hardware. In this presentation we will discuss some great AMD technologies that we are using in Thief. Most importantly, we will talk about Mantle and why it is something we have been waiting for for a long time. We are currently implementing Mantle inside of Unreal Engine 3 and will share some of our experiences. We will also cover additional AMD technologies we are supporting in Thief, like AMD TrueAudio, 4K Gaming, as well how are exploiting the power of the compute hardware in the GCN architecture. While we will primarily talk about our high-level view on these technologies and why we are excited about them, we will also share some helpful implementation tips from our programmers.

http://developer.amd.com/apu/home/agenda-sessions/#gaming-summit_528126ea83524

Can be found under Session and Gaming Summit.

Developing Thief with new AMD technology (read the abstract).

I guess this confirms Thief to have Mantle support.

 

[Paul98]

http://developer.amd.com/apu/home/agenda-sessions/#gaming-summit_528126ea83524

Can be found under Session and Gaming Summit.

Developing Thief with new AMD technology (read the abstract).

I guess this confirms Thief to have Mantle support.

Also looks like we will get more info tomorrow

 

[Minkoff]

I was also looking at something else for several days...
November 13th, 9:00 am – 9:30 am. - Keynote, Brendan Iribe, - CEO, Oculus VR
Does it have more of Mantle or Kaveri in it? Or both?

 

[Saylick]

No idea.

I hope not but I'm not betting either way.

I don't know either, but like you say, we aren't the ones taking a bet here; AMD and Mantle-promoting developers are the ones who are taking a bet. You and I both know that we wouldn't take a bet if the odds don't look good in our favor, but AMD and the developers are currently taking up that bet and I'm sure they wouldn't bet big if the payout wasn't going to be good either. In other words, they have data that we don't, and for them to take up this bet tells you something about this data...

 

[Spidre]

I don't know either, but like you say, we aren't the ones taking a bet here; AMD and Mantle-promoting developers are the ones who are taking a bet. You and I both know that we wouldn't take a bet if the odds don't look good in our favor, but AMD and the developers are currently taking up that bet and I'm sure they wouldn't bet big if the payout wasn't going to be good either. In other words, they have data that we don't, and for them to take up this bet tells you something about this data...

This is a really bad way of thinking. Every bad game to date was thought to be a good idea.

 

[Minkoff]

This is a really bad way of thinking. Every bad game to date was thought to be a good idea.

Well, Mantle is currently thought as a bad idea (3DFX and similar) so I guess this might turn out to be a good game

 

[SiliconWars]

It's not a bet, they already know what kind of improvements Mantle makes possible. Nobody is adding Mantle to their game engines on a whim or a promise. Nobody is using words like dramatic, ridicule or vastly to describe sub 10% increases.

 

[Saylick]

This is a really bad way of thinking. Every bad game to date was thought to be a good idea.

In certain cases, yes, I would agree, but this isn't the case. No developer knows exactly how well a game will sell until the game actually launches. You can estimate it, but you will never be able to put a hard number on it until the game starts to sell (or doesn't sell, for that matter).

Mantle is different because you can quantify the performance benefits Mantle will provide with hard numbers. This isn't a guessing game for them. AMD and developers already have an idea of what this number is; we don't.

 

[Saylick]

It's not a bet, they already know what kind of improvements Mantle makes possible. Nobody is adding Mantle to their game engines on a whim or a promise. Nobody is using words like dramatic, ridicule or vastly to describe sub 10% increases.

Not trying to disagree with you here, because I'm not, but I do believe there is a bet here: AMD is betting that Mantle will make a big splash on the video game industry. In support of what you are saying, what AMD isn't betting on is the actual % boost that Mantle provides; like you said, that number is more-or-less known to them already.