The AMD Mantle Thread

[Stuka87]

Again, Mantle is not free. Nor, will any NV low level, hardware specific API be.

Source? Post up your source or stop spouting off misinformation.

 

[DamnedLife]

Again, Mantle is not free. Nor, will any NV low level, hardware specific API be.

Hahahahahah laughing is best answer to your pathetic trolling.

 

[cytg111]

Hahahahahah laughing is best answer to your pathetic trolling.

srly! .. will cost big $$$'ers .. but it will be worth it if you want the best of the best.

Prolly like 7 buncks.

 

[smackababy]

Source? Post up your source or stop spouting off misinformation.

This has already been explained. Just because it is free to the user doesn't mean it doesn't come without a price. Additional development time (no matter how small) costs money.

 

[Erenhardt]

Source? Post up your source or stop spouting off misinformation.

Probably the same reasoning why freesync isn't free. You still need to buy the game - in the end of the day you are paying them money. This could be applied to every free thing in this world, but whatever

 

[cytg111]

This has already been explained. Just because it is free to the user doesn't mean it doesn't come without a price. Additional development time (no matter how small) costs money.

Correct but it is really time to shake up the directx monoploy on api's for games, so I for one welcome the extra cost for now. In the end, we, the consumers, will be the winners.

 

[OCGuy]

180fps compared to.....160? 165?

I really don't think that is what a "low level API" is meant for....lol.

 

[Enigmoid]

If it's running at 180FPS, I can guarantee it is very CPU bound and was likely running at maybe 100-120FPS on DX. Biggest improvements at high FPS (CPU bound).

Thanks for the link.

First level implies Singleplayer.

Singleplayer is hardly CPU bound

 

[krumme]

Sp is uninteresting whatever the result.

 

[Stuka87]

This has already been explained. Just because it is free to the user doesn't mean it doesn't come without a price. Additional development time (no matter how small) costs money.

I am fully aware of what you meant. But what you said was that somebody would have to pay AMD money. You did not say "but the developer has to implement it". This is a given, and is true for dang near anything when it comes to implementing any feature.

If the latter is what you intended, than you need to be more clear and stop posting things that look very much like trolling.

 

[VulgarDisplay]

My prediction for mantle is 15-20% on max fps and 30-45% increase on minimums.

I really think mantle is going to have a huge impact on minimum fps, and that is going to be awesome.

Also I truly can't wait for the reactions from everyone in this thread.

 

[chimaxi83]

This has already been explained. Just because it is free to the user doesn't mean it doesn't come without a price. Additional development time (no matter how small) costs money.

Yet still, even when financials have been mentioned 9000 times in this thread.... nobody cares. Why should anybody who is not AMD or a game developer care?

 

[RampantAndroid]

I am fully aware of what you meant. But what you said was that somebody would have to pay AMD money. You did not say "but the developer has to implement it". This is a given, and is true for dang near anything when it comes to implementing any feature.

If the latter is what you intended, than you need to be more clear and stop posting things that look very much like trolling.

With all due respect, you read into what he posted. It costs the end user in QC, development time or attention to DX if that happens to suffer.

 

[Imouto]

With all due respect, you read into what he posted. It costs the end user in QC, development time or attention to DX if that happens to suffer.

Yeah, we should make a petition to remove those wicked coffee machines in their offices.

 

[RampantAndroid]

Yeah, we should make a petition to remove those wicked coffee machines in their offices.

Huh? Where do you get that?

 

[Gloomy]

Huh? Where do you get that?

It costs money that could be used toward DX.

We should also remove their chairs and replace them with milk crates.

 

[RampantAndroid]

It costs money that could be used toward DX.

We should also remove their chairs and replace them with milk crates.

So, you know when you're going through a software cycle (I don't care if you're doing scrum or waterfall or whatever else) you generally "cost" the work out. Meaning you figure out how much time each task will take. Though I'm going to guess you don't really know much about software development cycles (and that's not an insult. It's a guess on my part.)

Generally speaking from my own experience, the cost usually is way too low to try to avoid cutting features early...and in the end you either end up cutting features way too late to make a difference, slipping your dates or slipping on quality.

Cost does not equate to money per se. It can if your development cycle takes longer to ship the same product as you did before.

Quit being so literal and obtuse.

 

[Carfax83]

That quote was in reference to Battlefield 4 performance though. The 290X is already faster than a Titan right now in DX11 in Battlefield 4 in a lot of benchmarks. Not by much, but still faster.

What benchmarks are those? The thing about benchmarks, is that they are subject to patches and driver updates.

The Radeons had a good lead on the NVidia cards when BF4 first became available, but after patch and driver updates, NVidia has all but closed the gap in SP, and actually has a large lead in multiplayer thanks to the aggressive multithreading implementation in their drivers:

NVidia drivers seem to be much better at leveraging multicore CPUs for performance. AMD is way behind NVidia in that front, and that's probably why they need something like Mantle, which is more geared towards multithreading at the foundation.

 

[Grooveriding]

I've seen you post your personal theory on driver differences and multicore CPUs before, but I'm not going to derail this thread down that rabbit hole. If you disagree on the benchmark front you can look around at some of the big sites and you'll easily find benchmarks, recent ones, showing the 290X with a small advantage at times as I said.

Here you posted one showing the Titan with a small advantage, my guess is because there is no MSAA being used and the benches I pay attention to are ultra+4xMSAA. Nvidia takes a bigger performance hit in BF4 from MSAA than AMD does, hence your linked bench.

edit, yup, looked that up, leaving out MSAA is giving nvidia a best case for BF4:

Nvidia has a harder time handling 4xMSAA in BF4. Here they are about on parity.

 

[OCGuy]

Wow NV does show strong now in BF4.....that wouldn't be good if the 780ti was king in a AMD optimized game that is also using a AMD proprietary API.

 

[Gloomy]

Those results seem... low. Is China Rising that demanding?

 

[bystander36]

My prediction for mantle is 15-20% on max fps and 30-45% increase on minimums.

I really think mantle is going to have a huge impact on minimum fps, and that is going to be awesome.

Also I truly can't wait for the reactions from everyone in this thread.

I was predicting 10-20% on averages, but I predict similar minimum improves to what you predict.

 

[Grooveriding]

Those results seem... low. Is China Rising that demanding?

BF4 is demanding at high resolutions. I had my 780 Classifieds clocked high enough that they were about as fast as moderately overclocked 780tis. Trying to play Ultra+4xMSAA on a single one of them I would call unplayable for multiplayer.

 

[Carfax83]

If you disagree on the benchmark front you can look around at some of the big sites and you'll easily find benchmarks, recent ones, showing the 290X with a small advantage at times as I said.

And I bet they are all in the single player campaign, as AMD is better able to compete because it's far less CPU intensive than the multiplayer..

Here you posted one showing the Titan with a small advantage, my guess is because there is no MSAA being used and the benches I pay attention to are ultra+4xMSAA. Nvidia takes a bigger performance hit in BF4 from MSAA than AMD does, hence your linked bench.

The reason why I did that was to show the CPU dependency that I was referring to and how crippled AMD's current drivers are in that regard. That's the whole purpose of creating Mantle, to make exploiting multicore processors easier and more effective.

It's a pity Mantle is destined to fail, because two APIs cannot co-exist indefinitely. One will eventually be eaten, and I can guarantee you it won't be Direct3D, as Direct3D despite it's flaws, has two massive advantages; cross architectural and backward compatibility. Enhanced multithreading capability can be added over time, as they did with deferred contexts.

 

[blackened23]

Well, nvidia did increase BF4 performance a ton through a couple of driver iterations:

Ultra 2X MSAA 2560x1600. I see nothing unplayable here on the GTX 780 vanilla. Fact is, carfax83 is correct: nvidia substantially increased BF4 performance through drivers. I personally remember how it performed at launch, and it DEFINITELY performs way way better now.