The AMD Mantle Thread

[krumme]

Wow, I'm surprised the mods let the blatant marketing continue. We have had 253 pages of BS and overstating the case about the next great thing, the results prove it, and now the crap continues?
Now that Mantle has been proven to be "not all that", we are told, just wait it will get better?
This was brilliant marketing move by a floundering company to sell their latest and greatest, based on promises. and it worked.
AMD will always be #2 in CPU and GPU and no amount of phoney promises will change that.

3770K + 290X, 64 players

This is real world testing. Not useless sp bm.

Mantle is 50% faster where and when it matters most.

 

[Sisyphean]

Reading some posts in here you'd think lives literally depended on Mantle failing...

Warning issued for thread crapping.
-- stahlhart

 

[Erenhardt]

Phenom II X6 users will be happy.....

http://www.anandtech.com/show/7728/battlefield-4-mantle-preview

Much value, such purchase!
Shame no Phenom II X4 on that graph. A lot of people use it for this day. Mine is feeding 1,2GHz HD7870 so I'm obviously interested.

 

[Paul98]

So there we have it. Unless you are running a crappy Intel i3 or AMD APU, there are very small gains with Mantle over DX. Even i5 that cost around $200 is not being bottlenecked with a powerful GPU like 290x.

Good news for low end and those who crossfire 290x or run 780Ti SLI. But that needs to be verified.

Or are playing multiplayer where you are using a lot more CPU, or running lower settings which will be the majority of players.

 

[rtsurfer]

Wow, I'm surprised the mods let the blatant marketing continue. We have had 253 pages of BS and overstating the case about the next great thing, the results prove it, and now the crap continues?
Now that Mantle has been proven to be "not all that", we are told, just wait it will get better?
This was brilliant marketing move by a floundering company to sell their latest and greatest, based on promises. and it worked.
AMD will always be #2 in CPU and GPU and no amount of phoney promises will change that.

Mantle does show the performance increases and AMD GPUs were still competitive with Nvidia GPUs even before Mantle.
780 didnt drop in price just for nothing.

Although I know you cant see that with your Nvidia Goggles while sipping your Intel Cool-Aid.

Infraction issued for personal attack.
-- stahlhart

 

[Sisyphean]

Mantle does show the performance increases and AMD GPUs were still competitive with Nvidia GPUs enen before Mantle.
780 didnt drop in price just for nothing.

Although I know you cant see that with your Nvidia Goggles while sipping your Intel Cool-Aid.

The average BF4 player with an i3/i5 or AMD similar coupled with a mid range AMD card will see a huge performance increase with Mantle and yet somehow it's, "not all that". I don't think anyone was seriously expecting over 10% increase for enthusiast level rigs, but from what we can see in the previews Mantle is delivering on its promise. When a lot of these gamers see their game perform around 30% better I think they are going to be very happy. For the common gamer with an AMD card Mantle seems to be win. Now we just need to see how wide Mantle will be adopted. IMHO, even though it looks good I'm going to hold reservations for the adoption rate from developers.

 

[Will Robinson]

Wow, I'm surprised the mods let the blatant marketing continue. We have had 253 pages of BS and overstating the case about the next great thing, the results prove it, and now the crap continues?
Now that Mantle has been proven to be "not all that", we are told, just wait it will get better?
This was brilliant marketing move by a floundering company to sell their latest and greatest, based on promises. and it worked.
AMD will always be #2 in CPU and GPU and no amount of phoney promises will change that.

"I find your lack of faith...disturbing" ^_^
# Vader

Multiplayer 64 results

Ultra Details (Core i7-3770K, 16 GB DDR3-1333, Radeon R9 290X @ 1.000/2.500 MHz, Windows 8.1 x64, Catalyst 14.1 Beta)

 

[SiliconWars]

Wow, I'm surprised the mods let the blatant marketing continue. We have had 253 pages of BS and overstating the case about the next great thing, the results prove it, and now the crap continues?
Now that Mantle has been proven to be "not all that", we are told, just wait it will get better?

You're surprised the mods would allow a post pointing out what the senior GPU writer wrote about Mantle?

 

[krumme]

"I find your lack of faith...disturbing" ^_^
# Vader

Multiplayer 64 results

Ultra Details (Core i7-3770K, 16 GB DDR3-1333, Radeon R9 290X @ 1.000/2.500 MHz, Windows 8.1 x64, Catalyst 14.1 Beta)

The graph basicly shows what is needed:

At GPU bound situation mantle is 6.7% faster. (4k)
At CPU bound situation mantle is >48% faster (1080)

What is perfectly clear from the results is Mantle is extremely good to use cpu power and thats its very important in a game like bf4.

Everyone knows that min fps in mp decides what settings you use. The 48% is important here. Its an huge jump for performance imho.

 

[ASM-coder]

So there we have it. Unless you are running a crappy Intel i3 or AMD APU, there are very small gains with Mantle over DX. Even i5 that cost around $200 is not being a bottleneck with a powerful GPU like 290x.

I'm pretty sure I read(and I think it was from Dice), that if you are running with ultra high settings, it makes
sense that you could hit GPU bottlenecks. So as you reduce any CPU bottleneck, you start hitting the GPU ceiling. Since the i7 isn't so constrained, it's already much closer to the GPU limits.
A more interesting chart will be to see if and how much FPS stats
converge with various CPU configurations.

Once again, less to be gained in a GPU constrained scenario.
But that i5 number seems to be curious.

 

[Shakabutt]

Wonder if Bulldozer users will get a boost from Mantle ^_^

 

[AnandThenMan]

Using a 4960X for the preview makes sure that Mantle will show its advantages as little as possible by brute-forcing the issue. It would be nice to see Ryan using a processor that many people actually have in their rigs.

 

[Arkadrel]

Using a 4960X for the preview makes sure that Mantle will show its advantages as little as possible by brute-forcing the issue. It would be nice to see Ryan using a processor that many people actually have in their rigs.

^ Think so too.

The question is why Anandtech uses a 1,050$ CPU + overclock, to show off Mantle.

Your avg joe that games doesnt own a i7 - 4960x.


Then again your avg joe probably doesnt own a 290x either.

 

[Erenhardt]

Using a 4960X for the preview makes sure that Mantle will show its advantages as little as possible by brute-forcing the issue. It would be nice to see Ryan using a processor that many people actually have in their rigs.

What it actually showed is that even paying huge premium for server chips doesn't make the single gpu run at its full capabilities - it is hold back by a cpu.

Now going by the improvements graph provided by amd, weak apu gets 40% uplift, while the a lot faster CPUs about 10%, which means, the difference between cheap and expensive CPUS is much lessened.

 

[AnandThenMan]

Then again your avg joe probably doesnt own a 290x either.

True. But I bet there are way more people running a 290x with a $300-$400 CPU than with a $1050 one.

 

[Venomous]

So where are these Tahiti tests at?

 

[omeds]

I'm very impressed. This would be amazing for 120/144hz users.

 

[SiliconWars]

To be fair, that's what is used in the graphics cards reviews. You can't go changing the test setup just to show Mantle in a better light than normal - what might need looking at now is the automatic choice of using the highest end, max overclocked CPU for that.

Mantle will gain so little from an overclocked CPU in terms of performance that it's probably worthless and the power savings at stock clocks should be taken instead. I've suspected for a while that this is what many reviewers will do - turn Mantle into a power benefit instead of a performance benefit.

I mean look at the article again - there is zero benefit between a 2 GHz i7 and 4.2GHz i7 for Mantle at max settings.

 

[Spjut]

Using a 4960X for the preview makes sure that Mantle will show its advantages as little as possible by brute-forcing the issue. It would be nice to see Ryan using a processor that many people actually have in their rigs.

I agree, but he did actually test it with 4 cores at 2ghz, and 2 cores with hyperthreading at 3 ghz, making it representative of an i5 and i3.

He also included AMD's own benchmark, although it would of course have been more interesting seeing it done by himself

 

[Noctifer616]

http://www.hardwareluxx.de/index.ph...nisse-im-kampf-mantle-vs-directx.html?start=2

Crossfire frame time results.

 

[AnandThenMan]

To be fair, that's what is used in the graphics cards reviews. You can't go changing the test setup just to show Mantle in a better light than normal -

It would not be better than normal, it would actually be "normal" as in a CPU price range most people buy. I get your point though.

 

[cytg111]

Looks like Repi noticed too and fled.

While we have fanboys trolling left and right, the only thing that will give you a warning is typing of [redacted] .. just saying that maybe the moderation-ship could be put to good use elewhere, - otherwise.
The nature of anandtech forums(tm) chased off a constructive IAA discussion with a key figure of the topic in question.

Sir, please clean up house.

Infraction issued for moderator callout.
-- stahlhart

 

[el etro]

i7-3960X + R9 290X. Seems you do only get 10% when GPU limited, give or take a bit. Not bad, not really anything amazing. I want to see frametime graphs though.

3960x costs double of 290x. Its was intended to sell and to be paired with dual 290x and stronger setups...

4770k is a 8 thread(the biggest number of threads BF4 engine can handle well) strong processor for 290x. You could start to look overclocked 4770k results with a pair of 290x...

Difference can be done by a bit of overclocking.Still there is no difference.

Do it and put a second 290x on the rig. Mantle shines anyway.

As long as there is no comparision to nVidia it's so hilarious to read the hype material.

Then bring your own tests to the table.

Guru3d's preview

http://www.guru3d.com/articles_pages/amd_mantle_preview,1.html

Guru3D test problem is that they paired a $150 CPU with a $550GPU in one test and a $1000 CPU with a $550 GPU in another test. Will not show any tangible gain this way.

As i doubt DICE benchmark that said CF had 58% performance increase is turly for AMD marketing.
http://www.hardwareluxx.de/index.ph...te-ergebnisse-im-kampf-mantle-vs-directx.html

There really not much difference in CF.

Read golem.de tests?

Mid range gpu-bound system:

Since we're talking about a gaming CPU only feature they could tests with machines where at least the GPU costs Double of the CPU.

This does not surprise. I think i5 systems stand to see the best gains, since they are lacking in BF4 multiplayer. I would of preferred to see at least a 270 used though.

5Ghz 8350 and i5 is very good tests for mainstream/High-End gaming machines. Don't forget 8320 too.

Who cares. What matter is Mantle is aprox 50% faster where it matters most at the big maps. What is then also surprising is actually how cpu bound we often are at bf4. Its aparently at times a damn taxing game on your cpu.

Like BF3 MP was. Mantle is what MP gamers were waiting.

With gains like this on i7 CPUs, i5 gains should be even larger.

i5@stock gains may match i7@stock gains.

No, because here you can reproduce the scene.
With active players is more a gambit than anything else.

A benchmark should always be reproducible. Otherwise you end with something like the Star Swarm test which is quite useless for benchmarks.

Even the Mantle developer says it in a beyond 3D post linked a few posts behind.

So, your problem is that the numbers from the MP part is not in favour for AMD?

Or its because it is only site that shows tremendous results for Nvidia, and it is the only site that who is bashing Mantle is using to argue their posts?

There are none or rare people that pair i3 with GTX 780 or R9 290X.

FX8320 can overclock to FX9370 levels and cost(In Newegg) the same as most of i3 models.

LOL, you post a review from LAST YEAR, shortly after the game launched?!! D:

Driver updates and patches have changed a lot of things since last year dude!

Last week test.

I find those 780Ti scores really hard to believe. We all know that the 780Ti beats a 290X, but it doesn't beat it with a ridiculous margin like that.

Its all core clock based, knowing the awesome performance scaling of the two chips. Game changes if we are comparing a 880Mhz 290x with a 963Mhz 780Ti, or if you are comparing a 1140Mhz GTX 780 with a 900Mhz R9 290. That's why boost clock/powertune blows reviews.

 

[AnandThenMan]

I agree, but he did actually test it with 4 cores at 2ghz, and 2 cores with hyperthreading at 3 ghz, making it representative of an i5 and i3.

Correct me if I'm wrong, but disabling cores in the BIOS does not bring a 4960X down to the performance level of an i3 for example. 5x more cache for starters.

 

[sontin]

Or its because it is only site that shows tremendous results for Nvidia, and it is the only site that who is bashing Mantle is using to argue their posts?

It's the only site which compared both.
Until there is a another site which compares nVidia and AMD/Mantle i take them more serious than people who are hyping mantle for whatever reason.