The AMD Mantle Thread

[ASM-coder]

There!
https://twitter.com/repi/status/429645531045199872

Interesting, and welcome!
Congrats on the release and I think a lot of us are looking forward to the future possibilities.

 

[desprado]

Where do you get the idea that the server is empty? Pure speculation on your part? The wording of the page roughly translated seems to indicate that there were in fact 64 players on the server. Nowhere was it stated that it was an empty server.

IMO the reviews did not tell that Severs was empty or full .

 

[rtsurfer]

Where do you get the idea that the server is empty? Pure speculation on your part?

More of an assumption.

I actually havent read the reviews* I just woke up) but i thought some sutws tested on empty servers to make the results reproducible. That would be the wise thing to do.

 

[cytg111]

Ill chime in with a few questions

1. What sort of performance benchmarks have you done inhouse? The up to 45% mark seems to have been rather conservativly set.

2. On the note on 1. Thinking high end builds, take the best case scenario (overclocked 4960X), adding on GPU power (crossfire, tri, quad), are you setting a new bar for the "Enthusiast" build here? (going where no frames have gone before) Is it a test you have done?

Thanks.

 

[krumme]

BTW:
nVidia cards are much faster in the MP pf BF4:

http://pclab.pl/art55953.html.

Nonsense. I knew 2 months ago we would have trouble with that stupid site that makes results nobody else can duplicate. We dont know anything about the testing.
How many players? 0 or 64? - i guess 0
What route for testing? - hmm??

And your selected quoting from the site doesnt help either. Try the next pages for a starter.

 

[Barbatruuk]

Wow, Mantle seems to stir the pot all over the internet atm. I'm impressed with the initial signs of performance and can't wait to read some detailed reviews on Mantle BF4 and Mantle drivers.

I'm running a stock 8350 and stock R9 290. I'm hoping this setup will be used in a couple of reviews as it's not an uncommon setup.

 

[sontin]

Lol?
That site shows numbers from the SP and MP. And in the SP the 290X is faster than the GTX780GHz edition.

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

 

[Gloomy]

Though I'm not here for PR, just noticed that this was the most active forum discussing Mantle and a lot of good posts here, though also some with some misconceptions or errors on them as well that I can try and help correct.

Why does 4x MSAA incur a 40% performance hit? Is this typical?

I'm guessing the DirectX result weren't hit as much because it was CPU-limited anyway.

 

[desprado]

Though I'm not here for PR, just noticed that this was the most active forum discussing Mantle and a lot of good posts here, though also some with some misconceptions or errors on them as well that I can try and help correct.

Sir it will be really kind of u answer that why there so much difference in ur CF result where DICE claim 58% of improvement but here it is a different story.

http://www.hardwareluxx.de/index.ph...te-ergebnisse-im-kampf-mantle-vs-directx.html
http://battlelog.battlefield.com/bf4/news/view/bf4-mantle-live/

 

[ASM-coder]

Dumb question perhaps, but why not?

I was thinking CF or nothing. Wrong thinking. I edited my post.

 

[Erenhardt]

BTW: Pclab.pl wrote that they used a 64-man server.
Google-translation:

I speak Polish. They use 64 player game mode, but they do not mention the number of players on the server.
Given their 20% differences between the runs (same hardware and settings) the number of players varies wildly.

Canned bot BF4 MP demo would be welcome. There is too much variability in human controlled environment, and empty sever testing is pointless.

 

[Despoiler]

Where do you get the idea that the server is empty? Pure speculation on your part? The wording of the page roughly translated seems to indicate that there were in fact 64 players on the server. Nowhere was it stated that it was an empty server.

An now where does it say it had anyone, but the tester on it? "64 man mode", the word mode indicates server configuration. It doesn't tell anyone that it was 64/64.

 

[parvadomus]

yes.................but it is advantage to Nvidia.

Their performance numbers respect to Nvidia have ZERO credibility. In fact, they bias is so clear that I don't doubt they made up that charts for their "NV drivers are better campaign".
Explain this:
An i3 is having the same performance as the i7 @ MP with MANTLE. Despite that bm being a highly threaded one, and the i3 does not even have turbo.
And the funniest part: they show the 780 being faster over a dual-core G3220 when they pointed in a previous review that NV cards didnt do well on dual cores ROFL. Link

 

[Slomo4shO]

BTW:
nVidia cards are much faster in the MP pf BF4:

http://pclab.pl/art55953.html

A GTX780TI should be 50% faster than a 290X. I said it in the Star Swarm thread:
Mantle is more a fix for AMD's poor DX driver quality than a fix to DX at all.

You seemed to have missed this multiplayer review which completely contradicts your statement:

 

[sontin]

Oh, and here is another benchmark from Hardwareluxx on a 64-man server - the same map Golem.de used:

http://www.hardwareluxx.de/index.ph...nisse-im-kampf-mantle-vs-directx.html?start=1

That's the problem with MP.

 

[blackened23]

Lol?
That site shows numbers from the SP and MP. And in the SP the 290X is faster than the GTX780GHz edition.

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

Yeah, I was also about to point that out. The 290X was faster in SP while the GTX 780 was quite a bit faster in MP. I'd attribute that to the heavily multi threaded DX11 drivers that NV has, and AMD doesn't.

But it is funny, that in the XFX HardOCP 290X review people were called out for questioning the results which were quite simply, ridiculous - these were cards not overclocked and for some reason mysteriously outperformed reference by a mile. Yet here we are. Full circle. With some people questioning results from PCLab, a reputable website, just because they don't favor the side they prefer. Sounds pretty hypocritical to me, pot kettle black and such, but whatever. Just shows that there is no moral high ground here when it comes to benchmark results.

Again, I think the Mantle results at that website are in line with everything else we've seen. Big benefits for APUs, and the GTX 780 takes the lead in MP testing due to nvidia's DX11 multi threaded drivers. I'd point out that AMD does not have MT DX11 drivers, unless something changed very recently.

 

[f1sherman]

Though I'm not here for PR, just noticed that this was the most active forum discussing Mantle and a lot of good posts here, though also some with some misconceptions or errors on them as well that I can try and help correct.

Hey that's great. We are truly honored to have you here.

1) Can you confirm if Mantle will work out of box in Thief and Plants vs. Zombies: Garden Warfare?
2) Mirrors Edge 2 - WHEN and will it have Mantle?
3) Was it really necessary to create another API? I don't have to tell you that Intel/NV will never support competition's ecosystem. Any word from Nvidia/Intel re:Mantle/BF4?

 

[desprado]

Their performance numbers respect to Nvidia have ZERO credibility. In fact, they bias is so clear that I don't doubt they made up that charts for their "NV drivers are better campaign".
Explain this:
An i3 is having the same performance as the i7 @ MP with MANTLE. Despite that bm being a highly threaded one, and the i3 does not even have turbo.
And the funniest part: they show the 780 being faster over a dual-core G3220 when they pointed in a previous review that NV cards didnt do well on dual cores ROFL. Link

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

 

[Paul98]

Looks great so far, I can't wait to try it out myself. Maybe dice will return the draw distance to where it was pre patch and re-enable environment physics. That would make the numbers even better.

 

[DaZeeMan]

Though I'm not here for PR, just noticed that this was the most active forum discussing Mantle and a lot of good posts here, though also some with some misconceptions or errors on them as well that I can try and help correct.

@ repi
Welcome, and thanks for dropping in!

250 Pages and counting, and this thread OFTEN figures prominently in Google searches on the subject. Quite the discussion!

---

Well, one thing is for sure, we ARE seeing increases in performance. Keep the benches coming!

 

[SiliconWars]

Lol?
That site shows numbers from the SP and MP. And in the SP the 290X is faster than the GTX780GHz edition.

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

Where else have you seen a 780 Ti outperform a 290X by almost 50%? The last results we saw(on [H]) have the 290X thumping the Ti.

Look at their results again - do you really believe that Mantle only give 1.8 fps from an i3 to i7? It's hard to get consistent MP results but if you use the logic it should be obvious enough that when the going gets tough, Mantle will prove itself vastly superior. That's what you're seeing on the german results - I guarantee that's a long, protracted battle with many players that's been benchmarked then.

The polish guys must have been running around empty space while looking at the sky on the 780 Ti. No other results we've seen recently support theirs.

 

[Erenhardt]

There non or rare people that pair i3 with GTX 780 or R9 290X.

Yep, and suddenly it can become a viable choice for gamers. Shifting a lot of $ from CPU to GPU purchase. leading to increased visuals quality.

Now, console APU makes a whole lot more sense. Given early mantle numbers I can see why small CPU cores were a choice for the next gen. Seems like it is not asymmetric at all. PS4 runs BF4 MP 64-man parties!

 

[SiliconWars]

Yeah, I was also about to point that out. The 290X was faster in SP while the GTX 780 was quite a bit faster in MP. I'd attribute that to the heavily multi threaded DX11 drivers that NV has, and AMD doesn't.

Right...so DX11's multi-threading is supposed to be better than Mantle's now?

Edit- DICE have made an amazing job of their DX11 threading btw, but there's no way it's as good as Mantle. I believe we're seeing almost worst-case for Mantle in BF4 because the game does such an immense job of CPU multi-threading as it is.

 

[ColonelBlimp]

I've learnt loads just reading this thread if I cut through the willy waving stuff.
Although I'm not technical enough to contribute my layman's understanding is that I should benefit a reasonable amount with my Phenom 965@3.8ghz and 280x.
Hope so, my board is AM3 only otherwise I would have popped an fx in there by now...

 

[desprado]

Right...so DX11's multi-threading is supposed to be better than Mantle's now?

So far there is only 1 review that has compared R9 290X using Mantle vs GTX 780 .So answer is yes for now.