How much does Mantle benefit low end CPU setups, lets run a few tests.

[humbug]

+1

If people look closely at the results there is a big increase in performance using 2 cores and HT on with Mantle over DX11, this is exactly the place in the market that Mantle is supposed to benefit most.

Like an i3.

It should also be noted that there is a huge increase for people with much older CPU's, i'm seeing a 50% increase in performance on a Phenom II.
Its gone from a big bottleneck to no bottleneck at all.

I would have thought people with an FX-63## would also see a huge increase.

 

[Ian Evey]

If anyone does not believe how bad a 4930k is on 2 cores and 4 threads @3.4ghz (memory @1600mhz).

Here is the proof

Go on Humbug you should have no trouble beating my score.

I think humbug really underestimates his amd cpu and seems to think it's the worst cpu ever which it isn't.

 

[Kaapstad]

Like an i3.

It should also be noted that there is a huge increase for people with much older CPU's, i'm seeing a 50% increase in performance on a Phenom II.
Its gone from a big bottleneck to no bottleneck at all.

I would have thought people with an FX-63## would also see a huge increase.

In the first set of figures with the CPU @3.5, 2 Cores and 2 Threads (similar to an i3) Mantle had a 22.8% gain over DX11.

If your mobo allows it you should experiment with turning off cores to see the difference, in theory the less cores the bigger the improvement with Mantle.

 

[humbug]

2 Cores and 2 Threads your performance is less than my DX, and MUCH less than my Mantle ^^^^

I think humbug really underestimates his amd cpu and seems to think it's the worst cpu ever which it isn't.

I own it, i don't think its a bad CPU at all.

Contrary to what some may believe it does not Bottleneck Games like Crysis 3 or even Sniper Elite V2.

It does however struggle with the likes of Planet Side 2 and other MMO's, but i don't like them anyway.

Never the less i know its an old CPU, i also know a modern Intel CPU is much stronger.

 

[drunkenmaster]

I think what is rather funny from the numbers in this thread, on a hex core AMD chip or a high end intel chip crippled to only 2c+ht beats the numbers, significantly, that Nvidia was touting for a 290x and hexcore Intel chip setup, 58fps they claimed for Mantle and 54 I believe with DX....

From Kaps numbers a dual core with HT beats that easily in Mantle and a 4c no HT beatst that in DX. The actual CPU Nvidia was using, their numbers were so completely bogus.

I think from everything I've seen of BF4/Thief benchmarking, minimums are the most interesting area to benchmark.
I think conclusions 3 and 4 are a little, overstated. DX faster than mantle in half the results at 4.5Ghz. Really? The first 2 results are a 3.5/6fps gain, the other 4 are margin or error numbers.

Same for HT/on off, margin of error stuff there I would say. Again in both circumstances, what happens with minimums included, maximums are useful to see where the gains are coming and to see difference between min/max, smallest difference would give relatively speaking the smoothest overall experience.

IE 50average with 25mins and 100max would be less smooth than 50 average with 40 mins and 60 max.

 

[drunkenmaster]

I think humbug really underestimates his amd cpu and seems to think it's the worst cpu ever which it isn't.

While it's by no means a bad processor it's a hexcore AMD chip that came out in April 2010, and the 4930k is a significantly better chip that launched in Q3 2013. It's 3 and a half year newer tech, when Intel was faster than the AMD chip 3.5 years before as well.

 

[humbug]

I think what is rather funny from the numbers in this thread, on a hex core AMD chip or a high end intel chip crippled to only 2c+ht beats the numbers, significantly, that Nvidia was touting for a 290x and hexcore Intel chip setup, 58fps they claimed for Mantle and 54 I believe with DX....

From Kaps numbers a dual core with HT beats that easily in Mantle and a 4c no HT beatst that in DX. The actual CPU Nvidia was using, their numbers were so completely bogus.

I think from everything I've seen of BF4/Thief benchmarking, minimums are the most interesting area to benchmark.
I think conclusions 3 and 4 are a little, overstated. DX faster than mantle in half the results at 4.5Ghz. Really? The first 2 results are a 3.5/6fps gain, the other 4 are margin or error numbers.

Same for HT/on off, margin of error stuff there I would say. Again in both circumstances, what happens with minimums included, maximums are useful to see where the gains are coming and to see difference between min/max, smallest difference would give relatively speaking the smoothest overall experience.

IE 50average with 25mins and 100max would be less smooth than 50 average with 40 mins and 60 max.

Well, i'm up 88% on the minimum.

For example. AMD Phenom II x6 1090T @ 4Ghz, R9 290 @ 1075 / 1400

DirectX: Min: 31.4
Mantle: Min: 57.1 (+88%)

DirectX: Max: 73.5
Mantle: Max: 89.6 (+20%)

DirectX: Avr: 45.4
Mantle: Avr: 67.8 (+50%)

 

[Kurisu Makise]

there is a big increase in performance using 2 cores and HT on with Mantle over DX11, this is exactly the place in the market that Mantle is supposed to benefit most.

Like an i3.

It should also be noted that there is a huge increase for people with much older CPU's, i'm seeing a 50% increase in performance on a Phenom II.
Its gone from a big bottleneck to no bottleneck at all.

not everyone though, my mantle benchmarks show much worse results than DX11 in Theif and BF4.

its not worth touching when its crippling my performance in games down to 20fps average.

hoping to be much happier with an nvidia card and i7 by the end of this year.

 

[Calin Banc]

Is this part of the point of Mantle, to overcome the fact AMD haven't made a decent gaming CPU in years?

Hold on, so if AMD and Intel were competitive just as nVIDIA and AMD, that would mean it's ok to have an inefficient API/software just because it can be overcome by brut force? Whatever the reasons were, it's definitely the right move and hopefully DX12 can seal the deal.

Anyway, that AMD CPU may be quite weak per core compared to a newer Intel one, hence the increase. If you look at Tom's article, AMD are getting the biggest increase, even with higher clock/core than Intel, efficiency is not good.

 

[LtMatt]

Here's how Mantle benefits my low end cpu.


1080P Ultra Preset +x4AA
70 FOV
2700k @ 5ghz
290P Crossfire @ 1000/1250

DX11.1

Spoiler

Mantle

Spoiler

Mantle 33% Faster Average FPS
Ten million times smoother
No Vsync input lag - Zero - Vsync on or off feel the same

 

[JediFragger]

I think humbug really underestimates his amd cpu and seems to think it's the worst cpu ever which it isn't.

You've got to have a hole in the head to get an AMD CPU though

And Matt, every time I turn around your 290s are running a lower core speed! For Gods sake pick a number and stick with it

 

[LtMatt]

You've got to have a hole in the head to get an AMD CPU though

And Matt, every time I turn around your 290s are running a lower core speed! For Gods sake pick a number and stick with it

Lol do you mean a higher core speed? 4.8ghz was my 24/7 but ive now created a 5Ghz 24/7 profile and tweaked my secondary timings to perfection.

 

[Marine-RX179]

not everyone though, my mantle benchmarks show much worse results than DX11 in Theif and BF4.

its not worth touching when its crippling my performance in games down to 20fps average.

hoping to be much happier with an nvidia card and i7 by the end of this year.

I've read that Mantle is only currently benefiting the 290s and is not optimised for HD7000 series yet. You'd better stick with using dx until there's confirmation that optimisation of Mantle has been made for 7000 series.

Also I find your logic is a bit strange. Mantle is something "extra" that AMD has to offer at the moment which Nvidia doesn't...unless you have have issue with dx on your AMD card, I don't see the reasoning on jumping boat to Nvidia, especially when they tends to be more expensive and give you less vram, and has lower memory bandwidth than the equivalent AMD cards. The latters can be ignored if Nvidia's pricing was the same as AMD's, but they are not...they are almost always more expensive.

Your i3 would be bottlenecking your 7950 on dx11, and dx12 ain't gonna be out till end of 2015. So between now and then, Mantle is actually your best chance to reduce the CPU bottleneck (despite it is not working for 7000 series at this point), unless of course you are upgrading to a overclocked i5/i7.

 

[LtMatt]

I've read that Mantle is only currently benefiting the 290s and is not optimised for HD7000 series yet. You'd better stick with using dx until there's confirmation that optimisation of Mantle has been made for 7000 series.

Also I find your logic is a bit strange. Mantle is something "extra" that AMD has to offer at the moment which Nvidia doesn't...unless you have have issue with dx on your AMD card, I don't see the reasoning on jumping boat to Nvidia, especially when they tends to be more expensive and give you less vram, and has lower memory bandwidth than the equivalent AMD cards. The latters can be ignored if Nvidia's pricing was the same as AMD's, but they are not...they are almost always more expensive.

Your i3 would be bottlenecking your 7950 on dx11, and dx12 ain't gonna be out till end of 2015. So between now and then, Mantle is actually your best chance to reduce the CPU bottleneck (despite it is not working for 7000 series at this point), unless of course you are upgrading to a overclocked i5/i7.

Read between the lines with this post...

After some research by myself i believe that the problem lies with the BF4 application itself and Dice/EA, rather than AMD. 7xxx series cards get nice gains on Thief and Starswarm but their gains on BF4 Mantle are more limited. You have to ask yourself why that is? Especially now after AMD have launched several more drivers. The problem is is DIce/EA and its an application issue. Hopefully it gets fixed in BF4, but the game has so many different issues on going.

However 7xxx still see gains on Mantle, nice gains in my HD7770 rig.

 

[Kaapstad]

Read between the lines with this post...

After some research by myself i believe that the problem lies with the BF4 application itself and Dice/EA, rather than AMD. 7xxx series cards get nice gains on Thief and Starswarm but their gains on BF4 Mantle are more limited. You have to ask yourself why that is? Especially now after AMD have launched several more drivers. The problem is is DIce/EA and its an application issue. Hopefully it gets fixed in BF4, but the game has so many different issues on going.

However 7xxx still see gains on Mantle, nice gains in my HD7770 case.

Does CF work with Starswarm, if it does I may have to look into it.

 

[LtMatt]

Does CF work with Starswarm, if it does I may have to look into it.

Not when i last checked and good point. I'm going to get on to AMD about that. However with Mantle crossfire, its up to game devs to implement it, its not something AMD do at driver level like in DX. Gives the devs more freedom to get better performance, however as we've seen with BF4 and Thief they haven't had it ready at launch which is disappointing. I hope its just a case of its new and its going to take a while to optimize it perfectly or such, rather than this becoming the norm for Mantle games.

 

[Gregster]

Good testing Kaap and comparing your results to Humbugs really does show how bad AMD CPU's are

Mantle making AMD CPU's look better... Who created Mantle? Nuff said

How does crossfire scale with Mantle chaps?

sorry, I couldn't resist

 

[Martini1991]

Lol do you mean a higher core speed? 4.8ghz was my 24/7 but ive now created a 5Ghz 24/7 profile and tweaked my secondary timings to perfection.

Derp

 

[robertchalmers1]

Good testing Kaap and comparing your results to Humbugs really does show how bad AMD CPU's are

Mantle making AMD CPU's look better... Who created Mantle? Nuff said

How does crossfire scale with Mantle chaps?

sorry, I couldn't resist

was that not the whole idea of mantle though, to give weaker cpus a boost in games?????

 

[Martini1991]

It's kinda irrelevant than Mantle benefit AMD CPU's more, I have no problem with that, they're boosting everyone, the fact that AMD CPU's see a higher gain isn't a problem at all.