Chris Roberts on Mantle:DX12 is basically the same

Rusty0611 · 20 Apr 2014 at 14:20

humbug said:
Your being far too disingenuous with your attempt at down playing it there, people with overclocked i7s are seeing significant gains with Mantle.

Only if they're multi-GPU. The gains for your average overclocked i7 / single GPU user are nothing to write home about. Sure any gain is of course a good thing but it's not game changing.

Ian Evey · 20 Apr 2014 at 14:26

shankly1985 said:
And I was waiting for you to reply back with max fps.
Tells me you don't fully understand what mantle does/offers the user over directx.

I'll leave this here for you information it's not about max frame per second.

That response tells me you don't understand that on bf4 ultra settings most nvidia cards on dx 11 run bf4 smoothly and looking nice. Unless mantle increases the IQ aswell which I have seen does not.

I am talking about a 64 player server also.

So of I can play 64 player servers on ultra with a nice smooth frame rate, and have some fun, why would I need mantle. I DONT.

Martini1991 · 20 Apr 2014 at 14:28

Ian Evey said:
That response tells me you don't understand that on bf4 ultra settings most nvidia cards on dx 11 run bf4 smoothly and looking nice. Unless mantle increases the IQ aswell which I have seen does not.

I am talking about a 64 player server also.

So of I can play 64 player servers on ultra with a nice smooth frame rate, and have some fun, why would I need mantle. I DONT.

That's a very weird viewpoint.

In that very slanted frame of mind, AMD owners don't need Mantle either, as they're able to do the same thing.

Ian Evey · 20 Apr 2014 at 14:31

LtMatt said:
They really do. These are how things look in DX for AMD and Nvidia. Fine for single card but multi card is not perfect for either side.

This is how it looks on Mantle. Just look at that impeccable variance. Trust me, you notice that.

Hold on a minute, No one mentioned sli or xfire.

As I only have 1 card I cannot comment on that at all. All I do know is that currently mantle has 2 games that make 0 difference for nvidia users currently, This would be a massive point for mantle if nvidia users had a poor gaming experience but they don't.

LtMatt · 20 Apr 2014 at 14:37

Ian Evey said:
Hold on a minute, No one mentioned sli or xfire.

As I only have 1 card I cannot comment on that at all. All I do know is that currently mantle has 2 games that make 0 difference for nvidia users currently, This would be a massive point for mantle if nvidia users had a poor gaming experience but they don't.

Ian Evey said:
Because nvidia users don't need it

You didn't mention single card, I assumed you were generalising. So are you now admitting that Nvidia users could benefit from Mantle, if it were supported?

AMD's frame times are as good, if not better than Nvidia's in BF4 and have been since release. Probably due to the fact its a GE title. The difference between DX-Mantle is not so noticeable on a single card. On dual card its night and day and it makes you realise how poor an experience DX11 offers in comparison. Its one of those things you have to try though before you can really comment on it. You can think your gameplay is smooth, then you try something smoother and you realise what you've been missing out on. I experienced something similar back before and after amd introduced frame pacing for Tahiti crossfire last August. Was mostly oblivious to it before, but afterwards i realise how much smoother it made games.

shankly1985 · 20 Apr 2014 at 14:37

Ian Evey said:
That response tells me you don't understand that on bf4 ultra settings most nvidia cards on dx 11 run bf4 smoothly and looking nice. Unless mantle increases the IQ aswell which I have seen does not.

I am talking about a 64 player server also.

So of I can play 64 player servers on ultra with a nice smooth frame rate, and have some fun, why would I need mantle. I DONT.

Its not just about getting 100fps etc have you ever asked yourself why does 30fps on Console feel so different to 30fps on PC? The clue is in the frame times and Mantle does frame times so much better than DirectX. You would not know this because you haven't used Mantle.
For example before you used 120hz, 60hz would feel so good, you then use 120hz and 60hz shows its weakness.

Dont dis something till you have tried it.

Ian Evey · 20 Apr 2014 at 14:37

Martini1991 said:
That's a very weird viewpoint.

In that very slanted frame of mind, AMD owners don't need Mantle either, as they're able to do the same thing.

Exactly amd users don't need it, But they have it and like it mostly.

That does not mean nvidia users need it which was my point. It's as if people are saying you cannot play bf4 if you don't have mantle as dx is atrocious and the constant "you can't use it hah" attitude by some is very strange when no one needs it to get playable frame rates "currently".

Ian Evey · 20 Apr 2014 at 14:40

shankly1985 said:
Its not just about getting 100fps etc have you ever asked yourself why does 30fps on Console feel so different to 30fps on PC? The clue is in the frame times and Mantle does frame times so much better than DirectX. You would not know this because you haven't used Mantle.
For example before you used 120hz, 60hz would feel so good, you then use 120hz and 60hz shows its weakness.

Dont dis something till you have tried it.

I have not dissed it apart from it's not going to be open, merely said nvidia users don't need it.

uksoldierboy · 20 Apr 2014 at 15:09

Rusty0611 said:
Only if they're multi-GPU. The gains for your average overclocked i7 / single GPU user are nothing to write home about. Sure any gain is of course a good thing but it's not game changing.

I'd disagree Rusty, i've seen great gains on BF4 using mantle, especially online, not only in raw FPS but the smoothness while playing.

LtMatt · 20 Apr 2014 at 15:10

uksoldierboy said:
I'd disagree Rusty, i've seen great gains on BF4 using mantle, especially online, not only in raw FPS but the smoothness while playing.

With your cpu at that speed you should really be adding in a second card. If you're at 1080P you'll be at 200 fps on Mantle with 290X's at 1150/1400.

Rusty0611 · 20 Apr 2014 at 15:12

uksoldierboy said:
I'd disagree Rusty, i've seen great gains on BF4 using mantle, especially online, not only in raw FPS but the smoothness while playing.

I saw the same on my mates PC when going from AMD 290 DX to Mantle but my PCs DX was smoother than the AMD DX. Mantle was about the same in 'smoothness' to my nVidia set up. Tested using his 290.

He borked his windows install and fried his motherboard so did a rebuild for him

.

uksoldierboy · 20 Apr 2014 at 15:13

Lead me not into temptation Matt!

, When funds allow it...

LtMatt · 20 Apr 2014 at 15:14

Rusty0611 said:
I saw the same on my mates PC when going from AMD 290 DX to Mantle but my PCs DX was smoother than the AMD DX. Mantle was about the same in 'smoothness' to my nVidia set up. Tested using his 290.

He borked his windows install and fried his motherboard so did a rebuild for him .

That was on Mantle launch drivers though i think i recall you saying, plus Dice have released dozens of patches since then. Times have changed and now we have the frame time data to back up the claims.

uksoldierboy said:
Lead me not into temptation Matt!, When funds allow it...

Mantle crossfire will be smoother than single card DirectX. The frame variance does not lie.

uksoldierboy · 20 Apr 2014 at 15:17

Rusty0611 said:
I saw the same on my mates PC when going from AMD 290 DX to Mantle but my PCs DX was smoother than the AMD DX. Mantle was about the same in 'smoothness' to my nVidia set up. Tested using his 290.

He borked his windows install and fried his motherboard so did a rebuild for him .

Strange, BF4 felt the same for me using DX on my 780Ti and the 290X :confused:

I dunno, could be the difference in systems maybe.
BF4 does for me at least, "feel" far better on my 290X now, than it ever did on the 780Ti, using mantle.

Rusty0611 · 20 Apr 2014 at 15:17

LtMatt said:
That was on Mantle launch drivers though i think i recall you saying, plus Dice have released dozens of patches since then. Times have changed and now we have the frame time data to back up the claims.

Indeed it was but you were all saying how much smoother Mantle was back then as well. I think the massive increase in perceived smoothness was just down to DX being a bit borked in those release drivers.

We didn't roll back to pre-Mantle drivers to see if the DX on those was comparable to my own set up so that's probably it.

uksoldierboy said:
Strange, BF4 felt the same for me using DX on my 780Ti and the 290X
I dunno, could be the difference in systems maybe.
BF4 does for me at least, "feel" far better on my 290X now, than it ever did on the 780Ti, using mantle.

Unless we've got a 'smoothometer'

p) to measure it (and I'm not just referring to frame times) then there's little point debating it as you've got human error factoring in so it's just anecdote versus anecdote.

uksoldierboy · 20 Apr 2014 at 15:20

LtMatt said:
Mantle crossfire will be smoother than single card DirectX. The frame variance does not lie.

Indeed

hopefully i can find another Asus Reference card though so it at least matches my current.

LtMatt · 20 Apr 2014 at 15:21

Rusty0611 said:
Indeed it was but you were all saying how much smoother Mantle was back then as well. I think the massive increase in perceived smoothness was just down to DX being a bit borked in those release drivers.

We didn't roll back to pre-Mantle drivers to see if the DX on those was comparable to my own set up so that's probably it.

DX was borked on 14.1 (fixed in 14.2 i believe) so id say that explains it because there is no comparison between them now for 290 users, particularly if you have more than one gpu.

Rusty0611 said:
Unless we've got a 'smoothometer' p) to measure it (and I'm not just referring to frame times) then there's little point debating it as you've got human error factoring in so it's just anecdote versus anecdote.

You've got me and im smooth. Veet'd my areas last night.

uksoldierboy said:
Indeed hopefully i can find another Asus Reference card though so it at least matches my current.

Yeah gotta match.

uksoldierboy · 20 Apr 2014 at 15:22

Rusty0611 said:
Unless we've got a 'smoothometer' p) to measure it (and I'm not just referring to frame times) then there's little point debating it as you've got human error factoring in so it's just anecdote versus anecdote.

Completely agree

uksoldierboy · 20 Apr 2014 at 15:29

LtMatt said:
Yeah gotta match.

Oh yes, would drive me crazy if they didn't, not that i spend my time gazing into my case, but i'd know, and that would be enough lol.

Rusty0611 · 20 Apr 2014 at 15:31

Matt has to have matching cards to draw attention away from his mismatching Beadle hand.