Overclock.net › Forums › Industry News › Hardware News › [pcper] AMD Mantle and NVIDIA 337.50 Scaling Demonstrated with Star Swarm on AM1
New Posts  All Forums:Forum Nav:

[pcper] AMD Mantle and NVIDIA 337.50 Scaling Demonstrated with Star Swarm on AM1 - Page 4

post #31 of 48
Well greatest limitation on the AM1 is it's x4 pcie slot. Sticking in anything more than a 7790 would start to choke it.
Power Tower
(22 items)
 
SteamBox
(9 items)
 
Doge Miner
(7 items)
 
CPUMotherboardGraphicsRAM
Ryzen 1700X AX370-Gaming 5 AMD Radeon R9 200 Series G.Skill DDR4-2400 
RAMRAMRAMHard Drive
G.Skill DDR4-2400 G.Skill DDR4-2400 G.Skill DDR4-2400 Samsung 840 Pro 
Hard DriveHard DriveHard DriveHard Drive
CX300 Crucial 480GB Toshiba 4TB Toshbia 4TB Western Digital Black 1TB 
CoolingOSMonitorMonitor
h110i Windows 10 42" LG TV 20" Digitizer ASUS 
KeyboardPowerCaseMouse
Corsair Vengeance Mechanical Keyboard  850watt Vampire Gold Rated NZXT S340 Elite Corsair RGB FPS Mouse 
Mouse PadAudio
Borderlands Mousepad Realtek HD 
  hide details  
Reply
Power Tower
(22 items)
 
SteamBox
(9 items)
 
Doge Miner
(7 items)
 
CPUMotherboardGraphicsRAM
Ryzen 1700X AX370-Gaming 5 AMD Radeon R9 200 Series G.Skill DDR4-2400 
RAMRAMRAMHard Drive
G.Skill DDR4-2400 G.Skill DDR4-2400 G.Skill DDR4-2400 Samsung 840 Pro 
Hard DriveHard DriveHard DriveHard Drive
CX300 Crucial 480GB Toshiba 4TB Toshbia 4TB Western Digital Black 1TB 
CoolingOSMonitorMonitor
h110i Windows 10 42" LG TV 20" Digitizer ASUS 
KeyboardPowerCaseMouse
Corsair Vengeance Mechanical Keyboard  850watt Vampire Gold Rated NZXT S340 Elite Corsair RGB FPS Mouse 
Mouse PadAudio
Borderlands Mousepad Realtek HD 
  hide details  
Reply
post #32 of 48
Quote:
Originally Posted by Alatar View Post

337.50 and mantle help in the same kinds situations, mantle helps more but only has a limited scope of games to work with.

this is the only point that really matters. good point.
post #33 of 48
I don't really think this is a fair comparison imo between the i7 & athlon with a 750ti. Intel's IPC alone goes up to 100% without that new 337.5 driver.
post #34 of 48
Quote:
Originally Posted by Ha-Nocri View Post

y not testing real games?

Probably because the only Mantle game is BF4 and so many people don't trust that game for benching due to bugs...though I cannot comment if that is true or not. I know I've seen a lot of people complain about testing done on a "broken game" though. What little I've seen of it the game seems fine. Thief is the only other game to use Mantle that I know of and from what benches I've seen the Mantle benefits were not very apparent. The only real boost was on a very low powered system where the frame rate went from unplayable to less unplayable (or barely playable depending on how you look at it).

The Starswarm bench is the only thing around that actually shows a strong improvement under Mantle that people will trust. So it is probably the best proving ground on which to bench.

The improvements to Nvidia GPUs with the 337.50 driver have been very close to the Mantle gains. Even my 580's here got the love.
Edited by Vagrant Storm - 4/22/14 at 7:26am
My System
(16 items)
 
  
CPUMotherboardGraphicsGraphics
2600K @ 4.7GHz Asus P8P67 B3 GTX 580 EVGA Hydro Copper 2 GTX 580 EVGA  
RAMHard DriveOptical DriveOS
4x4GB G.Skill Ripjaws bunch of 'em Blu-Ray For movies Windows 8.1 
MonitorMonitorKeyboardPower
24.4" Hans G HH251 X2 Yamakasi DS270  Blah 1000watt Super Flower 
CaseMouseMouse PadAudio
Built into Desk Microsoft SideWinder X8 Comfy one... Creative Extreme Gamer 
  hide details  
Reply
My System
(16 items)
 
  
CPUMotherboardGraphicsGraphics
2600K @ 4.7GHz Asus P8P67 B3 GTX 580 EVGA Hydro Copper 2 GTX 580 EVGA  
RAMHard DriveOptical DriveOS
4x4GB G.Skill Ripjaws bunch of 'em Blu-Ray For movies Windows 8.1 
MonitorMonitorKeyboardPower
24.4" Hans G HH251 X2 Yamakasi DS270  Blah 1000watt Super Flower 
CaseMouseMouse PadAudio
Built into Desk Microsoft SideWinder X8 Comfy one... Creative Extreme Gamer 
  hide details  
Reply
post #35 of 48
Quote:
Originally Posted by Vagrant Storm View Post

Probably because the only Mantle game is BF4 and so many people don't trust that game for benching due to bugs...though I cannot comment if that is true or not. I know I've seen a lot of people complain about testing done on a "broken game" though. What little I've seen of it the game seems fine. Thief is the only other game to use Mantle that I know of and from what benches I've seen the Mantle benefits were not very apparent. The only real boost was on a very low powered system where the frame rate went from unplayable to less unplayable (or barely playable depending on how you look at it)

Thief would be a great game to test it on.
Power Tower
(22 items)
 
SteamBox
(9 items)
 
Doge Miner
(7 items)
 
CPUMotherboardGraphicsRAM
Ryzen 1700X AX370-Gaming 5 AMD Radeon R9 200 Series G.Skill DDR4-2400 
RAMRAMRAMHard Drive
G.Skill DDR4-2400 G.Skill DDR4-2400 G.Skill DDR4-2400 Samsung 840 Pro 
Hard DriveHard DriveHard DriveHard Drive
CX300 Crucial 480GB Toshiba 4TB Toshbia 4TB Western Digital Black 1TB 
CoolingOSMonitorMonitor
h110i Windows 10 42" LG TV 20" Digitizer ASUS 
KeyboardPowerCaseMouse
Corsair Vengeance Mechanical Keyboard  850watt Vampire Gold Rated NZXT S340 Elite Corsair RGB FPS Mouse 
Mouse PadAudio
Borderlands Mousepad Realtek HD 
  hide details  
Reply
Power Tower
(22 items)
 
SteamBox
(9 items)
 
Doge Miner
(7 items)
 
CPUMotherboardGraphicsRAM
Ryzen 1700X AX370-Gaming 5 AMD Radeon R9 200 Series G.Skill DDR4-2400 
RAMRAMRAMHard Drive
G.Skill DDR4-2400 G.Skill DDR4-2400 G.Skill DDR4-2400 Samsung 840 Pro 
Hard DriveHard DriveHard DriveHard Drive
CX300 Crucial 480GB Toshiba 4TB Toshbia 4TB Western Digital Black 1TB 
CoolingOSMonitorMonitor
h110i Windows 10 42" LG TV 20" Digitizer ASUS 
KeyboardPowerCaseMouse
Corsair Vengeance Mechanical Keyboard  850watt Vampire Gold Rated NZXT S340 Elite Corsair RGB FPS Mouse 
Mouse PadAudio
Borderlands Mousepad Realtek HD 
  hide details  
Reply
post #36 of 48
Quote:
Originally Posted by SpeedyVT View Post

Quote:
Originally Posted by Vagrant Storm View Post

Probably because the only Mantle game is BF4 and so many people don't trust that game for benching due to bugs...though I cannot comment if that is true or not. I know I've seen a lot of people complain about testing done on a "broken game" though. What little I've seen of it the game seems fine. Thief is the only other game to use Mantle that I know of and from what benches I've seen the Mantle benefits were not very apparent. The only real boost was on a very low powered system where the frame rate went from unplayable to less unplayable (or barely playable depending on how you look at it)

Thief would be a great game to test it on.

I think the results would be interesting...

Here is one I found that is testing 337 on Thief. I don't typically go to this site for benches though so I don't know how trustworthy it is. Looks to be good SLI gains, but some of that is probably an update to the SLI profiles for Thief. 5% on a single high end GPU though. Doesn't have Mantle results on the bench, but just compare them to your favorite Mantle bench for Thief. I am kind of curious as to why there hasn't been more testing in Thief with the 337 driver. Probably because the Starswarm bench is a bit more exceptional in its gains...and is free and been out for a while. Thief is probably too new yet for most "professional" benchers to trust unless they were singling out Mantle. If you want to test Mantle there are not a lot of options.

Though I don't like the tests done with high end GPUs...they almost always are using a 780ti vrs a 290X...and I think a plain 780 should be used since it is priced closer to the 290X...really it should be a plain 780 vrs a plain 290 since those are close to being the exact same price. A 780ti is typically $100 more than a 290x so it should be expected to get a little more performance on average.
Edited by Vagrant Storm - 4/22/14 at 7:53am
My System
(16 items)
 
  
CPUMotherboardGraphicsGraphics
2600K @ 4.7GHz Asus P8P67 B3 GTX 580 EVGA Hydro Copper 2 GTX 580 EVGA  
RAMHard DriveOptical DriveOS
4x4GB G.Skill Ripjaws bunch of 'em Blu-Ray For movies Windows 8.1 
MonitorMonitorKeyboardPower
24.4" Hans G HH251 X2 Yamakasi DS270  Blah 1000watt Super Flower 
CaseMouseMouse PadAudio
Built into Desk Microsoft SideWinder X8 Comfy one... Creative Extreme Gamer 
  hide details  
Reply
My System
(16 items)
 
  
CPUMotherboardGraphicsGraphics
2600K @ 4.7GHz Asus P8P67 B3 GTX 580 EVGA Hydro Copper 2 GTX 580 EVGA  
RAMHard DriveOptical DriveOS
4x4GB G.Skill Ripjaws bunch of 'em Blu-Ray For movies Windows 8.1 
MonitorMonitorKeyboardPower
24.4" Hans G HH251 X2 Yamakasi DS270  Blah 1000watt Super Flower 
CaseMouseMouse PadAudio
Built into Desk Microsoft SideWinder X8 Comfy one... Creative Extreme Gamer 
  hide details  
Reply
post #37 of 48
Quote:
Originally Posted by SpeedyVT View Post

Well greatest limitation on the AM1 is it's x4 pcie slot. Sticking in anything more than a 7790 would start to choke it.

You are wrong here...

If it is 3.0 it would take faster than a 7970 before it chokes.
Intel Killer
(14 items)
 
  
CPUMotherboardGraphicsRAM
AMD 8320 4.4ghz Gigabyte 990fx XFX R9 290 (1100/1250) 8GB DDR3 1866 
Hard DriveCoolingOSMonitor
750GB Momentus XT Corsair H80i Windows 8 3x 23" 1080p 
KeyboardPower
Logitech Mk 710 OCZ 1000 watts 
  hide details  
Reply
Intel Killer
(14 items)
 
  
CPUMotherboardGraphicsRAM
AMD 8320 4.4ghz Gigabyte 990fx XFX R9 290 (1100/1250) 8GB DDR3 1866 
Hard DriveCoolingOSMonitor
750GB Momentus XT Corsair H80i Windows 8 3x 23" 1080p 
KeyboardPower
Logitech Mk 710 OCZ 1000 watts 
  hide details  
Reply
post #38 of 48
I got an average performance increase of 5.8% across 12 benchmarks in my testing going from 331.82 to 337.50. Those benchmarks weren't chosen to best demonstrate the difference, either. Mantle is way more effective, that's for sure, but until it is actually widely utilized, it doesn't do a whole lot for anyone who isn't focusing on a very specific scenario. Take me for example. I will never, ever go anywhere near BF4. I could have been interested in Thief if the game was well executed. Since it isn't, I'm not. The next Civ is a bit more interesting, but hardly a make-or-break game. So far it seems that Mantle will become relevant to me perhaps when the new Battlefront is released some time in 2015. Until then, a well optimized DX11 path gives me way more real life benefits. Then there is DX12 coming in 2015...

So while I give Mantle credit for being a technically superior API to DX11, I'm not exactly touched by it. DX11 is what my games run now and will run until H2 2015. It remains to be seen if the head start Mantle has on DX12 is enough to be a factor. The Crytek partnership is certainly a big win but then again, all the big game engines will go DX12 very quickly once it is released.
The Air Tunnel
(10 items)
 
  
CPUMotherboardGraphicsRAM
i7-4820K (4500MHz@1.28V) P9X79 Sapphire R9 290 Tri-X New Edition Team Group Vulcan 4x4GB 2133MHz 
Hard DriveCoolingOSPower
2x240GB SSD @RAID 0 Noctua NH-D14 SE2011 Windows 10 Pro Corsair AX750 
CaseAudio
Rosewill Armor EVO hiFace+AudioGD+DT770Pro 
  hide details  
Reply
The Air Tunnel
(10 items)
 
  
CPUMotherboardGraphicsRAM
i7-4820K (4500MHz@1.28V) P9X79 Sapphire R9 290 Tri-X New Edition Team Group Vulcan 4x4GB 2133MHz 
Hard DriveCoolingOSPower
2x240GB SSD @RAID 0 Noctua NH-D14 SE2011 Windows 10 Pro Corsair AX750 
CaseAudio
Rosewill Armor EVO hiFace+AudioGD+DT770Pro 
  hide details  
Reply
post #39 of 48
Quote:
Originally Posted by specopsFI View Post

I got an average performance increase of 5.8% across 12 benchmarks in my testing going from 331.82 to 337.50. Those benchmarks weren't chosen to best demonstrate the difference, either. Mantle is way more effective, that's for sure, but until it is actually widely utilized, it doesn't do a whole lot for anyone who isn't focusing on a very specific scenario. Take me for example. I will never, ever go anywhere near BF4. I could have been interested in Thief if the game was well executed. Since it isn't, I'm not. The next Civ is a bit more interesting, but hardly a make-or-break game. So far it seems that Mantle will become relevant to me perhaps when the new Battlefront is released some time in 2015. Until then, a well optimized DX11 path gives me way more real life benefits. Then there is DX12 coming in 2015...

So while I give Mantle credit for being a technically superior API to DX11, I'm not exactly touched by it. DX11 is what my games run now and will run until H2 2015. It remains to be seen if the head start Mantle has on DX12 is enough to be a factor. The Crytek partnership is certainly a big win but then again, all the big game engines will go DX12 very quickly once it is released.

Civilization VI smile.gif
Power Tower
(22 items)
 
SteamBox
(9 items)
 
Doge Miner
(7 items)
 
CPUMotherboardGraphicsRAM
Ryzen 1700X AX370-Gaming 5 AMD Radeon R9 200 Series G.Skill DDR4-2400 
RAMRAMRAMHard Drive
G.Skill DDR4-2400 G.Skill DDR4-2400 G.Skill DDR4-2400 Samsung 840 Pro 
Hard DriveHard DriveHard DriveHard Drive
CX300 Crucial 480GB Toshiba 4TB Toshbia 4TB Western Digital Black 1TB 
CoolingOSMonitorMonitor
h110i Windows 10 42" LG TV 20" Digitizer ASUS 
KeyboardPowerCaseMouse
Corsair Vengeance Mechanical Keyboard  850watt Vampire Gold Rated NZXT S340 Elite Corsair RGB FPS Mouse 
Mouse PadAudio
Borderlands Mousepad Realtek HD 
  hide details  
Reply
Power Tower
(22 items)
 
SteamBox
(9 items)
 
Doge Miner
(7 items)
 
CPUMotherboardGraphicsRAM
Ryzen 1700X AX370-Gaming 5 AMD Radeon R9 200 Series G.Skill DDR4-2400 
RAMRAMRAMHard Drive
G.Skill DDR4-2400 G.Skill DDR4-2400 G.Skill DDR4-2400 Samsung 840 Pro 
Hard DriveHard DriveHard DriveHard Drive
CX300 Crucial 480GB Toshiba 4TB Toshbia 4TB Western Digital Black 1TB 
CoolingOSMonitorMonitor
h110i Windows 10 42" LG TV 20" Digitizer ASUS 
KeyboardPowerCaseMouse
Corsair Vengeance Mechanical Keyboard  850watt Vampire Gold Rated NZXT S340 Elite Corsair RGB FPS Mouse 
Mouse PadAudio
Borderlands Mousepad Realtek HD 
  hide details  
Reply
post #40 of 48
Quote:
Originally Posted by specopsFI View Post

I got an average performance increase of 5.8% across 12 benchmarks in my testing going from 331.82 to 337.50. Those benchmarks weren't chosen to best demonstrate the difference, either. Mantle is way more effective, that's for sure, but until it is actually widely utilized, it doesn't do a whole lot for anyone who isn't focusing on a very specific scenario. Take me for example. I will never, ever go anywhere near BF4. I could have been interested in Thief if the game was well executed. Since it isn't, I'm not. The next Civ is a bit more interesting, but hardly a make-or-break game. So far it seems that Mantle will become relevant to me perhaps when the new Battlefront is released some time in 2015. Until then, a well optimized DX11 path gives me way more real life benefits. Then there is DX12 coming in 2015...

So while I give Mantle credit for being a technically superior API to DX11, I'm not exactly touched by it. DX11 is what my games run now and will run until H2 2015. It remains to be seen if the head start Mantle has on DX12 is enough to be a factor. The Crytek partnership is certainly a big win but then again, all the big game engines will go DX12 very quickly once it is released.

This is my feelings on it as well. If I can get a 15% boost with no change made to the games...as in just a driver update. I will take that over a 30% boost for two specific games when ran on specific hardware. The only thing I am little upset over is that Nvidia hasn't had their drivers set up like this for a long time. I mean, even Fermi got a boost. So whatever they are doing they are tapping potential that has been in these GPUs for a long time. I know, I know...development takes time, but still...I could have been using Ubersampling on The Witcher 2 for a long time tongue.gif

Then when DX12 comes out...if it even does half of what Mantle has the potential to do I think it will be the cloud on Mantle's sunny day. I just don't see Mantle surviving unless AMD starts churning out more and more versions of it for more GPUs and I don't know if that is possible for them to do with GPUs they don't make. Though even if they did that the developers would have to make a different version of the game for each mantle version...I think...in theory that is what should need to happen, but it is hard to tell at this point. It might be possible to add the extra libraries for other GPUs into one version, but then I think you will be loosing a lot of the benefit of Mantle because it won't be as low level any more and need to be detecting what library to access and when. I don't know...all guess work at this point.
My System
(16 items)
 
  
CPUMotherboardGraphicsGraphics
2600K @ 4.7GHz Asus P8P67 B3 GTX 580 EVGA Hydro Copper 2 GTX 580 EVGA  
RAMHard DriveOptical DriveOS
4x4GB G.Skill Ripjaws bunch of 'em Blu-Ray For movies Windows 8.1 
MonitorMonitorKeyboardPower
24.4" Hans G HH251 X2 Yamakasi DS270  Blah 1000watt Super Flower 
CaseMouseMouse PadAudio
Built into Desk Microsoft SideWinder X8 Comfy one... Creative Extreme Gamer 
  hide details  
Reply
My System
(16 items)
 
  
CPUMotherboardGraphicsGraphics
2600K @ 4.7GHz Asus P8P67 B3 GTX 580 EVGA Hydro Copper 2 GTX 580 EVGA  
RAMHard DriveOptical DriveOS
4x4GB G.Skill Ripjaws bunch of 'em Blu-Ray For movies Windows 8.1 
MonitorMonitorKeyboardPower
24.4" Hans G HH251 X2 Yamakasi DS270  Blah 1000watt Super Flower 
CaseMouseMouse PadAudio
Built into Desk Microsoft SideWinder X8 Comfy one... Creative Extreme Gamer 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Hardware News
Overclock.net › Forums › Industry News › Hardware News › [pcper] AMD Mantle and NVIDIA 337.50 Scaling Demonstrated with Star Swarm on AM1