Drazhar
CLASSIFIED Member
- Total Posts : 2370
- Reward points : 0
- Joined: 2008/12/07 02:14:51
- Status: offline
- Ribbons : 4
|
zildjian75
FTW Member
- Total Posts : 1261
- Reward points : 0
- Joined: 2007/04/03 16:47:53
- Location: Chas, SC
- Status: offline
- Ribbons : 1
Re: PC World tests DX12 with Futuremark
2015/03/26 06:31:05
(permalink)
Interesting Indeed! Looks like I won't have to upgrade my 980's for a good 10-15 years!! Riiiigggghhhht!!! There is no way it will be that much better. But I will keep my fingers crossed b/c it's going to take something besides a powerful discreet gpu to run the 3x surround 4K gaming rig of the future.
"The wise know their weakness too well to assume infallibility; and he who knows most, knows best how little he knows." Thomas Jefferson RIG#1 (Home) - i7 4790k @ 4.4ghz w/ CM V8 GTS - EVGA Z97 Classified - Win10 64bit - Corsair Dominator Platinum 16GB (2x8GB) @ 2133mhz - 2 x EVGA GTX 980SC in SLI - Corsair AX850 - 250GB Samsung Evo (main) - 500GB Samsung Evo (games) - 1TB WD Black (misc) - Aerocool Strike-X (Cheesy... But effective!)
|
knightsilver
Omnipotent Enthusiast
- Total Posts : 9430
- Reward points : 0
- Joined: 2008/11/21 13:42:06
- Location: Nebraska
- Status: offline
- Ribbons : 14
Re: PC World tests DX12 with Futuremark
2015/03/26 08:40:09
(permalink)
If true, native DX12/DP 1.2a-1.3 GPU's will be a nice upgrade!
|
mistermister
CLASSIFIED ULTRA Member
- Total Posts : 5804
- Reward points : 0
- Joined: 2008/03/29 02:38:09
- Location: San Diego
- Status: offline
- Ribbons : 13
Re: PC World tests DX12 with Futuremark
2015/03/26 09:13:09
(permalink)
So.... I need an i7 now lol.
AMD 3700x / X-570 Aorus Ultra / RTX-3090 FTW3
|
ledzppln6
iCX Member
- Total Posts : 286
- Reward points : 0
- Joined: 2011/11/12 08:39:08
- Location: Florida
- Status: offline
- Ribbons : 5
Re: PC World tests DX12 with Futuremark
2015/03/26 11:46:11
(permalink)
I'm very interested in seeing how well this actually gets implemented into games. It's promising to say the least.
| EVGA Z97 FTW| i7-4790K @ 4.6 | XSPC Raystorm D5 AX360 | EVGA GTX 980 Ti Hydro Copper | Cooler Master HAF-X | 16GB G. Skill Trident X 2400 | Samsung 840 Pro 256GB SSD | Corsair AX1200 | ACER XB270HU 144Hz 1440P G-SYNC |
|
Sajin
EVGA Forum Moderator
- Total Posts : 49131
- Reward points : 0
- Joined: 2010/06/07 21:11:51
- Location: Texas, USA.
- Status: offline
- Ribbons : 199
Re: PC World tests DX12 with Futuremark
2015/03/26 12:00:28
(permalink)
ledzppln6 I'm very interested in seeing how well this actually gets implemented into games. It's promising to say the least.
+1
|
seth89
CLASSIFIED ULTRA Member
- Total Posts : 5290
- Reward points : 0
- Joined: 2007/11/13 11:26:18
- Status: offline
- Ribbons : 14
Re: PC World tests DX12 with Futuremark
2015/03/28 13:39:44
(permalink)
Here are my results, but I dont have Windows 10 Beta installed yet... ...so no DX12. But my mantle test is in there. http://www.3dmark.com/aot/9377 Ran the test at 1080p60 AMD Radeon R9 290(3x) and Intel Core i7-3820 Processor DX11 Multi-threaded draw calls per second576553 DX11 Single-threaded draw calls per second575838 DX12 draw calls per second0.0 Mantle draw calls per second8655325 All well over 9,000!
|
knightsilver
Omnipotent Enthusiast
- Total Posts : 9430
- Reward points : 0
- Joined: 2008/11/21 13:42:06
- Location: Nebraska
- Status: offline
- Ribbons : 14
Re: PC World tests DX12 with Futuremark
2015/03/28 13:43:45
(permalink)
Seth89, how's Crossfire gameplay?
|
seth89
CLASSIFIED ULTRA Member
- Total Posts : 5290
- Reward points : 0
- Joined: 2007/11/13 11:26:18
- Status: offline
- Ribbons : 14
Re: PC World tests DX12 with Futuremark
2015/03/28 14:07:18
(permalink)
knightsilver Seth89, how's Crossfire gameplay?
With mantle on? I get the same FPS as DX11 in bf4 but i think thats b/c im already running the game at ultra and 200% scaling. With a single card I get better FPS using mantle vs dx11.
|
knightsilver
Omnipotent Enthusiast
- Total Posts : 9430
- Reward points : 0
- Joined: 2008/11/21 13:42:06
- Location: Nebraska
- Status: offline
- Ribbons : 14
Re: PC World tests DX12 with Futuremark
2015/03/28 14:14:12
(permalink)
Was mainly asking about (with Crossfire) tearing, stuttering, and just smoother game vs SLI(aka poorly made SLI drivers). Thx for the reply.
post edited by knightsilver - 2015/03/28 14:18:36
|
seth89
CLASSIFIED ULTRA Member
- Total Posts : 5290
- Reward points : 0
- Joined: 2007/11/13 11:26:18
- Status: offline
- Ribbons : 14
Re: PC World tests DX12 with Futuremark
2015/03/28 14:26:32
(permalink)
knightsilver Was mainly asking about (with Crossfire) tearing, stuttering, and just smoother game vs SLI(aka poorly made SLI drivers). Thx for the reply.
no stuttering for me, but if I move super fast I can sometimes see slight tearing. However most of the time I run at 75fps and my monitor refreshes at 75 so I think I luck out there.
|
seth89
CLASSIFIED ULTRA Member
- Total Posts : 5290
- Reward points : 0
- Joined: 2007/11/13 11:26:18
- Status: offline
- Ribbons : 14
Re: PC World tests DX12 with Futuremark
2015/03/28 14:54:31
(permalink)
@knightsilver Do you have Frame Pacing enabled under AMD CrossFire?
|
Brad_Hawthorne
Insert Custom Title Here
- Total Posts : 23174
- Reward points : 0
- Joined: 2004/06/06 16:13:06
- Location: Dazed & Confused
- Status: offline
- Ribbons : 39
Re: PC World tests DX12 with Futuremark
2015/03/28 16:12:57
(permalink)
All of our tests were performed at 1280x720 resolution at Microsoft's recommendation. Eh?
|
wmmills
CLASSIFIED ULTRA Member
- Total Posts : 5679
- Reward points : 0
- Joined: 2006/01/04 20:47:29
- Location: New Jersey
- Status: offline
- Ribbons : 40
Re: PC World tests DX12 with Futuremark
2015/03/28 16:42:44
(permalink)
Brad_Hawthorne
All of our tests were performed at 1280x720 resolution at Microsoft's recommendation. Eh?
Well, they have to make sure xbone can run it too. lol Good catch on that brad!!!
MOBO: EVGA x299 Dark, CPU: I9 10900X, RAM: Patriot Viper RGB 3600 32gb, SSD: Samsung 860 EVO 1TB, M2: Samsung 970 EVO+ 1TB, PSU: CoolerMaster M2 1500, CPU HSF: EVGA 240 CLC HSF~ P/P EK Furious Vardar, G-CARD:EVGA RTX3090 FTW3 Ultra Gaming w/Hybrid kit and Noctua IPPC 3000 P/P, CASE: LIAN-LI PC-V2010B w/ Window mod, OS: Windows 10 Pro 64bit, MON: Alienware AW3821DW New EVGA product? Register it NOW with this link: http://www.evga.com/register/default.asp?affiliatecode=4QFQRAMOII Help Our Vets From K-2! [link=https://strongholdfreedom
|
knightsilver
Omnipotent Enthusiast
- Total Posts : 9430
- Reward points : 0
- Joined: 2008/11/21 13:42:06
- Location: Nebraska
- Status: offline
- Ribbons : 14
Re: PC World tests DX12 with Futuremark
2015/03/28 16:43:59
(permalink)
I'm looking at AMD For my next build Seth89 and considering Crossfire down the road...
|
Drazhar
CLASSIFIED Member
- Total Posts : 2370
- Reward points : 0
- Joined: 2008/12/07 02:14:51
- Status: offline
- Ribbons : 4
Re: PC World tests DX12 with Futuremark
2015/03/28 18:32:47
(permalink)
I was figuring the recommended resolution was to ensure it wasn't going to be GPU bound?
|
Vlada011
Omnipotent Enthusiast
- Total Posts : 10257
- Reward points : 0
- Joined: 2012/03/25 00:14:05
- Location: Belgrade-Serbia
- Status: offline
- Ribbons : 11
Re: PC World tests DX12 with Futuremark
2015/03/28 19:11:42
(permalink)
knightsilver I'm looking at AMD For my next build Seth89 and considering Crossfire down the road...
If you buy R9-290X CF you will get same performance as TITAN X SLI in DX12. AMD just try to explain us that, we need to hurry, when other figure out performance of R9-290X CF in DX12 than will be too late, All supplies will be cleaned from people ready to feel power of AMD in DX12. Mantle is nothing, that's only mistake, that's AMD's plan to Microsoft launch DX12 and to beat most power full NVIDIA there. Jesus 70-80% better performance per 1W, TITAN X vs AMD, PhysX, tessellation, TXAA, 3D...gaming world is our.
|
seth89
CLASSIFIED ULTRA Member
- Total Posts : 5290
- Reward points : 0
- Joined: 2007/11/13 11:26:18
- Status: offline
- Ribbons : 14
Re: PC World tests DX12 with Futuremark
2015/03/29 06:34:45
(permalink)
knightsilver I'm looking at AMD For my next build Seth89 and considering Crossfire down the road...
Wait for the 380 if you want a 290x, it will be a modified 290x. The 390x will be the card to own this year, so I would wait untill the summer. AMD stated that all GNC cards (the 290 and up) will be able to run DX12. So if you buy a card now the best option is the R9 295x2. The 295x2 is the fasted card (i think) and its water cooled. The 290s and up should all be water cooled imo. I notice scaling has been great with the my 290s. AMD drivers are finally at Nvidia quality too. In addtion to building a crossfire system you can be an early adaptor of freesync, so far the reviews are good. If you went the nvidia route the 980 with the built in water loop looks awesome and performs very well. ( but the 295x2 is only $100 more).
|
rjohnson11
EVGA Forum Moderator
- Total Posts : 102038
- Reward points : 0
- Joined: 2004/10/05 12:44:35
- Location: Netherlands
- Status: offline
- Ribbons : 84
Re: PC World tests DX12 with Futuremark
2015/03/29 11:02:23
(permalink)
I ran the API Overhead Feature Test using DX12 on the Windows 10 Preview. One thing I discovered is that the DX12 part of the test doesn't run if you have SLI enabled, (GTX 970 SSC in SLI) running WDDM driver 349.90. You have to disable SLI and then run the test.
|
seth89
CLASSIFIED ULTRA Member
- Total Posts : 5290
- Reward points : 0
- Joined: 2007/11/13 11:26:18
- Status: offline
- Ribbons : 14
Re: PC World tests DX12 with Futuremark
2015/03/29 11:16:37
(permalink)
rjohnson11 I ran the API Overhead Feature Test using DX12 on the Windows 10 Preview. One thing I discovered is that the DX12 part of the test doesn't run if you have SLI enabled, (GTX 970 SSC in SLI) running WDDM driver 349.90. You have to disable SLI and then run the test.
can you post your results? I will run the test with one R9 290 using mantle
|
seth89
CLASSIFIED ULTRA Member
- Total Posts : 5290
- Reward points : 0
- Joined: 2007/11/13 11:26:18
- Status: offline
- Ribbons : 14
Re: PC World tests DX12 with Futuremark
2015/03/29 11:47:22
(permalink)
single card at 1080p60 http://www.3dmark.com/aot/10416API OVERHEAD TEST 1.0 VALID RESULT AMD Radeon R9 290(1x) and Intel Core i7-3820 Processor DX11 Multi-threaded draw calls per second733912 DX11 Single-threaded draw calls per second814828 DX12 draw calls per second0.0 Mantle draw calls per second7345523
post edited by seth89 - 2015/03/29 11:50:35
|
rjohnson11
EVGA Forum Moderator
- Total Posts : 102038
- Reward points : 0
- Joined: 2004/10/05 12:44:35
- Location: Netherlands
- Status: offline
- Ribbons : 84
Re: PC World tests DX12 with Futuremark
2015/03/29 12:17:51
(permalink)
seth89
rjohnson11 I ran the API Overhead Feature Test using DX12 on the Windows 10 Preview. One thing I discovered is that the DX12 part of the test doesn't run if you have SLI enabled, (GTX 970 SSC in SLI) running WDDM driver 349.90. You have to disable SLI and then run the test.
can you post your results? I will run the test with one R9 290 using mantle
OK here it is:
|
seth89
CLASSIFIED ULTRA Member
- Total Posts : 5290
- Reward points : 0
- Joined: 2007/11/13 11:26:18
- Status: offline
- Ribbons : 14
Re: PC World tests DX12 with Futuremark
2015/03/30 04:40:00
(permalink)
Very cool, I will run mine at 720p too later today.
|
warrior10
FTW Member
- Total Posts : 1039
- Reward points : 0
- Joined: 2011/03/12 02:29:09
- Status: offline
- Ribbons : 0
Re: PC World tests DX12 with Futuremark
2015/03/30 17:38:33
(permalink)
Probably a dumb question, but I have and Intel i7 3770K which is a good CPU, but not the i7 4900 series. I also have 2 680s which are DX 11 i bet. Would Windows 10 with DX12 help my 2 680s?? Probably not, they probably have to be a 900 series, which i was planning to get, right? I wonder How the new AMD 390's will fare myself, as opposed to Nvidia. Had to ask.
post edited by warrior10 - 2015/03/30 17:42:45
My new https://www.gigabyte.com/...70-Gaming-K7-rev-10#kfHAF-X Coolermaster: 4 Fans New Rig:Ryzen 1700 CPU-H100i Liquid arctic Cooled. 16gigs Corsair Vengeance DDR4-2400 180Ti Hybrid Blu-ray drive SCSI CDRW-RW drive SCSI Black WD Black caviar 2-1TB's and 1 500gig WD Blue 1 TB 1 Sand disk SSD 240 for OS Corsair PSU 1020 SWTOR Mouse and Headset Razer CHROMA Keyboard Dell 1444p G-Sync Monitor @144htz to 160htz
|
Drazhar
CLASSIFIED Member
- Total Posts : 2370
- Reward points : 0
- Joined: 2008/12/07 02:14:51
- Status: offline
- Ribbons : 4
Re: PC World tests DX12 with Futuremark
2015/03/30 20:26:59
(permalink)
warrior10 Probably a dumb question, but I have and Intel i7 3770K which is a good CPU, but not the i7 4900 series. I also have 2 680s which are DX 11 i bet. Would Windows 10 with DX12 help my 2 680s?? Probably not, they probably have to be a 900 series, which i was planning to get, right? I wonder How the new AMD 390's will fare myself, as opposed to Nvidia. Had to ask.
I believe NVIDIA indicated that all of their stuff Fermi and beyond would support most of DX12, and AMD has stated all GCN will support it?
|
Vlada011
Omnipotent Enthusiast
- Total Posts : 10257
- Reward points : 0
- Joined: 2012/03/25 00:14:05
- Location: Belgrade-Serbia
- Status: offline
- Ribbons : 11
Re: PC World tests DX12 with Futuremark
2015/03/30 21:13:42
(permalink)
No matter on card we will need 6GB+ and cards at least strong as TITAN X overclocked for DX12 and Windows 10. I strong believe in GM200 6GB with 1150-1200MHz base clock on summer for arround 750$. AMD will force NVIDIA to launch that card and ask only 100$ more than R9-390X 8GB. Off course I would not wait such card but TITAN X is too expensive hardware for normal humans. Special in Europe, NVIDIA couldn't ask more than 1000$ because that would be 1500e in Europe, and no matter on that they have again more expensive card than TITAN Black. Again, every year tradition continue. Price in Europe on some place is so how for TITAN X that people in USA could buy 3 cards for same money as European 2. Need to pay 150$ only. 1300e = 1420$. Or one TITAN X for free on every 2.5.
post edited by Vlada011 - 2015/03/30 21:30:47
|
lehpron
Regular Guy
- Total Posts : 16254
- Reward points : 0
- Joined: 2006/05/18 15:22:06
- Status: offline
- Ribbons : 191
Re: PC World tests DX12 with Futuremark
2015/03/30 23:16:26
(permalink)
wmmills
Brad_Hawthorne
All of our tests were performed at 1280x720 resolution at Microsoft's recommendation. Eh? Well, they have to make sure xbone can run it too. lol Good catch on that brad!!! No, all 3DMark Performance presets were done at 1280x720, the Extreme presets were all done at 1920x1080. Reason is obvious: To inflate the CPU scores by increasing CPU overhead deliberately, something Intel has always benefited from. In other words, if this DX12 preview was done at a higher resolution, the score would depend more on the GPU than the CPU; you might as well have an older and slower CPU in there as the score wouldn't change, it would be almost completely GPU-bound. Why use an i7, go get a Celeron and pair it with a Titan-X, it wouldn't make a difference if you ran the benchmark at 4K or higher resolutions. The logic goes against PC enthusiast status quo that wants to continue to believe that they are getting the right components or the job, when they aren't. The whole point of DX12 and Mantle is to reduce CPU overhead, to relieve you from having to get the best CPU just to play a game at high resolutions and details. If anything DX12 and Mantle benefits consoles and IGP systems much more than those with dedicated cards, because they generally have slow CPUs anyway. But the marketing is aimed at those people that make the most noise, i.e. free advertising due to being excited about gains = fans with high-end hardware.
post edited by lehpron - 2015/03/30 23:21:18
|
rjohnson11
EVGA Forum Moderator
- Total Posts : 102038
- Reward points : 0
- Joined: 2004/10/05 12:44:35
- Location: Netherlands
- Status: offline
- Ribbons : 84
Re: PC World tests DX12 with Futuremark
2015/03/30 23:52:20
(permalink)
lehpron
wmmills
Brad_Hawthorne
All of our tests were performed at 1280x720 resolution at Microsoft's recommendation. Eh? Well, they have to make sure xbone can run it too. lol Good catch on that brad!!! No, all 3DMark Performance presets were done at 1280x720, the Extreme presets were all done at 1920x1080. Reason is obvious: To inflate the CPU scores by increasing CPU overhead deliberately, something Intel has always benefited from.
In other words, if this DX12 preview was done at a higher resolution, the score would depend more on the GPU than the CPU; you might as well have an older and slower CPU in there as the score wouldn't change, it would be almost completely GPU-bound. Why use an i7, go get a Celeron and pair it with a Titan-X, it wouldn't make a difference if you ran the benchmark at 4K or higher resolutions. The logic goes against PC enthusiast status quo that wants to continue to believe that they are getting the right components or the job, when they aren't.
The whole point of DX12 and Mantle is to reduce CPU overhead, to relieve you from having to get the best CPU just to play a game at high resolutions and details. If anything DX12 and Mantle benefits consoles and IGP systems much more than those with dedicated cards, because they generally have slow CPUs anyway. But the marketing is aimed at those people that make the most noise, i.e. free advertising due to being excited about gains = fans with high-end hardware.
In other words (if I interpret this correctly) how many times have we posted in the forums that a weak CPU would cause drop in framerates even with a strong GPU? DX12 is engineered to reverse that psychology.
|
lehpron
Regular Guy
- Total Posts : 16254
- Reward points : 0
- Joined: 2006/05/18 15:22:06
- Status: offline
- Ribbons : 191
Re: PC World tests DX12 with Futuremark
2015/03/31 00:57:56
(permalink)
rjohnson11 In other words (if I interpret this correctly) how many times have we posted in the forums that a weak CPU would cause drop in framerates even with a strong GPU? DX12 is engineered to reverse that psychology. In part sure, but many of us compensate by overclocking the CPU, to make the weak strong, and a better overclocker was always recommended. True litmus of whether DX12 works is whether anyone can tell the difference between and Intel or AMD CPU while playing certain games, historically, a GPU-bound game doesn't care while a CPU-bound game would care. I don't know the statistics, but I figure those playing MMOs justify their CPU choices as being more CPU-bound, even with the details turned up. CPU has to communicate with the ethernet controller and handle all traffic beyond just running the game and telling the graphics card what to draw. But most reviews never cover multiplayer gaming benchmarks, because there isn't a single test that represents all scenarios with different players. So in effect, any game benchmark is biased towards single-player, and that makes them more GPU-bound. We must beware of DX12 bias, they will not test CPU-bound in order to inflate GPU-bound scores. That being said, I'm curious what effect DX12 has on multi-player, I honestly don't know. Ultimately, multi-player is the reign of consoles.
post edited by lehpron - 2015/03/31 01:01:00
|
Drazhar
CLASSIFIED Member
- Total Posts : 2370
- Reward points : 0
- Joined: 2008/12/07 02:14:51
- Status: offline
- Ribbons : 4
Re: PC World tests DX12 with Futuremark
2015/03/31 08:20:48
(permalink)
At least they did a test where they disabled several cores and turned off HT and progressively expanded to show the influence of more cores?
|