[Various] Ashes of the Singularity DX12 Benchmarks - Page 117 - Overclock.net - An Overclocking Community

Forum Jump: 

[Various] Ashes of the Singularity DX12 Benchmarks

 
Thread Tools
post #1161 of 2682 (permalink) Old 08-28-2015, 03:07 PM
New to Overclock.net
 
Mahigan's Avatar
 
Join Date: Aug 2015
Location: Ottawa, Canada
Posts: 1,749
Rep: 874 (Unique: 233)
Quote:
Originally Posted by PontiacGTX View Post

PCIE 2.0 is fine fpr a single gpu


Factor in the amount of draw calls increasing by the full use of Multi-Core CPUs sending traffic over the PCIe 2.0 bus in DX12.

Previous tests were conducted using DX11

"Once you eliminate the impossible, whatever remains, no matter how improbable, must be the truth." - Arthur Conan Doyle (Sherlock Holmes)
Mahigan is offline  
Sponsored Links
Advertisement
 
post #1162 of 2682 (permalink) Old 08-28-2015, 03:10 PM
New to Overclock.net
 
Noufel's Avatar
 
Join Date: Apr 2012
Location: Constantine, Algeria
Posts: 1,548
Rep: 54 (Unique: 40)
Quote:
Originally Posted by Mahigan View Post

Quote:
Originally Posted by PontiacGTX View Post

PCIE 2.0 is fine fpr a single gpu


Factor in the amount of draw calls increasing by the full use of Multi-Core CPUs sending traffic over the PCIe 2.0 bus though.

Previous tests were conducted using DX11
So with sli or cfx configuration on z77 and above with only 16 pcie 3.0 lanes and 8 lane for gpu that equals pcie 2.0 16 lanes we will see similar bottleneck ???
Noufel is offline  
post #1163 of 2682 (permalink) Old 08-28-2015, 03:14 PM
New to Overclock.net
 
Forceman's Avatar
 
Join Date: Sep 2012
Posts: 10,869
Rep: 794 (Unique: 514)
Quote:
Originally Posted by Mahigan View Post

Factor in the amount of draw calls increasing by the full use of Multi-Core CPUs sending traffic over the PCIe 2.0 bus in DX12.

Previous tests were conducted using DX11

Should be easy to test since most Intel boards (edit: or at least the Gigabyte ones) let you change the PCIe speed. Switch it to gen 2 and test to see the difference - if draw calls are that big an impact even the Futuremark API bench should show it. I'd check it but I'm off to work.

Forceman's Law: Any AMD/Nvidia GPU thread, no matter what the topic, will eventually include a post referencing the GTX 970.
Forceman is offline  
Sponsored Links
Advertisement
 
post #1164 of 2682 (permalink) Old 08-28-2015, 03:28 PM
New to Overclock.net
 
Mahigan's Avatar
 
Join Date: Aug 2015
Location: Ottawa, Canada
Posts: 1,749
Rep: 874 (Unique: 233)
Quote:
Originally Posted by Noufel View Post

I don't think it will help there are several latency problems with the PCIE 3.0 controler of the sabertooth gen3.

Hmm... well that doesn't help.

if I was in Canada I'd test it on my Intel rig with PCIe 2.0 vs PCIe 3.0.

"Once you eliminate the impossible, whatever remains, no matter how improbable, must be the truth." - Arthur Conan Doyle (Sherlock Holmes)
Mahigan is offline  
post #1165 of 2682 (permalink) Old 08-28-2015, 03:35 PM
New to Overclock.net
 
Themisseble's Avatar
 
Join Date: Oct 2013
Posts: 2,001
Rep: 38 (Unique: 29)
Quote:
Originally Posted by Mahigan View Post

Factor in the amount of draw calls increasing by the full use of Multi-Core CPUs sending traffic over the PCIe 2.0 bus in DX12.

Previous tests were conducted using DX11

- OneB1T wrote that Warning: Spoiler! (Click to show)
Here is something:
Saw phenom x4/x6 doing better than FX CPUs - PCIE2, worst memory controller...
see this
[email protected]

==Shot long shot 3 ==================================
Total Time: 5.008700
Avg Framerate : 32.314194 ms (30.946156 FPS)
Weighted Framerate : 32.396713 ms (30.867331 FPS)
CPU frame rate (estimated framerate if not GPU bound): 24.420563 ms (40.949097 FPS)
Percent GPU Bound: 99.271950%
Driver throughput (Batches per ms): 3960.520752
Average Batches per frame: 47504.597656


==Shot high vista ==================================
Total Time: 4.970572
Avg Framerate : 35.759510 ms (27.964590 FPS)
Weighted Framerate : 35.872990 ms (27.876127 FPS)
CPU frame rate (estimated framerate if not GPU bound): 26.463301 ms (37.788181 FPS)
Percent GPU Bound: 99.365089%
Driver throughput (Batches per ms): 3962.454102
Average Batches per frame: 50500.582031

[email protected]

==Shot long shot 3 ==================================
Total Time: 4.985974
Avg Framerate : 33.462917 ms (29.883827 FPS)
Weighted Framerate : 33.534695 ms (29.819862 FPS)
CPU frame rate (estimated framerate if not GPU bound): 29.043478 ms (34.431137 FPS)
Percent GPU Bound: 93.482292%
Driver throughput (Batches per ms): 5927.758789
Average Batches per frame: 47757.062500

==Shot high vista ==================================
Total Time: 4.998882
Avg Framerate : 33.776230 ms (29.606619 FPS)
Weighted Framerate : 33.864346 ms (29.529583 FPS)
CPU frame rate (estimated framerate if not GPU bound): 28.254892 ms (35.392101 FPS)
Percent GPU Bound: 98.633514%
Driver throughput (Batches per ms): 6525.858398
Average Batches per frame: 49477.691406

see the batches per ms figure 4k for i5 vs 6k for FX

Diver throughput? 50% more for FX? WHY?
SO i7 5960X same 4K batches per ms... also on old AMD CPU (not FX line) was benchmark which showed 4K batches per ms.
Themisseble is offline  
post #1166 of 2682 (permalink) Old 08-28-2015, 03:42 PM
Hey I get one of these!
 
KyadCK's Avatar
 
Join Date: Aug 2011
Location: Chicago
Posts: 7,209
Rep: 301 (Unique: 212)
Quote:
Originally Posted by Mahigan View Post

Quote:
Originally Posted by Themisseble View Post

@Mahigan

Anything new about CPU performance?

What if it wasn't the AMD FX CPU. What if it was the PCI Express 2.0 bus being the bottleneck?

If someone has an Asus Sabertooth 990fxgen3 r2 motherboard... I'd be curious to see their results...

I say this because the CPU Frame rate is 42.8FPS here using an AMD FX-9370 @4.7 and an R9 290 @1050 core and 1350 memory. We know the R9 can do better. Why such a low frame rate.

System Specs:
AMD FX-9370 @4.7GHz
ASUS Sabertooth 990FX Rev 1.0
CMZ8GX3M2A1866C9 8GB Corsair Ram 9-10-9-27
Sapphire R9 290
Seagate 2TB HD
Seagate 3TB HD
S27A950D 27" 120 Hz 3D Samsung

Result:



Now we know an R9 290x can do 48.9FPS at these same settings:



Since there's so much information being transferred over the PCIe bus... 2.0 might be the bottleneck.

HyperTransport. Add it to your list of things to read up on, even if it will go away with Zen. It's incredibly important to the current AM chipsets, PCI-e isn't the only thing standing in the way.

Forge
(18 items)
Forge-LT
(7 items)
CPU
AMD Threadripper 1950X
Motherboard
Gigabyte X399 Designare
GPU
EVGA 1080ti SC2 Hybrid
GPU
EVGA 1080ti SC2 Hybrid
RAM
32GB G.Skill TridentZ RGB (4x8GB 3200Mhz 14-14-14)
Hard Drive
Intel 900P 480GB
Hard Drive
Samsung 950 Pro 512GB
Power Supply
Corsair AX1200
Cooling
EK Predator 240
Case
Corsair Graphite 780T
Operating System
Windows 10 Enterprise x64
Monitor
2x Acer XR341CK
Keyboard
Corsair Vengeance K70 RGB
Mouse
Corsair Vengeance M65 RGB
Audio
Sennheiser HD700
Audio
Sound Blaster AE-5
Audio
Audio Technica AT4040
Audio
30ART Mic Tube Amp
CPU
i7-4720HQ
Motherboard
UX501JW-UB71T
GPU
GTX 960m
RAM
16GB 1600 9-9-9-27
Hard Drive
512GB PCI-e SSD
Operating System
Windows 10 Pro
Monitor
4k IPS
▲ hide details ▲
KyadCK is offline  
post #1167 of 2682 (permalink) Old 08-28-2015, 03:44 PM
new to OCN?
 
PontiacGTX's Avatar
 
Join Date: Aug 2011
Location: Venezuela
Posts: 26,368
Rep: 1536 (Unique: 924)
Quote:
Originally Posted by Mahigan View Post

Factor in the amount of draw calls increasing by the full use of Multi-Core CPUs sending traffic over the PCIe 2.0 bus in DX12.

Previous tests were conducted using DX11
if thats the issue mantle running on bf4 doesnt increases the draw calls aswell? and people gets noticeable improvement even on PCIE2.0
PontiacGTX is offline  
post #1168 of 2682 (permalink) Old 08-28-2015, 03:50 PM
New to Overclock.net
 
sugarhell's Avatar
 
Join Date: Apr 2012
Location: 38.051289,23.709791
Posts: 9,325
Rep: 563 (Unique: 286)
Quote:
Originally Posted by PontiacGTX View Post

if thats the issue mantle running on bf4 doesnt increases the draw calls aswell? and people gets noticeable improvement even on PCIE2.0
it would be interesting to Test ocing HT on that bench biggrin.gif

No it doesnt increase the draw calls. How can an API increase the draw calls?

Quote:
Originally Posted by BiG StroOnZ
Warning: Spoiler! (Click to show)
No I said if Vega is clocked @ 1600MHz out of the box, I will eat my shoe on Twitch.tv.

Sound good? wink.gif
Warning: Spoiler! (Click to show)
Quote:
Originally Posted by Juicin View Post
Not in games, which is what 99.9999 percent of us care about

And unless you own a business, you get paid for your time. So your compute power just makes more work for you not less

Ryzen is a joke to the vast majority of the market who would consider buying their product.
sugarhell is offline  
post #1169 of 2682 (permalink) Old 08-28-2015, 03:51 PM
New to Overclock.net
 
Mahigan's Avatar
 
Join Date: Aug 2015
Location: Ottawa, Canada
Posts: 1,749
Rep: 874 (Unique: 233)
Quote:
Originally Posted by PontiacGTX View Post

if thats the issue mantle running on bf4 doesnt increases the draw calls aswell? and people gets noticeable improvement even on PCIE2.0
it would be interesting to Test ocing HT on that bench biggrin.gif

Check it out smile.gif


Specs:
AMD FX-9370 @4.7GHz
ASUS Sabertooth 990FX Rev 1.0
CMZ8GX3M2A1866C9 8GB Corsair Ram 9-10-9-27
Sapphire R9 290 @ 1050MHz
Seagate 2TB HD
Seagate 3TB HD
S27A950D 27" 120 Hz 3D Samsung


Ars Technica Settings Result in this:


pcgameshardware Result in this:



PCGamesHardware

1. Did not perform the test in an objective manner.
or
2. Didn't know what they were doing.

"Once you eliminate the impossible, whatever remains, no matter how improbable, must be the truth." - Arthur Conan Doyle (Sherlock Holmes)
Mahigan is offline  
post #1170 of 2682 (permalink) Old 08-28-2015, 03:54 PM
New to Overclock.net
 
sugarhell's Avatar
 
Join Date: Apr 2012
Location: 38.051289,23.709791
Posts: 9,325
Rep: 563 (Unique: 286)
Who the heck use 12 TAA duration? This is so blur. Its impractical and none use this huge amount

Quote:
Originally Posted by BiG StroOnZ
Warning: Spoiler! (Click to show)
No I said if Vega is clocked @ 1600MHz out of the box, I will eat my shoe on Twitch.tv.

Sound good? wink.gif
Warning: Spoiler! (Click to show)
Quote:
Originally Posted by Juicin View Post
Not in games, which is what 99.9999 percent of us care about

And unless you own a business, you get paid for your time. So your compute power just makes more work for you not less

Ryzen is a joke to the vast majority of the market who would consider buying their product.
sugarhell is offline  
Closed Thread

Quick Reply
Message:
Options

Register Now

In order to be able to post messages on the Overclock.net - An Overclocking Community forums, you must first register.
Please enter your desired user name, your email address and other required details in the form below.
User Name:
If you do not want to register, fill this field only and the name will be used as user name for your post.
Password
Please enter a password for your user account. Note that passwords are case-sensitive.
Password:
Confirm Password:
Email Address
Please enter a valid email address for yourself.
Email Address:

Log-in



Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may post new threads
You may post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off