PC What's your specs and stats?!?

Cadamier

Cadamier

New member
I've seen a few posts now and am thinking that this could help TFP and other players. So I'll start it off with all this:

DXDIAG.exe results via pastbin

I think this is PassMark9 with my results here.

Here is the first block of DXDiag

------------------

System Information

------------------

Time of this report: 6/14/2019, 09:09:21

Machine name: DESKTOP

Machine Id:

Operating System: Windows 10 Pro 64-bit (10.0, Build 17134) (17134.rs4_release.180410-1804)

Language: English (Regional Setting: English)

System Manufacturer: ASUS

System Model: All Series

BIOS: 2801 (type: UEFI)

Processor: Intel® Core i7-4790K CPU @ 4.00GHz (8 CPUs), ~4.0GHz

Memory: 32768MB RAM

Available OS Memory: 32706MB RAM

Page File: 14486MB used, 18220MB available

Windows Dir: C:\WINDOWS

DirectX Version: DirectX 12

DX Setup Parameters: Not found

User DPI Setting: 96 DPI (100 percent)

System DPI Setting: 96 DPI (100 percent)

DWM DPI Scaling: Disabled

Miracast: Available, with HDCP

Microsoft Graphics Hybrid: Not Supported

DxDiag Version: 10.00.17134.0001 64bit Unicode

Here are the pasted passmark9

General Information

PerformanceTest Version 9.0.1031

Baseline ID 1228402

Operating System Windows 10 Professional Edition build 17134 (64-bit)

Submitted Date 14th of June, 2019

Benchmark Scores

Passmark Rating 5,637

CPU Mark 12,062

2D Graphics Mark 965

3D Graphics Mark 12,305

Memory Mark 3,138

Disk Mark 4,331

Hardware

CPU Type Intel Core i7-4790K @ 4.00GHz

CPU Measured Speed 4.00 GHz [Turbo: 4.40 GHz]

Motherboard SABERTOOTH Z97 MARK 1/USB 3.1

Memory 32GB , G Skill Intl F3-2666C12-8GTXD

Video Card Tested GeForce GTX 980 Ti

Hard Drive Tested Samsung SSD 840 PRO Series (512 GB)

Weird thing with PassMark is that while

it said I got penalized %60 for not being able to run in 4k BOO!!!!!!!!! Not sure what that's about but I don't like it lol
Then I grabbed FPS Monitor and did a few videos to look at the lowest game setting and ultra.

On the lowest game settings 'in the open' I can get about an average of 140 reaching up towards 200.

Heading into a large town it drops towards 120. Fighting a wandering horde in the town drops it towards 100.

In the games Ultra settings trying to run the same route I get from 81 to 61... But then I saw the big shamway foods building and spun as fast as I could and it got me towards 65...

Then I decided to do it with a new game on the games ultra setting...

So about 92 FPS. Sure there are more factors than can be noted. From these stats and comparing them with recommended settings I might just go with the basic ultra settings from now on. Here's an idea... Use the Navezgane map - dedicated route!

I surely don't know everything there is about FPS, but this could possibly help TFP and other players as a comparison, maybe other things too; Donno. With FPS Monitor I can select what you want to display, I dropped everything except FPS info. Changed the font and size and moved it to the upper left of the videos. I donno, is this worth the effort?!?

 
Not exactly sure how the benchmarking will assist with anything, because we already know the client is CPU-bound.

I'm happy to provide info if it would help though, so here goes.

DxDiag

------------------System Information

------------------

Time of this report: 6/15/2019, 09:05:54

Machine name: SYLENTHUNDER

Machine Id: {053825C6-AFF5-4788-AB09-EA1A9240DB6F}

Operating System: Windows 10 Pro 64-bit (10.0, Build 17763) (17763.rs5_release.180914-1434)

Language: English (Regional Setting: English)

System Manufacturer: System manufacturer

System Model: System Product Name

BIOS: 4701 (type: UEFI)

Processor: Intel® Core i7-3930K CPU @ 3.20GHz (12 CPUs), ~3.2GHz

Memory: 16384MB RAM

Available OS Memory: 16322MB RAM

Page File: 8311MB used, 22347MB available

Windows Dir: C:\WINDOWS

DirectX Version: DirectX 12

DX Setup Parameters: Not found

User DPI Setting: 96 DPI (100 percent)

System DPI Setting: 96 DPI (100 percent)

DWM DPI Scaling: Disabled

Miracast: Available, with HDCP

Microsoft Graphics Hybrid: Not Supported

DxDiag Version: 10.00.17763.0001 64bit Unicode

Passmark

General InformationPerformanceTest Version 9.0.1031

Baseline ID 1228778

Operating System Windows 10 Professional Edition build 17763 (64-bit)

Submitted Date 15th of June, 2019

Benchmark Scores

Passmark Rating 4,498

CPU Mark 12,690

2D Graphics Mark 625

3D Graphics Mark 8,323

Memory Mark 2,323

Disk Mark 5,115

Hardware

CPU Type Intel Core i7-3930K @ 3.20GHz

CPU Measured Speed 4.04 GHz [Turbo: 4.30 GHz]

Motherboard SABERTOOTH X79

Memory 16GB , Crucial Technology BLS8G3D18ADS3.16FE

Video Card Tested GeForce GTX 1060

Hard Drive Tested Samsung SSD 850 EVO 250GB (250 GB)

I will note that I've just restarted the PC after installing the Win10 June update. I'll have to test again later after I've reset all my overclocks. I don't imagine it will change anything other than the memory rank really though.

I average 80FPS in-game pretty much everywhere, but it can drop into the mid 50's when loading a large building or during a horde.

 
What I've seen since A17 is that the CPU can fluctuate (I could be wrong but most of the time not going much over %60) using most if not all cores but GPU is almost always peaked out in usage and ram. Albeit I didn't look at this stuff before but from a player standpoint overall game play seems to be much better. I've also noticed that client ram usage seems better too. A16 it would gobble a lot, but now it starts around 6GB but increases some what slowly towards 9GB. If I remember correctly what I saw in A16 is that it would go up to 12GB and more if you had it. I also saw that if you quit a game and not shut down the client that ram usage would increase again (memory leak?) pushing it up further. But if you close the client and then start the game it would be back towards 6GB. Again I didn't look at all this with A16, just a cursory look... OH! This could also help people having FPS issues, if they have similar setups then they should get similar values. I haven't looked at all the numbers but it looks like I get from 70FPS going up towards 100FPS at these settings:

Preset: Custom

Anti-Aliasing:Yes

Texture Quality: Full

UMA Texture Quality: High

Reflection Quality: Low

Reflected Shadows: No

Shadows Distance: Off

Water Quality: Low

Water Particles: %11

View Distance: High

LOD Distance: 100

Tree Quality: Ultra

Grass Distance: Far

Bloom: Yes

Depth of Field: No

Motion Blur: Off

SSAQ: Yes

SS Reflections: No

Sun Shafts: No

Texture Streaming: Yes

OH! My setup:

OS = C: (SSD)

Game = D: (SSD)

Steam = E: (HD)

WEIRD! I thought I had my paging file set, but at the moment there isn't one...

 
....Preset: Custom

Anti-Aliasing:Yes

Texture Quality: Full

UMA Texture Quality: High

Reflection Quality: Low

Reflected Shadows: No

Shadows Distance: Off

Water Quality: Low

Water Particles: %11

View Distance: High

LOD Distance: 100

Tree Quality: Ultra

Grass Distance: Far

Bloom: Yes

Depth of Field: No

Motion Blur: Off

SSAQ: Yes

SS Reflections: No

Sun Shafts: No

Texture Streaming: Yes

OH! My setup:

OS = C: (SSD)

Game = D: (SSD)

Steam = E: (HD)

WEIRD! I thought I had my paging file set, but at the moment there isn't one...
Click to expand...
Just for comparison, my setup is...

Preset: Custom

Anti-Aliasing:Yes

Texture Quality: Full

UMA Texture Quality: High

Reflection Quality: Low

Reflected Shadows: No

Shadows Distance: Near

Water Quality: High

Water Particles: %50

View Distance: High

LOD Distance: 100

Tree Quality: Ultra

Grass Distance: Near

Bloom: Yes

Depth of Field: No

Motion Blur: Off

SSAQ: Yes

SS Reflections: No

Sun Shafts: Yes

Texture Streaming: Yes

For my drives

OS = C: SSD

Game = D: SSD

Steam and save data = E: HDD's in RAID-0

Both SSD's are 250GB Samsung Evo 850's, and the HDD's are Hitachi Travelstar 7200RPM laptop drives.

Paging file is auto, because that is the most efficient atm.

 
So, with an optimized OC that puts my core clock speed at 4.25GHz, here's the Passmark scores.

Benchmark Scores

Passmark Rating 4,655 (103.7%)

CPU Mark 13,262 (104.5%)

2D Graphics Mark 636 (101.7%)

3D Graphics Mark 8,575 (103%)

Memory Mark 2,546 (109.6%)

Disk Mark 5,011 (97.96%)

Hardware

CPU Type Intel Core i7-3930K @ 3.20GHz

CPU Measured Speed 4.25 GHz [Turbo: 4.80 GHz]

Motherboard SABERTOOTH X79

Memory 16GB , Crucial Technology BLS8G3D18ADS3.16FE

Video Card Tested GeForce GTX 1060

Hard Drive Tested Samsung SSD 850 EVO 250GB (250 GB)

Now with those changes, my new average FPS is 110, and the average drop in towns with large buildings is in the mid 80's.

All I did was bump the CPU clock up a little.

The findings are very similar to a test I did back in a16.

Low frames with 1080 TI - less than 20

 
Last edited by a moderator:
So, with an optimized OC that puts my core clock speed at 4.25GHz, here's the Passmark scores.
Benchmark Scores

Passmark Rating 4,655 (103.7%)

CPU Mark 13,262 (104.5%)

2D Graphics Mark 636 (101.7%)

3D Graphics Mark 8,575 (103%)

Memory Mark 2,546 (109.6%)

Disk Mark 5,011 (97.96%)

Hardware

CPU Type Intel Core i7-3930K @ 3.20GHz

CPU Measured Speed 4.25 GHz [Turbo: 4.80 GHz]

Motherboard SABERTOOTH X79

Memory 16GB , Crucial Technology BLS8G3D18ADS3.16FE

Video Card Tested GeForce GTX 1060

Hard Drive Tested Samsung SSD 850 EVO 250GB (250 GB)

Now with those changes, my new average FPS is 110, and the average drop in towns with large buildings is in the mid 80's.

All I did was bump the CPU clock up a little.

The findings are very similar to a test I did back in a16.

Low frames with 1080 TI - less than 20
Click to expand...
That's right about where I'm at. Even on Horde Night

.
 
I've got an i7 3.5GHz, 32GB RAM, NVIDIA GTX 1080.

The game has trouble going over 45FPS unless I'm staring straight into the sky or the ground. I've only ever gotten it to go over 60FPS by switching it to lowest possible settings.

 
Not exactly sure how the benchmarking will assist with anything, because we already know the client is CPU-bound.
I'm happy to provide info if it would help though, so here goes.

DxDiag

------------------System Information

------------------

Time of this report: 6/15/2019, 09:05:54

Machine name: SYLENTHUNDER

Machine Id: {053825C6-AFF5-4788-AB09-EA1A9240DB6F}

Operating System: Windows 10 Pro 64-bit (10.0, Build 17763) (17763.rs5_release.180914-1434)

Language: English (Regional Setting: English)

System Manufacturer: System manufacturer

System Model: System Product Name

BIOS: 4701 (type: UEFI)

Processor: Intel® Core i7-3930K CPU @ 3.20GHz (12 CPUs), ~3.2GHz

Memory: 16384MB RAM

Available OS Memory: 16322MB RAM

Page File: 8311MB used, 22347MB available

Windows Dir: C:\WINDOWS

DirectX Version: DirectX 12

DX Setup Parameters: Not found

User DPI Setting: 96 DPI (100 percent)

System DPI Setting: 96 DPI (100 percent)

DWM DPI Scaling: Disabled

Miracast: Available, with HDCP

Microsoft Graphics Hybrid: Not Supported

DxDiag Version: 10.00.17763.0001 64bit Unicode

Passmark

General InformationPerformanceTest Version 9.0.1031

Baseline ID 1228778

Operating System Windows 10 Professional Edition build 17763 (64-bit)

Submitted Date 15th of June, 2019

Benchmark Scores

Passmark Rating 4,498

CPU Mark 12,690

2D Graphics Mark 625

3D Graphics Mark 8,323

Memory Mark 2,323

Disk Mark 5,115

Hardware

CPU Type Intel Core i7-3930K @ 3.20GHz

CPU Measured Speed 4.04 GHz [Turbo: 4.30 GHz]

Motherboard SABERTOOTH X79

Memory 16GB , Crucial Technology BLS8G3D18ADS3.16FE

Video Card Tested GeForce GTX 1060

Hard Drive Tested Samsung SSD 850 EVO 250GB (250 GB)

I will note that I've just restarted the PC after installing the Win10 June update. I'll have to test again later after I've reset all my overclocks. I don't imagine it will change anything other than the memory rank really though.

I average 80FPS in-game pretty much everywhere, but it can drop into the mid 50's when loading a large building or during a horde.
Click to expand...
I miss my Sabertooth x79 with my Intel Core i7-3960X with 16GB DDR3 1600.

Now I have:

Intel i7 860 @ 3.3ghz turbo off - I like that setting, I figure turbo going over the stock turbo would be more strain than a modest static 3.3ghz overclock.

Biostar T5XE CFX-SLI

16Gb DDR3 1333 (PC3 10666)

GTX 1080 stock clocked

OS SSD

Steam Games SSD

16 more hard drives, being a mix SSD and Magnetic including 10 3TB drives, most are USB, I had Esata but my controller card is not worth using Esata due to cable length and data loss on power outage (really). I have a boatload of ripped Movies and TV Shows and a Game Storage Drive and a Regular Storage Drive. I also Have drive dedicated to drive images, a VMWare Workstation 12.5 drive for DOS RPG games, A Retro Games drive With Amiga, C64 and DOS games and some drives I use to record and remove commercials from my Sling TV shows.

I don't have passmark 9, the last benchmark I owned was 3DMark 11 or 3DMark Vantage whichever one came out after the other, I don't remember.

I get bad frame rate spikes in the bigger city and this is probably due to my own settings that could be turned down especially the draw distance I have now.

 
Last edited by a moderator:
AMD Ryzen 7 2700X Eight-Core Processor

nVidia 1060 with proprietary drivers

32GB ram

MSI MS-7A33

3 SSDs, no physical hard drive.

Slackware 14.2+ 4.19.68 kernel

I think I'm running without a swap file - with Linux you can do this if you know what you are doing.

Switching from physical hard drives to SSDs made no difference. Doubling the memory only helped when loading 16K worlds, as those are memory pigs and will happily consume more than 16GB. For Navezgane, for example, doubling memory made no difference. Neither did running off of SSDs.

cpu usage is not noticeably high. What made a *huge difference was upgrading from my old trusty nVidia 660 to a 1060. And even with that, performance in areas with a lot of dynamic lights, with shadows enabled, is very poor. Turn off all shadows, and I get 80fps+ with ease. Even in cities, performance is good - so long as shadows are all turned off.

With my old 660, performance overall was decent but I had to use half size textures. With the 1060, I can use full size textures.

 
AMD Ryzen 7 2700X Eight-Core ProcessornVidia 1060 with proprietary drivers

32GB ram

MSI MS-7A33

3 SSDs, no physical hard drive.

Slackware 14.2+ 4.19.68 kernel

I think I'm running without a swap file - with Linux you can do this if you know what you are doing.

Switching from physical hard drives to SSDs made no difference. Doubling the memory only helped when loading 16K worlds, as those are memory pigs and will happily consume more than 16GB. For Navezgane, for example, doubling memory made no difference. Neither did running off of SSDs.

cpu usage is not noticeably high. What made a *huge difference was upgrading from my old trusty nVidia 660 to a 1060. And even with that, performance in areas with a lot of dynamic lights, with shadows enabled, is very poor. Turn off all shadows, and I get 80fps+ with ease. Even in cities, performance is good - so long as shadows are all turned off.

With my old 660, performance overall was decent but I had to use half size textures. With the 1060, I can use full size textures.
Click to expand...
I remember in A16 setting shadows from near to off was the biggest performance boost I could do with a lot of spotlights at my base.

I made a couple of videos for this demonstration:

I should have used fraps. you can barely see the FPS meter.

I really notice the performance boost and the much prettier 3D engine the devs put in for A17, but man in the city it still just taxes all my resources and I probably get 20 FPS. I have all of the post processing turned off except bloom. I should probably turn off shadows if I plan to loot in the big cities.

As far as your AMD Ryzen 7 2700X Eight-Core Processor, may I ask why you did not go with Intel for your gaming-rig. I had a K6-2 350 (garbage), Duron 800 (not bad), XP 2700 (good), Athlon FX 4000 (very good) and an Athlon X2 4800 (very good). After I built a PC with the Core 2 Quad I never thought of AMD again. I also never had any ATI or AMD Video cards just NVIDIA from TNT 2 up to my GTX 1080 and a very long list in between.

 
Last edited by a moderator:
A-17 is the same. If you have a lot of torches or other dynamic lights close together, you pretty much have to turn shadows off.

 
Is there any other large performance killing setting. I know that my draw distance and LOD distance could be lower. I turned up my textures to high seeing how my GTX 1080 has 8GB Vram so having it set to medium must have been a default setting I did not pay attention to. The textures now have really nice bump mapping so I am glad Green Dragon had that setting in his post or I would have never seen the difference.

 
Last edited by a moderator:
<snip>...

As far as your AMD Ryzen 7 2700X Eight-Core Processor, may I ask why you did not go with Intel for your gaming-rig. I had a K6-2 350 (garbage), Duron 800 (not bad), XP 2700 (good), Athlon FX 4000 (very good) and an Athlon X2 4800 (very good). After I built a PC with the Core 2 Quad I never thought of AMD again. I also never had any ATI or AMD Video cards just NVIDIA from TNT 2 up to my GTX 1080 and a very long list in between.
Click to expand...
I got my first PC about 1989 or so, it had an NEC V20 2.3MHz cpu, 512KB ram. :smile-new: . 30 years ago. I've used both Intel and AMD over the course of the last three decades, but I found my self favoring AMD for cpus. I ran servers on socket-A boards for years as the Athlon was running circles around the Pentium. I think I decided to leave Intel behind toward the end of the Pentium life cycle, when they were pushing the pipeline out to 30+ cycles just to get a higher clock speed to boast about. It was a pig with lipstick.

I went back to school about ten years ago, and one of my classmates used to work for Intel. While there, he was tasked to find out why the Athlon was faster than the Pentium. His report? The pentium was a kludge, internally, with too many long traces running back and forth, much longer than then what was inside the Athlon. That is thy the pipeline for the pentium was so long, the traces were long. When you ramp up clock speed, which Intel did, it takes more clock cycles to get from here to there and that means a longer pipeline. The Athlon didn't have the clock speed, but the pipeline was so much shorter it could get more done with a slower clock than the pentium could with a higher clock. Anyhow I digress...

My old Phenoms, followed by a FX4000, followed by several Ryzens do more than I need. I never looked back towards intel.

My ATI Rage Fury was garbage. I followed it with an nVidia TNT2-Ultra which was most excellent but back then nVidia was having some serious driver problems, so I eventually switched back to ATI cards for a few years. Somewhere along the line AMD dropped driver support for all their older cards on newer kernels, and I found myself with a couple thousand dollars of AMD/ATI video cards sitting on my desk that I couldn't use. Back to nVidia, this time with a 660 that gave excellent performance and lasted me for several years. Now I have a 1060 that I'm very happy with.

So yeah, today it's AMD cpus and nVidia video cards. For now.

 
My first PC was a C64 in 1988, running CRPG like wasteland, Might and Magic and Ultima 3. My first build was a Pentium 4 and it, or my build, or both had such overheating problems that it lasted 20 seconds and shut down, next I went with AMD Athlon 64 4000+ and I loved it (I still have the CPU in my computer toolbox), then to the Athlon X2 4800 and loved that. So after the Pentium 4 Disaster (which was fixed with a massive CPU Fan-Heatsink) I used AMD Twice, until the Core 2 Quad.

When ATI came out with the Radeon Series they finally had something that NVIDIA may have been worried about, but like you said the ATI rage and similar class GPUs were not known for their performance. Even still since the TNT 2 I have been a NVIDIA fanboy and never really thought about ATI or AMD GPUs

 
Last edited by a moderator:
My first was a Timex Sinclair TS1000, purchased .. 1983? Albertsons was dumping them for $20. I couldn't afford a C=64 at the time, and didn't want the Vic-20.

I actually cut my teeth on a Sperry Univac V76 mini series, those were the classic 6 foot tall monsters with blinking lights, right out of a sci fi movie...<sigh>...oh to be young again....

 
My first was a Timex Sinclair TS1000, purchased .. 1983? Albertsons was dumping them for $20. I couldn't afford a C=64 at the time, and didn't want the Vic-20.
I actually cut my teeth on a Sperry Univac V76 mini series, those were the classic 6 foot tall monsters with blinking lights, right out of a sci fi movie...<sigh>...oh to be young again....
Click to expand...
Back in the early 90s when I had my Amiga I would have done anything for a SGI workstation and some 3D apps like Wavefront or Softimage. I use to read magazines about them and have more info mailed to me just because I was bored with my Amiga at the time and I wanted to use Sculpt Animate 4D but there was no way I could afford it driving a taxi back then. I had brochures on VAX mini computers, Evens and Sutherland Workstations, HP workstations, IBM workstations I even had a brochure on a cray supercomputer that was submerged in water or oil.

I remember back in the early to mid 80s when I knew a guy who had a C64 and I thought it was the dumbest thing in the world, just typing test and seeing nothing but text. My opinion changed in 1988 when I got bored playing Defender on the Atari 2600 and my electronic chess set cheated on me - yes it cheated. I bought my C64 used and got a boatload of games and another lot of CRPGs and I was hooked.

I bought my Amiga 500, 501 expansion and 1084 stereo monitor brand new a little later, at about $1,400.00. I wanted a 386 back then but they ran about $3,000.00.

Oh to be young again is right, I am 51 going on 15 in November as they tell me.

I would love to build a Workstation / Gamig-Rig with a 18 core i9, 64GB DDR4 (3200 or higher for overclocking and tightening the timings) and with a pair of GTX 1080 ti SLI or RTX 2080 ti SLI (the later for fast GPU Rendering). I would probably overclock to about 3.8 - 4.0 Ghz and turn off turbo. I really do not like turbo on the i series CPUs I would rather have a static overclock than risk turbo going well beyond 4Ghz.

 
Last edited by a moderator:
Everything on and at high/ultra gets you under 60FPS regulary with 3700X/2070Super/32GB Ram oc´ed at 3600Mhz CL16, at a resolution of 2560x1440p

Can´t wait for beta and optimizing.

@bobrpggamer you wouldn´t get better with that system you would love to build. The 18core i9 isn´t really for gaming either.

 
Everything on and at high/ultra gets you under 60FPS regulary with 3700X/2070Super/32GB Ram oc´ed at 3600Mhz CL16, at a resolution of 2560x1440p
Can´t wait for beta and optimizing.

@bobrpggamer you wouldn´t get better with that system you would love to build. The 18core i9 isn´t really for gaming either.
Click to expand...
Well for Gaming / Workstation with 3D content creation software when I start to work on some of my designs and ideas, 18 cores will be great to have for rendering and displaying large scenes. Its just a dream now and with the price for commercial software use, the Gaming / Workstation would be the cheapest part of dream.

 
Passmark 5114 90% Lost 47% overall score cos it couldnt do its full 4k only 3/4 struggle to get 20fps on 7 days in medium lol & since the Alpha 18 update its crashed 20+ times to desktop, GFX has crashed about the same....

CPU Type AMD Ryzen 7 2700X

CPU Measured Speed 4.10 GHz

Motherboard X470 GAMING PRO CARBON (MS-7B78)

Memory 32GB , Corsair CMK32GX4M2B3000C15

Video Card Tested Radeon RX 580

Hard Drive Tested Samsung SSD 970 EVO 1TB (1 TB)

 
You must log in or register to reply here.
Back
Top