Jump to content

Recommended Posts

Posted

Hey All,

Like about 99% of the people who game and own a PC I am thinking about a GPU upgrade.  Luckily, in some ways, the scalped prices have slowed me down and given me more time to consider what I really need.  However, I do not fully understand how the GPU and DCS with multi-monitors function. 

I was hoping to get educated here as I know that allot of you have much more experience than I do.

This is my current display set-up, and frankly it is all that I physically have room for on my desk (and space that I have).  I currently only run two MFCD's on screen 2.  I run two displays because my eyes are kind of bad and I can not easily read the MFCD's on the main monitor.

kneeboard 02.gif

Screen 1 is running at 2560 x 1080 and screen 2 is running at 1920 x 1080.  I have the DCS settings set up as high as I could so that my system will run stable and minimize stutter.  I do limit the GPU to 40 fps (might dip below that near the ground etc...).  I watch the GPU utilization via MSI Afterburner in percent and use that number for bench marking.  The card mostly runs in the 60-80% range and around 40-50C.  My GPU runs a whole lot cooler at 40 fps vs unrestricted.

My screen 1 is capable of 3440 x 1440 (144 hz), but with my current GPU it is a "no deal".  So I could go to a higher resolution on the main screen if I wanted to (or could) do so.

Notice that my GPU currently is a RTX 2060 at 6MB.  MSI Afterburner (also Task Manager) reports that is at or very close to maximum memory utilized when in flight.  I do not understand how DCS stacks memory usage up on the card.  Or, how much memory the card really should have and not be utilizing shared GPU memory (which mine does do at times). 

Questions:

  • Is the GPU running both screens in 3D mode?
  • Does the cockpit run in 3D mode?
  • How much memory should the card load at 2560 x 2180?
  • How much memory should the card have to run at 3440 x 2520 and not use shared memory?

Thanks in advance.

Caldera

Happy 2022!

 

Posted
4 hours ago, Caldera said:

Hey All,

Like about 99% of the people who game and own a PC I am thinking about a GPU upgrade.  Luckily, in some ways, the scalped prices have slowed me down and given me more time to consider what I really need.  However, I do not fully understand how the GPU and DCS with multi-monitors function. 

I was hoping to get educated here as I know that allot of you have much more experience than I do.

This is my current display set-up, and frankly it is all that I physically have room for on my desk (and space that I have).  I currently only run two MFCD's on screen 2.  I run two displays because my eyes are kind of bad and I can not easily read the MFCD's on the main monitor.

kneeboard 02.gif

Screen 1 is running at 2560 x 1080 and screen 2 is running at 1920 x 1080.  I have the DCS settings set up as high as I could so that my system will run stable and minimize stutter.  I do limit the GPU to 40 fps (might dip below that near the ground etc...).  I watch the GPU utilization via MSI Afterburner in percent and use that number for bench marking.  The card mostly runs in the 60-80% range and around 40-50C.  My GPU runs a whole lot cooler at 40 fps vs unrestricted.

My screen 1 is capable of 3440 x 1440 (144 hz), but with my current GPU it is a "no deal".  So I could go to a higher resolution on the main screen if I wanted to (or could) do so.

Notice that my GPU currently is a RTX 2060 at 6MB.  MSI Afterburner (also Task Manager) reports that is at or very close to maximum memory utilized when in flight.  I do not understand how DCS stacks memory usage up on the card.  Or, how much memory the card really should have and not be utilizing shared GPU memory (which mine does do at times). 

Questions:

  • Is the GPU running both screens in 3D mode?
  • Does the cockpit run in 3D mode?
  • How much memory should the card load at 2560 x 2180?
  • How much memory should the card have to run at 3440 x 2520 and not use shared memory?

Thanks in advance.

Caldera

Happy 2022!

 

Can't answer your first two questions but as far as VRAM, DCS will schedule everything you got. 

Modules: F-14A/B | F-15C | F-16C | F/A-18C | SU-33 | Spitfire Mk IX | AH-64D | UH-1 | Super Carrier | Combined Arms | Persian Gulf | Syria | NTTR

Setup: VKB Gunfighter Mk.III F-14 CE HOTAS | Thrustmaster TWCS Throttle | MFG Crosswind V3 | Custom switch panel | Tek Creations F14 Display Panel | Custom F14 Left Vertical Console | Custom IR Tracker | Custom butt kicker

PC: i7 11700K | 64GB G-Skill DDR4 3600MHz | EVGA GeForce RTX 3080Ti FTW3 | DCS dedicated 2TB M.2 NVMe SSD | 3440x1440 144hz 34" ultrawide

Posted (edited)

Hector,

Thanks for the reply!

I got to looking at MSI Afterburner a bit more.  It has at least two memory usage variables that can be monitored.  One is called RAM usage and the other is called RAM usage /process, for both DRAM and VRAM.

So I am thinking that it would be it safe to assume that:

  • Ram usage is the ram allocated for use by the application
  • RAM usage /process is the ram that is currently in use by the application

For example on my system right now using MSI Afterburner running DCS:

  • GPU      Temp   Usage
  • VRAM    Usage  Usage /process
  • DRAM    Usage  Usage /process
  • Framerate
  •  
  • GPU      73C     54%
  • VRAM   6035   4203
  • DRAM  13356  7449
  • 40 fps

The memory values seem to vary by a few percent constantly all except the VRAM usage value, which stays pretty darn constant.  It would be interesting to see the difference of these variables being monitored on another card.  For example, a 3080 TI running on a higher resolution, such that is on your system.

 

It seems that maybe that a first glance at all this requires another look? 

If what I suspect is true, then things don't really look as marginal as I had thought earlier in regards to my system performance.

Caldera

 

Edited by Caldera
Posted

I'll need to download MSI afterburner, im currently using Precision X1. I'm will to check it out though, for consistency sake we should use same module, same map, same airfield. How about the Hornet, Caucasus at Sochi?

Modules: F-14A/B | F-15C | F-16C | F/A-18C | SU-33 | Spitfire Mk IX | AH-64D | UH-1 | Super Carrier | Combined Arms | Persian Gulf | Syria | NTTR

Setup: VKB Gunfighter Mk.III F-14 CE HOTAS | Thrustmaster TWCS Throttle | MFG Crosswind V3 | Custom switch panel | Tek Creations F14 Display Panel | Custom F14 Left Vertical Console | Custom IR Tracker | Custom butt kicker

PC: i7 11700K | 64GB G-Skill DDR4 3600MHz | EVGA GeForce RTX 3080Ti FTW3 | DCS dedicated 2TB M.2 NVMe SSD | 3440x1440 144hz 34" ultrawide

Posted

You have allocated Vram, this is what DCS will make available for the game to use. So all of it.

Then there is actual usage at that time, that is the lower number. 

A good test for this is the F-16 over a built up area with lots of buildings. That's a good test for max Vram usage. Syria map will probably show the most.

DCS seems to perform better on Nvidia. Get the best one you can. Your 3440x1440 monitor will run at 60fps and at that resolution using a 3080 easily. However those are like unicorns at the moment.

 

Posted (edited)

Hey Guys,

I did some more testing. 

So below is allot of pictures of two different screen configurations.  These were taken sitting on the tarmac at Incirlik on the Persia map.  Just for the sake of consistency all were under the same conditions.  The view was default forward in a A-10C II, no TrackIR and the TGP was off. 

So one new thing that I did discover, at least the way MSI Afterburner reports, is that the DRAM memory usage values would drop over time.  So I took a first screen shot as soon as I saw the GPU temperature stabilize and I took a second screen shot at 15 minutes after that time.  The FPS moves a bit so just depending on when I took the screen shot the differences can be noticed.

DCS @ 3440x2520 2 screens

S2 RH  01.gif

S2 RH  02.gif

DCS @ 3440x1440 1 screen

S1 RH  01.gif

S1 RH  02.gif

At high resolution the FPS stayed very steady I would call it right around 50FPS.   Notice that the DRAM usage is lower at higher resolution.  Anyone care to explain that?

 

DCS @ 2560x2180 2 screens

S2 RL  01.gif

S2 RL  02.gif

DCS @ 2560x1080 1 screen

S1 RL  01.gif

S1 RL  02.gif

At the lower resolution the FPS did not stay so steady.  In fact it would constantly dip down below the level of the higher resolution pictures that you see above.  I could not predict the average FPS and it was varying from 45-75 FPS.  The GPU never seemed to settle down with temperature or usage.  The GPU was all over the map.

All in all I am decidedly...    

...less certain.

Caldera

 

 

Edited by Caldera
Posted (edited)

Hello again,

I did some more wondering, as the instability in FPS at the low resolution bothered me.  The below screen shots show some of the highs and lows that I saw.

DCS @ 2560x2180 2 screens

S1 RL  01.gif

S1 RL  01a.gif

DCS @ 2560x1080 1 screen

S2 RL  01.gif

S2 RL  01a.gif

 

The below is a screen shot under the same circumstance as the above pictures taken at Batumi on the Caucasus map.

DCS @ 2560x2180 2 screens

S2 RL CAUC  01.gif

So this is much more what I had expected and a huge difference from the above pictures.  The FPS was really pretty stable and the GPU was running full throttle. 

There must be quite a difference in GPU performance vs resolution on different maps?  In the case of Caucasus vs Persia, a weaker GPU would prefer to play on Caucasus.

Caldera

Edited by Caldera
Posted

Hector,

8 hours ago, Hector45 said:

I'll need to download MSI afterburner, im currently using Precision X1. I'm will to check it out though, for consistency sake we should use same module, same map, same airfield. How about the Hornet, Caucasus at Sochi?

I can do that when you are ready.

Caldera

Posted

HI Caldera,

While resolution will have a small effect on the Vram required for DCS. The texture settings in DCS graphics section will have the most impact. The amount of Vram "IN USE" will also fluctuate as you move around the map of the game, but not that much. The numbers you posted make sense regardless of the resolution or if your using the 2nd monitor or not. Your using about 5gb of Vram on average and the whole amount is in the allocated section.

What I do see is that your GPU is running at maximum usage in some of those screenshots. So I am guessing you are NOT using Vsync or a framerate cap.

I can tell that to use DCS at 60fps with high settings at 3440x1440 resolution while also running a 2nd monitor you would want a 3070 minimum. Preferably a 3080 or even an AMD 6800xt. As for Vram its a good idea to have as much as you can really. 8gb of Vram is a good starting point.

If you look on Youtube for video's by "PointBlank" he has some optimisation videos for DCS using 3440x1440 resolution.

Posted (edited)

Boss,

Thanks!

Yes, I pulled the bridle off of my GPU and let it run at max FPS for testing purposes and I use G-Sync (compatible) in combination with 144Hz. 

I really did expect the VRAM usage to go up more than it did for higher resolution and for running two screens.  I expected my GPU to be choked due to not enough VRAM. 

It just isn't...

My current plan is to get a 3080 ti when I can find one near MSRP.  I do not intend to reward the scalpers.

 

This comment really launched me like a Falcon after a Pidgeon:

On 1/2/2022 at 6:05 PM, Hector45 said:

Can't answer your first two questions but as far as VRAM, DCS will schedule everything you got. 

 

In my case the only thing that really changed noticeably in all of my testing, was the FPS.  I have a FPS test flight set-up near Senaki on the Caucasus map.  It uses the A-10C II at tree top level with the TGP on and no other activity at all.  With the FPS capped at 40 in 1080 resolution my GPU runs a solid 40 FPS at around 50C.  When at 1440 resolution it struggles and runs often in the low 20 FPS at 80C. 

These are the settings that I have "Settled on" that hit the right amount of fidelity that I use for my current best enjoyment in playing.

Settings 01.gif

To be very honest, DCS at 1080P and 40 FPS is not in any way terrible.  In my case at 1440P, I notice that the cockpit textures / UI text seem sharper and the sky to be a remarkably better visual effect.  Also, there is a significant improvement on distant aircraft textures especially external lights that are on.

I have read / YouTubed that DCS, for spotting, is better at 1080P.  Is there is still any truth to that?

Caldera

 

Edited by Caldera
Posted (edited)
1 hour ago, Caldera said:

Yes, I pulled the bridle off of my GPU and let it run at max FPS for testing purposes and I use G-Sync (compatible) in combination with 144Hz. 

I really did expect the VRAM usage to go up more than it did for higher resolution and for running two screens.  I expected my GPU to be choked due to not enough VRAM. 

It just isn't...

Careful, with DCS it isn't a "one case fits all".  You need to see which module does what, because it can vary, and a lot.
Your impression will imediately change, believe it.

For example, try the F-14 (A & B) Tomcat or F-16C Viper modules, and in the Syria or The Channel maps.
Even better, do that in Multiplayer.

If on 1440P and higher, you'll definitely need 8GB+ VRAM on a GPU for that combination to be fluid.
DCS is a hog and will take whatever memory resources you have, be it VRAM or RAM (and pagefile).

 

1 hour ago, Caldera said:

To be very honest, DCS at 1080P and 40 FPS is not in any way terrible.  In my case at 1440P, I notice that the cockpit textures / UI text seem sharper and the sky to be a remarkably better visual effect.  Also, there is a significant improvement on distant aircraft textures especially external lights that are on.

I have read / YouTubed that DCS, for spotting, is better at 1080P.  Is there is still any truth to that?

As someone who jumped from 1080P to 4K, I confirm that it is easier for spotting on a lower screen resolution (as the less number of pixels bloat the wee dots at distance).
But the gains with image quality and clarity more than make up for it with the higher resolutions (such as 1440P and 4K).

 

Edited by LucShep

CGTC - Caucasus retexture  |  A-10A cockpit retexture  |  Shadows Reduced Impact  |  DCS 2.5.6 - a lighter alternative 

DCS terrain modules_July23_27pc_ns.pngDCS aircraft modules_July23_27pc_ns.png 

Spoiler

Win10 Pro x64  |  Intel i7 12700K (OC@ 5.1/5.0p + 4.0e)  |  64GB DDR4 (OC@ 3700 CL17 Crucial Ballistix)  |  RTX 3090 24GB EVGA FTW3 Ultra  |  2TB NVMe (MP600 Pro XT) + 500GB SSD (WD Blue) + 3TB HDD (Toshiba P300) + 1TB HDD (WD Blue)  |  Corsair RMX 850W  |  Asus Z690 TUF+ D4  |  TR PA120SE  |  Fractal Meshify-C  |  UAD Volt1 + Sennheiser HD-599SE  |  7x USB 3.0 Hub |  50'' 4K Philips PUS7608 UHD TV + Head Tracking  |  HP Reverb G1 Pro (VR)  |  TM Warthog + Logitech X56 

 

Posted (edited)

Luc,

1 hour ago, LucShep said:

Careful, with DCS it isn't a "one case fits all".  You need to see which module does what, because it can vary, and a lot.
Your impression will imediately change, believe it.

OK fair enough.  I only can test my system and I currently have no other points of reference to evaluate any thing else.

For example, take this screen shot:

  • S2 RL CAUC  01.gif

This picture shows VRAM allocated @ 6044 and usage @ 4758.

 

Now, considering that I have a 6 MB GPU the value of .787 or 78% is the ratio for allocated / used.  I do not actually know what the format over head is for VRAM.  But in case above it looks to be about 98%.  12 MB = 12288 x .98 = 12042 KB available on the GPU and 12042 x .8 = 9633 KB DCS usage potential.

If I consider the GPU to operate under a soft ceiling, in the above example then the margin would be 1286 KB and the soft ceiling would be 10756 KB.

All this would seem OK to me as "the system" (DCS or NVidia) should not risk over run. 

 

Say, I now have a GPU with 12 MB of VRAM and if I were to show the same picture take under the exact same parameters.

  • Would it show VRAM allocated @ 12042 and usage @ 4758?    
  • Would it show VRAM allocated @ 12042 and usage @ 9633?     80% usage
  • Would it show VRAM allocated @ 12042 and usage @ 10756?    soft ceiling

 

Thanks Guys for all the feed back,

Caldera

Edited by Caldera
Posted

Hi Caldera,

I can confirm that Lucshep really knows what he is on about with regards DCS and how it will use up your system resources. I would simply be repeating what he said. To help put this into perspective. I use a 6800xt at 1440p with a 60fps cap on a Freesync monitor. I use these settings and see my Vram go upto an average of 14500mb usage in Radeon overlay. DCS will use what is available but it is dependent on the memory speed too. 

DSC aircraft spotting is easier at lower resolution, I have a TV to try 4k on, any aircraft are much harder to spot. I find 1440p a great balance between visual sharpness, performance and playability. 

Finally if you plan to buy a 3080ti then all your worries about performance are pretty much out of the window. I have had the privileged of checking a 3080ti out and it performs pretty much perfect for DCS using 1440p and 60fps butter smooth. I bet it would be perfect for 4k too. Much better than my current 6800xt and my previous 2080ti.

 

DCS Settings.PNG

Posted

Bosco,

Thanks!

It seems to me that DCS is probably using some kind of a soft ceiling that, as you say, is dependent solely upon how much VRAM the card has.

 

Using my above assumptions:

16 MB is 16384 KB so 16384 x .98 = 16056 VRAM available.

16056 x .80 = 12845 which is not that close to the usage that you describe.

16056 - 1286 = 14770 which is pretty close to the usage that you describe.

 

It seems to near impossible to get around the current scalping strangle hold, at least in the near future.  Too bad, but that is how it is...

Considering a purchase from pre-built PC company, I see these cards listed in the options with respectable delivery dates.  Do you happen to know, is this a for real thing or a fantasy?

Caldera

Posted (edited)

I had both, the RX 6800 XT with 16 GB and now the RTX 3080 with 10 GB. Before I had a GTX 1080 Ti with 11 GB. With all GPUs the RAM was mainly completely filled at a crowded server. I need to mention that I run 3x 27" with 2560x1440, means a total of 7680x1440, so a whole bunch of pixels. I never saw a difference in fluent gameplay, even with the RTX having 10 GB "only". It performs much better compared to the RX 6800 XT at that resolution. At 2560x1440 it gives a different picture performance wise and also RAM allocation. If I remember right the RAM never was filled completely at that resolution.

I hope this helps,

Xoxen

Edited by xoxen

AMD Ryzen 7 5800X3D, MSI MPG X570 Gaming Plus, 64GB Crucial Ballistix DDR4-3600 CL16, Asus TUF Gaming RTX 4080 OC, Windows 11 64bit Home Premium, TrackIR 5 with TrackClip: Pro!, Virpil MongoosT-50CM3 Base + TM Warthog Stick + 7cm extension + WINWING Orion 2 with F-15EX grips, Cougar MFDs with 8" displays, Saitek Rudder Pedals, Samsung Odyssey G9 49" 5120x1440 @120 Hz

Posted (edited)

Hey All,

I took some pictures while playing the A-10C on the SoCal - Through the Inferno server.  I would guess that there were about 20 guests and quite a few AWACS / Tankers in flight.  The results some what surprised me considering the comments made above.

I have increased my FPS limiter to 45 and I have added fan speed monitoring.  My GPU fan speed minimum speed setting is about 30%.


This is in flight north of Sochi-Adler looking at at the ground with the TGP off.  Notice that GPU utilization is 38%.

MP Stats 01 .gif

This is at Sochi-Adler on the tarmac right after I parked looking straight forward with the TGP off.  Notice that GPU utilization is 98%.

MP Stats 02 .gif

This is the same shot as above taken 15 minutes after doing nothing.  Notice that GPU utilization is 45% and that VRAM usage has gone up by about 10%.  DRAM has also dropped for both allocated and usage.

MP Stats 03 .gif


I have to wonder...  The only really difference I see in all of my testing is basically changes in FPS / GPU utilization.

My frames can get pretty stuttery in multi player at times.  That could also be CPU utilization and / or server ping times as well as the GPU is struggling to keep up.  I have not monitored either.  But, DRAM and VRAM usage seems more or less constant for single and multi player modes.

Caldera

Edited by Caldera
Editor lost it's mind...
Posted

Hey All,

These pictures were taken at pretty low graphic settings, resolutions are  at 3440x1440 and 1920x1080 with the TGP off.  The only setting that I changed was MSAA between the two pictures.  The FPS is capped at 58.

This is MSAA off

Rez 01 .gif

This is MSAA 4x

Rez 02 .gif

My discovery, is that the screen resolution effect on memory usage is very minor in all the ways that I have tested.  Basically, one could almost ignore the effect of different resolutions and multiple monitors.  However, the GPU loading will increase  (FPS drops) as it has more pixels to manage.

But finally, now I see a setting that causes a step change in memory usage.  I am sure that the memory usage would go up for every graphic option that I turned on or turned up.

Maybe I am starting to get the idea...

Caldera

Posted (edited)

Well, I'm happy for you if you're OK with running 40-45 FPS with 2D monitor(s), 'cause I sure can't and find it unacceptable.

What you're realizing (or will, if you keep testing), is that the higher resolution you go for the same given settings, the more and more outrageous will be the requirements to keep good performance.

Yes, I can run 4K with no AA (which I do) and have similar GPU usage and framerate as at 1080P with 2x MSAA + 1.5x SSAA (which I did before) or 4x MSAA.
Sure, this is not really unique to DCS, it also happens with many other sims/games.

The difference, or more precisely the problem, is how extremely heavy the graphics options turn out to be (MSAA, SSAA, DOF, SSAO, SSLR, Clouds, Shadows, Water and Heat Blur, etc) how badly they impact GPU usage, as well as VRAM, and how it further impacts when scaling up the resolution.

Add that to the absurd textures resolution for everything, including less important objects, while at the same time there is a really visible struggle with VRAM management.
DCS uses a gazillion of files, and extremely large, oversized textures - 8K and 4K diffuse maps, normal maps, specular maps, and other whatever maps, and too many of these are enormous 32-bit (not DXT1/3/5 8-bit, like many of us wished), used on most of the things you see in game - it puts a huge toll on any system. I don't know who ever thought this would be a good idea but, alright...
Users with RTX3090 24GB and RX6800/6900XT 16GB GPUs have been mentioning 14GB+ of VRAM usage. It's absurd, shouldn't even be acceptable.

Yes, I know people will say that there are other problematic sims and games but DCS is, let's say, "kinda special" (as to not be unkind to it). 🙂

We're at a point in DCS like never before, where to keep using the latest version we either choose to, a) sell a kidney to get a 2000,00+ USD/EUR/GBP graphics-card in the current market (which would never cost more than 700,00 otherwise) or, b) keep using the good old one we still have, while sacrificing graphics options to the point where the game looks like arse but the GPU is still near-barbecue (due to really, really high GPU usage), even with a framerate cap at 60 FPS.

This is why users have asked for the implementation of Nvidia's DLSS (which unfortunately only works with NVIDIA GPUs) or Intel's upcoming XeSS (which is certain to be GPU agnostic, i.e universal), using AI or deep learning technology to upscale images from a lower resolution to a higher one, while maintaining (almost) the same level of visual quality (so, a win-win, for everybody). To at least try getting some performance back to older and mid-range GPUs.

We can, at least, roll back to previous versions of DCS and gain back some performance with far more pleasant graphics settings - and I just hope we can keep on doing that.

PS: sorry for venting, it's just that sometimes it's truly puzzling to me how this sim/game can be so good and so bad at the same time.

Edited by LucShep

CGTC - Caucasus retexture  |  A-10A cockpit retexture  |  Shadows Reduced Impact  |  DCS 2.5.6 - a lighter alternative 

DCS terrain modules_July23_27pc_ns.pngDCS aircraft modules_July23_27pc_ns.png 

Spoiler

Win10 Pro x64  |  Intel i7 12700K (OC@ 5.1/5.0p + 4.0e)  |  64GB DDR4 (OC@ 3700 CL17 Crucial Ballistix)  |  RTX 3090 24GB EVGA FTW3 Ultra  |  2TB NVMe (MP600 Pro XT) + 500GB SSD (WD Blue) + 3TB HDD (Toshiba P300) + 1TB HDD (WD Blue)  |  Corsair RMX 850W  |  Asus Z690 TUF+ D4  |  TR PA120SE  |  Fractal Meshify-C  |  UAD Volt1 + Sennheiser HD-599SE  |  7x USB 3.0 Hub |  50'' 4K Philips PUS7608 UHD TV + Head Tracking  |  HP Reverb G1 Pro (VR)  |  TM Warthog + Logitech X56 

 

Posted

Luc,

Yes, the market for GPU's is depressing...  It is the reason that I don't have a better one.

For ED's part, I notice that they are a pretty dedicated team to their product.  I am sure that the very topics you discuss are on the table, but maybe not on any ones plate just yet.  If I were a guest there for dinner I am sure some person present would say "These changes that you propose.  Do they make us any or any more money?".  I would just have to wonder what that answer might be. 

The cost of development is high and always balanced against providing paychecks for employees and profitable sales for the company.  More competition to a like game might just drive it. 

I usually don't worry about any of that stuff, and I just play the game.  I will say, for the equipment I have purchased, this is the most expensive single video game I ever played...

Caldera

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...