Tried the config change in P3dv2.3 and there was no change in fps in three separate test scenarios. No one's set up is exactly the same, so what works for one simmer may not for another. The change, however, did not decrease fps so I'll leave in and see how it goes in other flights.

No change here.

 

gb.

I tried every conceivable TBM combo with P3D quite some time ago.

 

Absolutely no difference. I too have a Titan Black.

 

YMMV

No change, no worse either.

I didnt notice any difference. No scientific test though.

I changed mine in FSX years ago on advice of NickN and have added it to all my systems since. A lot depends on how well your system is balanced, IIRC in my conversations with Nick, it is more like a "final" kick once everything else is in order.

 

But, as noted, just like every other tweak - YMMV.

 

BTW, a trick that SE uses to affect FPS in FSX:SE- reduce your default zoom - you'll get an increase. 

 

Vic

A lot depends on how well your system is balanced

 

Yep. Don't get your hopes up on this. As I wrote in the FSX-SE compatibility threads, for most setups, this tweak won't improve performance. Rob has two Titan Black GPUs and that may explain his experience.  Anyone with either a GTX 980, GTX 970, Titan or GTX 780 ti should test this in P3d and report back.

 

 

TEXTURE_BANDWIDTH_MULT=160

 

You might want to go higher, like to 400 or 800, since the Titan Black has a bandwidth of 336 GB/s, as does the GTX 780 ti. Probably any card either near or above 200 GB/s bandwidth might see an improvement as long as the setup is not CPU-limited.

GTX 780 here; no noticeable difference.

Keep in mind that when it comes to GPU bandwidth, there are two types. The first is how fast the system communicates with the GPU. This is determined by the PCIe version (mostly 2.0 and 3.0 these days) and the slot type (8x, 16x, etc.). This communication consists of sending DirectX commands to the GPU and is generally not the limiting factor unless one has a fast GPU and a slow CPU. The second type is how fast the GPU itself communicates with its VRAM. Those GPU bandwidth numbers in my previous post are for that type of bandwidth. This bandwidth affects how fast information is sent by the GPU to the screen.

 

Although no one (except possibly someone from ACES) quite understands what this setting does, most people online have assumed that it somehow impacts the second type of bandwidth. I could be way off on this, as I'm only regurgitating what I have read over the years.

 

Also, probably people should also point out their CPU and clock speed when reporting results.

 

" -- UPDATE SP1 -- (from Phil Taylor's blog) 
The mysterious sounding TEXTURE_BANDWITH_MULT is our first target. This is a setting in the [DISPLAY] section of the file, formatted like this: 
[DISPLAY] TEXTURE_BANDWIDTH_MULT=n 
Where n can range from 10 to some reasonable value that is related to your frame rate limit. 
From Rafael Cintron, part of the FS Graphics and Terrain team, comes this description: 
“The TEXTURE_BANDWIDTH_MULT option in the Graphics section is the target frame rate use for calculating texture bandwidth. The higher you set this value the more textures we will allocate and copy per frame to the graphics card. The lower you set this value, the less we will allocate and copy up to a minimum limit. As an example, the default rate in the “high” perf bucket setting is 40. The lowest perf bucket setting is 10. 
Higher settings on this flag can cause stutters on frames where the terrain system has finished compositing lots of textures. Lower settings can reduce stutters on busy frames and spread out the load across multiple frames“ 
So thinking this thru, if the value you set is 40, and your frame rate limit is 30, then we will send 40/30 or 4/3 as much textures per frame. 
Moving this value to 400, like I have seen some users post in the forums, is probably *not* what you want to do since that increases the texture load on the graphics card by 10x, eg 400/40 = 10x. And after talking about this to Raf, setting it to 0 is ignored and can be validated by setting it to 10 which should give no different results. "

I've had mine set at that for some time now.  Works like a charm and combined with the FFT being at 0.01 I get great performance with no blurred textures on my setup.  If you are seeing little or no change I think its because your video card is already doing all it can with what it's got.

If I recall correctly, LM's default on this setting is 30. They also mention in their docs that increasing this value is recommended for photoscenery. There is no recommendation for setting this value higher for anything else, if I remember correctly.

 

Referencing a post above about Nick, with FSX he did mention going higher on this number if you had the appropriate card to handle the bandwidth but didn't explicitly explain what the benefits if doing so would be.

 

Mark

Keep in mind that at some point, the amount of VRAM that the GPU has, could be a limiting factor. . A Titan/Titan Black card has 6 GB of onboard VRAM. Most other high end cards have 3 or 4 GB VRAM. What happens if the GPU needs more VRAM that it has onboard? It uses the much slower system RAM. The amount that can be used is visible to the user by clicking "Screen Resolution" and then "Advanced Settings".

 

And any improvement also depends on your output resolution. Back a year or two, pretty much everyone was running their flight sim at 1920x1080, but now there are people running 1440p and 2160p monitors.

I've had mine set at that for some time now. Works like a charm and combined with the FFT being at 0.01 I get great performance with no blurred textures on my setup. If you are seeing little or no change I think its because your video card is already doing all it can with what it's got.

Or, you have SSDs across the board making video card capability much less a factor.

I've done a lot more testing and unfortunately my results are inconsistent.  I've been testing with 337.88 drivers (NOT using nVidia Inspector).

 

I need to dig into this further ... I found what I think is a driver version mis-match.  Is there is anyone that has NOT installed (ever) any nVidia drivers beyond 337.88?  If so, can you tell me what version of nvGameSR.DLL you have?   You can find this in the nVidia Control Panel | Help | System Info ... click on the Components tab.

 

My mis-match maybe a result of me going back from 347.09 to 337.88.

 

So as of right now, I'm going to retract this 40% claim from the TBM=160 until I can figure out exactly why it's not consistently a 40% gain.

 

Sorry if I got anyone's hopes up and I'll try to find a consistent solution.

 

Cheers, Rob.

I've done a lot more testing and unfortunately my results are inconsistent.  I've been testing with 337.88 drivers (NOT using nVidia Inspector).

 

I need to dig into this further ... I found what I think is a driver version mis-match.  Is there is anyone that has NOT installed (ever) any nVidia drivers beyond 337.88?  If so, can you tell me what version of nvGameSR.DLL you have?   You can find this in the nVidia Control Panel | Help | System Info ... click on the Components tab.

 

My mis-match maybe a result of me going back from 347.09 to 337.88.

 

So as of right now, I'm going to retract this 40% claim from the TBM=160 until I can figure out exactly why it's not consistently a 40% gain.

 

Sorry if I got anyone's hopes up and I'll try to find a consistent solution.

 

Cheers, Rob.

Hello Rob

 

If I understand right, you are mentioning about to be using an older version of driver?

Well, I'll tell a similar thing, but in a completely different system, and old.

My video card is an HD6950 2GB.

Well drivers 13.1 or older has a much better performance in P3D V2.4, drivers 13.4 or newer has a much lower performance in P3D V2.4, but for FSX or P3D V1.4 does not happen the same, the performance is equal .

 

João Alfredo