F1 2019 3D Fix Helixvision/3DFM

[The_Nephilim]

Hey guys I fixed F1 2019 and it does not look to bad. I will go back and try and correct the remaining issues but it is a W.I.P so with out further ado here is the link to the Blog:



https://helixmod.blogspot.com/2022/07/f ... n-fix.html

[Shift-E]

nice! I'm not a big F1 guy, but still VERY cool to see a racing game get a 3D fix around here. nice job

[3DNovice]

...

[The_Nephilim]

Instead of asking the end user to swap profiles, why not leverage 3DMigoto to assign the needed stereo settings to the existing game profile?
It's possible that the existing settings that Nvidia implemented for 2D are useful for performance or fidelity.

I see that StereoProfile = 1 is uncommented in the d3dx.ini, but it is not being used for anything

You might want to tell them that the fix requires them to approve a UAC prompt to install the new profile settings, but most everyone knows this by now.

Well we are switching profiles as I bleieve there is more then just a stereo enabled in the Max Payne 3 profile which in my book is easier then trying to pick through it manually and figure what bits I need.. a Profile swap is quite painless..

Yes I had it enabled before guess I forgot to comment it back out no harm as it is not being used anyhow but I can turn it off next update..

Well if they are not aware of the UAC by now well guess they got no business messing with these fixes.. but if it really is a bother I can add it to the blog..

[3DNovice]

...

[The_Nephilim]

OK well I learned something new today thank you.. I was not aware I could insert the profile like that in the ini file.. I guess I would just copy the important part of the profile in the d3dx.ini file ? I did that and it says error, hwo do I get this profile bits into the d3dx.ini file??

I will implement it just looking to see how to do it and that way it will make it a bit easier for the end user.. haha!!


EDIT: well I have a big problem again the game is not launching in DX11 mode for some reason.. I am unsure why it is not working I did like I said to do and it still launches the game in DX12 really wierd,, working the problem.

Solved the issue of the exe not launching in dx11, I had to switch the named exe's. then put in Helixvision launch from exe and it seems to be reliablly launching in DX11 mode now..

[3DNovice]

...

[The_Nephilim]

well how do I know what bits from the Max Payne Profile I need to import into the new profile for F1? is it trial and error? as something in there fixed some issues.

Also I am having major teething issues with Helixvision reliably launching in DX11 mode, for some reason even thos I am using the dx11 exe it still launches dx12 for some mad reason.. really wierd still investigating..

[3DNovice]

...

[The_Nephilim]

Well thank you for the info I used those 3 and it looked like everything was fine as it was when using the max payne profile.

I also think I solved the game launching in DX11 mode vs dx12, I had to put the argument " -options 2 " in Helixvision.. so hopefully the fix is good now. now I am going to try and get Project cars 2 working as I ran into an issue with Helixvision..

[spin1]

Just saw this now. I am going to check this and Project Cars 2 out over the next couple weeks or whenever it's ready. I'm going to send you money later. I think you're the same person who I was speaking with previously about fixes for racing games such as F1 games? Yes?

I can't send a lot of money as I'm not a particularly wealthy person but I'll send more than just a coffee or case of beer

[The_Nephilim]

Just saw this now. I am going to check this and Project Cars 2 out over the next couple weeks or whenever it's ready. I'm going to send you money later. I think you're the same person who I was speaking with previously about fixes for racing games such as F1 games? Yes?

I can't send a lot of money as I'm not a particularly wealthy person but I'll send more than just a coffee or case of beer

Well it is as ready as it is gonna be.

[maurizioclaudio21]

Thanks for the wip
if you could fix the shadows a bit it would be great.
without shadows is a little unrealistic
To start the game in dx11 when you press play in steam, it asks you what option you want to do if in dx12 or dx11
Hoping that some "wizard" will help you with the shadows
thanks

[The_Nephilim]

do the shadows look broke to you?? did you disable them? I am not launching the game from Steam so that is not an option for me. I found a way to get it working tho. do you have 3d vision gear?

[Losti]

do the shadows look broke to you?? did you disable them? I am not launching the game from Steam so that is not an option for me. I found a way to get it working tho. do you have 3d vision gear?

Make a batch file and add gameexename.exe -dx11 to try start the game in dx11

[maurizioclaudio21]

do the shadows look broke to you?? did you disable them? I am not launching the game from Steam so that is not an option for me. I found a way to get it working tho. do you have 3d vision gear?

Hi


I send you two photos
the shadows are all broken if you run with the sun. With rain and cloudy it is fine.
I play with it but I have eliminated all the shadows from all the circuits
the only way was to skip all the broken shadows
I play in cm then I use 3d only for the award ceremony and the boxes.
I was hoping your fix solved the problem.

[The_Nephilim]

Yes I could of swore they where ok. I just checked it a minute ago they are indeed broke.. here is the shader I think is any shaderhacker available to help with this shadow shader?

[Losti]

Yes I could of swore they where ok. I just checked it a minute ago they are indeed broke.. here is the shader I think is any shaderhacker available to help with this shadow shader?

Dont look like its the right Shader for fixing. Looks like a general draw Shader. Shadows can also be in cs, shadows while hunting can freeze not disable so its worth moving camera to track down the right one. Seldom cases but sometimes it can be a vs needs correction.

[The_Nephilim]

yes I kinda thought it might not be the correct one I will try hunting it in Pink mode.. thank you Losti

[The_Nephilim]

hmmm got a lead here on this VS perhaps it is the correct one alot of indicators say it is:

[The_Nephilim]

OK I was just looking through this and while I am unsure what to do even tho Losti helped me out with a bit of info on how to fix things in ASM, I am not too sure if what I was taught will work here as it is a shadow fix..

I do know they can be tricky but I will go back and look at what Losti told me and see if I can figure something out here but no promises as I am pretty thick about these fixes LOL!!


Here is the modded file I am going to try and use in F1 2019.. hopefully it is right or does something and I will know I am on the correct path..

[Losti]

OK I was just looking through this and while I am unsure what to do even tho Losti helped me out with a bit of info on how to fix things in ASM, I am not too sure if what I was taught will work here as it is a shadow fix..

I do know they can be tricky but I will go back and look at what Losti told me and see if I can figure something out here but no promises as I am pretty thick about these fixes LOL!!


Here is the modded file I am going to try and use in F1 2019.. hopefully it is right or does something and I will know I am on the correct path..

I dont think that those fix there is working, because you have no depth there. But anyway, your ASM fix has issues:

Code: Select all

mad o0.xy, r2.xyxx, cb4[r0.w + 132].xyxx, cb4[r0.w + 132].zwzz
mad r30.xyzw, r2.xyxx, cb4[r0.w + 132].xyxx, cb4[r0.w + 132].zwzz 

DONT do any other operation for outputs than the original one!! And better to uncomment the original, its useless operation in this case.

Code: Select all

//mad o0.xy, r2.xyxx, cb4[r0.w + 132].xyxx, cb4[r0.w + 132].zwzz
mad r30.xy, r2.xyxx, cb4[r0.w + 132].xyxx, cb4[r0.w + 132].zwzz 

Code: Select all

mul r31.x, r31.x, r33.x
add r30.x, r31.x, r30.x

r31 is not used before, with the mul r31.x, r31.x, r33.x you multiply r31.x with stereo.x, but r31.x is zereo. If you simply want to add stereo to the output you just need

Code: Select all

add r30.x, r33.x, r30.x

If you want to use some multiplication, like l(0.5) for making half stereo.x for example or by separation for full depth, use

Code: Select all

mul r31.x, l(0.5), r33.x
add r30.x, r31.x, r30.x

OR this is the same

Code: Select all

mad r30.x, l(0.5), r33.x, r30.x

For use separation, you just need to replace the l(0.5) by r33.y..

Than you forgot to move the things back to the output!

Code: Select all

mov o0.xy, r33.xyzw

For the synthaxes you just have to follow the ASM shader refs:

https://docs.microsoft.com/en-us/window ... sl-sm4-asm

[The_Nephilim]

OK I made some adjustments how does this look:

Code: Select all

vs_5_0
dcl_globalFlags refactoringAllowed
dcl_constantbuffer cb4[159], dynamicIndexed
dcl_constantbuffer cb1[1], immediateIndexed
dcl_constantbuffer cb0[1], immediateIndexed
dcl_resource_buffer (float,float,float,float) t16
dcl_input v0.xy
dcl_input v1.xy
dcl_input_sgv v2.x, instance_id
dcl_output_siv o0.xyzw, position
dcl_output o1.xyz
dcl_output o2.xyz
dcl_output o3.xy



//dcl_temps 4

dcl_temps 36
dcl_resource_texture2d (float,float,float,float) t125
ld_indexable(texture2d)(float,float,float,float) r33.xyzw, l(0, 0, 0, 0), t125.xyzw



imad r0.x, v2.x, cb0[0].x, cb0[0].y
iadd r0.xyz, r0.xxxx, l(2, 3, 4, 0)
ld_indexable(buffer)(float,float,float,float) r1.xz, r0.xxxx, t16.wxyz
mov r2.x, r1.z
ld_indexable(buffer)(float,float,float,float) r1.yz, r0.yyyy, t16.xwyz
ld_indexable(buffer)(float,float,float,float) r2.zw, r0.zzzz, t16.xzyw
mov r2.y, r1.z
mov r1.z, r2.w
add r0.xyz, r1.xyzx, -cb4[55].xyzx
dp3 r0.w, r2.xyzx, r2.xyzx
sqrt r0.w, r0.w
div r1.xyz, r2.xyzx, r0.wwww
mul r2.xy, r0.wwww, v0.xyxx
mul r3.xyz, r1.yzxy, cb4[34].zxyz
mad r3.xyz, cb4[34].yzxy, r1.zxyz, -r3.xyzx
dp3 r0.w, r3.xyzx, r3.xyzx
max r0.w, r0.w, l(0.000010)
rsq r0.w, r0.w
mul r3.xyz, r0.wwww, r3.xyzx
mul r2.xzw, r2.xxxx, r3.xxyz
mad r2.xyz, r2.yyyy, r1.xyzx, r2.xzwx
add r0.xyz, r0.xyzx, r2.xyzx
imul null, r0.w, l(5), cb4[156].x
mad r1.w, r0.z, cb4[r0.w + 133].z, cb4[r0.w + 133].w
mad r1.w, r0.y, cb4[r0.w + 133].y, r1.w
mad r2.x, r0.x, cb4[r0.w + 133].x, r1.w
mad r1.w, r0.z, cb4[r0.w + 134].z, cb4[r0.w + 134].w
mad r1.w, r0.y, cb4[r0.w + 134].y, r1.w
mad r2.y, r0.x, cb4[r0.w + 134].x, r1.w


//mad o0.xy, r2.xyxx, cb4[r0.w + 132].xyxx, cb4[r0.w + 132].zwzz


mad r30.xyzw, r2.xyxx, cb4[r0.w + 132].xyxx, cb4[r0.w + 132].zwzz 

add r30.x, r31.x, r30.x
mov o0.xy, r33.xyzw



mov o0.w, l(1.000000)
mad r1.w, r0.z, cb4[r0.w + 135].z, cb4[r0.w + 135].w
mad r1.w, r0.y, cb4[r0.w + 135].y, r1.w
mad o0.z, r0.x, cb4[r0.w + 135].x, r1.w
add o1.xyz, r0.xyzx, cb4[55].xyzx
mul r0.xyz, r1.yyyy, cb4[29].xyzx
mad r0.xyz, cb4[30].xyzx, r1.zzzz, r0.xyzx
mad o2.xyz, cb4[28].xyzx, r1.xxxx, r0.xyzx
utof r0.x, cb1[0].x
div r0.x, l(1.000000, 1.000000, 1.000000, 1.000000), r0.x
mul o3.x, r0.x, v1.x
mov o3.y, v1.y
ret
// Approximately 43 instruction slots used

[The_Nephilim]

Well I goofed again I had readjusted the above code and will check to see if that fixed it but I had to add this:

add r30.x, r33.x, r30.x instead of this: add r30.x, r31.x, r30.x

will check later for now I am taking a break..

[Losti]

You initial output temp Operation is wrong. Read above.

[The_Nephilim]

You initial output temp Operation is wrong. Read above.

umm it is this line:

ld_indexable(texture2d)(float,float,float,float) r33.xyzw, l(0, 0, 0, 0), t125.xyzw


r33.xyzw should be r30.xyzw..

I am gonna run the game again and see if anything changed and try some experimenting..

[The_Nephilim]

I set it too the following and the result was the shadow sperates but it is still in mono I think I am close here.. here is the current code:

Code: Select all

 //dcl_temps 4

dcl_temps 9
dcl_resource_texture2d (float,float,float,float) t125
ld_indexable(texture2d)(float,float,float,float) r33.xyzw, l(0, 0, 0, 0), t125.xyzw



imad r0.x, v2.x, cb0[0].x, cb0[0].y
iadd r0.xyz, r0.xxxx, l(2, 3, 4, 0)
ld_indexable(buffer)(float,float,float,float) r1.xz, r0.xxxx, t16.wxyz
mov r2.x, r1.z
ld_indexable(buffer)(float,float,float,float) r1.yz, r0.yyyy, t16.xwyz
ld_indexable(buffer)(float,float,float,float) r2.zw, r0.zzzz, t16.xzyw
mov r2.y, r1.z
mov r1.z, r2.w
add r0.xyz, r1.xyzx, -cb4[55].xyzx
dp3 r0.w, r2.xyzx, r2.xyzx
sqrt r0.w, r0.w
div r1.xyz, r2.xyzx, r0.wwww
mul r2.xy, r0.wwww, v0.xyxx
mul r3.xyz, r1.yzxy, cb4[34].zxyz
mad r3.xyz, cb4[34].yzxy, r1.zxyz, -r3.xyzx
dp3 r0.w, r3.xyzx, r3.xyzx
max r0.w, r0.w, l(0.000010)
rsq r0.w, r0.w
mul r3.xyz, r0.wwww, r3.xyzx
mul r2.xzw, r2.xxxx, r3.xxyz
mad r2.xyz, r2.yyyy, r1.xyzx, r2.xzwx
add r0.xyz, r0.xyzx, r2.xyzx
imul null, r0.w, l(5), cb4[156].x
mad r1.w, r0.z, cb4[r0.w + 133].z, cb4[r0.w + 133].w
mad r1.w, r0.y, cb4[r0.w + 133].y, r1.w
mad r2.x, r0.x, cb4[r0.w + 133].x, r1.w
mad r1.w, r0.z, cb4[r0.w + 134].z, cb4[r0.w + 134].w
mad r1.w, r0.y, cb4[r0.w + 134].y, r1.w
mad r2.y, r0.x, cb4[r0.w + 134].x, r1.w


//mad o0.xy, r2.xyxx, cb4[r0.w + 132].xyxx, cb4[r0.w + 132].zwzz


mad r30.xyzw, r2.xyxx, cb4[r0.w + 132].xyxx, cb4[r0.w + 132].zwzz 

add r30.x, r33.x, r30.x
//mad r30.x, l(1), r33.x, r30.x
mov o0.xy, r30.xyzw          //changed r30 from r33 shadows seperate but in mono..



mov o0.w, l(1.000000)

[Losti]

The original Code is o0.xy so use r30.xy not r30.xyzw

[The_Nephilim]

I did not even catch the xyzw to xy only thank you for that. I did the following code and I am not sure but the distant shadows move when I press F9 to reload original shaders but the closest shodow just stays in mono and doesnt move when reloading orig. shaders.. I also tried many multiples of xy on the R30 I seen but nothing seems to work.. is there something I am missing here like a line of code or similar.. I thought I had an idea but something eludes me, I seem as thick as always haha!!

So I still do not think it is working. I will go back and reread your lesson you gave me a while back.. but here is what I currently have for the code:

Code: Select all

 //dcl_temps 4

dcl_temps 8
dcl_resource_texture2d (float,float,float,float) t125
ld_indexable(texture2d)(float,float,float,float) r33.xyzw, l(0, 0, 0, 0), t125.xyzw



imad r0.x, v2.x, cb0[0].x, cb0[0].y
iadd r0.xyz, r0.xxxx, l(2, 3, 4, 0)
ld_indexable(buffer)(float,float,float,float) r1.xz, r0.xxxx, t16.wxyz
mov r2.x, r1.z
ld_indexable(buffer)(float,float,float,float) r1.yz, r0.yyyy, t16.xwyz
ld_indexable(buffer)(float,float,float,float) r2.zw, r0.zzzz, t16.xzyw
mov r2.y, r1.z
mov r1.z, r2.w
add r0.xyz, r1.xyzx, -cb4[55].xyzx
dp3 r0.w, r2.xyzx, r2.xyzx
sqrt r0.w, r0.w
div r1.xyz, r2.xyzx, r0.wwww
mul r2.xy, r0.wwww, v0.xyxx
mul r3.xyz, r1.yzxy, cb4[34].zxyz
mad r3.xyz, cb4[34].yzxy, r1.zxyz, -r3.xyzx
dp3 r0.w, r3.xyzx, r3.xyzx
max r0.w, r0.w, l(0.000010)
rsq r0.w, r0.w
mul r3.xyz, r0.wwww, r3.xyzx
mul r2.xzw, r2.xxxx, r3.xxyz
mad r2.xyz, r2.yyyy, r1.xyzx, r2.xzwx
add r0.xyz, r0.xyzx, r2.xyzx
imul null, r0.w, l(5), cb4[156].x
mad r1.w, r0.z, cb4[r0.w + 133].z, cb4[r0.w + 133].w
mad r1.w, r0.y, cb4[r0.w + 133].y, r1.w
mad r2.x, r0.x, cb4[r0.w + 133].x, r1.w
mad r1.w, r0.z, cb4[r0.w + 134].z, cb4[r0.w + 134].w
mad r1.w, r0.y, cb4[r0.w + 134].y, r1.w
mad r2.y, r0.x, cb4[r0.w + 134].x, r1.w


//mad o0.xy, r2.xyxx, cb4[r0.w + 132].xyxx, cb4[r0.w + 132].zwzz


mad r30.xyzw, r2.xyxx, cb4[r0.w + 132].xyxx, cb4[r0.w + 132].zwzz 

//mad r30.x, l(.5), r33.x, r30.x
add r30.x, r33.x, r30.x

mov o0.xy, r30.xy

[Losti]

If a simple Stereo correction is not working in the vs than you have to fix in PS often, or in both that also can happen.

[The_Nephilim]

If a simple Stereo correction is not working in the vs than you have to fix in PS often, or in both that also can happen.

Well I will see if I can track down the correct PS for the shadows. I just hought shadow fixes needed special attention or cause certain issues when trying to fix them..

[The_Nephilim]

All I could find was this hopefull shader for shadows in the PS..
I only found 1 other and it was the one from before that you said was not it. so it must be this one here:

[Losti]

All I could find was this hopefull shader for shadows in the PS..
I only found 1 other and it was the one from before that you said was not it. so it must be this one here:

Have you checked the peer shaders for the VS?
When dump_usage=1 is enabled, than you should get a shaderusage.txt that you can look for.

The shader you gave me at last looks good, but without having the game its hard to check what fix we need.

i can only make suggestions

[Losti]

Im not as good as in HLSL like in ASM so i hope there is no expression error. May be you need some sign poop like -= or negative positive depth or negative or pos convergence, all things that has to be tried out^^. NOTE in some examples i use 1920 as the width resolution, without analythe the CBs i dopnt know where its stored ^^

1.

Code: Select all

  r1.xyzw = scene_view.invViewProjection._m10_m11_m12_m13 * -r0.wwww;
  r1.xyzw = r0.yyyy * scene_view.invViewProjection._m00_m01_m02_m03 + r1.xyzw;
  r0.xyzw = r0.xxxx * scene_view.invViewProjection._m20_m21_m22_m23 + r1.xyzw;
  r0.xyzw = scene_view.invViewProjection._m30_m31_m32_m33 + r0.xyzw;
  r0.xyz = r0.xyz / r0.www;
  
float4 stereo = StereoParams.Load(0);
r0.x -= stereo.x*(-r0.z - stereo.y)/scene_view.ViewProjection._m00; 

2.

Code: Select all

//
 float4 r10,r11;
float4 stereo = StereoParams.Load(0);
//
  r11.xyzw = scene_view.invViewProjection._m10_m11_m12_m13 * -r0.wwww;
  r11.xyzw = r0.yyyy * scene_view.invViewProjection._m00_m01_m02_m03 + r11.xyzw;
  r10.xyzw = r0.xxxx * scene_view.invViewProjection._m20_m21_m22_m23 + r11.xyzw;
  r10.xyzw = scene_view.invViewProjection._m30_m31_m32_m33 + r10.xyzw;
//
r0.y+= stereo.x*((stereo.y*r10.w-1)*1920.0)*0.5);
//
  r1.xyzw = scene_view.invViewProjection._m10_m11_m12_m13 * -r0.wwww;
  r1.xyzw = r0.yyyy * scene_view.invViewProjection._m00_m01_m02_m03 + r1.xyzw;
  r0.xyzw = r0.xxxx * scene_view.invViewProjection._m20_m21_m22_m23 + r1.xyzw;
  r0.xyzw = scene_view.invViewProjection._m30_m31_m32_m33 + r0.xyzw;
  r0.xyz = r0.xyz / r0.www;
//  

3.

Code: Select all

  r1.xyzw = scene_view.invViewProjection._m10_m11_m12_m13 * -r0.wwww;
  r1.xyzw = r0.yyyy * scene_view.invViewProjection._m00_m01_m02_m03 + r1.xyzw;
  r0.xyzw = r0.xxxx * scene_view.invViewProjection._m20_m21_m22_m23 + r1.xyzw;
  r0.xyzw = scene_view.invViewProjection._m30_m31_m32_m33 + r0.xyzw;
  r0.xyz = r0.xyz / r0.www;

float4 stereo = StereoParams.Load(0);

r0.x+=stereo.x*(r0.z-stereo.y);
  

34

Code: Select all

float4 stereo = StereoParams.Load(0);

r0.y+=stereo.x*(r0.x-stereo.y);

  r1.xyzw = scene_view.invViewProjection._m10_m11_m12_m13 * -r0.wwww;
  r1.xyzw = r0.yyyy * scene_view.invViewProjection._m00_m01_m02_m03 + r1.xyzw;
  r0.xyzw = r0.xxxx * scene_view.invViewProjection._m20_m21_m22_m23 + r1.xyzw;
  r0.xyzw = scene_view.invViewProjection._m30_m31_m32_m33 + r0.xyzw;
  r0.xyz = r0.xyz / r0.www;
  

No more blind guesses sorry May this helps. Please note again: things like this r0.x+=stereo.x*(r0.z-stereo.y); can also be r0.x-=stereo.x*(r0.z-stereo.y); or r0.x+=stereo.x*(r0.z+stereo.y); or r0.x-=stereo.x*(-r0.z+stereo.y); and so on.....

[The_Nephilim]

Hey I am not too good with HLSL either, I was wondering tho the ASM code in that shader is no good? it is there just close to the bottom.. I had put a post here I guess I did not click submit haha!!

Thank you tho for the HLSL code I seen those used before so maybe one will work I will give them a try thnx again Losti

Code: Select all

 ps_5_0
dcl_globalFlags refactoringAllowed
dcl_constantbuffer cb4[212], immediateIndexed
dcl_constantbuffer cb1[2], immediateIndexed
dcl_constantbuffer cb0[9], immediateIndexed
dcl_sampler s6, mode_default
dcl_sampler s15, mode_default
dcl_resource_texture2d (float,float,float,float) t0
dcl_resource_texture2d (float,float,float,float) t4
dcl_input_ps linear sample v1.xyw
dcl_output o0.xyzw
dcl_output o1.xyzw
dcl_output o2.xyzw
dcl_temps 2
max r0.x, v1.w, l(0.000010)
rcp r0.x, r0.x
mul r0.xy, r0.xxxx, v1.xyxx
mad r0.xy, r0.xyxx, l(0.500000, 0.500000, 0.000000, 0.000000), l(0.500000, 0.500000, 0.000000, 0.000000)
add r0.z, -r0.y, l(1.000000)
mad r0.yw, r0.xxxz, l(0.000000, 2.000000, 0.000000, 2.000000), l(0.000000, -1.000000, 0.000000, -1.000000)
sample_indexable(texture2d)(float,float,float,float) r0.x, r0.xzxx, t4.xyzw, s6
mul r1.xyzw, -r0.wwww, cb4[25].xyzw
mad r1.xyzw, r0.yyyy, cb4[24].xyzw, r1.xyzw
mad r0.xyzw, r0.xxxx, cb4[26].xyzw, r1.xyzw
add r0.xyzw, r0.xyzw, cb4[27].xyzw
div r0.xyz, r0.xyzx, r0.wwww
mul r1.xyz, r0.yyyy, cb0[5].xyzx
mad r0.xyw, cb0[4].xyxz, r0.xxxx, r1.xyxz
mad r0.xyz, cb0[6].xyzx, r0.zzzz, r0.xywx
add r0.xyz, r0.xyzx, cb0[7].xyzx
mad r0.w, r0.y, l(2.000000), l(-1.000000)
add r1.xyz, -|r0.xwzx|, l(1.000000, 1.000000, 1.000000, 0.000000)
mad r0.xy, r0.xzxx, l(0.500000, 0.500000, 0.000000, 0.000000), l(0.500000, 0.500000, 0.000000, 0.000000)
mad r0.xy, r0.xyxx, cb0[8].xyxx, cb0[8].zwzz
lt r1.xyz, r1.xyzx, l(0.000000, 0.000000, 0.000000, 0.000000)
or r0.z, r1.y, r1.x
or r0.z, r1.z, r0.z
discard_nz r0.z
deriv_rtx_coarse r0.zw, r0.xxxy
mul r0.zw, r0.zzzw, cb4[211].yyyy
deriv_rty_coarse r1.xy, r0.xyxx
mul r1.xy, r1.xyxx, cb4[211].yyyy
sample_d_indexable(texture2d)(float,float,float,float) r0.xyzw, r0.xyxx, t0.xyzw, s15, r0.zwzz, r1.xyxx
add r1.x, r0.w, -cb1[1].x
mul o0.xyzw, r0.xyzw, cb1[0].xyzw
lt r0.x, r1.x, l(0.000000)
discard_nz r0.x
mov o1.xyzw, l(0,0,0,0)
mov o2.xyzw, l(0,0,0,0)
ret
// Approximately 36 instruction slots used

//~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/

I did notice it appears when I have the VS Enabled it moves when I press f9 to load origianl it just might be it is almost working just in mono?? so maybe something amiss in the VS?? OR will the above ASM code be better to manipulate?

[Losti]

Yeah it is possible that the vs needs full depth and the ps the depth alignment. Sorry but if easy things not working i have to check it myselfe. But no time this year.

[The_Nephilim]

Well that is OK Losti I am thankful for the help you provided.. I will have to ponder this for a bit and see what I can come up with.