Main > Main Forum
Anyone adjusting HLSL on a per game basis?
BadMouth:
--- Quote from: PL1 on November 09, 2013, 12:30:36 am ---
--- Quote from: BadMouth on November 08, 2013, 10:02:20 pm ---Has anyone made up .lay files with a shroud?
--- End quote ---
Have you looked at MameLayPlus?
Scott
--- End quote ---
I will check it out. I thought it just put marquees on the sides when using a widescreen monitor.
B2K24, I'm not sure my settings are going to translate to very may other setups.
I originally set the shadowmask to the same dimensions as the original resolution per everything I've come across online.....and it looked ok, but looked wrong to me.
The mesh of the shadowmask looked too big IMO. This is one of the reasons that I wanted to go photograph real machines.
It's doubtful that the monitor in the ms pacman machine above was original, but it looks like the power pill in the pic is divided into 13 sections vertically.
With the shadow mask set to the original resolution, I only get 9 sections. So following the math, I change the shadow mask to 414x322.
It looks right to me now, but the way I arrived at it bothers me. I was hoping to find settings that "are correct" based on original resolutions and maybe have to adjust for it being stretched to 1080.
I'm running this on a 27" rotated monitor, so the rendered screen area is around 22".
Not rotating the game and having a smaller image makes these settings look like complete crap.
Anyone else with a big monitor want to give them a try and let me know what you think?
(the scanlines are probably light for most people's taste, but these are only for the 288x224 games)
I've been testing with ms pacman
--- Code: ---# DIRECT3D POST-PROCESSING OPTIONS
#
hlsl_enable 1
hlslpath hlsl
hlsl_ini_read 0
hlsl_ini_write 0
hlslini %g
hlsl_prescale_x 4
hlsl_prescale_y 4
hlsl_preset -1
hlsl_write
hlsl_snap_width 2048
hlsl_snap_height 1536
shadow_mask_alpha 0.300000
shadow_mask_texture aperture.png
shadow_mask_x_count 414
shadow_mask_y_count 322
shadow_mask_usize 0.093750
shadow_mask_vsize 0.093750
curvature 0.000000
pincushion 0.025000
scanline_alpha 0.200000
scanline_size 0.830000
scanline_height 0.500000
scanline_bright_scale 1.000000
scanline_bright_offset 1.000000
scanline_jitter 0.000000
defocus 0.900000,0.900000
converge_x 0.200000,0.000000,0.000000
converge_y 0.000000,0.000000,0.000000
radial_converge_x 0.200000,0.000000,0.000000
radial_converge_y 0.000000,0.000000,0.000000
red_ratio 1.000000,0.000000,0.000000
grn_ratio 0.000000,1.000000,0.000000
blu_ratio 0.000000,0.000000,1.000000
saturation 1.050000
offset 0.000000,0.000000,0.000000
scale 1.020000,1.020000,1.020000
power 1.300000,1.300000,1.300000
floor 0.020000,0.020000,0.020000
phosphor_life 0.3000000,0.3000000,0.3000000
--- End code ---
PL1:
--- Quote from: BadMouth on November 09, 2013, 12:11:55 pm ---
--- Quote from: PL1 on November 09, 2013, 12:30:36 am ---
--- Quote from: BadMouth on November 08, 2013, 10:02:20 pm ---Has anyone made up .lay files with a shroud?
--- End quote ---
Have you looked at MameLayPlus?
--- End quote ---
I will check it out. I thought it just put marquees on the sides when using a widescreen monitor.
--- End quote ---
For horizontal WS setups it has cocktail layouts and for vertical WS setups, it has the marquee (1024 pixels wide) at the top, a bezel around the monitor area, and instruction card. (default layout images attached)
Scott
BadMouth:
adjusted sf2 based on the above pics (mostly the damage meter)
I'm very happy with the results.
Applied to all games of the same resolution.
settings are attached as a txt file
THESE PICS ARE HLSL
McHale:
I'm only using HLSL for vector games. But I do set them on a per game basis.
ABACABB:
Hey BadMouth, I USED to use HLSL with my previous flatscreen. Once I scored the arcade monitor from you there's no looking back, for now. Agree HLSL does look pretty good on a flatscreen but the real thing is unbeatable IMHO.
I'm guessing this is for your NEO GEO slim cab where a CRT would never fit anyways? Cool idea. Hope it works out for you.
Navigation
[0] Message Index
[*] Previous page
Go to full version