Mednafen PCE-fast with overscan visibility

[Tatsuya79]

Updated the alternate version on 1st post with adjustable overscan width: 300 to 352 pixels wide selectable in core options.

That should fix some problems with AMD and Intel cards too.

[makecram1]

Hi, you did a great job fixing overscan and adjustable settings. Could you fix remaining “pixel aspect ratio” numbers for theoretical perfection? It’s very noticeable in “main menu” and “linearity 320x240 test” in 240p test suite compared to real TV.

Dot clock rates

256px-wide low mode = 8:7 par ntsc 320px-wide mid mode = 6:7 par ntsc 512px-wide high mode = 4:7 par ntsc

This emulator handles all cropping options correctly (pixel aspect ratio => display ratio)

0ldsk00l/nestopia/blob/master/libretro/libretro.cpp

Thank you!

[wasabi]

Hello, I have a little problem with the scanlines in the bettle pce fast emulator, I can no longer show the 2 missing pixel lines at the bottom. Even if you put the initial scanline at 5, the 2 missing lines no longer appear.On the photo above we can see the problem. Normally on the bottom left circle, the vertical line of the cross exceeds the circle by 2 pixels. Before, I managed to make these 2 pixel lines appear by removing 2 lines at the top with the scanline options, but it doesn’t work anymore, it looks like it’s broken.

I don’t know how to solve the problem. It seems that the problem comes from the last scanline that I can not go beyond 242, I think we should be able to go up to 244

[negativeExponent]

if the problem is the bottom scanlines, then adjust last_scanline option… dunno if there option changes since because i dont use this core but a similar option should be there. initial scanlines only removes top scanlines.

wait what? 244? why limit there… lets just draw the full scanlines then…

[wasabi]

I tried all different method, but the I do not succeed, impossible to have the two missing pixel line. In some games the bottom lines are more useful than the top one. On blazing lazers we have 3 lines at the top which can be deleted (we see in a stage that they are too much) but there are two lines at the bottom necessary . I don’t see how to put the scanline complete because by default I always miss 2 lines at the bottom. 244 would likely correspond to the 2 missing lines. But maybe some games have even more horizontal lines. I took the example of blazing lazers because it really surprised me

[Cyber]

@hunterk

If you’re going to automatically check for resolution changes each frame, I don’t think there’s any way to avoid spamming “set_geometry” each frame, as well.

@Tatsuya79

I thought perhaps checking for a video change before applying it (or not) would be wiser / keep the logging clean / softer on processing cost?

Could this be contributing to the audio cut outs I’ve been experiencing during CD games which switch resolutions while the same CD Audio track is playing when I’m using some heavier shaders like CRT-Guest-Advanced?

It happens in Ys Book I & II and Shadow Of The Beast intros.

I’ve documented my findings in the CRT-Guest-Advanced thread.

New CRT shader from Guest + CRT Guest Advanced updates

Greetings @guest.r! I’ve been having an issue with Shadow of The Beast for TurboGrafx-CD. During the attract mode several full screen still images are cycled while some awesome CD background music is played, however during the transitions between stills the music pauses briefly and it totally messes up the timing and immersion of this simple but enjoyable part of this game. This does not occur on real hardware. I’ve tested it with redump images both CHD as well as uncompressed. Turned CD Cac…

New CRT shader from Guest + CRT Guest Advanced updates

I tested them all. Here are my findings: Percentages are max GPU usage and CPU usage observed respectively during my test runs. nnedi3-nns16-2x-luma.slangp 40%, 3% Not a hitch nnedi3-nns16-2x-rgb.slangp 34%, 3% Not a hitch nnedi3-nns16-4x-luma.slangp 40%, 3% Not a hitch nnedi3-nns32-2x-rgb-nns-4x-luma.slangp 49%, 3% Very slight dips could be noticed in the audio but they stopped almost completely after a couple cycles of the images. nnedi3-nns32-4x-rgb.slangp 91%, 3% Hardly any d…

@Tatsuya79

Updated both cores to allow CD Volume up to 200. It’s allowed by Mednafen to make up for the usually low CD volume.

CDDA and ADPCM (digital voices with a CD-ROM Unit chip) can be modified directly, PSG (internal PC Engine sound chip) needs restart. 200 / 150 / 130 seemed nice on games I tested.

Hi are these settings accurate when compared to real hardware? If not can we finally setup a test bed using an SPL meter or app and maybe sound test in a few popular SuperCD-ROM games to put this issue with volume mixing to rest once and for all?

Is there any other PC-E Emulator that has this set accurately enough that I can use as a reference?

@Tatsuya79

Following this hint I made the CD PSG volume change immediately too. All the CD volume settings are applied during the game now.

I updated the link at the end of 1st post.

Thanks for your efforts in improving the emulation of my favorite video game system family.

There’s more I would like to chat with you about for example getting compatibility of Super Darius on par with EmuHawk and faux scaling games like Art of Fighting on par with Ootake.

Do you think it’s possible?

[hunterk]

If it’s having to reconstruct the image, and the shader has a long compile time, it could be involved, yeah.

[Cyber]

@hunterk

If it’s having to reconstruct the image, and the shader has a long compile time, it could be involved, yeah

Thanks, I think @Duimon tested it using an i7 6700K and a GeForce GTX 1070 using the HSM__ADV presets, and saw/heard no sign of stuttering.

I use 2 x 1070s in SLi but only one is utilized in RetroArch with a Ryzen 5 5600X. I recently tested it using a GTX 970 at 1080p with a Ryzen 5 3600X and the issue was there as well.

Both @Duimon and my previous tests were performed at 4K using default audio drivers and HDMI audio.

Anyone interested in trying to improve or resolve this?

[hunterk]

Unfortunately, if that’s the case, there’s nothing to be done about it, aside from switching to shaders that compile faster.

[Cyber]

Either that or brute force with more powerful hardware.

Okay, concerning the audio mixing improvements that were spoken about (200/150/130). Are these settings also baked into the Beetle PCE-Fast core currently?

How would one go about getting accurate mixing levels calibrated and baked into future Beetle Cores?

Also, do you know of any other PCE emulator that has the mixing levels accurate enough to use as a reference in the absence of real hardware?