BVM-A20F1M Sega consoles Sync problems

[fernan1234]

Well, picture is skewed with the AES

Not surprising, though there are cheap solutions available to fix these skewing issues, but nothing yet for the off-spec refresh rates. It's a real shame indeed.

I agree that 480p is not that satisfying, especially if you add scanlines. Linedoubling + scanlines does give you a picture very similar to the original 240p, but 480p is generally more dull than 240p/480i on these monitors, and on top of that the scanlines reduce brightness, so the picture lacks the punch of the real thing. BTW, I also noticed that the MiSTer's 480p mode also has sync issues if vsync adjust is set to low lag.

I actually found that 720p using the MiSTer's scanline filters gives a better result than linedoubling, closer to how it should be, but still a compromise.

Getting another monitor is also a good solution, but let's face it, these ones are the best of the best. We probably really just need the 240p signals adjusted closer to spec. I bet there are some old video processors out there that can do this already, we just don't know yet.

[daty2k1]

It's true that I prefer the picture on the A20 compared to the M4 but at least I can fallback to it for arcade games.

I think some Extron processors are supposed to fix most issues but I didn't get the chance to test this yet. And there's also the solution that Bob was teasing but I'm not sure if there was any progress on this front.

[jwrose]

I use a Extron rgb 203rxi to fix skewing (flagging) with my AES and PC Engine and Turbo. Doesn’t help with SMS though.

[fernan1234]

All Extron units with DDSP and SERR switches can fix the skewing/flagging issue. It used to be believed that SMS needed the 580xi in particular for the comp sync switch, but I don't think that's the case. When I connected my Megadrive directly to a 203Rxi it did have problems syncing with it, but when I put a GBS-Control in passthrough mode in between it displayed perfectly and fixed the flagging issue on SMS games.

Looks like none of these Extron units can do anything to help with oddball refresh rate sources though, sadly.

[Fudoh]

Looks like none of these Extron units can do anything to help with oddball refresh rate sources though, sadly.

can be solved by doubling the resolution and then down scaling to 240p again as long as either processor normalizes the output refresh rate. Doubling can be done lag free what's the cost of down conversion? Half a frame using a GBS-C unit?

I mean, it's not perfect, but anybody has a BVM-A sitting at home already and wants to his oddball refresh rate sources, it's not impossible.

[fernan1234]

can be solved by doubling the resolution and then down scaling to 240p again as long as either processor normalizes the output refresh rate. Doubling can be done lag free what's the cost of down conversion? Half a frame using a GBS-C unit?

I mean, it's not perfect, but anybody has a BVM-A sitting at home already and wants to his oddball refresh rate sources, it's not impossible.

Yeah I really want to find a satisfactory solution to this as soon as I can, but in regard to this particular approach, I already tried it and the results are just OK, would definitely not call it "solved" this way. Specifically, linedoubling 240p to 480p and then adding scanlines is an alright compromise as I mentioned before, but when it comes to the GBS-C it it's working with a linedoubled picture it will add scanlines in the right place but as we know this gives a less bright and vibrant picture, and on top of that the GBS-C's scanlines have a flaw where color bleeds for certain colors, most notoriously blue (recently went over this on the GBS-C topic). The OSSC's scanlines would do much better, just don't have one right now, but the disadvantages of a 480p picture pretending to be 240p would still be there.

As to the GBS-C's actual 15khz downscaling, that would be an excellent solution based on results I've seen, but the GBS-C can only do this downscaling to true 240p from an interlaced 15khz signal. So basically what we would need is some device that can convert or "break" 240p into 480i, without further damage to the picture quality, and then feed that 480i into the GBS-C so it can downscale it properly to a, hopefully, normalized 240p signal for the BVM-As (and probably also the picky Ikegamis and JVCs).

So now even more than before I want to know what's the best way to "break" 240p RGB into 480i (which can be either RGB or YPbPr since the GBS-C takes both as inputs).

[Ryoandr]

GBSC can absolutely do 31k 480p > 15k 240p downscale.



you have to feed it combined sync, which I did with the 74LS86 circuit.


On the picky JVC DT-V I could also use the GBSC in 15k passthrough to get rid of Neogeo / PCE upperskew, then using RGBHV directly from the GBS to the DT-V. If I sync combine, I get the upper skew back. I don't have an A serie BVM so I can't help on this.

[fernan1234]

GBSC can absolutely do 31k 480p > 15k 240p downscale.



you have to feed it combined sync, which I did with the 74LS86 circuit.


On the picky JVC DT-V I could also use the GBSC in 15k passthrough to get rid of Neogeo / PCE upperskew, then using RGBHV directly from the GBS to the DT-V. If I sync combine, I get the upper skew back. I don't have an A serie BVM so I can't help on this.

Oh really? Your monitor detects a "480i" 15khz input?

I'll have to try again then to double check, but when I tried with the MiSTer set to 31k 480p it didn't work, and I'm pretty sure the MiSTer is also set to output combined sync.

Actually I'll do that check now and come back to confirm.

[Ryoandr]

Oh really? Your monitor detects a "480i" 15khz input?

Absolutely !

[fernan1234]

Absolutely !

Maybe it works with your PC outputting a nice well-behaved 60hz signal. Do you have a MiSTer that you could test the GBS-C's 15khz downscale with? I spent a good while trying different settings and could not get it to work.

I tried setting the MiSter to 480p (720x480) as well as 640x480, which the GBS-C consider "low resolution", and so I tried enabling and disabling upscaling for "low res" on the GBS-C as well. Tried enabling/disabling force scandoubler, force vga scaler, combined sync, vsync adjust, and combinations of these. The GBS-C won't downscale 480 to 240 15khz, and it won't improve the sync issues with the oddball refresh rate cores either.

But if you or anyone else with a MiSTer + GBS-C manages to get better results, please let me know. It could be that it is working but this monitor in particular can't sync to the downscaled output.

[Fudoh]

The GBS-C does definitely downscale and normalize a 480p COMPONENT signal. I didn't try VGA in, since some people said VGA was buggy (or maybe it was because of the composite sync requirement). You can get YUV output on the MiSTer easily right ? Give this another try. I don't see why it should not be working.

[fernan1234]

The GBS-C does definitely downscale and normalize a 480p COMPONENT signal. I didn't try VGA in, since some people said VGA was buggy (or maybe it was because of the composite sync requirement). You can get YUV output on the MiSTer easily right ? Give this another try. I don't see why it should not be working.

I imagine that Ryoandr's PC is outputting RGB and not YPbPr, but I can give that a try just to exhaust all possibilities. However, I don't think that downscaling is not working either way, but it doesn't necessarily solve the off-spec refresh rate problem, since the MiSTer still seems to preserve the core's original refresh rate even with linedoubled/scaled output, and the GBS-C just passes it through, or itself fails to sync with it, but I'll test with different settings again to be sure.

Another possibility I'll explore is using a custom video mode on MiSTer.

[Ryoandr]

Note that GBSC currently has problems with some real MVS / AES boards, it can't sync properly to them. I have several of them, and I sent one to Rama in hope he manages a breakthrough.

Would be kinda unlucky that MiSTer core reproduces these peculiar boards behavior

[Fudoh]

but it doesn't necessarily solve the off-spec refresh rate problem, since the MiSTer still seems to preserve the core's original refresh rate even with linedoubled/scaled output, and the GBS-C just passes it through, or itself fails to sync with it, but I'll test with different settings again to be sure.

MiSTer preserves the original refresh, but the GBS doesn't. AFAIK (I really don't use a GBS regularly) the GBS converts everything to 59.94 or 60Hz.

[fernan1234]

OK the GBS-C is a dead end for the MiSTer, but using custom video modes is by far the most optimal solution anyway. I found a few really satisfactory options. Of course, all of these normalize the refresh rate so fast horizontally scrolling scenes will have cyclic stuttering. This is the one drawback that will be unavoidable in these monitors (side note: ironically, while the A BVMs can sync to these, the D BVMs apparently can't sync to most). When not needing a custom aspect ratio, you'll want to set aspect ratio to "full screen" in all cases.

Code: Select all

1. video_mode=720,16,63,59,224,11,21,6,13500

This gives you a nice SDTV-like 720x224, but for cores like Neo Geo or anything with 320x224 resolution you'll want either a custom aspect ratio (custom_aspect_ratio_1=640:224) in your ini file to use and avoid shimmering during horizontal scrolling, or to use a scaler filter to minimize it (I got really nice sharp results with interpolation (sharp), or laczos2_15).

For arcade cores like CPS the only option is to use scaler filters, or

Code: Select all

2. video_mode=768,72,72,144,224,11,21,6,16632

This one is perfect for CPS and other 384x224 resolutions, needing no custom aspect ratio or scaler filters. An equivalent 640x224 mode could be used with the same advantages for Neo Geo, etc.

Code: Select all

3. video_mode=1920,152,247,280,224,11,21,6,40860

In a way this one is ideal as 1920 is perfectly divisible by both 384 and 320, however this is based on HDMI output for a Raspberry Pi and both h and v sync are positive polarity, so for the BVM you'll need to invert the sync either on your DAC if it supports it or on a sync combiner like the Extron RGB interfaces. This also means that when using the regular direct video mode that uses negative sync polarity you'll have to switch back, so this inconvenience sadly keeps this from being the perfect solution.
edit: actually this was only problematic with my HDMI-DAC setup, via IO board analogue it's perfectly fine.

Code: Select all

4. video_mode=1440,38,124,114,240,4,3,15,27027

This is based on the CEA mode 8 standard, and uses negative polarity so it doesn't have any of the drawbacks of the above option. But you'll want custom aspect ratios like these for 320 and 384 horizontal resolutions:

Code: Select all

custom_aspect_ratio_1=1280:224
custom_aspect_ratio_2=1152:224

Since you have more horizontal pixels to work with, while not needing to switch sync polarity, this may be the overall best option to accommodate a larger range of resolutions.
edit: actually the custom aspect ratios above may not be needed, setting the core's AR to full screen seems to avoid scrolling shimmering while filling more of the screen like with direct video.


I was able to figure all of this out using info from the following sites:
https://www.improwis.com/tables/video.webt
https://github.com/Xilinx/embeddedsw/bl ... gs_table.c
https://android.googlesource.com/kernel ... ee7f2%5E!/
https://misterfpga.org/viewtopic.php?f= ... 9&start=50


Lots of great info there and probably can help others figure out different and even better options for other monitors. What I would really appreciate now is if someone could help figure out a 1920x224 or 1920x240 mode that uses negative sync polarity, as that would be the best of all worlds. If anyone knows how to do this, please let us know here!

[RGB0b]

I'm having trouble understanding the MiSTer issues - I just tested my setup into my D9H and SMS works either from the analog output, or "Direct Mode" via HDMI-to-VGA (outputting RGBs).

An original SMS console directly connected definitely has the same "sync curl" problem we've all been discussing.

If you update your MiSTer, does it not work for you? I posted my MiSTer.ini file below (sorry for the wall of text) and here's links to direct mode and the converters I use:

https://www.retrorgb.com/mister-240p-over-hdmi.html

https://github.com/theypsilon/Update_All_MiSTer

https://www.amazon.com/shop/retrorgb?li ... 72N6BL06SV (I tested with the Portta one)

Code: Select all

[MiSTer]
key_menu_as_rgui=0     ; set to 1 to make the MENU key map to RGUI in Minimig (e.g. for Right Amiga)
forced_scandoubler=0   ; set to 1 to run scandoubler on VGA output always (depends on core).
ypbpr=0                ; set to 1 for YPbPr on VGA output.
composite_sync=1       ; set to 1 for composite sync on HSync signal of VGA output.
vga_scaler=0           ; set to 1 to connect VGA to scaler output.
hdmi_audio_96k=0       ; set to 1 for 96khz/16bit HDMI audio (48khz/16bit otherwise)
keyrah_mode=0x18d80002 ; VIDPID of keyrah for special code translation (0x23418037 for Arduino Micro)
volumectl=0            ; enable audio volume control by multimedia keys
recents=1
vscale_mode=0          ; 0 - scale to fit the screen height.
                       ; 1 - use integer scale only.
                       ; 2 - use 0.5 steps of scale.
                       ; 3 - use 0.25 steps of scale.
vscale_border=0        ; set vertical border for TVs cutting the upper/bottom parts of screen (1-99)
;bootscreen=0          ; uncomment to disable boot screen of some cores like Minimig.
;mouse_throttle=10     ; 1-100 mouse speed divisor. Useful for very sensitive mouses
rbf_hide_datecode=0    ; 1 - hides datecodes from rbf file names. Press F2 for quick temporary toggle
menu_pal=0             ; 1 - PAL mode for menu core

; USER button emulation by keybaord. Usually it's reset button.
; 0 - lctrl+lalt+ralt (lctrl+lgui+rgui on keyrah)
; 1 - lctrl+lgui+rgui
; 2 - lctrl+lalt+del
; 3 - same as 0 (lctrl+lalt+ralt on keyrah)
reset_combo=0

fb_size=0              ; 0 - automatic, 1 - full size, 2 - 1/2 of resolution, 4 - 1/4 of resolution.
fb_terminal=0          ; 1 - enabled (default), 0 - disabled

dvi_mode=0             ; set to 1 for DVI mode. Audio won't be transmitted through HDMI in DVI mode.


; 0 - 1280x720@60
; 1 - 1024x768@60
; 2 - 720x480@60
; 3 - 720x576@50
; 4 - 1280x1024@60
; 5 - 800x600@60
; 6 - 640x480@60
; 7 - 1280x720@50
; 8 - 1920x1080@60
; 9 - 1920x1080@50
;10 - 1366x768@60
;11 - 1024x600@60
;
; custom mode: hact,hfp,hs,hbp,vact,vfp,vs,vbp,Fpix_in_KHz
;   video_mode=1280,110,40,220,720,5,5,20,74250
video_mode=8


; set to 1-10 (seconds) to display video info on startup/change
video_info=2

; Set to 1 for automatic HDMI VSync rate adjust to match original VSync.
; Set to 2 for low latency mode (single buffer).
; This option makes video butter smooth like on original emulated system.
; Adjusting is done by changing pixel clock. Not every display supports variable pixel clock.
; For proper adjusting and to reduce possible out of range pixel clock, use 60Hz HDMI video
; modes as a base even for 50Hz systems.
vsync_adjust=2


[menu]
fb_size=1
video_mode=8

[daty2k1]

On my side, no issue with SMS on MiSTer but I get a curved picture on real hardware, even with the gscartsw fix...

[RGB0b]

Also, regarding the GBS-C, I just tested this:

- Plug an SMS into the front VGA port via RGBs (you might need a sync stripper device for it to work)

- Set the GBS-C to 15KHz passthrough

- Set the output to component video

- Connect it to the BVM with a "VGA to BNC" cable and only connect RGB.

SMS worked without a sync issue. I didn't have time to test it with light guns or anything, but can someone else test and confirm this with a BKM-68x?

[fernan1234]

I'm having trouble understanding the MiSTer issues - I just tested my setup into my D9H and SMS works either from the analog output, or "Direct Mode" via HDMI-to-VGA (outputting RGBs).

The MiSTer issues were not about the SMS core. I mentioned that that particular core has no problems, in contrast with the original hardware. The issues were with even more off-spec refresh rate cores like arcade cores or the Neo Geo core that uses MVS timings. For such cores, the alternative video modes I posted seem to be the best solution.

- Set the output to component video

Your suggestion also works well as mentioned earlier for original hardware SMS on these monitors, but it is not necessary to set output to component video, RGBHV works too, you'll just need to combine the sync as usual.

[RGB0b]

I'm having trouble understanding the MiSTer issues - I just tested my setup into my D9H and SMS works either from the analog output, or "Direct Mode" via HDMI-to-VGA (outputting RGBs).

The MiSTer issues were not about the SMS core. I mentioned that that particular core has no problems, in contrast with the original hardware. The issues were with even more off-spec refresh rate cores like arcade cores or the Neo Geo core that uses MVS timings. For such cores, the alternative video modes I posted seem to be the best solution.

Okay. I misunderstood and thought it was an issue with all cores on the MiSTer. What about outputting MiSTer in direct mode (or from the I/O's dsub port) via RGBHV into an Extron sync combiner? Anything with the "SERR" function could work. I've just trying to spare people the hassle of custom timings when you shouldn't have to worry about any of this with CRT's.

- Set the output to component video

Your suggestion also works well as mentioned earlier for original hardware SMS on these monitors, but it is not necessary to set output to component video, RGBHV works too, you'll just need to combine the sync as usual.

Right, but I was hoping to skip the last piece needed and just use the GBS-C. I have other people helping me test this now and SMS seems to work, but not PCE.

PCE - > GBS-C -> Extron 192 does work, so at least that's a solution. As always, one device would be better though.

I'll get back to working on the sync regeneration box as soon as there's time.

[fernan1234]

What about outputting MiSTer in direct mode (or from the I/O's dsub port) via RGBHV into an Extron sync combiner? Anything with the "SERR" function could work. I've just trying to spare people the hassle of custom timings when you shouldn't have to worry about any of this with CRT's.

The serration pulses from Extron RGB interfaces only help for top picture flagging/skewing issues like the one on the SMS, PCE RGB mods, and Neo Geo AES. They do not make a difference for those odd refresh rates that are too far away from standard levels, which the circuitry of this monitor is incapable of syncing to. The only solution seems to be using video processing that can scale the refresh rate to something more standard (around 60Hz), with the known compromise which that entails. At least it's a bit more tolerant than most flat panels in this regard. It's a CRT, but it was not really made with analogue needs in mind, but more for the new digital age.

edit: and since this is a timings sync capability issue, it's not unique to the 68X, but composite and S-video inputs would also have the same limitation (and of course even more so SD-SDI).

[jwrose]

The serration pulses from Extron RGB interfaces only help for top picture flagging/skewing issues like the one on the SMS, PCE RGB mods, and Neo Geo AES. .

SERR hasn’t helped with my SMS. That has some to do with horizontal slices (probably a bad paraphrase)

[fernan1234]

SERR hasn’t helped with my SMS. That has some to do with horizontal slices (probably a bad paraphrase)

Actually SMS could be unique in particular in not only needing SERR but also the composite sync dip switch that the 580xi has, as retrorgb found a long time ago. But as also mentioned, the GBS-C's passthrough is a good solution for SMS without needing that particular Extron unit, which can still be found cheap though not very often nowadays. I also don't know if there may be a difference between actual SMS vs. Mega Drive/Genesis in SMS mode for this.

[PeterWar]

Has further progress been done on solving this issue?

[jwrose]

Has further progress been done on solving this issue?

Not that I've heard. I'm still curious, too. In my setup I could route my SMS to my BVM and/or a Sony WEGA, so I just game on the WEGA with SMS- but would love to have a solution for the BVM as well.

[fernan1234]

The upcoming 68X clone card by skumulus is likely to fix the sync issues, at least some of them.

[jwrose]

The upcoming 68X clone card by skumulus is likely to fix the sync issues, at least some of them.

Interesting! Are there updates posted somewhere on that? I googled and didn't see anything.

[fernan1234]

The upcoming 68X clone card by skumulus is likely to fix the sync issues, at least some of them.

Interesting! Are there updates posted somewhere on that? I googled and didn't see anything.

Skum has made some posts on reddit about it, and there was also a long post on his blog back in March: https://immerhax.com/?p=624

Last I heard he's really close to finishing it up, needs to test some more of the problematic sources.

[RGB0b]

Has further progress been done on solving this issue?

Yes. I promise some solid info will be released within a month...probably sooner.

[jwrose]

Has further progress been done on solving this issue?

Yes. I promise some solid info will be released within a month...probably sooner.

Excellent!