BVM-A20F1M Sega consoles Sync problems

[daty2k1]

Last I heard he got SMS working without any issue and also NeoGeo AES as well if I remember correctly. He was waiting to receive a MiSTer so he could test arcade cores. On my monitor, none of them are working except SectionZ.

[skum]

Yeah, the BKM-68X "clone" (I prefer to call it an alternative) seems to definitely solve basically *all* sync issues, at least on consoles that I've been able to try. Note that I have actually not done anything to actively combat the sync issues. BVM-A runs HVSYNC internally, and I simply create that via a ISL59885 sync separator.

The systems I've been able to try up until now are:
SNES (2CHIP US) 60Hz (CSYNC)
SNES (1CHIP FAT US) 60Hz (CSYNC)
SNES (2CHIP EU) 50Hz (Sync on luma)
SNES (1CHIP EU, SNSP-1CHIP-01) 50Hz (Sync on luma)
Genesis (US) 60Hz (CVBS sync)
Genesis 2 (US) 60Hz (CSYNC)
Mega Drive (EU) 50Hz (CVBS sync)
Mega Drive 2 (EU, DFO) 60Hz (CSYNC)
Mega Drive 2 (EU, DFO) 50Hz (CSYNC)
Master System (on Genesis through Power Base) 60Hz (CVBS sync)
Master System (EU, DFO) 60Hz (CSYNC)
Master System (EU, DFO) 50Hz (CSYNC)
Master System (EU, VA3, no DFO) 60Hz (59,38Hz) (CSYNC)
Master System (EU, VA3, no DFO) 50Hz (CSYNC)
Nintendo 64 (NUSJ) 60Hz (THS7314 RGB) (CSYNC)
PSone SCPH-101 (EU) (MM3+MFO) 60Hz (RGC CSYNC)
PSone SCPH-101 (EU) (MM3+MFO) 50Hz (RGC CSYNC)
Wii (EU) 576i 50Hz (YPbPr)
Wii (EU) 480i 60Hz (YPbPr)
Wii (EU) 480p 60Hz (YPbPr)
PS3 (EU) 576i 50Hz (YPbPr)
PS3 (EU) 576p 50Hz (YPbPr)
PS3 (EU) 720p 60Hz (YPbPr)
PS3 (EU) 1080i 60Hz (YPbPr)
Xbox 360 (EU) 576i 50Hz (YPbPr)
Xbox 360 (EU) 480i 60Hz (YPbPr)Making it
Xbox 360 (EU) 576p 50Hz (YPbPr)
Xbox 360 (EU) 480p 60Hz (YPbPr)
Xbox 360 (EU) 720p 50Hz (YPbPr)
Xbox 360 (EU) 720p 60Hz (YPbPr)
Xbox 360 (EU) 1080i 60Hz (YPbPr)
PC Engine (White), External RGB mod (THS7374, Genesis 2 plug), 60Hz (CSYNC)
Analogue Nt Mini 2.0 60Hz RGBS
Amiga 500 RGBS (EU/PAL) 50Hz (CSYNC)

A whole bunch of arcade cores through MiSTer (VGA IO Board 6.1, CSYNC) including DoDonPachi (57.2Hz), and various CPS games.

A whole bunch of Neo Geo games through MiSTer (VGA IO Board 6.1, CSYNC) both AES and MVS cores.

The biggest oddity I've encountered is DoDonPachi core which runs at some 57.2Hz VSYNC. On the original, the image rolls, but the OSD is steady. On mine, the picture is steady but OSD rolls. I suspect this is because the original ends up thinking its a 50Hz sync, while mine ends up reporting 60Hz.

The current issue revolves around the worldwide electrical component shortage. I've made some redesigns to *try* to accomodate for it, but basically when I release this, nobody will be able to build it unless they source the components from a bunch of places, claming to have stock, where some of them seems more than a little shady. The "normal" places seems to have stock around the end of this year. It blows.

[daty2k1]

Incredible, can't wait

[fernan1234]

The current issue revolves around the worldwide electrical component shortage. I've made some redesigns to *try* to accomodate for it, but basically when I release this, nobody will be able to build it unless they source the components from a bunch of places, claming to have stock, where some of them seems more than a little shady. The "normal" places seems to have stock around the end of this year. It blows.

Even if these only start getting made around the end of the year, that's pretty soon! Thank you so much for designing this, as well as your other alternative cards!

Maybe down the line there can even be alternatives for the less expensive but almost as hard to find 61D card for NTSC/PAL/SECAM decoding. Same goes for the JVC DT-V monitor cards, I was just talking with someone on that topic about how the IF-C01PNG card is actually much harder to find loose than the IF-C01COMG, for which thankfully now we have your alternative.

[skum]

Even if these only start getting made around the end of the year, that's pretty soon! Thank you so much for designing this, as well as your other alternative cards!

Maybe down the line there can even be alternatives for the less expensive but almost as hard to find 61D card for NTSC/PAL/SECAM decoding. Same goes for the JVC DT-V monitor cards, I was just talking with someone on that topic about how the IF-C01PNG card is actually much harder to find loose than the IF-C01COMG, for which thankfully now we have your alternative.

Yeah well, hopefully a production run of the 68X can be done. That will be announced when/if possible, because soldering them up by hand takes forever, plus there are some rather "advanced" IC's requiring hot-air and solder-paste.

The 61D and IF-C01PNG is also something I'd like to look into. The precursor for that would be an open source composite/S-Video to YPbPr/RGB circuit, which would open up for both a 61D, IF-C01PNG and BKM-127W replacements. Essentially though, at the moment, the Koryuu *can* be used along with the IF-C01COMG, BKM-129X, and BKM-68X ("clone" and original) as an alternative, but internal cards are of course cooler

[fernan1234]

The 61D and IF-C01PNG is also something I'd like to look into. The precursor for that would be an open source composite/S-Video to YPbPr/RGB circuit, which would open up for both a 61D, IF-C01PNG and BKM-127W replacements. Essentially though, at the moment, the Koryuu *can* be used along with the IF-C01COMG, BKM-129X, and BKM-68X ("clone" and original) as an alternative, but internal cards are of course cooler

Awesome to hear you're interested in those. Besides being cooler, the internal cards would also offer the original's comb filter options. I don't know about the IF-C01PNG as I haven't had the chance to use one, but the 61D has comb filters that yield very impressive results. And I'm sure there can be other differences in the pictured compared to using an external RGB/YPbPr conversion.

[skum]

Well the internal cards would only feature comb filtering if it was actually implemented as they all convert to YPbPr in the end, thus it's done before "hitting" the monitor itself. It almost looks like the ADV7802 would be an "easy" solution to use, along with some DACs...

[RGB-A]

I can confirm the operation of PC ENGINE with an external RGB Boost PCB with RGB cable of Genesis 2 in the BVM-A with BKM-68X.

The console is directly connected to Sync Strike without any extra device.

[nissling]

The biggest oddity I've encountered is DoDonPachi core which runs at some 57.2Hz VSYNC. On the original, the image rolls, but the OSD is steady. On mine, the picture is steady but OSD rolls. I suspect this is because the original ends up thinking its a 50Hz sync, while mine ends up reporting 60Hz.

I've had this very same experience on my BVM-A24 when playing Pier Solar on Dreamcast in its native widescreen mode. It outputs an oddball resolution at 416p and while the image is stable, OSD just rolls. Resolution is recognized as 480/60p by the monitor so I suspect the OSD is displayed with a certain scanline interwall and when the resolutions don't match yet the image syncs, this happens.

Does the PC-Engine normally really have sync issues on the A-series? I've got a Core Grafx 2 with internal RGB mod (THS7374) and it have always synced perfectly on my A24 with 68X with no other hardware in use, yet the D24 and OSSC did get occasional drop outs with that console.

[PeterWar]

The biggest oddity I've encountered is DoDonPachi core which runs at some 57.2Hz VSYNC. On the original, the image rolls, but the OSD is steady. On mine, the picture is steady but OSD rolls. I suspect this is because the original ends up thinking its a 50Hz sync, while mine ends up reporting 60Hz.

I've had this very same experience on my BVM-A24 when playing Pier Solar on Dreamcast in its native widescreen mode. It outputs an oddball resolution at 416p and while the image is stable, OSD just rolls. Resolution is recognized as 480/60p by the monitor so I suspect the OSD is displayed with a certain scanline interwall and when the resolutions don't match yet the image syncs, this happens.

Does the PC-Engine normally really have sync issues on the A-series? I've got a Core Grafx 2 with internal RGB mod (THS7374) and it have always synced perfectly on my A24 with 68X with no other hardware in use, yet the D24 and OSSC did get occasional drop outs with that console.

My PC Engine Duo RX works just fine on the BVM-A20F1M+BKM-68x

[RGB-A]

Great then. I thought I spent both in the D series and in Serie A. Buy the RGB recently for PC Engine and gave me joy to see its perfect performance.

Is there any simple solution without depending on an extron to remove the bias of the top image on the Master System executed in Genesis?

[Fudoh]

Is there any simple solution without depending on an extron to remove the bias of the top image on the Master System executed in Genesis?

I had the impression that the solution posted a little while ago solved the issue (or does this only work for SMS, but not MS running through a MD base system?) viewtopic.php?p=1429842#p1429842

Should be pretty easy to put a stripper and a sync combiner onto a small single board.

[RGB0b]

Should be pretty easy to put a stripper and a sync combiner onto a small single board.

My only concern with that is safety: If you take a Genesis RGB cable that sync's on composite video, it'll probably be fine. One that sync's on csync should be fine too, as long as it's got the right resistor in it.

But what does the voltage look like, if the csync cable doesn't have the resistor on the sync line? Or what if it's one of those older cables with the circuits in them? Or what if people run it through a switch with a sync stripper, into this device? Or those GrafxBooster knockoffs that all have the wrong circuit on them?

Questions like this are the only reason we haven't slapped together a new box with SCART in and RGBs-out labeled "For 68x Only": I want to make sure there isn't a scenario that we're sending crazy voltage to a $2000+ card. I'd even be okay if it was unsafe for some monitors, but totally safe for the 68x...but we need to confirm that.

I'm certainly not criticizing the suggestion, just explaining why we haven't already made one like that. Also, we're only a few days away from something much cooler

[daty2k1]

Also, we're only a few days away from something much cooler

Can't wait. My monitor has been collecting dust since I sold my card...

[RGB0b]

Here you go, the moment many of you were waiting for is here: https://www.retrorgb.com/reverse-engine ... m-68x.html

[daty2k1]

Excellent work ! I see several revisions mentioned. Do you think there will be more in the future ? Is the current one firmware updatable ? Thanks

[skum]

Excellent work ! I see several revisions mentioned. Do you think there will be more in the future ? Is the current one firmware updatable ? Thanks

Most of these revisions are my internal ones. The production one will be the one which contains all the things we've found during testing (which actually isnt much, but still issues that should be resolved). There are of course no guarantees, but i will try to backport fixes if possible through firmware upgrades and add-on boards.

[daty2k1]

Excellent, thank you. And congratulations for this, it's an incredible achievement.

[PeterWar]

Just bought one even though I already own a BKM-68x. I really want to contribute on paying for the excelent R&D effort and will be happy to betatest for the community.

[PeterWar]

Apparently the BKM-68x replacement boards got delayed again, the wait is taking quite long.

[daty2k1]

Yes, targeting mid March now...

[PeterWar]

After many LONG months of wait, I'm glad to say the BKM 68X replacement board by Castlemania games has arrived, here are some results after quick testing:

The good:
The replacement board has fixed my Neo Geo AES (3-5 board) sync issues, now I can play using RGB.
It has also fixed sync issues with my RGB moded Atari 2600.
It seems to play 60hz Japanese games fine in my RGB moded NUS-001FR N64.
It has fixed 60hz gamemode on my switchless moded 50hz Saturn.

The bad:
It has not fixed my sync issues with my 50/60Hz switchless modded Master System II
It has improved Master system game sync on my unmoded 60hz Japanese Megadrive 1+everdrive, some sync issues remain.
It has not fixed my sync issues with my 50/60Hz switchless modded Megadrive 1.

The ugly:
It has added sync issues to Megadrive and mega cd games on my unmoded 60hz Megadrive 1, Sonic 1 gets out of sync.
It has added sync issues when playing Sega Dreamcast with PACKAPUNCH RGB 480p SCART cable, 15KHz/576i mode remains OK.

The neutral:
Playstation, PCE, SNES, NES

So right now, I don't see this as a replacement of the BKM-68x, but rather as a complement to the BKM-68x, with this card and the 68x I am now able to play virtually all consoles except some Master System games.

[skum]

After many LONG months of wait, I'm glad to say the BKM 68X replacement board by Castlemania games has arrived, here are some results after quick testing:

The good:
The replacement board has fixed my Neo Geo AES (3-5 board) sync issues, now I can play using RGB.
It has also fixed sync issues with my RGB moded Atari 2600.
It seems to play 60hz Japanese games fine in my RGB moded NUS-001FR N64.
It has fixed 60hz gamemode on my switchless moded 50hz Saturn.

The bad:
It has not fixed my sync issues with my 50/60Hz switchless modded Master System II
It has improved Master system game sync on my unmoded 60hz Japanese Megadrive 1+everdrive, some sync issues remain.
It has not fixed my sync issues with my 50/60Hz switchless modded Megadrive 1.

The ugly:
It has added sync issues to Megadrive and mega cd games on my unmoded 60hz Megadrive 1, Sonic 1 gets out of sync.
It has added sync issues when playing Sega Dreamcast with PACKAPUNCH RGB 480p SCART cable, 15KHz/576i mode remains OK.

The neutral:
Playstation, PCE, SNES, NES

So right now, I don't see this as a replacement of the BKM-68x, but rather as a complement to the BKM-68x, with this card and the 68x I am now able to play virtually all consoles except some Master System games.

Have you tried switching off 75 Ohm termination on the sync line? SEGA consoles seemingly have atrocious signal-to-noise ratio, which can trigger the ISL59885 sync separator I use on the board (and since such is present, any pre-monitor sync separators should be avoided seemingly). Anyways, what I've noted, is that if the sync termination is removed, the SnR goes up dramatically, especially on SEGA consoles, which seems to eliminate further issues. How big a problem it is seemingly varies with console revision and of course caps age and whatnot... Also keeping it off doesn't seem to cause issues with other consoles...

During development of the card, when I first hooked up SMS, it sync'ed like shit. I found out my cable wasnt wired for CSYNC, so the sync was CVBS. I then added the color filter mentioned in the ISL59885 datasheet, and it suddenly worked. Running without the color filter (and using CSYNC) have always worked on all my SEGA consoles (SMS native and modded, Genesis and Mega Drive 1/2)

[daty2k1]

On my side, it has fixed every single issue I had with the original card.

The only downside is the Dreamcast in 480p as mentioned. The picture is warped at the top.

Right now the output is connected to a Retrotink 5X.

[skum]

On my side, it has fixed every single issue I had with the original card.

The only downside is the Dreamcast in 480p as mentioned. The picture is warped at the top.

Right now the output is connected to a Retrotink 5X.

Dreamcast 480p seems to have two possible fixes, at least that I know of: The RetroAccess cable with some active sync magic circuitry or the discrete V-Sync add-on board for the 68X card (and then run RGBHV).

[PeterWar]

Have you tried switching off 75 Ohm termination on the sync line? SEGA consoles seemingly have atrocious signal-to-noise ratio, which can trigger the ISL59885 sync separator I use on the board (and since such is present, any pre-monitor sync separators should be avoided seemingly). Anyways, what I've noted, is that if the sync termination is removed, the SnR goes up dramatically, especially on SEGA consoles, which seems to eliminate further issues. How big a problem it is seemingly varies with console revision and of course caps age and whatnot... Also keeping it off doesn't seem to cause issues with other consoles...

During development of the card, when I first hooked up SMS, it sync'ed like shit. I found out my cable wasnt wired for CSYNC, so the sync was CVBS. I then added the color filter mentioned in the ISL59885 datasheet, and it suddenly worked. Running without the color filter (and using CSYNC) have always worked on all my SEGA consoles (SMS native and modded, Genesis and Mega Drive 1/2)

Thanks a lot for your prompt reply! and congratulations for designing this great card, I've tried removing the 75 Ohm termination on the sync line, this has made matters a bit better with my 60HZ japanese Megadrive 1, as I can now play megadrive games and everdrive master system games without major sync issues, but the picture looks skewed on Sonic 1 loading screens both in the Megadrive and the Master System versions. It is important to note that this only works while Sync regeneration on my gscartsw is turned ON.

[skum]

Thanks a lot for your prompt reply! and congratulations for designing this great card, I've tried removing the 75 Ohm termination on the sync line, this has made matters a bit better with my 60HZ japanese Megadrive 1, as I can now play megadrive games and everdrive master system games without major sync issues, but the picture looks skewed on Sonic 1 loading screens both in the Megadrive and the Master System versions. It is important to note that this only works while Sync regeneration on my gscartsw is turned ON.

Weird that sync regeneration should be needed. I have no such units and have zero issues of those sorts...