Neo Geo AES plus mods | A Treble Project

[Tomdominer]

Did my own mod of the Neo Geo joystick, uses 24mm black buttons with replacement coloured tops.

 

 

[Treble]

Poor old @Tomdominerhas landed in a pickle with the AES I supplied for him

 

Basically, some models of the AES don't play well with upscalers. It seems to be focussed around the sync, as these are quite well-exposed in the menu of the OSSC and tweaking settings solves it. I can say that with confidence, as my mate Mike got an AES off me (3-4 board) with the RGB bypass and applied these tweaks:

Spoiler

Sync OPT. > H-PLL Post-Coast 2 lines
Sync OPT. > Hsync Tolerance 8.30us

 

...and it's worked perfectly ever since. Mine worked perfectly right out of the gate (3-6 board, RGB mod).

 

Tom's got a Retrotink 5X and it's not playing nicely, which is a known issue. Mike Chi - the developer of the Retrotink - hasn't got a firmware fix yet. Hopefully he will, but it may need a revised device, possibly, with larger amount of control over the sync signal.

 

Some people have had luck changing the capacitance on the sync signal line, which is definitely worth a shot. If I had a 5X to test, I'd give it a try with a straight replacement like-for-like, then with different values until or if it produced a stable image. 

 

Anyway, I say all that to say this: I've created a chart to help understand which Neo Geo AES signals you can use with which devices. It's intended as a living document, so please comment against the Google Sheet if you have any input! I'm not expert and am more than willing to be educated

 

Click the image to be taken to the full chart inc. comments.

 

 

[Tomdominer]

From my limited reading, I think it's basically a crapshoot. The board revisions, combined with the variance or components used, means that you can't guarantee it'll work with the Tink without issues. I'm gonna try some minor revisions to the mods that Treble did in an attempt to fix the issues, and report back.

 

Basically fair warning here that you just don't know if a RGB modded Neo Geo AES console will play nice with the Retro Tink 5X. It might, it might not....

[Treble]

Yeah but just to be clear, I don't think it's the RGB bypass itself that causes an issue with the Retrotink. I think the stock signal is just as vulnerable

 

On the bright side, if it's just capacitor values we can share the details with the community (and Mike Chi) and help folks out

[AceGrace]

I must have got lucky because my AES and CD work fine with the RetroTink 5x Pro.

 

I didn't even know there was a problem.

[Treble]

2 hours ago, AceGrace said:

I must have got lucky because my AES and CD work fine with the RetroTink 5x Pro.

 

I didn't even know there was a problem.

 

Any idea which board revision you have? 

[AceGrace]

Not offhand. Would you be able to tell from the serial number?

[Treble]

9 hours ago, AceGrace said:

Not offhand. Would you be able to tell from the serial number?

 

Hmm, not really. Generally speaking the later boards will be somewhere between 3-4 and 3-6, ie serial numbers over 100K (source: fiddling with about 8 of these things over time!) but as with most electronics stuff, I think they probably chucked in whatever they had stock of at the time. 

 

Most board revisions are for consolidation and cost reduction, but if any earlier board is still stockpiled in a warehouse, they'll go through that backlog before adding a newer model. 

 

Anyway, it's not that important, I was just curious

[Tomdominer]

11 hours ago, AceGrace said:

Not offhand. Would you be able to tell from the serial number?

 

You can see it through the cartridge slot. Its around here on the board. Use a flash light (I used my mobile phone one) to read it.

 

 

http://www.jamma-nation-x.com/jammax/tutorials.html?show=aesrevision

[Treble]

l never thought to do that 😅

[phillv85]

Neither of my AES consoles (unmodded) play nice with my Sony PVMs. They’re fine over composite, but connect them to a BNC to SCART with its own sync and it won’t give a stable picture over RGB.

[Underwurlde]

Is there any point in doing the RGB bypass mod if you've already done this mod:

https://mmmonkey.co.uk/snk-neo-geo-aes-rgb-neo-aes3-6-vertical-lines/

 

Btw, the issue with the Retrotink 5x isn't soley related to the 3-6 revision.

My Neo Geo Mvs MV-1F also displays the exact same problem.

The Retrotink is the common dominator.

[Treble]

On 06/06/2022 at 21:50, Underwurlde said:

Is there any point in doing the RGB bypass mod if you've already done this mod:

https://mmmonkey.co.uk/snk-neo-geo-aes-rgb-neo-aes3-6-vertical-lines/

 

 

I'd have to test it myself (and will! Got two machines on the go as we speak) but I'd guess this is fine and the bypass mod would add very little. As usual, your board revision matters. The 3-6 has one of the worst signals - my daily driver is a 3-6, so I'm speaking from experience - and just about any mod will be an improvement. Short of destroying the circuit completely, it'd be tough to make it any worse

 

Spoiler

Btw, the issue with the Retrotink 5x isn't soley related to the 3-6 revision.

My Neo Geo Mvs MV-1F also displays the exact same problem.

The Retrotink is the common dominator.

 

On no, absolutely! People are trying to track down the exact way to mitigate it, so that Mike Chi can slap some firmware in where you can adjust the right settings. In all honesty, though, I don't think the 5X will ever get there. I expect it'll need a new circuit giving greater control over the sync values. There are so many variables in how the various AES revisions deal with their AV circuit, it's mind-boggling.

[Treble]

On 03/06/2022 at 12:30, phillv85 said:

Neither of my AES consoles (unmodded) play nice with my Sony PVMs. They’re fine over composite, but connect them to a BNC to SCART with its own sync and it won’t give a stable picture over RGB.

 

What are they syncing on? Pin 3 on the AES AV connector is composite video and pin 7 is composite sync. You might want to play with swapping them around, as PVMs may play better with one over t'other.

 

As an aside, the 'bypass' part of the RGB Bypass is a misnomer: when you do the mod, you're actually creating a new circuit. Yes, you 'bypass' what the AES is putting out and connect further down the line, but you also add new resistors, plus capacitance, so you're actually building a far better AV circuit.

 

[Treble]

So, the AES projects continue. I bought two junk ones on spec (I overpaid, it turns out, but that's a story for another time). Neither came with any accessories, so I decided to convert a good-quality stick into one that'd work with an AES.

 

There are loads of reasons for this, but the main ones are that:

i) sticks from the PS3 era onwards started shifting toward an 'as good as the arcade' build, especially from Hori.

ii) these are much better quality than the original AES sticks. Sorry, but it's true.

iii) Sticks were much more common in Japan than in the west at the time (circa mid- to late-2000s) so they're cheap to buy.

 

I settled on the Hori Fighting Stick V3 as it's a good, inexpensive stick with quality components. Aesthetically it's close enough to the Neo Geo 'look' to not frighten the horses as well

 

 

My plan was to make it as modular as possible for a home-baked project. To that end, I decided a panel-mount DB15 connector with a bunch of wires connecting internally to some project board would keep it tidy inside, and a Neo Geo extension lead could then attach/detach as needed on the outside.

 

Click the photos below to be taken to the hi-res Google Photos album.

 

 

Clockwise from top left: image 1 is my Persona stick, which is the same model as the V3. I used it prior to receiving the actual V3s so I could map out the wiring. The second pic is the board with start/select on it, so I traced that out onto the board in pic 3. You can also see that I've marked-up where the buttons and joystick inputs go. The bottom 3 pics show the wired-up DB15s. I had a spare ribbon cable for one, but I wouldn't use that again - I prefer my separate wiring: it's silicone, not plastic, so far more bendy but also more robust. It's good squeezy.

 

The final one are the little snippets of project board I used. This is easier than wiring direct, as you can do it in advance and it's easier to keep track of what goes where.

 

 

Above is the final stretch. That's one of the sets wired into the DB15 and project board, then you're looking at the extension leads. These were almost a CM too deep and wouldn't fit the DB15 connector! I had to cut away and smooth-off that, then also cut out some channels for the mounting screws (not pictured).

 

Finally, you can see the guts at the bottom-left, and the final product next to that. The USB connection isn't pictured, but I added that back - both work perfectly

 

 

[Tomdominer]

Great stuff @Treble. Where did you get the  Hori Fighting Stick V3. Kinda looks perfect for my needs!

[Tomdominer]

On 24/05/2022 at 15:55, Treble said:

Poor old @Tomdominerhas landed in a pickle with the AES I supplied for him

 

Basically, some models of the AES don't play well with upscalers. It seems to be focussed around the sync, as these are quite well-exposed in the menu of the OSSC and tweaking settings solves it. I can say that with confidence, as my mate Mike got an AES off me (3-4 board) with the RGB bypass and applied these tweaks:

  Reveal hidden contents

Sync OPT. > H-PLL Post-Coast 2 lines
Sync OPT. > Hsync Tolerance 8.30us

...and it's worked perfectly ever since. Mine worked perfectly right out of the gate (3-6 board, RGB mod).

 

Tom's got a Retrotink 5X and it's not playing nicely, which is a known issue. Mike Chi - the developer of the Retrotink - hasn't got a firmware fix yet. Hopefully he will, but it may need a revised device, possibly, with larger amount of control over the sync signal.

 

Some people have had luck changing the capacitance on the sync signal line, which is definitely worth a shot. If I had a 5X to test, I'd give it a try with a straight replacement like-for-like, then with different values until or if it produced a stable image. 

 

Anyway, I say all that to say this: I've created a chart to help understand which Neo Geo AES signals you can use with which devices. It's intended as a living document, so please comment against the Google Sheet if you have any input! I'm not expert and am more than willing to be educated

 

So, just to follow up on this. I (finally) picked up a neo geo stereo rgb cable from retro gaming cables, as they were out of stock for months (I was using my mates mono rgb rgc cable which I had borrowed) and the new cable seems to have completely eliminated the noise issue. We aren't sure whether the new geo cables have had a board revision to deal with the issue, or that the mono cable I was using was actually faulty, or perhaps separating the sound output makes it less noisy, or what.

I'm going to post the mono cable back, and my mate is gonna see whether he has issues with it, but I thought it was worth mentioning here, is anyone else is also having issues.

[Treble]

Most likely to be the resistors in the SCART head.