RGB bypass mod

Golgo13

Golgo13

n00b
Joined
Dec 9, 2007
Posts
5
Oh... Now i got it... SCART is the standard connection in Europe and it's RGB, and here in Brazil it's the composite... I thought it was all the same.

Then I think I'm a little alone now. Have to look for some modders here in Brazil to do the s-vid mod.

Oh well,

I would like to thank everybody for the support (chaoticjelly and MKL).

Thanks a lot... really! And excuse me if I bother in some way.
 
MKL

MKL

Basara's Blade Keeper
20 Year Member
Joined
Mar 8, 2002
Posts
3,686
The grid pattern/vertical striping is limited to revision 3-6 because on this revision there is an area called "NTSC only" and the RGB lines are split before they go into the RGB encoder and sent to this area:

http://img215.imageshack.us/img215/9176/aes36be1.jpg

The problem is that these traces run under the clock signal for the encoder and cause some weird interference that shows up as the infamous vertical striping. If these traces, which are totally useless anyway, are cut/lifted from the PCB the striping effect disappears and you get a flawless RGB picture. Earlier revisions don't have these split RGB traces and consequently no striping/checkerboard effect whatsoever:

http://img85.imageshack.us/img85/2975/aes35yl1.jpg

That gamesx article fails to explain the phenomenon and claims that revision 3-5 (the '126000 serial') has a partial striping effect ('checkerboard'), i.e. better RGB than 3-6 but worse than the preceding revisions, which is simply false.

Revisions 3-5 and 3-6 have a brighter RGB output as a result of different values for coupling caps and resistors but this obviously doesn't "dramatically decrease the RGB quality". In fact the new values (470uF/75ohm vs. 100uF/68ohm) are what Sony recommends. In the end the difference between the darker RGB of the earlier revisions (NEO-AES, AES3-3, AES3-4) and the brighter RGB of the later revisions (AES3-5, AES3-6) is trivial and can be compared to a supergun setup with adjustable RGB pots where some will prefer the picture a little darker while others a little brighter.
 
M

mmmonkey

Quiz Detective
Joined
Aug 6, 2008
Posts
83
I know it's resurecting an old thread. But after MKL pointed me at the previous post I tried it out, I reversed the RGB Bypass I had already performed and cut the 3 traces shown in the photo, the image quality is amazing!
 
JLZ

JLZ

Rugal's Thug
Joined
May 8, 2008
Posts
92
I am pretty new to the Neogeo, and I have a few questions.

I have an unmodded Jap AES, serial number of 186476, hooked up to a Samsung LCD TV. When I plug the scart into the 'av' scart socket, I get vertical distortion lines (for example, on the heath bar of most King of Fighters games)

If I plug it into the 'RGB' scart socket, I get a clearer, slightly darker picture, but with moving diagonal lines across the image.

If I perform the RGB bipass mod (which is cutting 3 traces, I believe), will this solve the 2 different types of distortion I get?

Is it true that after I perform the mod I will be unable to connect using a composite cable?

Many thanks
 
tony starks

tony starks

Starks is always out.
Joined
Dec 19, 2006
Posts
751
JLZ said:
I am pretty new to the Neogeo, and I have a few questions.

I have an unmodded Jap AES, serial number of 186476, hooked up to a Samsung LCD TV. When I plug the scart into the 'av' scart socket, I get vertical distortion lines (for example, on the heath bar of most King of Fighters games)

If I plug it into the 'RGB' scart socket, I get a clearer, slightly darker picture, but with moving diagonal lines across the image.

If I perform the RGB bipass mod (which is cutting 3 traces, I believe), will this solve the 2 different types of distortion I get?

Is it true that after I perform the mod I will be unable to connect using a composite cable?

Many thanks
Click to expand...

PM my man ChaoticJelly, he will be able to answer your questions.

Starks out.
 
C

chaoticjelly

Kabuki Klasher
Joined
Sep 30, 2006
Posts
128
JLZ said:
I have an unmodded Jap AES, serial number of 186476, hooked up to a Samsung LCD TV. When I plug the scart into the 'av' scart socket, I get vertical distortion lines (for example, on the heath bar of most King of Fighters games)

If I plug it into the 'RGB' scart socket, I get a clearer, slightly darker picture, but with moving diagonal lines across the image.

If I perform the RGB bipass mod (which is cutting 3 traces, I believe), will this solve the 2 different types of distortion I get?

Is it true that after I perform the mod I will be unable to connect using a composite cable?

Many thanks
Click to expand...

When you say distortion, do you mean wavy interference? or vertical white lines?

Same with RGB, is it actual distortion - wavy interference?

If it's just vertical white lines then RGB bypass should solve the problem.

If it is the picture that's appearing wavy, unfortunately you need to find another TV to play your Neo Geo on.

I have a Samsung LCD too and it does the same thing.

It depends how you do the mod as to whether you can use a composite AV cable afterwards. Besides, if you have an RGB SCART cable that doesn't really matter anyway.

Let me know if you'd like this doing, I've done it for quite a few NG.com members and I can fit one of Razoola's UniBios' and a power LED at the same time if you like.
 
Xian Xi

Xian Xi

JammaNationX,
15 Year Member
Joined
Dec 1, 2005
Posts
27,750
Just like Chaotic said, if the picture is wavy as meaning the picture is creating a wave traveling upward to the right, you need to use a different TV.
 
JLZ

JLZ

Rugal's Thug
Joined
May 8, 2008
Posts
92
In the 'av' socket it creates vertical rainbox coloured lines, but in the 'RGB' scart socket it shows waves traveling upward to the right.

Seems I need a new TV (I guess a CRT).

The strange thing is that this distortion as not always there, but is very noticeable when it is.

Would a mod still improve my picture quality?
 
Neo Alec

Neo Alec

Ned's Ninja Academy Dropout
20 Year Member
Joined
Dec 7, 2000
Posts
11,996
Tip of the day: Always try a different TV before you resort to modding an old console. :)
 
G

gtsamour

n00b
Joined
Oct 28, 2007
Posts
33
After reading everything I could find about the RGB fix, I'm a bit confused and would really appreciate some help...

Here's the deal...
I have an AES3-5 NeoGeo model thus no striping at all.
Although I think my RGB out image is very good (and I know about RGB images...), if it can get better I'd like to do it.

The problem is that I wouldn't want to lose the capability of composite out signal even if i have to tap it from somewhere else on board.
Is there a way to have both? (rgb bypass and composite)?
I also couldn't find any proceedure instructions for the AES3-5 model.

Can someone shed some light please?
 
Last edited:
M

mosdefben09

n00b
Joined
May 28, 2009
Posts
3
MKL said:
The grid pattern/vertical striping is limited to revision 3-6 because on this revision there is an area called "NTSC only" and the RGB lines are split before they go into the RGB encoder and sent to this area:

http://img215.imageshack.us/img215/9176/aes36be1.jpg

The problem is that these traces run under the clock signal for the encoder and cause some weird interference that shows up as the infamous vertical striping. If these traces, which are totally useless anyway, are cut/lifted from the PCB the striping effect disappears and you get a flawless RGB picture. Earlier revisions don't have these split RGB traces and consequently no striping/checkerboard effect whatsoever:

http://img85.imageshack.us/img85/2975/aes35yl1.jpg

That gamesx article fails to explain the phenomenon and claims that revision 3-5 (the '126000 serial') has a partial striping effect ('checkerboard'), i.e. better RGB than 3-6 but worse than the preceding revisions, which is simply false.

Revisions 3-5 and 3-6 have a brighter RGB output as a result of different values for coupling caps and resistors but this obviously doesn't "dramatically decrease the RGB quality". In fact the new values (470uF/75ohm vs. 100uF/68ohm) are what Sony recommends. In the end the difference between the darker RGB of the earlier revisions (NEO-AES, AES3-3, AES3-4) and the brighter RGB of the later revisions (AES3-5, AES3-6) is trivial and can be compared to a supergun setup with adjustable RGB pots where some will prefer the picture a little darker while others a little brighter.
Click to expand...

hello mkl did you be my hero ? i m new menber , and i have some questions for the revised aes 3.5 ,so the threads with explication for rgb by pass is not necessary for this serial ? or not , that technic where cut the line of pcb have work on 3.5 ? i realise a dream to get neo geo , and i don t want kill her , cause i not the expert for manipulate the fer a souder , sorry i don t find the word !thanks to answer me .i am french can you make tuto for mod model 3.5 ? on the rgb high quality or mail bemathis@wanadoo.fr serial number is 112 993 bye
strange effect rainbow with composit cable , and if i chose i prefered the cut technic ? cause i m not a best in manipulate the brasing
 
Last edited:
W

Whitesnake

n00b
Joined
May 2, 2010
Posts
8
MKL said:
The grid pattern/vertical striping is limited to revision 3-6 because on this revision there is an area called "NTSC only" and the RGB lines are split before they go into the RGB encoder and sent to this area:

http://img215.imageshack.us/img215/9176/aes36be1.jpg

The problem is that these traces run under the clock signal for the encoder and cause some weird interference that shows up as the infamous vertical striping. If these traces, which are totally useless anyway, are cut/lifted from the PCB the striping effect disappears and you get a flawless RGB picture. Earlier revisions don't have these split RGB traces and consequently no striping/checkerboard effect whatsoever:

http://img85.imageshack.us/img85/2975/aes35yl1.jpg

That gamesx article fails to explain the phenomenon and claims that revision 3-5 (the '126000 serial') has a partial striping effect ('checkerboard'), i.e. better RGB than 3-6 but worse than the preceding revisions, which is simply false.

Revisions 3-5 and 3-6 have a brighter RGB output as a result of different values for coupling caps and resistors but this obviously doesn't "dramatically decrease the RGB quality". In fact the new values (470uF/75ohm vs. 100uF/68ohm) are what Sony recommends. In the end the difference between the darker RGB of the earlier revisions (NEO-AES, AES3-3, AES3-4) and the brighter RGB of the later revisions (AES3-5, AES3-6) is trivial and can be compared to a supergun setup with adjustable RGB pots where some will prefer the picture a little darker while others a little brighter.
Click to expand...

I know this is an old thread and a Neo Geo forum but I've been experiencing the same issue on my Sega Nomad when using an RGB Scart and I'm guessing it is the same issue with the RGB Encoder.

Does anyone have such a guide to fix this like a simple trace cut like the Neo Geo fix?

I tried to PM MKL but it wouldn't let me and I also tried getting in touch with mmmonkey but his web email is down.

Thanks
 
MKL

MKL

Basara's Blade Keeper
20 Year Member
Joined
Mar 8, 2002
Posts
3,686
Whitesnake said:
I know this is an old thread and a Neo Geo forum but I've been experiencing the same issue on my Sega Nomad when using an RGB Scart and I'm guessing it is the same issue with the RGB Encoder.

Does anyone have such a guide to fix this like a simple trace cut like the Neo Geo fix?
Click to expand...

I have no experience with the console you mention but I find it very unlikely unless this is a widely (or universally if the console doesn't have more than one PCB revision) reported issue (as it is with the neo geo).
 
Xian Xi

Xian Xi

JammaNationX,
15 Year Member
Joined
Dec 1, 2005
Posts
27,750
So you want it spoon-fed to you?

You say it uses a Sony Encoder, open up the Nomad, download the data sheet for the Sony encoder, read the pinout, trace the RGB points, eliminate any components decreasing output. If needed build an RGB amp.

It isn't hard dude.

The RGB bypass mod for the Neo is different. The bypass mod takes the RGBS signals from a stronger source and removes any component that would decrease the signal strength.

If you are talking about the 3-6 trace cut with the Neo and how the Nomad has a similar issue then you need to provide pics of the internals of the Nomad that are high res front and back.
 
RGB32e

RGB32e

n00b
Joined
Dec 8, 2009
Posts
40
Whitesnake said:
Sega Nomad is a portable Sega Mega Drive 2 essentially and from what I know it is quite a well known issue on all Mega Drives. But I've yet to see a definitive fix.

I believe a lot of them also use a Sony RGB Encoder.

Here is two close up pictures from my LCD TV screen.

http://img199.imageshack.us/img199/3755/image078rq.jpg
http://img80.imageshack.us/img80/2424/image076f.jpg
Click to expand...

The jail bars (vertical stripes) on the Genesis 2 can be fixed by buffering the csync output. This is what I use for my Genesis 2 (short lead from the console to this):
fetch.php


The chip on the left is a LM1881 sync separator, which feeds the hex buffer on the right.

On a different topic, it seems the Sega Saturn systems have the same JB problem some AES systems have! From reading this thread, I think the solution for saturn systems is effectively the same as the AES.... remove the video encoder circuit from the equation... and that the subcarrier can cause interference!
 
Xian Xi

Xian Xi

JammaNationX,
15 Year Member
Joined
Dec 1, 2005
Posts
27,750
The JB on the AES isn't remedied by removing the encoder it is by cutting the RGB traces that run under the clock.
 
RGB32e

RGB32e

n00b
Joined
Dec 8, 2009
Posts
40
Xian Xi said:
The JB on the AES isn't remedied by removing the encoder it is by cutting the RGB traces that run under the clock.
Click to expand...

Right. Your picture is missing:
http://img215.imageshack.us/img215/9176/aes36be1.jpg

Do you have a copy to upload again?

Since there are quite a few internal revisions of the Saturn, each is different... but your picture would help me identify the issue on the SS.

I've also noticed that the csync line runs right underneath all of the EMI Ferrite beads (all but EM3 and EM14)... could this cause noise to be picked up?

fetch.php
 
Kiel

Kiel

All About Shooters Mod., If you fuck with me, I'll
20 Year Member
Joined
Oct 3, 2001
Posts
7,804
Bumping for some help.

I have this revision that MKL posted (his pic)
aes35yl1.jpg


Am i correct in assuming that if i cut these traces labeled RGB it will get rid of the checker board pattern my system displays?

Plus I am okay if I can not use composite anymore
 
Last edited:
Xian Xi

Xian Xi

JammaNationX,
15 Year Member
Joined
Dec 1, 2005
Posts
27,750
There is no checkerboard pattern on a 3-5, only 3-6.
 
Xian Xi

Xian Xi

JammaNationX,
15 Year Member
Joined
Dec 1, 2005
Posts
27,750
Did you open it to confirm it's a 3-5? And that just looks like composite noise to me.
 
Kiel

Kiel

All About Shooters Mod., If you fuck with me, I'll
20 Year Member
Joined
Oct 3, 2001
Posts
7,804
Yup, it's opened up right now. 3-5, I have it hooked up with a Japanese rgb cable.

If you want pics of the board I'll be happy to throw em up bit it won't be until midnight tonight.
 
Neo Alec

Neo Alec

Ned's Ninja Academy Dropout
20 Year Member
Joined
Dec 7, 2000
Posts
11,996
Show a black screen if you really want to see it. I could always see it the worst on the Unibios splash screen. If you're not seeing it on a black background, I don't think you've got it.
 
You must log in or register to reply here.
Top