XRGB-mini Framemeister

[RGB32E]

Also, the audio signals are routed over a shielded twisted pair (L and R are twisted together), so you loose some channel separation.

How do you avoid this then?

By building your own, buy the official cable that uses mini-coax (SFC and PS1), or wait a few weeks for some new surprises!

Then again, you can always build something better yourself! Especially if you want to connect directly to the 8MDIN RGB input without adapters or switches.

That is actually an amazing idea. I am more than capable of making my own cables, but I have no idea where to buy the parts (cable/8pin DIN connector/snes multi-out connector)

A number of the various proprietary connectors can be purchased from Assembler Games user Helder, or from Tim Worthington's shop.

My favorite 8MDIN plug when building RGB direct cables for the XRGB-mini is the CalRad 30-505-8. I'm surprised neither RGC or RCA use this to terminate their 8MDIN cables adapters!? Perhaps it's too expensive.

There are a couple of Belden models I really like that work really well despite not being mini-coax. When you use thicker insulation and conductors, and not the 28 AWG (practically Kynar!) cables RCA uses on their standard cables, you can avoid the need to use mini coax cable, as you don't get cross talk artifacts like checker-boarding on RGB video or buzz on audio.

[CkRtech]

I would also encourage you to make your own cables because it really isn't that difficult, and you have full quality control over the materials and the build.

I've cannibalized a couple of SNES composite cables in the past in order to build a SNES RGB cable, but I suppose I will go the Helder route in the future. Of course if you find yourself with a bunch of extra composite video cables taking up space, you can always repurpose the multiout connector (in the case of Nintendo multiouts like SNES/N64) and add your own wire.

Grab some heat shrink to help protect your joints from each other and make them clean.

[Ripthorn]

Also, the audio signals are routed over a shielded twisted pair (L and R are twisted together), so you loose some channel separation.

How do you avoid this then?

By building your own, buy the official cable that uses mini-coax (SFC and PS1), or wait a few weeks for some new surprises!

Then again, you can always build something better yourself! Especially if you want to connect directly to the 8MDIN RGB input without adapters or switches.

That is actually an amazing idea. I am more than capable of making my own cables, but I have no idea where to buy the parts (cable/8pin DIN connector/snes multi-out connector)

A number of the various proprietary connectors can be purchased from Assembler Games user Helder, or from Tim Worthington's shop.

My favorite 8MDIN plug when building RGB direct cables for the XRGB-mini is the CalRad 30-505-8. I'm surprised neither RGC or RCA use this to terminate their 8MDIN cables adapters!? Perhaps it's too expensive.

There are a couple of Belden models I really like that work really well despite not being mini-coax. When you use thicker insulation and conductors, and not the 28 AWG (practically Kynar!) cables RCA uses on their standard cables, you can avoid the need to use mini coax cable, as you don't get cross talk artifacts like checker-boarding on RGB video or buzz on audio.

Several scarts cables uses capacitors or other components on scart connector, If I'm gonna use a mini din 8, where I would put these components?

[RGB32E]

Several scarts cables uses capacitors or other components on scart connector, If I'm gonna use a mini din 8, where I would put these components?

That's always a tricky one. Generally speaking you can use surface mount tantalum capacitors instead of electrolytic to resolve the space constraint, and can make building 21 pin cables easier too! Cable builders with shops generally won't use tantalums due to them being significantly more expensive. However, if you're building a cable for yourself, the price difference is negligible.

For the SNES, 3 SMD tantalums will fit in a SNES plug. Just heat shrink the cap and you're good to go (just like CkRtech mentioned).

For Genesis 2 cables, you can put the resistors in the 9MDIN plug (works with the ones Tim sells), and use SMD tantalums for the caps. I did this with a G2 cable a couple of months ago and the audio quality using "quality" double shielded multi-conductor cable is excellent. Though, installing a Mega Amp was responsible for most of the improvement in audio quality, but the wrong cable can degrade it!

[Ripthorn]

Several scarts cables uses capacitors or other components on scart connector, If I'm gonna use a mini din 8, where I would put these components?

That's always a tricky one. Generally speaking you can use surface mount tantalum capacitors instead of electrolytic to resolve the space constraint, and can make building 21 pin cables easier too! Cable builders with shops generally won't use tantalums due to them being significantly more expensive. However, if you're building a cable for yourself, the price difference is negligible.

For the SNES, 3 SMD tantalums will fit in a SNES plug. Just heat shrink the cap and you're good to go (just like CkRtech mentioned).

For Genesis 2 cables, you can put the resistors in the 9MDIN plug (works with the ones Tim sells), and use SMD tantalums for the caps. I did this with a G2 cable a couple of months ago and the audio quality using "quality" double shielded multi-conductor cable is excellent. Though, installing a Mega Amp was responsible for most of the improvement in audio quality, but the wrong cable can degrade it!

What type of wire you recommend to build the RGB cable with a mini din 8 connector? Also, "heat shrink the cap" what does it mean?
Regarding the tantalum capacitor, I saw several with the same μF but with different voltage values, does it matter?

[FBX]

I just did a video capture recording of my SNES using the RGB upgraded cable offered by Retro_Console_Accessories, and then a 2nd video capture using the official S-Video cables from Nintendo. Not only was there ZERO loss of channel separation on the RGB upgraded cable, but it actually outperformed the Official S-Video cable in terms of video cross-talk shielding on those audio channels. Here's the download to listen for yourself (and more importantly decide for yourself instead of letting someone try to scare you away from getting the cable):

http://filetrip.net/dl?RdKd4x0vSZ

Notice when Link's house loads into view, you can actually hear significantly more video cross-talk in the audio on the S-Video cable than you can on the RGB upgraded cable. These comparison videos are also a good demonstration of picture quality going from S-Video to RGB, so that's an added bonus.

[viletim]

Several scarts cables uses capacitors or other components on scart connector, If I'm gonna use a mini din 8, where I would put these components?

Resistors connected to the video signal (either series or parallel) should always go into the console end of the cable. Series capacitors can go anywhere they will fit.

[CkRtech]

I just did a video capture recording of my SNES using the RGB upgraded cable offered by Retro_Console_Accessories, and then a 2nd video capture using the official S-Video cables from Nintendo. Not only was there ZERO loss of channel separation on the RGB upgraded cable,

TL;DR: Your recordings actually prove RGB32E's statements RE: crosstalk correct...and also aren't good examples of channel separation loss.

Long version -

Regardless of technical side of the argument, I am not certain that one can conclude that there is zero loss of channel separation using these scenes from LttP as a demo as the sounds are most likely not designed to be 100% isolated in the first place. It would probably be best to use something like a SNES Test cart (if it offers isolated sound tests) or some other cart that will do something along those lines in order to check for stereo crosstalk.

I say that because you can definitely hear the sword in the opposite side channel from the side of the screen from which it was used. If those sounds were supposed to be 100% isolated (which really isn't good sound mixing, imho), then there should not be any sound of a sword in the opposite channel. You could argue that Link can't technically stand 100% left or right on the screen, but I digress...

Rather than compare the Nintendo S-Video cable's audio to the RetroConAccess cable's audio using Link to the Past, the best demo would probably be to compare (if you are doing it by human ear alone - which you could argue is the only relevant thing) using the following:

1: A game or program that allows you to pan 100% and/or test only a single channel
2: Record the analog audio output using said test
3: Record the digital audio output using said test (for reference purposes - could be audio has "issues" before the DAC...however unlikely <shrug>)

It is quite possible that the analog output of the SNES has issues somewhere along the path from the DAC to the multiout - in which case the external cables used in the equation are irrelevant.

but it actually outperformed the Official S-Video cable in terms of video cross-talk shielding on those audio channels.

It shouldn't be a matter of one being better than the other. There shouldn't be any video noise in the audio. There is noise present in the audio of both recordings, and that is enough to argue that shooting for something better is worth attempting.

Also note that if what RGB32E said about the construction method of the pro cables is true and the audio L/R from the SNES is transported in twisted pair...that is actually improper design for unbalanced audio (at least from what I have read in the past).

I went the digital route with SNES, so I personally can't offer any of my custom cable work of analog audio cables for SNES as an example by providing a recording. Again, it is possible that there are issues with the audio prior to the cable inside the SNES.

[viletim]

There are a couple of Belden models I really like that work really well despite not being mini-coax. When you use thicker insulation and conductors, and not the 28 AWG (practically Kynar!) cables RCA uses on their standard cables, you can avoid the need to use mini coax cable, as you don't get cross talk artifacts like checker-boarding on RGB video or buzz on audio.

All you need to do is space the conductors apart a bit, especially between the audio and video signals.



Example A is the most common way. Use a 7 or 8 conductor shielded cable and just go for it. It's not ideal because there's capacitive coupling between the audio signal (AR, AL) and the video signals running through the adjacent conductors.

Example B gives a much better result. Use a shielded cable with some extra conductors. Connect them to ground (both ends) chose the ground conductors so it effectively divides the cable in half. Audio on one half, video on the other. A power supply rail will work as good as a ground connection under these circumstances (it's decoupled by capacitors inside the console). Now there is a considerable distance between the audio and video signals with negligible coupling between the two.

[RGB32E]

Tim, while it may seem at first like a contradiction to your examples, the thick insulated 24 AWG Belden 9932 and 9933 actually work extremely well in practice. The 9932 fits the 9 MDIN plugs you sell perfectly without modification.


Example wiring with Belden 9932. Ground on the center conductor, and green video next to red and sync.

Compared to the 22AWG JSC Cable 6663, the Belden 9932 uses much thicker insulation for each 24 AWG conductor.

JSC on the left, Belden on the right.

Comparing the 9932 to a 15 conductor cable using 28/30 AWG conductors with multiple ground conductors like the one RCA uses, you still end up getting noticeable cross talk on the 15 conductor cable (buzzing on the audio in during bright images and high volume levels), but not with the 9932! So, I've found that yes, arrangement matters, but so does conductor size and insulation thickness/material. Your examples are a fair comparison if the conductors and insulation are the same between the two examples, but things change when the conductors and insulation are thicker in practice.

Your suggestion to put the 75 ohm resistors on the 9 MDIN plug end for G2/MD2 RGB cables works wonders on reducing (eliminating?) RGB on to audio cross talk, even when using 9932 cable stock!

[FBX]

TL;DR: Your recordings actually prove RGB32E's statements RE: crosstalk correct...and also aren't good examples of channel separation loss.

Long version -

Regardless of technical side of the argument, I am not certain that one can conclude that there is zero loss of channel separation using these scenes from LttP as a demo as the sounds are most likely not designed to be 100% isolated in the first place.

My videos prove the opposite of RGB32E's statement. He tried to claim there was loss of stereo channel separation MORE SO than usual because of the design of the cable. My videos show there is the exact same amount of separation as the official S-Video cables, but with the added benefit of less video noise as well.

Case in point: I made an audio recording of each video demonstration that isolates the opposite channel from Link's position. In BOTH the S-Video and the Retro_Console_Accessories cables, the audio level of the sword beam was the same low level of loudness compared to the side Link is on. Keep in mind the SNES has a 256-point assignable position of any given sound effect or music note, and likely the programmers did not assign the sword beam to a max position (probably something like 16 left and 240 right). This is why the isolated opposite channel still plays the sound at a soft volume. This can be confirmed using an emulator since any notion of cross-talk is removed from the equation.

In summation: There is nothing to fear about Retro_Console_Accessories coaxial cables being any worse of quality than any other cable offered. In fact, they are universally praised for being of superior quality.

[CkRtech]

In BOTH the S-Video and the Retro_Console_Accessories cables, the audio level of the sword beam was the same low level of loudness compared to the side Link is on.

I think there are too many subjective items in this test. Too many variables and not enough controls. I don't think we can draw any conclusions from it at all.

In summation: There is nothing to fear about Retro_Console_Accessories coaxial cables being any worse of quality than any other cable offered. In fact, they are universally praised for being of superior quality.

"Superior quality" relatively speaking - essentially they are typically considered the lesser of two evils when compared with retro gaming cables (at least from what I've read in the past).

People praise them as being the best, but I think most consumers purchasing RGB cables only care that they get RGB with a sync signal that works with their display without any noticeable artifacts. As soon as that requirement is met, any issue with audio interference from the video is shrugged off.

Even better cables can be made than are currently provided from the major Internet sellers, but I think most people consider Retro Console Accessories to be good enough for their needs despite the flaws. Personally, I make my own & encourage people with the skills to do the same. It allows for a lot more flexibility as well as quality control.

[Smashbro29]

Even better cables can be made than are currently provided from the major Internet sellers, but I think most people consider Retro Console Accessories to be good enough for their needs despite the flaws. Personally, I make my own & encourage people with the skills to do the same. It allows for a lot more flexibility as well as quality control.

How much would it cost to make these top tier cables? I do not at all have the know how but it would be cool to have.

[FBX]

This will be my last word on the subject of the 'audio concerns' about Retro_Console_Accessories' work:

My point was simply to illustrate that RGB32E's scare tactic of chasing away potential customers of a great seller in the hobby was blowing things way out of proportion. As I said before, any actual measurable 'loss of channel separation' would be undetectable to the human ear. As such that people need not worry about finding themselves with inferior sound quality as a result of buying said cables. The whole argument is similar to "oh you better buy those gold-plated Monster cables, or your sound quality will be shite", when in reality, you wouldn't be able to tell a damn bit of difference when tasked in a blind test comparison. My videos speak for themselves in illustrating exactly this concept. Watch them and decide for yourselves just how 'big' the difference is.

Feel free to rip apart this post with whatever technicalities you can come up with, but I'm done debating it because it's quite frankly over nothing that should worry anyone in the first place.

[rtw]

As already hinted, this is all too subjective. The only way we can determine this is by dedicated test signals and scope measurements.

I would like to know if the "noise" emitted when the video signal is saturated is really common mode ? In that case we can get rid of it with a choke.

Can someone explain what the benefit is of 75 ohms resistors on the RGB signals at the start of the signal path ? Impedance matching ?

And the capacitors which I see people soldering inside their SCART contacts. What is their function and why is 220uF the selected value ?

Should the shield in the cable only be terminated at one end ?

There is too much superstition about all of this, we need definite answers

[RGB32E]

Can someone explain what the benefit is of 75 ohms resistors on the RGB signals at the start of the signal path ? Impedance matching ?

Here's the explanation. I've tested and confirmed that it is beneficial place the resistors at the Genesis side when building cables. I believe it is something that should be considered standard moving forward.

Now for the technical bit:

The video encoder in the Master System and Mega Drive can drive the video line directly with the addition of a 75 ohm series resistor and coupling capacitor (usually 220u but the value is not critical). The resistor serves two purposes. First is to match the source impedance with the load so there are minimal reflections along the transmission line (cable), second it isolates the low impedance output of the video driver from the cable capacitance. The capacitance of a long cables can be enough to make the video encoder oscillate. This is a characteristic of all high speed voltage feedback amplifiers. Capacitive loads are not allowed. You can find warnings in the better video encoder datasheets about this.

To make a passive RGB cable for these consoles you need to put the resistors into the console end not the SCART end. Even if you don't care for good quality video, the video signal couples into the audio signal worse if the resistor is placed at the SCART end. The capacitor can at either end and it's only blocking DC. If you leave it out (like Sega did), it forces the video driver (in the video encoder chip) into class A mode (instead of class AB) and it makes the video encoder get very hot.

And the capacitors which I see people soldering inside their SCART contacts. What is their function and why is 220uF the selected value ?

To remove the DC offset present, which results in AC coupled video. For example, if you don't include capacitors in a NTSC SNES RGB cable, the black level will be elevated in most cases and give a too bright/washed out image. You can search the terms for more info, or look at any of the RGB encoder datasheets for more information.

Should the shield in the cable only be terminated at one end ?

Are you referring to floating the shield? There are some audio applications where this can be useful, but I haven't encountered any benefit for doing so when it comes to game cables.

There is too much superstition about all of this, we need definite answers

Certainly! Making an objective statement about using a STP to transmit stereo audio not being ideal has resulted in hyper defensive posts and other shenanigans.

[RGB32E]

why is 220uF the selected value ?

Tim addressed this question a few weeks short of a decade ago over on the nfggames forum.

http://nfggames.com/forum2/index.php?PH ... 6#msg11546

[austin532]

RGB32E, did you change your mind about calling FBX the class clown of the forum?

[AndehX]

It's been a week now and my XRGB has only just arrived in customs (according to the tracking)
I hope it won't be much longer now for delivery, and I am praying to the higher powers that I do t get hit with an outrageous customs charge.

[FinalBaton]

Been meaning to ask something :
does RGB32E sell stuff or is he affilliated in some way with a seller?

[RGB32E]

Been meaning to ask something :
does RGB32E sell stuff or is he affilliated in some way with a seller?

No affiliations, and not actively selling. PM me if you have questions.

[rtw]

Thank you for your detailed answers RGB32E

I am starting to understand the background behind some of these designs.

However I still have a few questions if you could bear with me

It seems like each console in reality should have it's own cable. Is this a correct observation ?

Should the cables be changed for the FrameMeister ? I.e. is the input impedance of the FM the same as a SCART input ?

Yes I was referring to "floating the shield", i.e. shield only connected to ground on the console side.

Where is the easiest place to buy: 24 AWG Belden 9932 ?

[keropi]

interesting read, I'll have to check my sms1/md1 scart cables now

[BuckoA51]

I just got the first batch of Genesis/Megadrive RGB bypass amps in and for that you actually have to remove the caps from the SCART cable. Not sure if the resistors will cause a problem in this setup as I'm really not that technically minded, will try and find out for you all.

[keropi]

^ I missed them completely, where can I find more info on these amps?

[BuckoA51]

^ I missed them completely, where can I find more info on these amps?

Here:- http://www.videogameperfection.com/2015 ... ss-boards/
and here:- http://www.retrorgb.com/genesisrgbbypass.html

[keropi]

^ thanks

[Moonwalker]

A lot of problems solved, time for a new one.

I got myself a Carrier Airwing CPS 1 board which I cant get working, the sync will not workout no matter what I do. There is a graphics glitch when the portraits in the intro show up, so maybe the board is damaged? Rest looks ok though but is unplayable because of the sync issue. I tried everything in the sync menu and played around with the 5v which influenced the sync. Anyway, it will not work out. Are there other settings which I could try?

Sorry for posting many posts in a row, but they are totally unrelated.

Sorry for asking again...

[RGB32E]

I just got the first batch of Genesis/Megadrive RGB bypass amps in and for that you actually have to remove the caps from the SCART cable. Not sure if the resistors will cause a problem in this setup as I'm really not that technically minded, will try and find out for you all.

http://shmups.system11.org/viewtopic.ph ... 70#p809270 BuckoA51, you even posted right after Konsolkongen! Quite the coincidence.

What is insufficient about removing the subcarrier connection from the VDP to the CXA? It's a much simpler and effective approach. Blaming the CXA1145M for the jail-bars is misguided as it's possible to get really clean RGB output from this IC. CXA1145 pin 11 can be utilized for 75 ohm CSYNC output, which is handy for displays and the XRGB-mini that are 75 ohm terminated. The 2.2k pull up resistor through 220uF and 75 ohm resistor is an interesting sync circuit idea. Wonder what Tim has to say about that design?

A lot of problems solved, time for a new one.

I got myself a Carrier Airwing CPS 1 board which I cant get working, the sync will not workout no matter what I do. There is a graphics glitch when the portraits in the intro show up, so maybe the board is damaged? Rest looks ok though but is unplayable because of the sync issue. I tried everything in the sync menu and played around with the 5v which influenced the sync. Anyway, it will not work out. Are there other settings which I could try?

Sorry for posting many posts in a row, but they are totally unrelated.

Sorry for asking again...

Do you have any other CPS1 boards you could try to identify if this is specific to that particular PCB? Are you using a Sigma supergun? Those have circuitry that has been reported by several users to cause incompatibility issues with the XRGB-mini. A direct CSYNC connection to the XRGB mini works fine from CPS1 boards. Of course, you'd need series resistors on the RGB lines when connecting directly from the JAMMA harness to the XRGB-mini.

Thank you for your detailed answers RGB32E I am starting to understand the background behind some of these designs. However I still have a few questions if you could bear with me

No problem!

It seems like each console in reality should have it's own cable. Is this a correct observation?

Not necessarily? Only if the internal circuitry is different between consoles that share the same output connection, and/or you don't want to swap cables from console to console.

Should the cables be changed for the FrameMeister? I.e. is the input impedance of the FM the same as a SCART input?

In what way? The red, green, blue, and sync input pins on the mini are 75 ohm terminated (75 ohm resistor to ground on input).

Yes I was referring to "floating the shield", i.e. shield only connected to ground on the console side.

Only if there's a ground loop you're trying to resolve. As mentioned before, haven't encountered that scenario, so both ends are connected to shield/ground when I build cables.

Where is the easiest place to buy: 24 AWG Belden 9932 ?

Not familiar with the options to Norway. You could try ordering from Mouser? The 9933 provides an extra conductor that can be used for 5VDC if you require switching voltage for SCART, or integrating a LM1881/ect.

[cfx]

.