RGB/YPbPr Intellivision Thoughts

[the_crayon_king]

"For optimum performance, the analog outputs should each
have a source termination resistance to ground of 75 Ω (doubly
terminated 75 Ω configuration). This termination resistance
should be as close as possible to the ADV7125 to minimize
reflections."

 

Think about it. If the placement of the series resistor was not important then TVs could just have a 75 ohm series and a 75 ohm termination inside of them and every console/cable could just be a straight through.

 

What I am saying is putting a resistor 3 feet down in the SCART head is probably not "correct".

However, what is best practice and what will work are two very different things.

 

I'm more excited by the news that there is apparently a nice RGB mod for the coleco now.

 

 

@Icelvlan Anyway, if you want to remove those resistors and use ones inside your cable I say go for it. Just do not bridge those pads as they go to ground (because they are not series). You may have already known that but it bears repeating.

 

 

Edited October 29, 2020 by the_crayon_king

[Icelvlan]

So just let us know when your ready to start selling these!

[Icelvlan]

Any progress?

[the_crayon_king]

18 hours ago, Icelvlan said:

Any progress?

 

I still have a great number of parts coming from China and that is what I have been waiting for to do a legitimate batch (of 100). The goal is to have many I can pull from so I don't have to be concerned with putting these together on short notice.

 

Let me put some together so I have an actual count and then I will take some pictures and put together some info.

It looks like I have enough parts to put together 5-10 of these.

[+-^CrossBow^-]

Cool, I'm in for one for now to try and test internally until you have the parts and means for a larger order.

 

Thanks!

 

[the_crayon_king]

This is what you would get as a basic configuration:

 

RGB/YPbPr board

8 pin mini din output with breakout (you can use it for audio out but you probably shouldn't)

2 RCAs for audio.

 

Notice that the spot for capacitors on the output has 0ohms (000) instead of capacitors.

So if you need caps inside of the console (as in you don't have them in your cable just let me know).

Also if you need YPbPr RCAs let me know.

 

PAL/PALSEL sets/toggles between tan and yellowish for the color "tan". Tan is default.

COL/YPbPr sets/toggles between RGB and YPbPr. RGB is default.

SOG/SOG sets/toggles sync on green. No sync on green is default. < you would probably know if you need this.

 

The silkscreen on these boards is terrible and the black color is coming off near some pads (mostly the dac). Also I am hand soldering all of this so forgive the lack of uniformity. These issues are superficial just know that I am aware and can't do anything about it.

 

If you want to order I am going to set the price at 35$ for now (this includes first class shipping/tracking).

Just PM me with:

Email

Mailing Address

Anything you need set (such as if you want the board configured for the Aquarius, desire a toggle switch, YPbPr RCAs, or capacitors on the kit).

 

If you do not specify anything I will assume you want this configured for the Intellivision, without capacitors on the board, and without a toggle switch or YPbPr RCAs.

-------------------------------------------------

On another note if anyone has a Fairchild I could use someone to test a board I have. My Fairchild is no longer working but I am like 85% sure that the kit/code I made would work. You would need a Fairchild and an Oscope at a minimum but otherwise I could talk you through it.

 

On another another note The RCAs I got were cheap and they break if you over-tighten them so err on the side of caution when using them.

 

From my notes:

--MOD KIT >>  AY-3-8915 PIN NUMBER
--2            >> 4
--3            >> 5
--5            >> 6
--6            >> 7
--7            >> 8
--5V          >> 13
--CLK        >> 15
--GND       >> can be taken from anywhere.

 

Let me know if you guys have any questions I'm going back to making boards.

 

[+-^CrossBow^-]

In my case I tend to use Megadrive/Genesis 2 AV out cables so I wouldn't be using the 8pin din and would be using a 9pin din instead in most cases. Also I think we spoke about this before, but for Genesis 2 RGB cables I wouldn't need the caps and would need those outputs left at 0ohms just like I have to do with the TMS-RGB.

 

A hint with using the RCA jacks that will help. Get some 1/4" star washers to use. It will help with keeping the ground tab put on the jack as your tighten it and help keep the jack from rotation without having to crank down to tighten it. The star washer type doesn't really matter as you get similar results using inner or outer star washers. I've both on hand and just grab without one first from my bin.

 

[Icelvlan]

Is the board ready for Odyssey 2?

[the_crayon_king]

On 11/14/2020 at 1:36 AM, Icelvlan said:

Is the board ready for Odyssey 2?

If you mean the code for this CPLD based board I have tested it but I haven't went through and nitpicked everything. The main issue will be finding a usable clock and figuring out two palettes (that one should be able to fit two full palettes).

 

That non-programable O2 mod kit exists. I have it installed in my personal console.

If I thought there would be a significant demand I would populate some of them.

The limitation with that board is it can only do 4 different voltages for a color 0V,  .4V, 1V, and 1.4V.

The 1V and 0.4V can be anything but you want it to add up to 1.4 ish. You can create much more subtle mixtures of colors on a CPLD based design; the most "correct" looking colors might not even be feasible with a non programmable design.

[the_crayon_king]

So where I am at with everything:

The DACs I ordered from China were canceled so that set everything back a bit.

I can possibly cut the DAC out of the design.

 

The Intellivision is still more or less complete.

 

The Fairchild: code is complete but I don't have a way to test it. Could use another palette.

 

The Aquarius: needs another palette but is complete otherwise.

 

The Arcadia 2001/Magnavox O2: Both of these have clock issues. Instead of buffering the clock I think it would make more sense to just use a simpler method to get RGB. This method has the added benefit of being able to use Y/C, Composite, RGB, and YPbPr. The other method also either doesn't need or is far more tolerant of clocks.

 

The long and the short of it is that for these two particular consoles the voltages would be the same across a CPLD/non-CPLD mod. So there is no reason to use a CPLD mod here. I can make it work for a CPLD I just do not see a reason to do it over a buffer design since the buffer design has many added benefits.

 

 

 

[+-^CrossBow^-]

What is your current status on this? Very curious to see the final assembly of parts and to see how other testing goes!

 

[the_crayon_king]

13 hours ago, -^CrossBow^- said:

What is your current status on this? Very curious to see the final assembly of parts and to see how other testing goes!

 

Testing so far has gone well. Or well enough.

I have learned that the clock is a big source of noise.

https://i.imgur.com/NHkeG68.png

This is unterminated so the noise is 13mv p-p if the clock is using the CPLD

and 6.6mv p-p if the CPLD is not using the clock.

If you count the peaks out the frequency will match whatever the input clock frequency is which is how I know it is clock related.

 

Here is another example:

https://i.imgur.com/NwvWlLV.png

int dac clock: is just the clock going to the DAC I am using.

int zero clock: is using a R2R DAC (I had to hand solder to the CPLD pins ouch)

NOt my MOD: is another mod kit designed by someone else that also uses the same dac that I am. Might have helped to measure that one.

 

Anyway this noise is amplified if you use ground reference on the console's motherboard and not the mod kit. When installing the kit if you use ground off the console's motherboard and not the mod kit the noise will become significant enough to actually appear on the TV screen (something like jailbars). There must be an explanation for that but I cannot really guess as to what it is; maybe ground bounce?

From console to console the noise varies as well.

Look at the O2 for example: https://i.imgur.com/uYWRuIK.png

 

I don't think any of this is significant enough to worry about but I could use other opinions. The end device should filter this noise anyway. Even that "Not my mod" still looked good on the output.

 

Another issue is that I was supposed to be using a 530 ohm resistor and instead was using a 560ohm; this lowered to output maximum by about 20-30mv; or enough to cause white to become slightly offwhite.

Here is a comparison with the O2: https://i.imgur.com/yetKoVS.png

If you go into paint and click the dropper, click a color, and the hit edit color; you can actually more or less see the decimal values I programed in reflected on the "new" one.

Note I am actually using 536ohms since 530 is impossible to find. It does seem to more or less hit the full scale though. I might look for a resistor below 530 and shoot for 1.428/714mv instead.

 

Anyway, that is all information you can read or ignore. The main thing is I have been waiting for the last part to arrive (the DAC). There was an issue with the first seller so that set me back a while. I don't have an estimate for when they will get here other than "soon". I have everything else.

 

I am also trying to get a hold of a Fairchild as it is the only console out of the ones this kit should work for that I haven't tested yet.

 

 

 

[+-^CrossBow^-]

@the_crayon_king all excellent and interesting information! Thank you for taking the time to post that update! I've been watching this pretty closely as I'm sure others have as well.

[the_crayon_king]

So update. I have the ICS I needed and testing is going really well.

Somehow I have room for more palette adjustments than last time.

This is letting me set certain colors to a second palette. So far I have only managed to squeeze in light blue, tan, and brown.

 

This is just what I personally like and is an optional thing that you can click on or leave off. I picked these two games to hopefully kind of highlight why I am doing this.

The light blue should contrast better with blue and cyan for water.

Left is what I personally chose for my uses right is IMBezerk's palette.

Edited December 31, 2020 by the_crayon_king

[the_crayon_king]

Ok so these are out in the wild now on ebay.

I want to sell a small batch and wait to get feedback then sell the rest.

 

I am still writing the tutorial for this and I am not sure the preferred format. I will probably make a simplistic pictorial guide and then at a later date make a video guide.

If anyone is after the Arcadia 2001 kit it is done as well. Just leave a note you want it flashed with Arcadia code and not Intellivision if you want the Arcadia version.

[Icelvlan]

When are you receiving your test OSSC?

[+-^CrossBow^-]

On 1/2/2021 at 11:59 PM, the_crayon_king said:

Ok so these are out in the wild now on ebay.

I want to sell a small batch and wait to get feedback then sell the rest.

 

I am still writing the tutorial for this and I am not sure the preferred format. I will probably make a simplistic pictorial guide and then at a later date make a video guide.

If anyone is after the Arcadia 2001 kit it is done as well. Just leave a note you want it flashed with Arcadia code and not Intellivision if you want the Arcadia version.

Just placed my order off Ebay for one to test with. As I've stated before, I will be using mine with a 9-pin mini din as I have those on hand and breakout boards to solder onto as well for them. I might end up adding additional switches on my testing to try the different options to see what may or may not work best.

 

Thanks for making these up! I also have another kit coming from France I believe on the way, but not sure the status of that one in transit.

 

[the_crayon_king]

On 1/4/2021 at 9:26 AM, -^CrossBow^- said:

Just placed my order off Ebay for one to test with. As I've stated before, I will be using mine with a 9-pin mini din as I have those on hand and breakout boards to solder onto as well for them. I might end up adding additional switches on my testing to try the different options to see what may or may not work best.

 

Thanks for making these up! I also have another kit coming from France I believe on the way, but not sure the status of that one in transit.

 

I am curious how another kit might function. If it is non-programable solution I mean. I drew up another way to do this with non-programable parts but it was going to take like 10+ ICs instead of around 2 (like this board uses). Maybe it is a PAL only board since I think that board might have some differences with its outputs.

 

Make sure to knock off those 75 ohm resistors on the board. They are termination resistors so it should be fine to remove. Although, I did not test the configuration you guys are wanting to use. I do know typically resistors are placed as close as possible to the start and end of a line in order to reduce ringing or reflections.

 

I looked at the sync signals and it appears like the old method (what everyone else did) would be missing 9 sync pulses. 3 on either side of vsync and 3 inside vsync. I only restored the 3 inside of Vsync; and everything I have worked both with and without this addition. The three I restored exist in the inputs but are excluded if you just OR all the sync signals. The sync output from the encoder (if memory serves) is also missing those 3 pulses inside of vsync. So for whatever reason maybe they are not necessary and adding them back might have been a fruitless effort. The other 6 (if there are supposed to be 6 more) do not exist in the code at all. I cannot restore them with the CPLD and would need an external PLL and an OR/XNOR/XOR to combine them external to the CPLD. 

 

I am just selling 10 and waiting to get some feedback. I am satisfied with what I see on my end but I do not know how the kits will react on people's various setups.

I know my XRGB took some tweaking with the A/D setting to make it look good. The CRT (through YPbPr) has looked fine this entire time. We don't talk about that no name "HD" scaler but it worked on that as well. This may even work with the OSSC; I still do not have one to confirm/deny. I would assume if it doesn't work with one line doubler (like the OSSC) that it won't work with any of them.

 

I am clearly not an EE. I really just made this because I needed this for my own consoles. If there is some sort of unforeseen issue I will have to redesign the board because the code is complete IMO.

 

I am going to make a R2R DAC version of this board anyway which will has it's own list of pros/cons. Notably, that version should have composite, RGB, YPbPr (CRT), and S-Video. It will be easier to acquire parts for and most likely more tolerant of clocks. The con is that it will require soldering 100 or so resistors

Edited January 7, 2021 by the_crayon_king

[YannAros]

Hi,

 

I'm the "creator" of this other solution.

You can see the schematic and even downlod files from here : https://easyeda.com/yannick.erb/Intellivision-Secam-RH-RGB-Mod_copy

The inputs are taken from the STIC directly (V1 to V5) other inputs are Audio, GND and power, so nothing PAL specifc.

The aim was to create a very simple kit, not necessitating any specifc tools to build it.

 

The palette is however not easily changed, you need to modify resistances values and there is no fancy functionality on it.

The only specificity is that the sync signal can be either NOT(V5 AND V1) or NOT(V5), the second solution proved to be working with many modern screens where the usual solution was not working leading to loss of horizontal sync.

I have never tested (and I don't know of anyone else) with OSSC.

 

Best regards,

YannAros

[+-^CrossBow^-]

Well, I will certainly be able to let you both know how the kits turn out on the OSSC. If that doesn't work, I do have a SCART RGB to component converter box I can use although the results from that in the past were very lack luster on my setup. But that would be enough to at least get a signal to a display in this case.

 

[the_crayon_king]

12 hours ago, YannAros said:

Hi,

 

I'm the "creator" of this other solution.

You can see the schematic and even downlod files from here : https://easyeda.com/yannick.erb/Intellivision-Secam-RH-RGB-Mod_copy

The inputs are taken from the STIC directly (V1 to V5) other inputs are Audio, GND and power, so nothing PAL specifc.

The aim was to create a very simple kit, not necessitating any specifc tools to build it.

 

The palette is however not easily changed, you need to modify resistances values and there is no fancy functionality on it.

The only specificity is that the sync signal can be either NOT(V5 AND V1) or NOT(V5), the second solution proved to be working with many modern screens where the usual solution was not working leading to loss of horizontal sync.

I have never tested (and I don't know of anyone else) with OSSC.

 

Best regards,

YannAros

Very nice

So we started out the same; I was curious if it would start out the same or if it could be done in a different way.

I had mine private because it went through several embarrassing stages (and still probably doesn't work but here).

Some of the parts I used may be cheaper; I'll have to cross compare at some point.

 

At any rate I like the idea of having a non programmable solution out there as well so kudos.

 

8 hours ago, -^CrossBow^- said:

Well, I will certainly be able to let you both know how the kits turn out on the OSSC. If that doesn't work, I do have a SCART RGB to component converter box I can use although the results from that in the past were very lack luster on my setup. But that would be enough to at least get a signal to a display in this case.

 

It would be nice to know what all this can work on so be sure to try it on whatever you can. If you are using component then you are most likely going to need a capacitor on those lines.

 

The information below is not particularly for you or directed at anyone (I use this site as my personal notepad and it is time I wrote this down lest I forget).

V5 and V1 are the sync but just ORing/NORing them together will just give sync without those 3 pulses in vsync [which may or may not matter].

I guess I meant people are only using V5 and V1 for sync but there are other parts of sync inside of the 1XXX1.

 

Refer to the pictures attached. If you NOR together 10101 and 11101 you will get all the hsync pulses. (the reason I am ORing instead of NORing has to do with the way I wrote the code).

 

V5V4V3V2V1

 

1 X X 0 0 Blanking > so if V5 is ever active is can be used as an OE to turn off RGB

10001 = ?

10011 = ?

10101 = bulk of hsync

10111 = ?

11001 = vsync

11011 = ?

11101 = hsync pulses inside and shortly after vsync

11111 = test

 

 

 

 

[+-^CrossBow^-]

5 minutes ago, the_crayon_king said:

It would be nice to know what all this can work on so be sure to try it on whatever you can. If you are using component then you are most likely going to need a capacitor on those lines.

Pretty sure the SCART cables I have, have 220µf caps inside the SCART housing already along with the resistors. In reality, the resistors being left on would likely work still, but just result in a darker picture then intended as that is what happens on the TMS-RGB for the CV if I don't remove the 75R off the output lines. In fact, I wonder something else now...

 

Will let you know what I guess when I've had a time to install these kits into my Sears intelly.

 

[+-^CrossBow^-]

If only the USPS could be as fast with packages sent a few weeks ago as what they are able to do with newer sent stuff. Yeah, @the_crayon_king your kit arrived in the mail today already?! Not sure when I can get to testing as I have at least 3 consoles on the way that have to be finished up first before I can dedicate time to test this.

 

[YannAros]

Hi,

 

I've studied a bit the CSYNC stuff, and ended doing 1XXX1 (V5 to V1) that seemed to work with my LCD TVs, then I've found another TV that wasn't syncing good horizontaly.

Feedling around I could obtain a stable image by "delaying" the sync signal adding a 47nF cap between sync and GND, however as you delay the sync signal the image gets shifted towards the left.

Playing further I've found that using 1XXXX is working on this LCD TV, but not on my other ones working with 1XXX1. Later on I've found another LCD TV that had the same behaviour requiring 1XXXX and some other users of the kit had to do the same.

So my PCB comes with a jumper allowing to select either one of these methods.

 

I'm also using V5 as OE input for blanking as you suggest.

I took a look at your schematic, that from a high level should work, I didn't look into the details.

Mine is using separate resistors for each RGB of each colour, allowing more control over the end result.

 

One thing also regarding component selection, I've done it so that I can get the full PCB assembled and SMD component mounted by JLCPCB, this way, even if the component is a bit more expensive, I have no complex soldering to do.

At the end we are not talking about huge costs (assembled PCB for a batch of 30 should be around 4$ without shipment cost and custom fees), I'm adding connectors for convenience but you can solder wires directly on PCB to.

 

This is what it is ooking like in the French Intellivision model that has embedded power supply, I designed it so that it can be mounted where the power supply PCB usually sits:

 

Edited January 8, 2021 by YannAros

[Icelvlan]

So when are you two going to work on 7800 mod? Haha