NES/SNES 240p dejitter mod

The place for all discussion on gaming hardware
copy
Posts: 232
Joined: Fri Apr 29, 2016 10:38 pm

Re: NES/SNES 240p dejitter mod

Post by copy »

marqs wrote:For 1) you need to disconnect BH7236AF video encoder pin 6 and connect that to snes_dejitter CPLD pin 11 which provides clean subcarrier. I think there was some discussion about this a few pages back. The latest firmware from nes-fix branch must also be installed on the board.
Great, thank you marqs, that explains it. I did see the subcarrier discussion earlier but it went over my head at the time.
marqs wrote:2) sounds strange, are you sure your display has showed it correctly in that mode earlier? You might also want to take a look into NESRGB dejitter modification which might be easier in the end if you want to preserve cvbs & s-video.
Full disclosure, I have had trouble with 256x240 optim mode on the NES even before this. Most recently I couldn't find sampling settings that gave me a completely clean picture, but at least I was able to see the full width of the image. I suspected my dejitter rewiring was responsible for this setback, but perhaps it's unrelated. Maybe my NES and display are just not destined to be compatible with 256x240.

I was actually contemplating upgrading to the NESRGB 2.0 with built-in dejitter. I'll have to weigh that versus getting into the dejitter board firmware updating. I recall the earlier discussions about how complex that can be. Anyway, thanks again for your help.
Dochartaigh
Posts: 1518
Joined: Thu Mar 02, 2017 6:53 pm

Re: NES/SNES 240p dejitter mod

Post by Dochartaigh »

I honestly haven't been following this much, but do see people post about issues with NES/SNES and the OSSC, and the dejitter mod as a fix for it. I'm just curious about what percentage of people have problems with setups like these?

I'm getting more into learning how to properly upscale and tested my (RGB) NES and SNES 1-CHIP03 (with CSYNC mod) with an OSSC over the weekend. Maybe tested 5-7 games on each, for roughly 5-10 minutes each, and didn't have any issues in 4x and 5x mode (don't really use the lower multipliers) on my TCL 4K TV. Should it be safe to say that my exact consoles won't give me problems in the future? Are there certain games which more commonly give people problems?

I did notice that when I routed the NES and SNES from the OSSC to an Extron DSC 301 HD that the image was a flickery mess (could still see and read everything, just like black horizontal bar-things making the picture look like crap). Switched to Genesis, Saturn, and PS1 and it was fine with the OSSC + DSC combo. I'm assuming the DSC doesn't like the non-standard NES/SNES sync then, right? Honestly can't think of an instance when I would use non-480p consoles with the DSC, but who knows what the future may hold as I start experimenting more.
User avatar
Harrumph
Posts: 368
Joined: Tue Jan 19, 2016 10:06 pm
Location: Sweden

Re: NES/SNES 240p dejitter mod

Post by Harrumph »

Dochartaigh wrote:. Should it be safe to say that my exact consoles won't give me problems in the future? .
No because it’s not about the consoles (or the OSSC) it is about display tolerance. So if you switch display, it may then be incompatible due to the jitter.
nmalinoski
Posts: 1974
Joined: Wed Jul 19, 2017 1:52 pm

Re: NES/SNES 240p dejitter mod

Post by nmalinoski »

Dochartaigh wrote:I honestly haven't been following this much, but do see people post about issues with NES/SNES and the OSSC, and the dejitter mod as a fix for it. I'm just curious about what percentage of people have problems with setups like these?
It's not a problem with the NES/SNES and the OSSC, it's a problem with the NES/SNES and other connected equipment. The OSSC is one of the few devices that doesn't actually have a problem with the janky sync from these consoles; but I think people say it's a problem with the OSSC, because the OSSC doesn't smooth-over the problem for them, instead just passing on the janky sync to downstream AV devices.
Dochartaigh wrote:Honestly can't think of an instance when I would use non-480p consoles with the DSC, but who knows what the future may hold as I start experimenting more.
I don't use 240 directly with my DSC, but I'll run them through the OSSC. I'll have multipliers on 240p, but I'll let the DSC do scaling to my display's native 1360somethingx768; and I might have 480i set to either 3x or passthrough, so that the DSC will deinterlace and scale it.
copy
Posts: 232
Joined: Fri Apr 29, 2016 10:38 pm

Re: NES/SNES 240p dejitter mod

Post by copy »

marqs wrote:For 1) you need to disconnect BH7236AF video encoder pin 6 and connect that to snes_dejitter CPLD pin 11 which provides clean subcarrier.
Okay, I've decided I will try to do this. Can anyone confirm if the following would be a successful way to do it?

1) Remove the circled capacitor to (I think?) disconnect the original NESRGB subcarrier from the video encoder?

2) Jumper snes_dejitter CPLD pin 11 to the right pad of the removed NESRGB capacitor?

Image

I also found this helpful arthrimus video on how to update the firmware.
copy
Posts: 232
Joined: Fri Apr 29, 2016 10:38 pm

Re: NES/SNES 240p dejitter mod

Post by copy »

On further examination, it looks like the removing the circled resistor would disconnect the NESRGB subcarrier.

Then I could connect the dejittered subcarrier straight to pin 6 of the video encoder, or perhaps to the right end of the capacitor? Either way, should the capacitor be removed?

Image

(BTW, I should note that these images are not of my actual NESRGB, on which I've already wired the dejittered csync into the video encoder.)
User avatar
marqs
Posts: 1030
Joined: Sat Dec 15, 2012 12:11 pm
Location: Finland

Re: NES/SNES 240p dejitter mod

Post by marqs »

copy wrote:Then I could connect the dejittered subcarrier straight to pin 6 of the video encoder, or perhaps to the right end of the capacitor? Either way, should the capacitor be removed?
That trace indeed seems to lead there, but you should verify it with a multimeter continuity check. It should be ok to leave the cap there.
copy
Posts: 232
Joined: Fri Apr 29, 2016 10:38 pm

Re: NES/SNES 240p dejitter mod

Post by copy »

marqs wrote:That trace indeed seems to lead there, but you should verify it with a multimeter continuity check. It should be ok to leave the cap there.
Thanks again marqs. The continuity does check out on my multimeter. I will try this in the near future and report back.


Switching to the SNES: I successfully installed dejitter in my 1CHIP-02 this weekend. Regarding my earlier question on how best to connect csync with an RGB bypass board, I concluded it is best to send the dejitter CSYNC_o straight to pin 3 of the multiout. I also found yet another arthrimus video where he already discovered the same thing.

(TL;DR version: I first experimented with connecting dejitter CPLD pin 15 straight to the "CS" input on the RGB bypass, on the grounds that the RGB bypass is designed to accept the higher Vpp csync signal from the SNES CPU.

This gave mixed results: the OSSC accepted the resulting csync output from the RGB bypass board just fine. However, the Framemeister produced only audio and no video. I found that I could achieve a picture on the Framemeister by increasing its SYNC_LEVEL setting. But this worried me that the csync signal was maybe off spec.

I didn't want to risk having an improper csync that could cause any problems in the future with any devices. After connecting dejitter CSYNC_o to the multiout, both the Framemeister and OSSC worked with their default settings, so that way seems like it will be more reliable overall.)

Oh yeah, and I've decided I'm not going to worry about dejittered composite and s-video on SNES since it would require lifting pins, and I'm not confident I could do that.


Image
User avatar
marqs
Posts: 1030
Joined: Sat Dec 15, 2012 12:11 pm
Location: Finland

Re: NES/SNES 240p dejitter mod

Post by marqs »

Fron what I see from the RGB bypass board picture, CS is routed to THS7374 input through a voltage divider, and its output is routed to multiAV pin through a 470ohm resistor. If you wire 2.5Vpp sync from dejitter board to CS pin (instead of 5Vpp from the CPU/PPU) and use a cable that has another 470ohm resistor on sync like, then (depending on input voltage divider resistor values) you could end up with lower than nominal sync level which should still work if you respectively adjust sync threshold on OSSC/FM. If you hook up dejitter csync directly to multiAV, then I'd remove that 470ohm resistor from RGB bypass board output side to avoid multiple drive.
User avatar
Arthrimus
Posts: 106
Joined: Mon Apr 02, 2018 5:49 pm
Location: Arkansas

Re: NES/SNES 240p dejitter mod

Post by Arthrimus »

marqs wrote:If you hook up dejitter csync directly to multiAV, then I'd remove that 470ohm resistor from RGB bypass board output side to avoid multiple drive.
I can confirm that this should be done if you run the csync_o directly to the AV out. I ran into issues with one of my early RGB Bypass and dejitter installs when I failed to do this.
plus ça change,
plus c'est la même chose,
The more that things change,
The more they stay the same.- RUSH- Circumstances

I install and sell mods at arthrimus.com | SNES RGB Bypass+Dejitter available now! | Watch me live stream my work on YouTube
copy
Posts: 232
Joined: Fri Apr 29, 2016 10:38 pm

Re: NES/SNES 240p dejitter mod

Post by copy »

Thanks guys. I'll remove that 470 ohm resistor. I have also now closed the J3 jumper on the dejitter board.
sofakng
Posts: 216
Joined: Fri Sep 02, 2016 1:30 am

Re: NES/SNES 240p dejitter mod

Post by sofakng »

Do I need to use any special (shielded) cables if I wire a dejitter ON/OFF switch with the SNES?

Here is from the FAQ:
Is it possible to disable de-jitter functionality after installation?

When the board is installed so that MCLK_EXT_i and CLK_SEL_i are not used, a simple bypass can be added without firmware modifications. To do that, connect MCLK_EXT_i to CPLD pin 34 (MCLK_XTAL_o) and add a ON-OFF switch that connects MCLK_SEL_i to 5V.
Cloudmp
Posts: 3
Joined: Thu Sep 06, 2018 3:57 pm

Re: NES/SNES 240p dejitter mod

Post by Cloudmp »

Hi guys, i decided to update the firmware on my NESRGB 1.4 board and use the dejitter firmware, however in the instructions on Tim page maybe there is something wrong, it says to isolate the PPU pin 18 (clock) and to solder a wire between this pin and pad 2. It shouldn't be pad 3? after update to the dejitter firmware, pad 3 becomes PPU clock output, and pad 2 now is Sync input, and following the instructions, pad 2 already has a bridge between it and the CS# pad. Can someone give me a light, i don't wanna blow up my poor AV Famicom! :lol:
lilock3
Posts: 1
Joined: Sat Aug 24, 2019 2:18 pm

Re: NES/SNES 240p dejitter mod

Post by lilock3 »

Cloudmp wrote:Hi guys, i decided to update the firmware on my NESRGB 1.4 board and use the dejitter firmware, however in the instructions on Tim page maybe there is something wrong, it says to isolate the PPU pin 18 (clock) and to solder a wire between this pin and pad 2. It shouldn't be pad 3? after update to the dejitter firmware, pad 3 becomes PPU clock output, and pad 2 now is Sync input, and following the instructions, pad 2 already has a bridge between it and the CS# pad. Can someone give me a light, i don't wanna blow up my poor AV Famicom! :lol:
I've just finished doing this myself today, mine is a 1.4 board in an original Famicom. It took a bit of time to get it working right, so I though I would post the issues I encountered to help out anyone else trying to do the same. I also discovered that my board is one of the ones with the dodgy voltage regulators, so I took the opportunity to do the capacitor fix as Tim suggests on his website.

Yes you're are right, it's palette pad 3 that outputs the new clock signal to the PPU, not pad 2 - the instructions would appear to be wrong. Also, I found that J8 must be open (not mentioned in the instructions). The biggest issue I had though is that the 1.9A firmware does not work! It doesn't seem to produce a valid clock signal for the PPU and the console fails to operate. 1.9 works perfectly, and I'm happy enough sticking with the older palette, as that's what was in the 1.8 firmware anyway.

It's taken me a fair few hours of repeatedly re-flashing and investigating with a multi-meter to figure all this out and finally get it working properly, so I hope this info is of use to someone!
Cloudmp
Posts: 3
Joined: Thu Sep 06, 2018 3:57 pm

Re: NES/SNES 240p dejitter mod

Post by Cloudmp »

Cloudmp wrote:Hi guys, i decided to update the firmware on my NESRGB 1.4 board and use the dejitter firmware, however in the instructions on Tim page maybe there is something wrong, it says to isolate the PPU pin 18 (clock) and to solder a wire between this pin and pad 2. It shouldn't be pad 3? after update to the dejitter firmware, pad 3 becomes PPU clock output, and pad 2 now is Sync input, and following the instructions, pad 2 already has a bridge between it and the CS# pad. Can someone give me a light, i don't wanna blow up my poor AV Famicom! :lol:

I talked with Tim via email and yes, the tutorial was wrong, you need to solder pin 18 of the PPU on pad 3 of the NESRGB board, instead of pad 2. He already fixed this typo on his site.
Cloudmp
Posts: 3
Joined: Thu Sep 06, 2018 3:57 pm

Re: NES/SNES 240p dejitter mod

Post by Cloudmp »

lilock3 wrote:
Cloudmp wrote:Hi guys, i decided to update the firmware on my NESRGB 1.4 board and use the dejitter firmware, however in the instructions on Tim page maybe there is something wrong, it says to isolate the PPU pin 18 (clock) and to solder a wire between this pin and pad 2. It shouldn't be pad 3? after update to the dejitter firmware, pad 3 becomes PPU clock output, and pad 2 now is Sync input, and following the instructions, pad 2 already has a bridge between it and the CS# pad. Can someone give me a light, i don't wanna blow up my poor AV Famicom! :lol:
I've just finished doing this myself today, mine is a 1.4 board in an original Famicom. It took a bit of time to get it working right, so I though I would post the issues I encountered to help out anyone else trying to do the same. I also discovered that my board is one of the ones with the dodgy voltage regulators, so I took the opportunity to do the capacitor fix as Tim suggests on his website.

Yes you're are right, it's palette pad 3 that outputs the new clock signal to the PPU, not pad 2 - the instructions would appear to be wrong. Also, I found that J8 must be open (not mentioned in the instructions). The biggest issue I had though is that the 1.9A firmware does not work! It doesn't seem to produce a valid clock signal for the PPU and the console fails to operate. 1.9 works perfectly, and I'm happy enough sticking with the older palette, as that's what was in the 1.8 firmware anyway.

It's taken me a fair few hours of repeatedly re-flashing and investigating with a multi-meter to figure all this out and finally get it working properly, so I hope this info is of use to someone!

Yes, the firmware 1.9A was also wrong on Tim's site! He already corrected that (he told me by email), and i can confirm that now it works, because i did this process yesterday and it was all smooth for me. luckily the J8 was already open on mine, so i didn't encounter the problem you had, and as a side note, my board was also with the fake voltage regulator, and i fixed that too.

If you want to use the FBX smooth pallet, you can flash the 1.9A without any problems now, only be sure to download the correct one, that is on Tim's site since 08/26
User avatar
nullifer
Posts: 29
Joined: Fri Aug 30, 2019 3:57 pm
Location: Italy

Re: NES/SNES 240p dejitter mod

Post by nullifer »

Could you guys reccomend me some resources, or practice projects that I could do to get good at soldering before moving onto a sensitive project such as this?
RyleFury
Posts: 1
Joined: Fri Sep 06, 2019 8:41 pm

Re: NES/SNES 240p dejitter mod

Post by RyleFury »

Hello, I put the dejitter chip on a PAL SNES for a friend who have the Super CIC and it really works great ^^
I tested before and after soldering the dejitter with the Framemeister. Even if this is pretty subtle, we can see that the image lacks of stability without the dejitter. After soldering the dejitter, the image is perfectly stable, so this is really a great find, thank you.

However, I was surprised to see there was not a diagram for the PAL SNES in the installation folder in Github. I came here to summit my own diagram. I tested it today and there is no mistake. Also, there are the two possibilities when the user can have the csync in the luma or cvideo signal, or the two of them. Also, you can note that SNSP-CPU-01 and SNSP-CPU-02 are almost identical, so your instructions and the diagram work for the two of them.

Image

Edit: I made a pull request on the Github project during the approval of my message.
copy
Posts: 232
Joined: Fri Apr 29, 2016 10:38 pm

Re: NES/SNES 240p dejitter mod

Post by copy »

For fun this weekend, I "upgraded" one of my v1.2 dejitter boards to be effectively a 1.3. I ordered the new R9, R10, and R15 resistors from the BOM, swapped out R9 and R10, and soldered R15 across the MCLK_EXT_i and adjacent GND pad.

It's working great so far in my SHVC. Of course, it was already working fine before, but I just like knowing that the CSYNC_o voltage is closer to the stock value now.
SkydivinGirl
Posts: 6
Joined: Thu Nov 28, 2019 1:57 am

Re: NES/SNES 240p dejitter mod

Post by SkydivinGirl »

Hey everyone,

I installed the Arthrimus RGB Bypass + Dejitter combo board in a SNES-CPU-1CHIP-01 console and I'm getting color distortions at the top of the screen The distortion seems worse through composite but I can still see some through RGB on the OSSC. If I disable the dejitter function, everything looks great.

https://youtu.be/XGFs3DSz21I

I saw the 'pink line' issue on an NES earlier in the thread and it seems like that's the closest I've seen to the issue I'm having. I used the JED with a date of 1/7/2019 to program the CPLD. I connected everything as shown in Arthrimus's 1CHIP install video, including connecting the Line Level CSync to the lifted PIN 7 of the chip that he mentions.

I'd appreciate any help anyone can provide in helping me resolve this issue. :)

Thanks!

Heather
SkydivinGirl
Posts: 6
Joined: Thu Nov 28, 2019 1:57 am

Re: NES/SNES 240p dejitter mod

Post by SkydivinGirl »

Hey everyone,

So, today I installed another one of these combo boards into an NTSC SNES Jr/Mini. RGB is perfect but the output on Composite and S-Video is even more messed up than the 1-Chip in my previous post. I followed both of the videos by Arthrimus for these exact PCBs and I know I did everything correctly so I'm completely at a loss as to why these aren't working. I've even tried them through a Retrotink-2X with the same results as my PVM. The only thing I can think of for the Mini is that I was not able to replace the R3 resistor with the 1.69K one he mentions at the end of his video because I don't have one right now. I don't know that it would have this effect though.

I even tried re-flashing the CPLD with the one release older code with the same results. I also double checked the RGB output on the 1-Chip and it is fine so this issue is only occurring on the S-Video and Composite outputs.

What am I missing? :(

Thanks,

Heather
User avatar
Arthrimus
Posts: 106
Joined: Mon Apr 02, 2018 5:49 pm
Location: Arkansas

Re: NES/SNES 240p dejitter mod

Post by Arthrimus »

SkydivinGirl wrote:Hey everyone,

So, today I installed another one of these combo boards into an NTSC SNES Jr/Mini. RGB is perfect but the output on Composite and S-Video is even more messed up than the 1-Chip in my previous post. I followed both of the videos by Arthrimus for these exact PCBs and I know I did everything correctly so I'm completely at a loss as to why these aren't working. I've even tried them through a Retrotink-2X with the same results as my PVM. The only thing I can think of for the Mini is that I was not able to replace the R3 resistor with the 1.69K one he mentions at the end of his video because I don't have one right now. I don't know that it would have this effect though.

I even tried re-flashing the CPLD with the one release older code with the same results. I also double checked the RGB output on the 1-Chip and it is fine so this issue is only occurring on the S-Video and Composite outputs.

What am I missing? :(

Thanks,

Heather
I've discovered the same behavior on certain individual SNES motherboards also. I'm still not sure what the root cause is, but in my testing it only happens if you lift pin 7 of the S-RGB encoder and wire it to the LLCS pad. As of right now my recommendation is to skip that step until a solution is found. You lose dejittered Composite and Luma sync but dejittered C-Sync will still work fine you don't wire up LLCS.
plus ça change,
plus c'est la même chose,
The more that things change,
The more they stay the same.- RUSH- Circumstances

I install and sell mods at arthrimus.com | SNES RGB Bypass+Dejitter available now! | Watch me live stream my work on YouTube
SkydivinGirl
Posts: 6
Joined: Thu Nov 28, 2019 1:57 am

Re: NES/SNES 240p dejitter mod

Post by SkydivinGirl »

Thanks for the reply Arthrimus! Does the original Marqs85 schematic also have the LLCS option? If so, do we know if this happens with those boards?

I'm happy to help with troubleshooting if you have any ideas and need to test solutions. I also have a scope if it will help look at the sync signal.

Thanks again!

Heather
User avatar
Arthrimus
Posts: 106
Joined: Mon Apr 02, 2018 5:49 pm
Location: Arkansas

Re: NES/SNES 240p dejitter mod

Post by Arthrimus »

SkydivinGirl wrote:Thanks for the reply Arthrimus! Does the original Marqs85 schematic also have the LLCS option? If so, do we know if this happens with those boards?

I'm happy to help with troubleshooting if you have any ideas and need to test solutions. I also have a scope if it will help look at the sync signal.

Thanks again!

Heather
No it doesn't. This was an idea that was suggested earlier in this thread by Marqs if I recall correctly. It was supposed to help people who are using HD Retrovision cables since they use composite video I stead of C-sync.

On an original dejitter board you would tap the LLCS signal directly from the CPLD output. In my board I simply added a pad for it so it's easier to install the mod. I'm assuming the same symptoms would be present on the standard dejitter board, although I have not tested it myself.
plus ça change,
plus c'est la même chose,
The more that things change,
The more they stay the same.- RUSH- Circumstances

I install and sell mods at arthrimus.com | SNES RGB Bypass+Dejitter available now! | Watch me live stream my work on YouTube
SkydivinGirl
Posts: 6
Joined: Thu Nov 28, 2019 1:57 am

Re: NES/SNES 240p dejitter mod

Post by SkydivinGirl »

Thanks Arthrimus! For now, I've added the switch to enable/disable the mod. This will let me keep the mod in place but turn it off if I use composite or S-Video. That will do for now. I do hope someone does a bit more research on this problem and can come up with a solution. Your combination board makes for a very clean install and I really like it!

Heather
rama
Posts: 1373
Joined: Wed Mar 08, 2017 3:15 pm

Re: NES/SNES 240p dejitter mod

Post by rama »

Just chiming in, with the same issue.
Composite Video has discolouration on top of the image when dejitter is enabled.
It looks like the TV chroma decoder is loosing lock, or maybe the S-RGB (1Chip SNES) has some issues.
User avatar
marqs
Posts: 1030
Joined: Sat Dec 15, 2012 12:11 pm
Location: Finland

Re: NES/SNES 240p dejitter mod

Post by marqs »

rama wrote:Just chiming in, with the same issue.
Composite Video has discolouration on top of the image when dejitter is enabled.
It looks like the TV chroma decoder is loosing lock, or maybe the S-RGB (1Chip SNES) has some issues.
For NES(RGB) the fix was to output uninterrupted subcarrier from CPLD to composite encoder. Probably the same solution works here, but I'm suprised to hear RGB is affected on some models too.
SkydivinGirl
Posts: 6
Joined: Thu Nov 28, 2019 1:57 am

Re: NES/SNES 240p dejitter mod

Post by SkydivinGirl »

My initial thought that the rgb was affected was incorrect. It's just composite and S-Video.

Could you point to the conversation about the NES fix or is it in this thread? Is it an additional wire from the CPLD to the composite encoder?

Thanks!

Heather
rama
Posts: 1373
Joined: Wed Mar 08, 2017 3:15 pm

Re: NES/SNES 240p dejitter mod

Post by rama »

Yeah, just Composite Video here as well. RGB worked fine on first attempt :)

I can't test any modifications as I did the mod for a friend, but the NES fix sounds like it would work here as well.
User avatar
maxtherabbit
Posts: 1763
Joined: Mon Mar 05, 2018 4:03 pm

Re: NES/SNES 240p dejitter mod

Post by maxtherabbit »

does this occur on all SNES if you feed the encoder dejittered sync? or is it just a 1-chip specific issue?
Post Reply