THS7374 vs THS7314 + 1 chip brightness fixing

The place for all discussion on gaming hardware
Post Reply
leonk
Posts: 1080
Joined: Sun Mar 13, 2011 9:29 pm
Location: Toronto, Canada

Re: THS7374 vs THS7314 + 1 chip brightness fixing

Post by leonk »

retrorgb wrote: When you filter an already filtered signal, it softens the image. That's why the THS7314 looks so good on RGB monitors and through OSSC/Framemeister with the filters turned off, but people say it's "blurry" in other situations. The stock S-RGB chip does not have a filter either, so using the 7316 or 7374 (filter off) will produce the same type of output as original, unmodded 1CHIP's....same with other consoles. It would be hard to have a setup with one SNES having a filter on and the rest with no filter. In most scenarios, it's better to deal with that on the head-end.
Here's my take on the whole thing. :)

Analog input devices (PVM):

- The unfiltered "noise" (as Tim W. put it) is very fast moving spikes, coupled into the video signal. These spikes are too fast for analog devices, and they just don't "see" them.

Scalars (OSSC/PVM):

- These sample the input analog signal. They sample the input at much higher frequency; they see these spikes and assume they are valid data. Hence you see jailbars in output. So we need to have a LPF before the analog sampler to remove these spikes.
- In the case of OSSC that's easy - the LPF is user controllable
- In the case of XRGB mini - it's not. The latest firmware DOES NOT allow one to turn on LPF without introducing other video processing (linear/bob deinterlace!) - GAME1 mode turns on the LPF in XRGB mini - it softens the picture to look worst than THS7314 based solutions!

Digital displays:
- I think this is a non-starter (unless you have a digital display that takes as input RGB signal). By the time the signal hits my LCD, it's already gone through the XRGB mini and been digitized. LPF does not apply.

Other consoles:
- N64RGB uses the THS7374 - the LPF is enabled by default
- NESRGB - I see no issues on XRGB in picture mode - I assume LPF is enabled on NESRGB
- NEC consoles - latest kits have switched to using THS7374 - what happens with 1CHIP can apply here as well
- SNES - see this thread. :)
- Sega Genesis & others - I don't see jailbars on XRGB mini - I assume there's some sort of LPF in console (could be wrong)

Dilema:
- XRGB mini in PICTURE mode is the "recommended" solution that most use (It's what FBX sets most his profiles to!)
- At this point, there are a lot more XRGB mini users than OSSC - a solution is needed that works for them
- Unless one can point to a way to turn only LPF in XRGB mini - turning off LPF in THS7374 is not a good option in my opinion
leonk
Posts: 1080
Joined: Sun Mar 13, 2011 9:29 pm
Location: Toronto, Canada

Re: THS7374 vs THS7314 + 1 chip brightness fixing

Post by leonk »

retrorgb wrote:I think we're seeing different things. Here's an example of the S-RGB chip directly into a capture card (no filter) on the left and then through a basic 9MHz LFP on the right:
Image
YES!! EXACTLY!!!

SNES Mini or SNES 1CHIP-03 with THS7374 LPF disabled and XRGB mini in PICTURE mode (most of FBX profiles) looks almost like the picture on the left on my TV!!
RGB0b
Posts: 543
Joined: Wed Dec 05, 2012 1:52 pm

Re: THS7374 vs THS7314 + 1 chip brightness fixing

Post by RGB0b »

...but that's using the S-RGB amp, not the 7374. That means you should see the same interference on any setup with an unfiltered amp on an unmodified console. The CXA's are unfiltered, so that means Genesis, Saturn, Neo Geo (CD & AES) and PS1. Also, other consoles like the Jaguar and any mod that's done with either the CXA chip or any of those other amps like Tim's AV Driver.

BTW, I think your description of why you see the interference on digital but not analog is spot-on.
Frankym2612
Posts: 30
Joined: Thu Dec 29, 2016 5:43 pm

Re: THS7374 vs THS7314 + 1 chip brightness fixing

Post by Frankym2612 »

Oh I thought voultar was saying that that interference was aliasing with no filter but I guess I read wrong I know in picture mode on a xrgb mini the filter is off but there still is some sort of filtering still I believe also i just want to try the other fix and possibly along with another board maybe to see a bunch of different options even if not needed
leonk
Posts: 1080
Joined: Sun Mar 13, 2011 9:29 pm
Location: Toronto, Canada

Re: THS7374 vs THS7314 + 1 chip brightness fixing

Post by leonk »

no replies from anyone? guess I'm the only THS7374 with XRGB mini user with latest firmware. ;)
Frankym2612
Posts: 30
Joined: Thu Dec 29, 2016 5:43 pm

Re: THS7374 vs THS7314 + 1 chip brightness fixing

Post by Frankym2612 »

I just got my 7374 from voultar I have a framemeister and a 1 chip 03 I will see pretty soon if there is any interference but there shouldn't be any also I may take pics right now I have the 7314 in it
RGB0b
Posts: 543
Joined: Wed Dec 05, 2012 1:52 pm

Re: THS7374 vs THS7314 + 1 chip brightness fixing

Post by RGB0b »

BTW Franky, you were right, it is technically aliasing. I confused myself for a moment there...

...in situations like this, while it's the correct technical term, I prefer to say "interference", as most people associate "aliasing" with the the jagged edges on 3D images (like with N64) and people think of "anti-aliasing" as the smoothing of those edges. People like me don't like to see anti-aliasing on N64, so that seems like a "bad" term.
Frankym2612
Posts: 30
Joined: Thu Dec 29, 2016 5:43 pm

Re: THS7374 vs THS7314 + 1 chip brightness fixing

Post by Frankym2612 »

Ya you were right too I was basically trying to say the aliasing found in the snes isn't the same "jailbar" interference as seen in the Genesis master system etc from poor board routing although it looks like bars it's a different kind from like you and voultar were saying from to high frequency from the d a conversion. I was wondering if voultars board with the filter on framemeister filter off is essentially like the 7314 with framemeister filter off just without attenuation?
User avatar
Voultar
Posts: 550
Joined: Fri Jan 10, 2014 8:29 pm
Location: USA

Re: THS7374 vs THS7314 + 1 chip brightness fixing

Post by Voultar »

Frankym2612 wrote:Ya you were right too I was basically trying to say the aliasing found in the snes isn't the same "jailbar" interference as seen in the Genesis master system etc from poor board routing although it looks like bars it's a different kind from like you and voultar were saying from to high frequency from the d a conversion. I was wondering if voultars board with the filter on framemeister filter off is essentially like the 7314 with framemeister filter off just without attenuation?
The THS7374 has a better low order filter than the 7314.
Xaranar
Posts: 112
Joined: Mon Mar 20, 2017 8:43 am

Re: THS7374 vs THS7314 + 1 chip brightness fixing

Post by Xaranar »

FYI, I'm using borti4938's latest board revision of the SNES RGB bypass, link here, and his board is designed to make use of either the 7373 or the 7374. The 7373 differs from the 7374 in that 3 of its 4 channels make use of an HD LPF, while the other uses an SD one, and the 7374 uses an SD filter on all 4 channels. I've tried it with both amps, and you get a better result with the 7374, the filtering on the 7373 isn't as strong it seems and still shows the jailbars.
leonk
Posts: 1080
Joined: Sun Mar 13, 2011 9:29 pm
Location: Toronto, Canada

Re: THS7374 vs THS7314 + 1 chip brightness fixing

Post by leonk »

Xaranar wrote:FYI, I'm using borti4938's latest board revision of the SNES RGB bypass, link here, and his board is designed to make use of either the 7373 or the 7374. The 7373 differs from the 7374 in that 3 of its 4 channels make use of an HD LPF, while the other uses an SD one, and the 7374 uses an SD filter on all 4 channels. I've tried it with both amps, and you get a better result with the 7374, the filtering on the 7373 isn't as strong it seems and still shows the jailbars.
It will be helpful if you provide more details on how you use it:

- lpf on rgb amp on/off
- output device (ossc? xrgb mini?)
- what settings you use there? (lpf on output device via ...)
Xaranar
Posts: 112
Joined: Mon Mar 20, 2017 8:43 am

Re: THS7374 vs THS7314 + 1 chip brightness fixing

Post by Xaranar »

leonk wrote:It will be helpful if you provide more details on how you use it:
Sure thing
leonk wrote:- lpf on rgb amp on/off
LPF on the RGB amp on
leonk wrote:- output device (ossc? xrgb mini?)
XRGB Mini
leonk wrote:- what settings you use there? (lpf on output device via ...)
I use FirebrandX's profile, so LPF off.
User avatar
zeruel85
Posts: 150
Joined: Mon Nov 30, 2015 3:47 pm
Location: Italy

Re: THS7374 vs THS7314 + 1 chip brightness fixing

Post by zeruel85 »

From what I know, THS7373 provides a higher cut-off frequency for its internal low-pass filter, so, even if you enable it, you don't see the effects on 15kHz SDTV signals (such as the THS7316, but in that case you cannot bypass the internal LPF).
In conclusion, if you use the THS7373 you need an external LPF, with the typical SDTV cut-off frequency.

Correct me if I'm wrong, please.
Xaranar
Posts: 112
Joined: Mon Mar 20, 2017 8:43 am

Re: THS7374 vs THS7314 + 1 chip brightness fixing

Post by Xaranar »

zeruel85 wrote:From what I know, THS7373 provides a higher cut-off frequency for its internal low-pass filter, so, even if you enable it, you don't see the effects on 15kHz SDTV signals (such as the THS7316, but in that case you cannot bypass the internal LPF).
In conclusion, if you use the THS7373 you need an external LPF, with the typical SDTV cut-off frequency.

Correct me if I'm wrong, please.
That sounds about right to me.
borti4938

Re: THS7374 vs THS7314 + 1 chip brightness fixing

Post by borti4938 »

That's correct. What you have is aliasing due to the ADC with your scaler device (at least I assume that as you didn't provide pictures of what you observed).

The HD filter has nearly no effect in the SD range. So it is just meant to use the THS7316 as a replacement for the THS7314 with 'no' lpf. The filtering for a ADC has to be applied in your ADC-device.
The THS7373 is just to complete the replacements: THS7316 for the THS7314 and THS7373. It might be useful in the case that you don't want to have a filtering in the SD range for the video signals but within your sync-signal (the THS7373 has on SD filter).
User avatar
zeruel85
Posts: 150
Joined: Mon Nov 30, 2015 3:47 pm
Location: Italy

Re: THS7374 vs THS7314 + 1 chip brightness fixing

Post by zeruel85 »

borti4938 wrote:That's correct. What you have is aliasing due to the ADC with your scaler device (at least I assume that as you didn't provide pictures of what you observed).

The HD filter has nearly no effect in the SD range. So it is just meant to use the THS7316 as a replacement for the THS7314 with 'no' lpf. The filtering for a ADC has to be applied in your ADC-device.
The THS7373 is just to complete the replacements: THS7316 for the THS7314 and THS7373. It might be useful in the case that you don't want to have a filtering in the SD range for the video signals but within your sync-signal (the THS7373 has on SD filter).
Hey borti, what about the bypass symbol in the functional block diagram for CVBS channel (I mean on the THS7373 datasheet)? Have you seen it? I supposed that only the 3 HD channels were bypassed... Is there a typo in the diagram?
Xaranar
Posts: 112
Joined: Mon Mar 20, 2017 8:43 am

Re: THS7374 vs THS7314 + 1 chip brightness fixing

Post by Xaranar »

borti4938 wrote:That's correct. What you have is aliasing due to the ADC with your scaler device (at least I assume that as you didn't provide pictures of what you observed).

The HD filter has nearly no effect in the SD range. So it is just meant to use the THS7316 as a replacement for the THS7314 with 'no' lpf. The filtering for a ADC has to be applied in your ADC-device.
The THS7373 is just to complete the replacements: THS7316 for the THS7314 and THS7373. It might be useful in the case that you don't want to have a filtering in the SD range for the video signals but within your sync-signal (the THS7373 has on SD filter).
Hey borti, I just received some of your SNES RGB bypass v4.1a boards from OSH park, and I have to say, you sir are a genius. I was previously using v3 of that board but recently upgraded, and I always had problems with visual noise on my SNES's output, like I would get a speckled effect on the picture that was seemingly random, but after time it would sort of arrange itself into diagonal patterns sloping from right to left upwards, at a very shallow angle, not like the diagonal effect you see when using composite video for sync. Anyway, I found it to be a common thing with Nintendo consoles as my N64 had the same problem, and I think something about the signal Nintendo consoles output is very susceptible to interference, as my Mega Drive has no such issues. I then took apart the JP21 adapter you get packed in with the Framemeister and noticed that its wires are individually shielded, so I rewired it for SCART and tried it out... the visual noise was reduced, but not eliminated. Then, after talking to retrorgb over on another thread, I told him I was using TTL sync levels because for some reason, 75 ohm termination on v3 of your board did not play nicely with the Framemeister, so when I got your new board revision I soldered parts on for TTL sync by default. Retrorgb suggested I try 75 ohm termination, so I did, and since I did that, the visual noise is completely eliminated from my SNES.

TL;DR: Thank you for your hard work and being so awesome!
borti4938

Re: THS7374 vs THS7314 + 1 chip brightness fixing

Post by borti4938 »

Great to read that and thank you for sharing your experience :) I always appreciate those kind of words :D
(I don't own a Framemeister)
zeruel85 wrote:
borti4938 wrote:That's correct. What you have is aliasing due to the ADC with your scaler device (at least I assume that as you didn't provide pictures of what you observed).

The HD filter has nearly no effect in the SD range. So it is just meant to use the THS7316 as a replacement for the THS7314 with 'no' lpf. The filtering for a ADC has to be applied in your ADC-device.
The THS7373 is just to complete the replacements: THS7316 for the THS7314 and THS7373. It might be useful in the case that you don't want to have a filtering in the SD range for the video signals but within your sync-signal (the THS7373 has on SD filter).
Hey borti, what about the bypass symbol in the functional block diagram for CVBS channel (I mean on the THS7373 datasheet)? Have you seen it? I supposed that only the 3 HD channels were bypassed... Is there a typo in the diagram?
I guess, it is a copy 'n' paste error here. The text clearly says that the bypass mode is only available for the three HD channels (although it should not make any differences). Also the pin 9 is labeled with 'HD Bypass'; so I assume that the text is correct here.
User avatar
zeruel85
Posts: 150
Joined: Mon Nov 30, 2015 3:47 pm
Location: Italy

Re: THS7374 vs THS7314 + 1 chip brightness fixing

Post by zeruel85 »

borti4938 wrote:I guess, it is a copy 'n' paste error here. The text clearly says that the bypass mode is only available for the three HD channels (although it should not make any differences). Also the pin 9 is labeled with 'HD Bypass'; so I assume that the text is correct here.
Agree with you! :mrgreen:
takeshi385
Posts: 26
Joined: Sat May 07, 2011 11:10 pm

Re: THS7374 vs THS7314 + 1 chip brightness fixing

Post by takeshi385 »

Hey borti I bought an rgb bypass board based on your design from videogameperfection and I am wondering if you have any insight on why the sides of my screen are darker than the middle of the picture. Best way I can describe it is that it looks like someone is cover both outermost parts of the picture with a rectangular dark clear plastic.
leonk
Posts: 1080
Joined: Sun Mar 13, 2011 9:29 pm
Location: Toronto, Canada

Re: THS7374 vs THS7314 + 1 chip brightness fixing

Post by leonk »

takeshi385 wrote:Hey borti I bought an rgb bypass board based on your design from videogameperfection and I am wondering if you have any insight on why the sides of my screen are darker than the middle of the picture. Best way I can describe it is that it looks like someone is cover both outermost parts of the picture with a rectangular dark clear plastic.
Not enough detail! Bori's rgb amp is the most versitile design one can get. you need to specify more details:

- 1chip or jr snes
- ttl sync or 75 ohm
- what rgb cable. any components?
- lpf on or off?
- video output device
takeshi385
Posts: 26
Joined: Sat May 07, 2011 11:10 pm

Re: THS7374 vs THS7314 + 1 chip brightness fixing

Post by takeshi385 »

leonk wrote:
takeshi385 wrote:Hey borti I bought an rgb bypass board based on your design from videogameperfection and I am wondering if you have any insight on why the sides of my screen are darker than the middle of the picture. Best way I can describe it is that it looks like someone is cover both outermost parts of the picture with a rectangular dark clear plastic.
Not enough detail! Bori's rgb amp is the most versitile design one can get. you need to specify more details:

- 1chip or jr snes
- ttl sync or 75 ohm
- what rgb cable. any components?
- lpf on or off?
- video output device
It's an snes jr with buffered csync. Low pass filters are off. I'm not sure about ttl sync or 75ohm. When I installed it I closed j3 and j1. I am using an xrgb mini with a csync ntsc cable. I am using this cable
Spoiler
because I was told I shouldn't use my official pal gamecube rgb cable.
Spoiler
Image
I'm not sure how well it will show up, but notice how from the "A" in mario all the way to the edge of the screen it is darker. It does this on both sides.
Dochartaigh
Posts: 1520
Joined: Thu Mar 02, 2017 6:53 pm

Re: THS7374 vs THS7314 + 1 chip brightness fixing

Post by Dochartaigh »

copy wrote: Sorry to keep repeating myself, but does this change the recommended resistor value for a simple 1CHIP brightness correction as detailed here (750 Ohm)?
Sorry to well, repeat somebody else who has repeated themselves before (lol), but I'm also looking for this same question to be answered.

I have a phat non-Mini/JR (full size) SNES with a 1CHIP-02. I was about to buy the resistors linked to above to mod mine as it really is too bright (and I'm getting tired of turning down my monitors every time I switch to SNES).

Is this still the best method to reduce that brightness, or should I try the (more complicated) method outlined on page 1? (which I would hate to snap that lead like RetroRGB did...). Thanks.
borti4938

Re: THS7374 vs THS7314 + 1 chip brightness fixing

Post by borti4938 »

@takeshi385:
First, I do not offer official support on these products. So the question in this forum is the right way in a first or second place but please do not adress me in special!

Second, to your problem: It could be also the case that the mod-cable combination is not suitable for the X-RGB mini. First, your eBay reference link does not show what's inside the sync-wire in your cable, and second I don't know how the X-RGB behaves on different sync-setups.
Try to short J2 as well. Also try to adapt the sync level on your X-RGB mini (if possible).
takeshi385
Posts: 26
Joined: Sat May 07, 2011 11:10 pm

Re: THS7374 vs THS7314 + 1 chip brightness fixing

Post by takeshi385 »

I've had this account forever and I finally made my first post last night. Ugh I gotta stop making accounts if all I'm going to do is lurk.

borti4938 wrote:@takeshi385:
First, I do not offer official support on these products. So the question in this forum is the right way in a first or second place but please do not adress me in special!

Second, to your problem: It could be also the case that the mod-cable combination is not suitable for the X-RGB mini. First, your eBay reference link does not show what's inside the sync-wire in your cable, and second I don't know how the X-RGB behaves on different sync-setups.
Try to short J2 as well. Also try to adapt the sync level on your X-RGB mini (if possible).

I appreciate the help guys. With regards to J2, when I originally closed that jumper I didn't get a picture, however I forgot to mess with sync settings. Also with regards to the cable, there isn't any components on the sync line. The sync line is just a passive wire in the cable.

Also um, address you special, I don't understand.
borti4938

Re: THS7374 vs THS7314 + 1 chip brightness fixing

Post by borti4938 »

"Hey Borti..." along with stating the problem sounds like you expect support or at least an answer by me in particular. I just wanted to clarify that this is not the case. ;)
Of course, if I read a problem and nobody else answered before I read the problem, I'm ALWAYS willing to help as good as I can.

Back to your problem:
can you replace the resistors R41 and R42. R41 by a 4.99k and R42 by a 4.3kohm resistor. Optionally short J2 once again and R43. I'm curious if this helps?
takeshi385
Posts: 26
Joined: Sat May 07, 2011 11:10 pm

Re: THS7374 vs THS7314 + 1 chip brightness fixing

Post by takeshi385 »

I don't have those resistors on hand, but I will definitely purchase them. Also I apologize if I came off too casual.
When I saw your reply I interpreted as:
Spoiler
Image
Anyhow the problem isn't too severe after I play with the settings and it is only apparent in certain games. I am currently waiting on the resistor for the brightness fix. One of the first things I did, (although I didn't expect it to fix my current issue) was I installed a 220uf 10v cap to the 7805 as a filter. However the white line issue isn't common with the snes mini.
leonk
Posts: 1080
Joined: Sun Mar 13, 2011 9:29 pm
Location: Toronto, Canada

Re: THS7374 vs THS7314 + 1 chip brightness fixing

Post by leonk »

I suggest you contact the seller of the RGB amp to get support.

Personally, I installed many of Borti's designed RGB amp with 0 problems. I even sent a copy to RetroRGB (who installed it into a 1CHIP SNES) and saw no issues.

It couple be bad amp, bad scart cable, or bad install. The design is solid.
takeshi385
Posts: 26
Joined: Sat May 07, 2011 11:10 pm

Re: THS7374 vs THS7314 + 1 chip brightness fixing

Post by takeshi385 »

leonk wrote:I suggest you contact the seller of the RGB amp to get support.

Personally, I installed many of Borti's designed RGB amp with 0 problems. I even sent a copy to RetroRGB (who installed it into a 1CHIP SNES) and saw no issues.

It couple be bad amp, bad scart cable, or bad install. The design is solid.
Yah the install was super easy. I think I spent about 15 minutes on it. Here are a few pics( my camera had a hard time focusing on the vias). No cold joints.
Spoiler
ImageImageImage
I also replaced the 7805. I don't think the cable is the issue, because there is a lack of components in it. I think it's something do with the snes mini. However the imperfection doesn't show itself when using an av cable. It also could be that I am using j3 instead of j2. When I get a chance I will try to get my xrgb mini to work with the j2 setting.

Also voultar I was out of ribbon cable so I'm sorry, I'm not worthy. However I have done the math and balanced the equations and found that I made a rookie mistake by not having globs of hot glue everywhere. Once again, I'm not worthy. Lol, actually not to be too cheeky, I am really big fan of your work voultar.

UPdate:

I figured out the issue and it is not what I expected. The backlight in my lcd tv is dying, and to a lesser extent my tv is dying. I tried the snes mini with the current setup and another tv and that was how I figured it out. I never thought the bypass board was the issue because my problem to my knowledge couldn't be caused by a sync issue.I thought it was my snes. I figured on the off chance that it was the board I should talk to you guys, because In edition to designing the rgb board I am using, you guys know every video issue with the snes even ones I still can't see.

Thanks for all the help, and I'm kinda sad the adventure ends with the issue being a that my vizio lcd is dying. It just seems anticlimactic, like a bad action film.
Last edited by takeshi385 on Tue Jun 13, 2017 7:44 am, edited 1 time in total.
User avatar
zeruel85
Posts: 150
Joined: Mon Nov 30, 2015 3:47 pm
Location: Italy

Re: THS7374 vs THS7314 + 1 chip brightness fixing

Post by zeruel85 »

@takeshi385: do you have 220uF into the SCART, for RGB signals?
Post Reply