Don't mind me, just adding a timestamped transcription here... (0:00) Hey, FirebrandX here with a video tutorial on finding the optimal timing for a given 240p console. Now what you’ll need for this is an Artemio 240p Test Suite. Now I’m using the Neo Geo here, and it has a port of his Test Suite. It’s not made by him, but it has the important test patterns that we need. And specifically what we’re gonna be using is the checkerboard pattern. And this will take all the guesswork out of trying to find optimal timings for your (0:30) consoles. I know a lot of people complain about, “I can’t get rid of the flickering pixels.” That’s just because you’re trying to guess, and you can’t guess. You have to know what you’re doing. And the checkerboard pattern will actually guide you there and make this really easy to do. So right now I have Neo Geo set up in Line 4x mode, but it’s in generic mode. So the first thing I’m gonna do is change the output mode from generic (1:00) to 320x240 optimal. Let me find that there. Line 4x mode is in generic, and it is now in 320x240 optimal. But obviously not perfectly aligned. And we’re going to take care of that. So first let me get the Neo Geo controller and move this down to checkerboard pattern. And you see here there’s all sorts of ugly, vertical, gray, wiggly streaks. (1:30) And those are effectively the timing edge of the signal, if you will. And you see how many there are on the screen. Our object is to adjust each sample rate until there’s just one of those instead of all those going across the screen. And I’ll start by going to - let’s see. Sampling Options. And because we’re now in Optimal Timing mode, we can go down in Sampling Options to (2:00) Advanced Timing for 320x240. And then in that menu, I’m gonna - first thing I’m gonna do is change the active from 240 to 224 because the Neo Geo has a resolution of 320x224, not 320x240. So let’s see here. V. Active. There we are. 240. Let’s change that down to 224. And I’m using a Datapath (2:30) Vision E1S card. I’m using the window right now to see what I’m doing. And after adjusting the V. Active to 224, I notice it’s shifted downward. It’s out of alignment. So now I’m gonna change the V. Backporch to fix that. Increase the value. Keep going, keep going. And one more. And that should do it. Now it’s perfectly aligned with the window. So the important thing now is to go to H. Sample Rate. And their (3:00) default value of 426 as you can see is way off. So I’m gonna start lowering it. And by lowering it, you’ll see that the position of those streaks starts to change. And as you keep going, you’ll notice the gap between them gets larger. So let me jump down to 400. All right, this is 400 now. Now you can see they’re much larger. The gaps between them are much larger. And this lets you know that you’re going in the right (3:30) direction. Now say you went the other way without realizing it. This same test pattern would let you know because those stripes are getting skinnier, and more of them. So then you need - you know you need to go in the other direction. So let me keep lowering it. Let’s go down to 390. And you see 390. Now there’s very few of them, and they’re much larger. The spaces between are much larger. And just keep going. 388. 387. (4:00) 386. 385. Now with 385 here, you can see there’s a stripe on the right you’re sort of picking up on, and a little bit of the one on the left. So it looks like one more value. 384 should do the trick. And you can see here, now all the middle of the stripe is pretty much in the middle of the picture, and that’s no big deal. If I were to keep going, you could see if I keep going lower like 381, 380, the stripes start (4:30) getting more numerous and the gaps smaller between them. And that lets you know that you’ve overshot the mark. So obviously for the Neo Geo, 384 is the target sample rate. And that’s how easy it is to find. But now what we need to do is clean up this picture. And how we do that is we use the Sampling Phase function, which by the way is in the same Sampling Options menu. So Sampling Options, Sampling Phase. And this is the default 180. (5:00) You can see this is really bad right off the bat, so obviously 180 is not correct for this. So I’m gonna increase it. This is 191. You see how it’s started to lighten up a little bit? And I’m increasing it more. It’s getting lighter and less flickery. You just keep doing this until you get a nice, solid, stable picture. And it looks like I lost sync a little bit there, but it’ll come back if I keep going. Yeah, there it came back. That looks pretty good right there. Let’s keep going. This is 303, 315. (5:30) And it’s starting to flicker a little bit as I keep going. It’s at 337. So let’s go back. That looks pretty good. 292. Somewhere around 292 or 303. But let’s say 303 right here. All right, my image is shifted to the left, so I need to adjust my H. Backporch from my Vision window, so let me go ahead and go to that. (6:00) I have to go back to Advanced Timing for 320x224. There it is. And H. Backporch. If I go higher to 50, that’s shifting it the wrong way so I need to adjust it the other direction. Keep going lower and lower. Let’s try 41. I got to - need to go way lower. Let’s try 32. Still lower. Keep going. Looks like one more to do it. All right. H. Backporch (6:30) at 25. Perfectly aligned with my Vision Datapath window. And that’s all there is to adjusting optimal timings for the console itself. Let me go ahead and back out of this test pattern here. There we are. And as you can see, everything’s nice and clean. Pixels are sharp. There’s no flicker anywhere. It looks fantastic. (6:53)

Thank you for this. I've heard the OSSC was quite customisable had no idea on what was needed to improve the video.

I feel sorta stupid that I didn't know you had a KZfaq channel till now. Just saw the tweet you put up for this video. You've just gotten a new subscriber. :)

Incredibly informative thankyou

this video just blew my mind never did I thought I could fine tune the OSSC to do this!! All I did was just plug and play and that was it

Very helpful, thanks!

Useful info, nice vid

thanks! worked great

A thing I like to do to find the best phase setting is to first find the worst - the one that gives the least contrast between pixels. Then I simply offset that by 180°.

Cool stuff..

It would be great if we could see your OSSC at the same time

Thanks for the video. My dumb self was using the overscan feature instead of checkerboard, which is a lot slower than your method. ^-^;

This is a great tutorial. As someone who isn't into flash carts, I wish I could have some other way of duplicating this process. It'd cost me a few hundred in flash carts just to adjust the timings to make my consoles look perfect, unfortunately.

Thanks for the great video, it will be a big help. I have the same capture card and what capture software do you use with it? Any help would be appreciated! Thanks again

Were you able to figure out the optimal timings for the PS2? I recently got an OSSC and just can't figure out the best settings... On the junkerhq wiki is an optimal timings page but it only lists NTSC resolutions for the PS2 and I'm having trouble getting PAL games (576i) to look right. The aspect ratio always seems off to me and I'm alrady getting crazy trying to tweak it by eye as something always looks off... There's also a black frame around the picture which doesn't happen in 480i when the junkerhq settings are used. Also, do you know if the optimal timings posted there can be used for 480p as well? I'm a noob when it comes to this stuff but I really want my games to look right.. Any help is appreciated!

I have to try to figure this out on a dream cast

I wish the N64 had one of these :(

Hi. Are these settings the same for all MVS boards? I have an 1FZ and OSSC but unfortunately no way of running test suite at the moment. Thanks!

There's just a **tad** amount of noise no matter what I do when I use the PS1 240p test suite. Could it be my cable or is this normal?

My Neo Geo AES displays vertical scrolling lines only when connected to OSSC via scart. Is this my console problem?

Awesome video! If the Generic 4x3 works well on a TV and looks good, what advantage is there to using the optimal mode that was dialed in? I guess I'm not seeing the difference.

Is there a test suite availiable for the Sony PlayStation 2 so I can dial in 480i and 480p?

Why they is no option foe some 288p content...i can set 240p/288p proc at lets say 2x and more down next to line2x mode i can only chose ntsc resolutions for example 256x240 optim. What should i chose if i use a n64 pal

Does anyone have a guide on how to get this done with an MS-DOS PC, VGA connection 3x Generic mode? I just can't get rid of the shimmering pixels in this mode. I'm able to do it in the 2x mode, though.

Hi Firebrand - thank you for the Neo Geo settings from your page. I have a couple questions: 1. On my Samsung QLED (Q80r to be exact). Is setting V.Active to 224 the optimal setting? Setting my V. Active down from 240 to 224 causes overscan so the picture makes the upper and bottom portions get cut. Adjusting the V. back porch won’t help as a result. I end up just setting V. Active back to 240. 2. On my QLED and Dell Ultrasharp (16:10) My Video randomly loses signal for about 2-4 seconds and restores itself. Would you happen to know why that happens? My AES is RGB bypassed. I am using both an HD Retrovision and a Scart cable from Retrocables UK.

Thanks but settings for Dreamcast connect to scart???

Hi KarbuncleX Awesome tutorial, best one I've found. I've come across an issue after setting the Megadrive up using your method (which is 320x224 like the NG), strange vertical bars - ibb.co/f7Cxem I followed your process to the letter, everything is aligned properly and the checkerboard test looks great. The picture now looks amazing, best i've ever had it but using the 320x224 optimal output, these vertical bars appear and are not present when using 4:3 generic. They are very noticeable in Sonic 1 in the sky. Any idea what setting I need to adjust to get rid of these? Thanks

Are these settings likely different for individual consoles of the same make and model? Is my AES going to require different timings than yours?

Is there maybe a test suite that would allow me to set up the horrible resolutions PS2 uses?

Hi I have a question about Sampling PHase. Is it only applicable to fine tuning on CRT, PVM,PVM,etc? At the moment I have my Nintendo Wii hooked up via component to my OSSC and then hdmi out to my Elgato HD60 capture card and then to my Asus PC monitor. When I tried doing the Wii test suite and the checkerboard pattern test, I noticed no differences whatsoever when changing the sampling phase.

What would you use as an alternative instead of the window from your datapath card ?

Awesome! Does this guide still apply to Lx5 mode on a 4K TV?

How would you do this for consoles without 240p Test suites?

Is there a way to put this test suite on my NeoSD?

Did you do the test on a tv or monitor?

How did you get your ossc to work with a datapath.card?

Man I love my OSSC I just got but there are too many options and settings and I cannot figure out how to set what up... This video apply to NES and SNES? I always get it looking decent but when moving on SNES everything becomes sharp..then reblurs a little when I stop moving..what is causing that??? I am using a recommended monitor VX3276

what a difference between ossc and emulators shaders on PC ???

Hi FBX, love your work. I've been unable to cut my fingers on razor sharp pixels like yours. I'm running an RGB modded SNES Jr. through the OSSC and into a startech USB3HDCAP. I've gone through you tutorial with a base of 256x240 optimized (as that makes more sense for a starting point on NES/SNES). I've adjusted the H Total to 256 and the V Total to 239, and adjusted the porches to fit the frame perfectly. Then I adjusted the sampling rate until there was one giant blurry bar, no problem, fix the phase. The picture looked great, but there are still some flaws that really bother me. 1. The checkerboard doesn't look razor sharp: cdn.discordapp.com/attachments/152242231665885184/388159066649788416/unknown.png Compression or screenshot might remove the effect that I see in person... 2. Sampling doesn't seem quite right in this picture, specifically look at the red text and how there are two shades of red (where the darker shade is less than the width of a normal pixel): cdn.discordapp.com/attachments/152242231665885184/388158936529764352/unknown.png 3. Color bleed test clearly shows imperfect sampling in red and green: cdn.discordapp.com/attachments/152242231665885184/388159155589873664/unknown.png Particularly issues 2 and 3 are quite noticeable on LCD. I hope you can take the time to take a look at this, hopefully we can both learn something! I disagree that -- "it's just that easy!"

Nes 240p test suite doesn't even have goddamn checkerboard pattern