RetroTINK 5x-Pro

Odolwa · Joined: 29 Apr 2021 Posts: 165

parodius wrote:

lmgtfy
https://www.retrorgb.com/retrotink-5x-f ... v2-73.html

That link doesn’t explain what genlock is.

Steven · Joined: 11 May 2021 Posts: 1358 Location: Tokyo

EnragedWhale wrote:

It doesn’t. It’s reporting the buffer lag, not total lag and is labelled as such. It says right there in the firmware update notes. I can’t imagine it’s possible to do triple buffer any faster. Everyone should be using genlock or frame lock anyway assuming their display can handle it. Triple buffer is a compatibility mode, but even then is still very low lag.

Does the triple buffer not cause the lag, seeing as the lag is specifically only there when the triple buffer is active? In that case it should definitely say that it has more than a few ms of lag with the triple buffer active.

EnragedWhale · Joined: 21 Jul 2014 Posts: 144 Location: UK

Steven wrote:

EnragedWhale wrote:

It doesn’t. It’s reporting the buffer lag, not total lag and is labelled as such. It says right there in the firmware update notes. I can’t imagine it’s possible to do triple buffer any faster. Everyone should be using genlock or frame lock anyway assuming their display can handle it. Triple buffer is a compatibility mode, but even then is still very low lag.

Does the triple buffer not cause the lag, seeing as the lag is specifically only there when the triple buffer is active? In that case it should definitely say that it has more than a few ms of lag with the triple buffer active.

The buffer lag readout is there at all times and doesn’t change when you switch between the different frame output modes. It’s the base lag that the RT5X requires to achieve the scaling you’ve selected. It’s the exact lag you’ll get in frame or gen lock modes. Remember the RT5X is a upscaler, not a line doubler so it needs a few more ms than an OSSC etc.

Triple buffer adds lags on top of this which is unavoidable due to the nature of how triple buffering works.

EnragedWhale · Joined: 21 Jul 2014 Posts: 144 Location: UK

parodius wrote:

EnragedWhale wrote:

apparently Genlock + VRR is immune to sync drops when changing resolutions on some displays.

Do you know if this applies to comp/s-vid only or RGB/YPbPr as well?

I don’t sorry. I’d test it for you but I have a launch RT5X without genlock.

EnragedWhale · Joined: 21 Jul 2014 Posts: 144 Location: UK

Shake Your Rump wrote:

This seems to be the only forum discussing this device outside of Reddit (or discord).

After all this time, anyone want to share their preferred CRT mask settings? I’ve tinkered a bit, but always come back to the default PVM600 settings. Currently playing Wild Arms on the PS1 and was surprised that I actually like the “consumer-2” settings.

What are you folks using for 240p mask effects?

I also like the PVM600 settings but I bump colour up by one. I even use it on 480p/576p sources as I like the texture it gives although it’s probably in no way authentic.

Steven · Joined: 11 May 2021 Posts: 1358 Location: Tokyo

EnragedWhale wrote:

Steven wrote:

EnragedWhale wrote:

It doesn’t. It’s reporting the buffer lag, not total lag and is labelled as such. It says right there in the firmware update notes. I can’t imagine it’s possible to do triple buffer any faster. Everyone should be using genlock or frame lock anyway assuming their display can handle it. Triple buffer is a compatibility mode, but even then is still very low lag.

Does the triple buffer not cause the lag, seeing as the lag is specifically only there when the triple buffer is active? In that case it should definitely say that it has more than a few ms of lag with the triple buffer active.

The buffer lag readout is there at all times and doesn’t change when you switch between the different frame output modes. It’s the base lag that the RT5X requires to achieve the scaling you’ve selected. It’s the exact lag you’ll get in frame or gen lock modes. Remember the RT5X is a upscaler, not a line doubler so it needs a few more ms than an OSSC etc.

Triple buffer adds lags on top of this which is unavoidable due to the nature of how triple buffering works.

That's what I'm saying; it says buffer lag, but it's not actually displaying the lag that is caused by the triple buffer, so perhaps it should be renamed to something else.

cfx · Joined: 02 Dec 2008 Posts: 413

Shake Your Rump wrote:

This seems to be the only forum discussing this device outside of Reddit (or discord).

After all this time, anyone want to share their preferred CRT mask settings? I’ve tinkered a bit, but always come back to the default PVM600 settings. Currently playing Wild Arms on the PS1 and was surprised that I actually like the “consumer-2” settings.

What are you folks using for 240p mask effects?

I greatly prefer Shadow Mask to any of the other presets. PVM600 is probably second for me, but the (black) scanlines are too heavy for my taste.

I haven't spent enough time playing with the individual settings to possibly get something I like more.

I don't like the look of actual BVM or PVM CRTs with the razor-sharp scanlines and large spacing, and prefer the look of an RGB arcade monitor or similar. The larger PVMs like the 25" and 27" are fine too but they don't look like what people generally mean when they say "PVM". I have a JVC DT-V1710CG and I hate the picture on it.

Shake Your Rump · Joined: 19 Jan 2023 Posts: 4

EnragedWhale wrote:

I also like the PVM600 settings but I bump colour up by one. I even use it on 480p/576p sources as I like the texture it gives although it’s probably in no way authentic.

For a while now, I’ve been using the settings someone posted here long ago that is essentially the PVM600 preset with colour up one and pre-emphasis down two. It’s pretty nice looking.

I’m a big fan of 1080(over), so I can’t use the setting for 480i/p PS2 games, as the image doesn’t scale properly. I actually went for a month or so using the PVM scanlines on my PS2 until I finally realized why my games had these weird bright lines going through the image. It was due to the uneven scaling with scanlines.

cfx wrote:

I greatly prefer Shadow Mask to any of the other presets. PVM600 is probably second for me, but the (black) scanlines are too heavy for my taste.

I haven't spent enough time playing with the individual settings to possibly get something I like more.

I don't like the look of actual BVM or PVM CRTs with the razor-sharp scanlines and large spacing, and prefer the look of an RGB arcade monitor or similar. The larger PVMs like the 25" and 27" are fine too but they don't look like what people generally mean when they say "PVM". I have a JVC DT-V1710CG and I hate the picture on it.

The thing about the PVM600 preset that got me using it is that, other than the pixel shape, it looks extremely similar to my 27” Sony FV310. The FV310 preset on the RetroTink nails the weird pixel shape, but the aperture grille and scanlines of the PVM600 preset more closely match the CRT.

I guess the screen size makes quite a difference. I mean, on a CRT, the scanlines would be different thicknesses based on physical screen size. The RetroTink doesn’t take this into consideration. It would be neat if a physical screen size could be entered into the settings, and the proper characteristics are calculated. I guess we just have to eyeball it.

Shake Your Rump · Joined: 19 Jan 2023 Posts: 4

Odolwa wrote:

parodius wrote:

lmgtfy
https://www.retrorgb.com/retrotink-5x-f ... v2-73.html

That link doesn’t explain what genlock is.

From what I have gathered:
Frame lock uses the clock signal from the console. Gen lock detects the clock signal, but creates its own, matching, clock signal instead. This has the benefit of being more stable.

My experience is that you can use "gen lock" instead of "triple buffer" for a rock-solid image. Even the 240p to 480i video switching is solid.

Paraphrasing from a recent interview with Mike: triple buffer is there if you absolutely need it, but gen lock is the ideal setting now.

bobrocks95 · Joined: 30 Apr 2012 Posts: 3090 Location: Kentucky

Steven wrote:

That's what I'm saying; it says buffer lag, but it's not actually displaying the lag that is caused by the triple buffer, so perhaps it should be renamed to something else.

I'd agree, if there's going to be a lag measurement displayed to the user ideally it should be the total lag of the entire processing chain. If that's not possible and it's always going to be right around 4ms I think we're going to have confused people using triple buffer and not understanding the downsides. Not that it's a huge deal for people who look into it enough to care- triple buffer is certainly the most compatible and I think the 5X ships with it on by default?

Odolwa · Joined: 29 Apr 2021 Posts: 165

Is Gen Lock ideal in most situations? Does it look better than Framelock? Would you use it in place of Framelock or just in place of Triple Buffer Mode? Why does Gen Lock not work on older versions of the RT5X?

SavagePencil · Joined: 11 Nov 2013 Posts: 394

Odolwa wrote:

Why does Gen Lock not work on older versions of the RT5X?

lmgtfy

Hardware revision for launch devices makes it incompatible.

EnragedWhale · Joined: 21 Jul 2014 Posts: 144 Location: UK

Odolwa wrote:

Is Gen Lock ideal in most situations? Does it look better than Framelock? Would you use it in place of Framelock or just in place of Triple Buffer Mode? Why does Gen Lock not work on older versions of the RT5X?

Mike said on Bob’s stream that most people should use Gen Lock. It’s functionally the same as frame lock but more compatible. If you used frame lock before with no issues then you won’t observe any difference, but it is more resistant to signal drops and may also enable faster re syncing on resolution changes depending on your display.

cfx · Joined: 02 Dec 2008 Posts: 413

Shake Your Rump wrote:

I guess the screen size makes quite a difference. I mean, on a CRT, the scanlines would be different thicknesses based on physical screen size. The RetroTink doesn’t take this into consideration. It would be neat if a physical screen size could be entered into the settings, and the proper characteristics are calculated. I guess we just have to eyeball it.

I don't know if that's true, but even if it is, it is not why the PVM-2530 or PVM-2950 look different than the others. They're not the same type of monitor, and were much less expensive costing only about $1500 when new--I know because I bought one back then, though I might not remember exactly. I don't know the specs but I'm sure the TVL ratings were lower. As far as I know, no one made a picture tube of the type used in BVMs or smaller PVMs in that large of a size, at least not until the HD CRTs came along.

768peeistrash · Joined: 01 May 2021 Posts: 7

which system/cable connection shines the brightest for y'all on the 5x pro and what basic settings?

generic or optimal phase, h sampling, and scanlines or not

fernan1234 · Joined: 14 Aug 2017 Posts: 1964

generic+scanlines or bust

SavagePencil · Joined: 11 Nov 2013 Posts: 394

@Odolwa, the RetroTink 5X wiki is surprisingly useful. And it does go into detail on the different lock features and mechanisms.

Triple Lei · Joined: 01 Apr 2005 Posts: 361

SavagePencil wrote:

@Odolwa, the RetroTink 5X wiki is surprisingly useful. And it does go into detail on the different lock features and mechanisms.

Thanks for the link, but man, tons of misinformation in the PSP section. Who wrote this crap?

Quote:

The specific model of PSP you're using will determine whether it's capable of Video Out, and to what extent:

The original PSP model (PSP-1000) and the PSP Street (PSP-E1000) do not support Video Out.
The PSP-2000 is capable of Video Out, but only using Component cables. PSone Classics can also not be played using Video Out on this model.
The PSP-3000 can do Video Out using Composite, S-Video and Component cables, including with PSone Classics.
The PSP Go is also capable of Video Out for all content, but only when using an official dock. Because of this, it requires a different cable.

Correct info & rant in spoiler tag:

Spoiler: show

Anyway, since we still have no "official" PSP preset from Mike, it seems a bunch of us are independently looking for the best PSP settings on the RT5X, especially if they need to be changed with each firmware as this dude suggests. I'm not the only one who's not completely satisfied with Wobbling Pixel's settings; I don't mind cropping the gameplay area a bit if it means no black borders at all, for example. Here are my old settings. I just took a screenshot of the menu in WipEout Pure with the OSSC preset first, and I tried to match my RT5X to that.

One thing I'm damn sure of is I'm keeping "V Size" at +60% no matter what. It might crop a little bit of the picture on the top and bottom, but that's the only size at which BVM scanlines still look correct (at least at Generic 4:3), which has to mean... something. I dunno, I got the idea from FirebrandX's OSSC video from 2017.

To do:

Find best zoom/crop/stretch values for all those PSP compilations with "pixel perfect" modes (I've found I'm going to need at least two of these)
Revisit Final Fantasy Tactics: War of the Lions 240p patch with RT5X

768peeistrash · Joined: 01 May 2021 Posts: 7

fernan1234 wrote:

generic+scanlines or bust

right but which scanline setting?

Josh128 · Joined: 16 Jan 2014 Posts: 1800

768peeistrash wrote:

right but which scanline setting?

Your username is incorrect.

evil_ash_xero · **Posted:** Tue Jan 31, 2023 4:44 am

I've read that this has bilinear interpolation. Is this the same as a bilinear filter on an emulator?

Seems like something like that would be pretty useful on PSX or Saturn games, especially.

BuckoA51 · Joined: 02 Oct 2010 Posts: 3288 Location: Ireland

Odolwa wrote:

Is Gen Lock ideal in most situations? Does it look better than Framelock? Would you use it in place of Framelock or just in place of Triple Buffer Mode? Why does Gen Lock not work on older versions of the RT5X?

On my dejittered SNES and my LG C9 I actually find regular frame lock to be better and genlock drops out occasionally.

Unseen · Joined: 25 May 2014 Posts: 695

evil_ash_xero wrote:

I've read that this has bilinear interpolation. Is this the same as a bilinear filter on an emulator?

Seems like something like that would be pretty useful on PSX or Saturn games, especially.

No, it will not make blocky textures look smooth.

Rock Man · Joined: 06 Mar 2010 Posts: 459 Location: Southtown

Unseen wrote:

evil_ash_xero wrote:

I've read that this has bilinear interpolation. Is this the same as a bilinear filter on an emulator?

Seems like something like that would be pretty useful on PSX or Saturn games, especially.

No, it will not make blocky textures look smooth.

I found this out the hard way. Perhaps a firmware update can be considered in the future if Mike is down for that!

bobrocks95 · Joined: 30 Apr 2012 Posts: 3090 Location: Kentucky

Texture filtering isn't something that can be done with a post-processing filter. Best you could do is blur the whole screen. Might as well just use composite at that point.

Rock Man · Joined: 06 Mar 2010 Posts: 459 Location: Southtown

bobrocks95 wrote:

Texture filtering isn't something that can be done with a post-processing filter. Best you could do is blur the whole screen. Might as well just use composite at that point.

What's your perspective of the mClassic? For me it tends to smooth stuff out but not to the degree I'd like. regardless there is some smoothing that takes place, it somehow manages to avoid making everything blurry.

bobrocks95 · Joined: 30 Apr 2012 Posts: 3090 Location: Kentucky

Rock Man wrote:

bobrocks95 wrote:

Texture filtering isn't something that can be done with a post-processing filter. Best you could do is blur the whole screen. Might as well just use composite at that point.

What's your perspective of the mClassic? For me it tends to smooth stuff out but not to the degree I'd like. regardless there is some smoothing that takes place, it somehow manages to avoid making everything blurry.

It's kind of like enabling FXAA on a PC game, right? Edge-detection post-processing anti-aliasing, maybe some contrast or edge enhancement on top? Not for me at all, but if it looks good to you more power to you. Me suggesting composite isn't really a joke either- if you really truly hate jaggies on low-res 3D consoles maybe you'll honestly like composite? I would rather have the polygons cut my eyes myself lol

orange808 · Joined: 20 Aug 2016 Posts: 2782

The OSSC had a filter firmware. Try that. I suggest adding a Darbeevision last in the chain before the display to sharpen the image a little. You'll have control over the output and it's better than the mClassic. Let's face it, the mClassic kinda sucks.

evil_ash_xero · **Posted:** Fri Feb 03, 2023 5:56 am

orange808 wrote:

The OSSC had a filter firmware. Try that. I suggest adding a Darbeevision last in the chain before the display to sharpen the image a little. You'll have control over the output and it's better than the mClassic. Let's face it, the mClassic kinda sucks.

Oh does it? What kind of filters does it have? Like Bilinear or Blargg's NTSC filter?

Rock Man · Joined: 06 Mar 2010 Posts: 459 Location: Southtown

orange808 wrote:

The OSSC had a filter firmware. Try that. I suggest adding a Darbeevision last in the chain before the display to sharpen the image a little. You'll have control over the output and it's better than the mClassic. Let's face it, the mClassic kinda sucks.

This gives me hope that the Tink could get such a feature.

It is currently Mon Mar 20, 2023 6:37 pm	View unanswered posts
Board index » Technical » Hardware	View active topics

shmups.system11.org

RetroTINK 5x-Pro

Who is online