Is there a better device than OSSC for 480p output?

[Fudoh]

It seems like allowing the monitor to handle the last mile (960p - 1080p) would make the scaler pointless.

it's not pointless. It's like using a prescale option in an emulator. Scaling doesn't hurt, BAD scaling hurts.

The point of using an expensive external scaler is specifically to avoid the blocky poorly scaled image created by almost every flat-screen monitor.

is it about blockiness ? It's usually more about less artefacting or smoother scaling.

With that in mind, what is the reason why people use 960p instead of scaling to fit the native res?

The OSSC isn't a full scaler, so you just don't have the option to scale 480p to 1080p using it.

[orange808]

With that in mind, what is the reason why people use 960p instead of scaling to fit the native res?

There are many.

If something in your chain uses chroma subsampling, you can increase the color resolution before processing to prevent damage to the signal.

You might also use 960p to bypass a smoothing filter that a device in your chain uses for 480p, provided the device handles 960p differently.

You also might scale to get compatibility with a device in your chain. For instance, I have a warp processor that uses a line buffer and dislikes 480p. That line buffer allows for all kinds of transformations with very low latency. For instance, I can feed the OSSC in, get a properly warped signal, and feed that into my Datapathx4 and get perfect tate rotation with scanlines, padding and CRT curve on the edges with just one frame of lag total. Holy grail.

[MidOrFeed2015]

I mean that the ossc can increase the IQ of 480p games (such as xbox, GC, DC, etc) by doing line2x. is this currently the best method? or is there another device that can do much better?

The best method for handing a 480p signal on a modern display is arguably to let the display do all the work. The scalers built-in to modern TVs really are fantastic. The only real exception I can think of are older 1080i-native HD CRTs, but that's really a special corner case. At this point the OSSC is basically a "dumb" ADC converter between the 480p YPbPr analog input and digital HDMI output.

Unfortunately, the OSSC is still flawed in the context of the PS2, as it currently can't automatically sense or react to the mode change from 15kHz RGBS to 31kHz+ RGsB.

With the side question of why would you want to? There's no reason to run RGBs out of a PS2 when YPbPr is just as good with a high-quality cable like the OEM Sony PS2/PS3 cable, HD Retrovision PS2/PS3 cable, or Monster cable.

I should say as well that my 4KTV has no issues displaying 960p (or although not related to topic: Line5x 240p).
direct to TV or Passthrough mode for me is a no go because it looks really really bad.

[thebigcheese]

At least on my TCL, there is a noticeable difference between letting the TV scale 480p vs scaling 960p. With 480p, it's noticeably less sharp, but, on the other hand, that's not necessarily a bad thing. With 960p, it seems to basically just do nearest neighbor scaling instead, so it's a much sharper (but blockier) image. I haven't really made up my mind which I prefer yet, but I believe that input lag is the same. Wish 480i also had the same input lag. Stupid PS2.

[energizerfellow‌]

I should say as well that my 4KTV has no issues displaying 960p (or although not related to topic: Line5x 240p).
direct to TV or Passthrough mode for me is a no go because it looks really really bad.

Define "really really bad".

[MidOrFeed2015]

I should say as well that my 4KTV has no issues displaying 960p (or although not related to topic: Line5x 240p).
direct to TV or Passthrough mode for me is a no go because it looks really really bad.

Define "really really bad".

the colors appear duller and darker, and there is a lot of blur. I much more prefer the sharper 960p look. the only thing that passthrough seems to do correctly is display at 4:3. I can't figure out why my ossc always outputs 960 in widescreen (but I guess that's another issue)

[NoAffinity]

I should say as well that my 4KTV has no issues displaying 960p (or although not related to topic: Line5x 240p).
direct to TV or Passthrough mode for me is a no go because it looks really really bad.

Define "really really bad".

the colors appear duller and darker, and there is a lot of blur. I much more prefer the sharper 960p look. the only thing that passthrough seems to do correctly is display at 4:3. I can't figure out why my ossc always outputs 960 in widescreen (but I guess that's another issue)

you can try changing the 480p in sampler setting between DTV and VESA, see if that changes how the TV handles the output, but it's likely just how the TV handles the output (the OSSC is not to blame ).

[BuckoA51]

You also might scale to get compatibility with a device in your chain. For instance, I have a warp processor that uses a line buffer and dislikes 480p. That line buffer allows for all kinds of transformations with very low latency. For instance, I can feed the OSSC in, get a properly warped signal, and feed that into my Datapathx4 and get perfect tate rotation with scanlines, padding and CRT curve on the edges with just one frame of lag total. Holy grail.

I wish to hear more about your setup, what's a warp processor?

[Dochartaigh]

At least on my TCL, there is a noticeable difference between letting the TV scale 480p vs scaling 960p. With 480p, it's noticeably less sharp, but, on the other hand, that's not necessarily a bad thing. With 960p, it seems to basically just do nearest neighbor scaling instead, so it's a much sharper (but blockier) image. I haven't really made up my mind which I prefer yet, but I believe that input lag is the same. Wish 480i also had the same input lag. Stupid PS2.

I have two TCL TV's which are new(ish) to me. Have you tried any 480p widescreen games on yours? Even with my OG Xbox set to widescreen, it's not showing up on my TCL's (via OSSC) in widescreen - and the TV's stretch or whatever setting doesn't seem to work without cutting off content.

[thebigcheese]

At least on my TCL, there is a noticeable difference between letting the TV scale 480p vs scaling 960p. With 480p, it's noticeably less sharp, but, on the other hand, that's not necessarily a bad thing. With 960p, it seems to basically just do nearest neighbor scaling instead, so it's a much sharper (but blockier) image. I haven't really made up my mind which I prefer yet, but I believe that input lag is the same. Wish 480i also had the same input lag. Stupid PS2.

I have two TCL TV's which are new(ish) to me. Have you tried any 480p widescreen games on yours? Even with my OG Xbox set to widescreen, it's not showing up on my TCL's (via OSSC) in widescreen - and the TV's stretch or whatever setting doesn't seem to work without cutting off content.

I think you have to change something on the OSSC to get it to work. The problem is that I think these older systems don't display a signal that a modern TV recognizes as widescreen, it just thinks it's all squished into 4:3. So when you stretch it, it looks normal. But I'm lazy and don't like dealing with having to change settings/profiles on the OSSC, so I just play everything in 4:3 instead. Seems more authentically retro, anyway.

[Xer Xian]

I wish to hear more about your setup, what's a warp processor?

.. Sorry

[orange808]

I wish to hear more about your setup, what's a warp processor?

.. Sorry

Nope.

That's the quick and easy way, but not the fastest.

[RocketBelt]

My understanding is that it is indeed a myth that its YPbPr output is poorer than RGB, and I've never noticed any visual problems with my SCPH-39001 paired with my first-party PS2 component cable.

I suspect that different PS2 revisions handle this differently. On my PAL v7 the component signal is plainly noisy compared to the rgb signal, using the same cable when switching between outputs. So we can't use the 'shitty cable' argument here.
Worse, interlaced screen modes on the component signal are significantly darker than they should be, (again using the same cable). Tested this on two different display devices as well. RGB has the correct brightness and no noise.
So I can confirm that it is not a myth. There may be no difference on your console, but you can't miss it on mine.

[energizerfellow‌]

I suspect that different PS2 revisions handle this differently. On my PAL v7 the component signal is plainly noisy compared to the rgb signal, using the same cable when switching between outputs. So we can't use the 'shitty cable' argument here.
Worse, interlaced screen modes on the component signal are significantly darker than they should be, (again using the same cable). Tested this on two different display devices as well. RGB has the correct brightness and no noise.
So I can confirm that it is not a myth. There may be no difference on your console, but you can't miss it on mine.

Using a PVM/BVM by chance? If so, you'll want to see if it has a YPbPr level setting with the options of N10/SMPTE, BETA 7.5, or BETA 0. This should take care of the component video brightness issue.

[Ed Oscuro]

My understanding is that it is indeed a myth that its YPbPr output is poorer than RGB, and I've never noticed any visual problems with my SCPH-39001 paired with my first-party PS2 component cable.

I suspect that different PS2 revisions handle this differently. On my PAL v7 the component signal is plainly noisy compared to the rgb signal[...]

I don't want to derail the thread, but I will comment that I did not notice it with slim PS2s (I forget the exact models I tested; I believe 8X / 90K series, one each in US and JPN regions). It might also be a PAL region thing, as I was asked to look for it by Fudoh. I will have to take another look with my new 30K model series PS2s.

[Lawfer]

Worse, interlaced screen modes on the component signal are significantly darker than they should be

Significantly? I haven't noticed that on my 39001 PS2. Of course, RGBS is what you should stick with if you want the best and most accurate output from a PS2.

[RocketBelt]

Using a PVM/BVM by chance?

No, a DVDO VP50 or a normal LG TV.
I don't want to derail this thread either, just wanted to make sure the ps2 component/rgb thing isn't written off as a myth.