OSSC Pro

The place for all discussion on gaming hardware
User avatar
orange808
Posts: 3196
Joined: Sat Aug 20, 2016 5:43 am

Re: OSSC Pro

Post by orange808 »

Satisfying black frame insertion requires an adjustable rolling strobe. 120Hz isn't fast enough to do that with video signal alone.

Doubling the video output frames from your favorite console (after the console produces the video) will reduce lag? How? The software running on the console doesn't know or care if the video output has been manipulated.

The useful "pie in the sky" feature I really want is a video processor that reports itself as a Freesync/HDMI VRR display. (Not to be confused with Nvidia GPU variable refresh, a completely different feature.) Emulators on PC/Linux that can't use CRT emudriver use "Freesync" as a quick and dirty hack. We all know the refresh cadence from a Super Nintendo isn't variable (sans a sync little jitter that isn't emulated, anyhow). An R-Type PCB outputs a stable refresh as well. So, a machine that would report itself as a "Freesync" display, lock to the signal, and output the native frame rate to a non-Freesync display would be very useful. That's the feature I'll pay a huge premium for. Add downscaling and we don't need CRT emudriver anymore (at all!)
Last edited by orange808 on Sun Jan 10, 2021 8:29 pm, edited 1 time in total.
We apologise for the inconvenience
User avatar
Konsolkongen
Posts: 2310
Joined: Fri May 16, 2008 8:28 pm
Location: Denmark

Re: OSSC Pro

Post by Konsolkongen »

When the TV receives a 120Hz signal the input lag is reduced. Halved in case of the LG CX.

https://youtu.be/NrRUTqfB_Ts?t=823
fernan1234
Posts: 2175
Joined: Mon Aug 14, 2017 8:34 pm

Re: OSSC Pro

Post by fernan1234 »

orange808 wrote:So, a machine that would report itself as a "Freesync" display, lock to the signal, and output the native frame rate to a non-Freesync display would be very useful. That's the feature I'll pay a huge premium for. Add downscaling and we don't need CRT emudriver anymore (at all!)
This certainly sounds interesting, and perhaps even feasible without changes to the currently planned FPGA model.
User avatar
orange808
Posts: 3196
Joined: Sat Aug 20, 2016 5:43 am

Re: OSSC Pro

Post by orange808 »

Konsolkongen wrote:When the TV receives a 120Hz signal the input lag is reduced. Halved in case of the LG CX.

https://youtu.be/NrRUTqfB_Ts?t=823
In order to understand this, we must be specific with terminology. Yes, display lag at 120Hz can be (as low as) 8 1/3 ms if the display has no video processing lag.

Display lag isn't input lag and I already explained why.
We apologise for the inconvenience
User avatar
Konsolkongen
Posts: 2310
Joined: Fri May 16, 2008 8:28 pm
Location: Denmark

Re: OSSC Pro

Post by Konsolkongen »

orange808 wrote: In order to understand this, we must be specific with terminology. Yes, display lag at 120Hz can be (as low as) 8 1/3 ms if the display has no video processing lag.

Display lag isn't input lag and I already explained why.
HDTVtest as well as Rtings calls this inputlag, I've never heard otherwise and it really don't matter as it's the same thing in the end. Less lag.
User avatar
orange808
Posts: 3196
Joined: Sat Aug 20, 2016 5:43 am

Re: OSSC Pro

Post by orange808 »

Konsolkongen wrote:
orange808 wrote: In order to understand this, we must be specific with terminology. Yes, display lag at 120Hz can be (as low as) 8 1/3 ms if the display has no video processing lag.

Display lag isn't input lag and I already explained why.
HDTVtest as well as Rtings calls this inputlag, I've never heard otherwise and it really don't matter as it's the same thing in the end. Less lag.
:lol: Have a great day.
We apologise for the inconvenience
User avatar
Konsolkongen
Posts: 2310
Joined: Fri May 16, 2008 8:28 pm
Location: Denmark

Re: OSSC Pro

Post by Konsolkongen »

What a weird condescending behaviour :(
Blacksheep
Banned User
Posts: 78
Joined: Thu Nov 28, 2019 10:49 pm

Re: OSSC Pro

Post by Blacksheep »

marqs wrote:Doubling the refresh rate induces both motion incoherency (2 strobes/scans per frame) and latency (half a frame at minimum) so it's not without tradeoffs. You can do motion interpolation to mitigate the former (remember the 100Hz CRT TVs) but that bumps the lag even more. That said, 240p 120+Hz output should be doable on the Pro model.
viewtopic.php?p=1395980#p1395980
marqs wrote:
H6rdc0re wrote:Would it be possible to double refreshrates? With new TV's we're getting 1080p 120Hz support and this cuts input lag in half.
1080p@120Hz pixel clock of 300MHz exceeds current HW capabilities but 720p@120Hz is still doable. Duplicating refresh rate will cause you min. half-frame latency on input side, though.
viewtopic.php?p=1422936#p1422936
tongshadow
Posts: 613
Joined: Sat Jan 07, 2017 5:11 pm

Re: OSSC Pro

Post by tongshadow »

Konsolkongen wrote:What a weird condescending behaviour :(
Watch this video, should clear up some things:
https://www.youtube.com/watch?v=msOWcvoIC8M

Tl;dr 120hz only means lower input lag when the source is also at 120fps, and higher. This is not revelant for the 60hz sources being doubled to 120hz, and yes games with unlocked frame rate.
User avatar
Josh128
Posts: 2124
Joined: Thu Jan 16, 2014 9:01 am

Re: OSSC Pro

Post by Josh128 »

Blacksheep wrote:
marqs wrote:Doubling the refresh rate induces both motion incoherency (2 strobes/scans per frame) and latency (half a frame at minimum) so it's not without tradeoffs. You can do motion interpolation to mitigate the former (remember the 100Hz CRT TVs) but that bumps the lag even more. That said, 240p 120+Hz output should be doable on the Pro model.
viewtopic.php?p=1395980#p1395980
marqs wrote:
H6rdc0re wrote:Would it be possible to double refreshrates? With new TV's we're getting 1080p 120Hz support and this cuts input lag in half.
1080p@120Hz pixel clock of 300MHz exceeds current HW capabilities but 720p@120Hz is still doable. Duplicating refresh rate will cause you min. half-frame latency on input side, though.
viewtopic.php?p=1422936#p1422936

^^

Im still hoping for this 240p120 option (+motion interpolation if possible) with the ability to output over analog to transform my VGA monitor into a BVM. I'll likely get a Pro just to play around with that feature-- that and being able to line triple 240p, inserting a scanline for one of the tripled lines, and outputting the resulting 720p in a centered, black bordered 768p for my plasma.
User avatar
orange808
Posts: 3196
Joined: Sat Aug 20, 2016 5:43 am

Re: OSSC Pro

Post by orange808 »

The OSSC Pro is a retro video game scaler. What retro machine outputs 120Hz? PCs and consoles outputting 120Hz should be plugged into a switch.

I wrote assembler on the metal in another life. For a trackball or dial, you might poll the controller more than once per frame. That means the position of the player is updated more often than the frame is refreshed. The controller lag is actually lower than users can see. The video update is the bottleneck, there; people can't see how quickly the software is updating the player position in that situation. That's an outlier, but it presents a case where the lag isn't straightforward.

It could be the other way. I may not have time to do all my logic and update the video frame. So, the on screen reaction to input has to wait.

If we throw out the specifics and assume polling the controller once per frame and immediate response, you are still dealing with a machine that outputs between 50 and 70 frames per second. The software polls the controller and updates the game state 50 to 70 times per second in this perfect imaginary hypothetical scenerio.

"Doubling" the video refresh cadence after the video signal comes out doesn't change the total input lag. The true lag is determined by the source's software logic and hardware (plus the amount of time it takes for the display to show a visible change). Our retro consoles don't refresh at 120Hz, so you're cloning frames. The game still updates at 50 to 70 times per second. The cloned frames have no new content. The console/logic won't run any faster. I don't have any new updated frames to display!

That's why your assertion doesn't make sense in this situation.
We apologise for the inconvenience
User avatar
Konsolkongen
Posts: 2310
Joined: Fri May 16, 2008 8:28 pm
Location: Denmark

Re: OSSC Pro

Post by Konsolkongen »

tongshadow wrote:
Konsolkongen wrote:What a weird condescending behaviour :(
Watch this video, should clear up some things:
https://www.youtube.com/watch?v=msOWcvoIC8M
I wasn't aware that the doubling the refresh on the OSSC Pro would increase the lag by half a frame. My assumptions in my previous posts was that this was a lagfree process.
strayan
Posts: 671
Joined: Sun Mar 19, 2017 8:33 pm

Re: OSSC Pro

Post by strayan »

Josh128 wrote:Im still hoping for this 240p120 option (+motion interpolation if possible) with the ability to output over analog to transform my VGA monitor into a BVM. I'll likely get a Pro just to play around with that feature-- that and being able to line triple 240p, inserting a scanline for one of the tripled lines, and outputting the resulting 720p in a centered, black bordered 768p for my plasma.
This is exactly what I’m most excited about. 2560x1440p would also be a killer feature because I’d almost certainly replace my 32 inch HDCRT with a 32 inch 1440p LCD. I’ve been eyeing the Zisworks monitors off for ages but this will widen my options.
Last edited by strayan on Sun Jan 10, 2021 10:58 pm, edited 1 time in total.
User avatar
Kez
Posts: 818
Joined: Thu Jul 20, 2017 7:09 am

Re: OSSC Pro

Post by Kez »

fernan1234 wrote:The OSSC Pro market would be the most demanding kind of user, or at least that's my impression. After all, the original OSSC will continue to be available for those with simpler needs and/or smaller budgets.
tongshadow wrote:Yea, if price is an issue stick to the original OSSC or buy the Retrotink/RADX machines. I understand the Pro has additional features over those machines, but if it's a device aimed towards the power user then why not go all the way and add exclusive features? I think that despite the price it would be very marketable to have the first video processor/scaler that outputs 120hz and linetriples 480p and linedoubles 720p.
I see your points, but there are already people in this thread who were interested and have been put off by the asking price. Being a "power user" is all well and good but most people have an upper limit on how much these features are worth to them and every price increase is going to cross that threshold for someone. I know people want this device to be like the ultimate solution but mostly I'd like for it to be very successful, and I think it will be better for everyone long-term if the device is more accessible to the masses. The more people buy it now, the more firmware updates we'll get and the more likely it is there will be another hardware update in the future.
Last edited by Kez on Mon Jan 11, 2021 8:47 am, edited 1 time in total.
User avatar
orange808
Posts: 3196
Joined: Sat Aug 20, 2016 5:43 am

Re: OSSC Pro

Post by orange808 »

Konsolkongen wrote:
tongshadow wrote:
Konsolkongen wrote:What a weird condescending behaviour :(
Watch this video, should clear up some things:
https://www.youtube.com/watch?v=msOWcvoIC8M
I wasn't aware that the doubling the refresh on the OSSC Pro would increase the lag by half a frame. My assumptions in my previous posts was that this was a lagfree process.
:lol:

We need a rimshot sound effect every time you post.
We apologise for the inconvenience
Blacksheep
Banned User
Posts: 78
Joined: Thu Nov 28, 2019 10:49 pm

Re: OSSC Pro

Post by Blacksheep »

tongshadow wrote:Watch this video, should clear up some things:
https://www.youtube.com/watch?v=msOWcvoIC8M

Tl;dr 120hz only means lower input lag when the source is also at 120fps, and higher. This is not revelant for the 60hz sources being doubled to 120hz, and yes games with unlocked frame rate.
That video demonstrates what happens when the frame rate is increased while the refresh rate is kept constant. We were writing about the opposite though.

Here is why doubling the refresh rate causes half a frame of latency:

After receiving half a frame, you begin to output that at double the speed. When the next frame begins, you output the previous frame again (of course still at double the speed) while buffering the current one. Rinse and repeat.
User avatar
NormalFish
Posts: 282
Joined: Tue May 26, 2015 3:35 pm

Re: OSSC Pro

Post by NormalFish »

Kez wrote:The more people buy it now, the more firmware updates we'll get and the more likely it is there will be another hardware update in the future.
Especially given the open source nature of the project, releasing the highest-possible-end unit out of the gate just doesn't make sense. If there are enough people in desperate need of a 4k OSSC Pro, then there's nothing stopping that from continuing after the release of this project. But if the project is largely disregarded by the broad community because it's too expensive, there won't be as many people around interested in working on iterative projects.
N64
Posts: 15
Joined: Sun Aug 16, 2020 4:37 pm

Re: OSSC Pro

Post by N64 »

Blacksheep wrote:Also note that 240p x6, 480p x3 and 720p x2 would already be possible with the current design (meaning without the beefier FPGA and HDMI chip) with an output resolution of 1920x1440.
That's only true for 4:3 content, 1280x720 can't be linedoubled to 1920x1440. I'm thinking of stuff like the OG Xbox, which can output 480p, 720p, or 1080i but not 1080p. I prefer to output 720p, but it looks a little blurry on a 4K TV, and most 4K TVs can take a 1440p signal well, and 1440p monitors are very common. 1920x1440 will certainly be a nice option for 240p and 480p 4:3 content, but it seems like you'll have no way of knowing until you try which displays will play nice with a 1920x1440 signal since it isn't as widely used.
Blacksheep
Banned User
Posts: 78
Joined: Thu Nov 28, 2019 10:49 pm

Re: OSSC Pro

Post by Blacksheep »

N64 wrote:That's only true for 4:3 content, 1280x720 can't be linedoubled to 1920x1440.
That is both incorrect.

First, you assume that the input signal has pixels, which is not the case with analog signals. You can sample an analog signal that's intended to be displayed at 16:9 as 4:3. The result will be too skinny, but could be stretched again to 16:9 at a later stage, e.g. by the display. You will lose detail by doing that. (Depending on the source bandwidth, you may not lose detail at all by doing that, if you sample it right.) [update2]

Furthermore, if the input is a 1280x720 digital signal, you can generate a 960x720 / line-doubled 1920x1440 signal out of that—depending on the approach it will result in loss of detail, a wrong aspect ratio, or missing parts of the picture. In both the analog and digital input case, you can skip samples/pixels/columns at e.g. an appropriate interval, e.g. every 4 and improve on that by interpolating. You can also discard portions at the beginning and/or end of lines so as to effectively crop the input to 4:3. This way, content pixel aspect ratio is preserved and what is visible is not reduced in detail. Or you can leave the horizontal untouched/use an horizontal factor of 1, just doing 1280x(720 * 2), i.e. 1280x1440 embedded into 1920x1440: then nothing is cut off and no reduction of detail occurs.

Lastly, there is anamorphic 16:9 DAR 960x720 (non-square pixels), look up e.g. DVCPRO HD.

(I should have added "up to 60Hz" and "isomorphic 16:9 720p with compromises" to the original post.) [update]
Last edited by Blacksheep on Mon Jan 11, 2021 4:27 pm, edited 2 times in total.
strumm
Posts: 5
Joined: Mon Jan 11, 2021 2:48 am

Re: OSSC Pro

Post by strumm »

For all the Commodore 64 users, it seems like it would be worthwhile to have S-Video as on board supported. pretty please!
User avatar
kitty666cats
Posts: 1270
Joined: Tue Nov 05, 2019 2:03 am
Location: Massachusetts, USA

Re: OSSC Pro

Post by kitty666cats »

tongshadow wrote:Cant wait to use 1440p straight to my CRT monitor on the VGA output without needing shitty HDMI->VGA converters that barely work. Also means I could have an AIO solution to connect my PS3 and other consoles to my CRT cleanly.
Derpity derp, I am confused here - is the Pro going to have VGA out? Or is this in reference to a (hypothetical or confirmed...?) add-on board?

I went back a few pages from this post but did not see anything definitive, may have missed it :o
tongshadow
Posts: 613
Joined: Sat Jan 07, 2017 5:11 pm

Re: OSSC Pro

Post by tongshadow »

It's gonna have VGA Out through the GPIO expansion port. I only need to figure out how to output analog audio now...
Blacksheep
Banned User
Posts: 78
Joined: Thu Nov 28, 2019 10:49 pm

Re: OSSC Pro

Post by Blacksheep »

tongshadow wrote:It's gonna have VGA Out through the GPIO expansion port.
No, there is no VGA going through the expansion port. schematic_v1p3.pdf, p. 15/18
port24
Posts: 5
Joined: Wed Dec 18, 2019 8:57 pm

Re: OSSC Pro

Post by port24 »

@marqs I was under the mistaken impression that someone had requested OLED. You having it by default is different.

I think being a pro product, price should be viewed as less of an issue. Having support for 720p scaling would be amazing.
User avatar
Josh128
Posts: 2124
Joined: Thu Jan 16, 2014 9:01 am

Re: OSSC Pro

Post by Josh128 »

Blacksheep wrote:
tongshadow wrote:It's gonna have VGA Out through the GPIO expansion port.
No, there is no VGA going through the expansion port. schematic_v1p3.pdf, p. 15/18
According to marqs himself:
2x20 pin GPIO connector for future expansion possibilities such as:
* composite & s-video input module
* secondary video output (e.g. VGA) module
* latency tester interface module
* game controller port module (for using OSSC Pro as a dedicated FPGA console)
Blacksheep
Banned User
Posts: 78
Joined: Thu Nov 28, 2019 10:49 pm

Re: OSSC Pro

Post by Blacksheep »

The question about a VGA out / whether an add-on board would be required was answered by the statement that the expansion port will have a VGA out, which is false. As it stands, you'll need an add-on board which has a digital to analog converter, as your quote implies.
User avatar
Josh128
Posts: 2124
Joined: Thu Jan 16, 2014 9:01 am

Re: OSSC Pro

Post by Josh128 »

Blacksheep wrote:The question about a VGA out / whether an add-on board would be required was answered by the statement that the expansion port will have a VGA out, which is false. As it stands, you'll need an add-on board which has a digital to analog converter, as your quote implies.
Not going to argue semantics. It will indeed be capable of VGA out (at some point) via an expansion module attached to the GPIO connector.
Blacksheep
Banned User
Posts: 78
Joined: Thu Nov 28, 2019 10:49 pm

Re: OSSC Pro

Post by Blacksheep »

Josh128 wrote:Not going to argue semantics. It will indeed be capable of VGA out (at some point) via an expansion module attached to the GPIO connector.
Semantics are important if we strive to understand each other. About the VGA out via an add-on board attached to the expansion port (I wouldn't call it GPIO connector for a few reasons), how do you know? If nobody ever makes one, then it won't. However, I do not surmise that it will pan out like that.
tongshadow
Posts: 613
Joined: Sat Jan 07, 2017 5:11 pm

Re: OSSC Pro

Post by tongshadow »

Blacksheep wrote:
Josh128 wrote:Not going to argue semantics. It will indeed be capable of VGA out (at some point) via an expansion module attached to the GPIO connector.
Semantics are important if we strive to understand each other. About the VGA out via an add-on board attached to the expansion port (I wouldn't call it GPIO connector for a few reasons), how do you know? If nobody ever makes one, then it won't. However, I do not surmise that it will pan out like that.
It looks common:
https://market.samm.com/gert-vga-666-gpio-to-vga
Blacksheep
Banned User
Posts: 78
Joined: Thu Nov 28, 2019 10:49 pm

Re: OSSC Pro

Post by Blacksheep »

tongshadow wrote:
Blacksheep wrote:
Josh128 wrote:Not going to argue semantics. It will indeed be capable of VGA out (at some point) via an expansion module attached to the GPIO connector.
Semantics are important if we strive to understand each other. About the VGA out via an add-on board attached to the expansion port (I wouldn't call it GPIO connector for a few reasons), how do you know? If nobody ever makes one, then it won't. However, I do not surmise that it will pan out like that.
It looks common:
https://market.samm.com/gert-vga-666-gpio-to-vga
Very good. Although that board was not specifically designed for the OSSC Pro (and it currently says "out of stock" in that webshop you linked, although I've seen that it's an open source design), a cursory glance suggests that it could be usable. I've seen that there also exists at least one design for a Raspberry Pi VGA Board for which you wouldn't have to discard 6 out of 24 RGB bits. However, it has to be noted that, before such a board is usable with the OSSC Pro, the OSSC Pro firmware must be extended so that the video data is sent to the pins of the expanson port—the current public revision cannot do this. I'm confident that this will happen eventually.
ross wrote:Earlier in this thread, it was said that the adaptive line multiplier in the current prototype can output any resolution up to a 180 MHz pixel clock, but are the specific active pixel values on each axis limited by and encompassed in these output resolutions?
marqs wrote:Speaking of combinations, below is table for planned (most already implemented) adaptive line multiplication output modes. Together with the pure LM modes (identical to oriiginal OSSC), they should cover a good amount of common output resolutions which are suitable for frame sizes resulting from line multiplication.

R= 240p proc & 288p proc & 480i proc & 576i proc & 480p proc & 576p proc R= 720x480 [Line2x] & 720x576 [Line2x] & 720x240 [240p restore] & 720x288 [288p restore] & 720x240 [line drop] & 720x288 [line drop] R= 1280x720 [Line3x] & 1920x1080i [Line4x] & 1280x1024 [Line4x] & 1920x1080i [Line4x] & 1280x1024 [Line2x] & 1920x1080i [Line2x] R= 1280x1024 [Line4x] & 1920x1080 [Line4x] & 1920x1080i [Line4x] & 1920x1080 [Line4x] & 1920x1080i [Line2x] & 1920x1080 [Line2x] R= 1920x1080i [Line4x] & 1920x1440 [Line5x] & 1920x1080 [Line4x] & & 1920x1080 [Line2x] & R= 1920x1080 [Line4x] & & 1920x1440 [Line6x] & & 1920x1440 [Line3x] & R= 1920x1080 [Line5x] & & & & & R= 1600x1200 [Line5x] & & & & & R= 1920x1200 [Line5x] & & & & & R= 1920x1440 [Line6x] & & & & &
So I assume something like 3840x240 is impossible to output, or 1440x1080 without black bars encoded in the signal?
What you're suggesting is not impossible to output, the clockgen should manage that. However, the purpose of the adaptive line multiplication mode is to generate compatible output resolutions. You're suggesting to diverge from compatible resolutions, which is at odds with that purpose. Perhaps another output mode would be better suited for the application you're thinking of (if you had a specific one in mind).
Post Reply