Should full scene RT be deprioritised until RT solutions are faster? *spawn

[IQandHDR]

A new patch for Spider-Man 2 has enabled RR to work with native resolution.

Marvel's Spider-Man 2 Update 1.212.1.0 Released, Patch Notes

Nixxes released a new patch for the PC version of Marvel's Spider-Man 2. So, let's see what this new update brings to the table.

www.dsogaming.com

Uh, it is almost like technology is progressing right in front of our eyes, who would have thaught

 

[Daozang]

Maybe we need a dedicated ray accelerator, by a company named "RTfx".
please don't kill me

 

[milk]

Maybe we need a dedicated ray accelerator, by a company named "RTfx".
please don't kill me

Dont say that near Microsoft or that will be the name of their next console.

 

[Cyan]

Microsoft says DirectX Raytracing 1.2 will deliver up to 2.3x performance uplift

Well, in the best-case scenario.

www.tomshardware.com

 

[davis.anthony]

I'm surprised we don't see more games using SVOGI as it seems like a nice middle ground in terms of frame time cost and quality.

 

[trinibwoy]

I'm surprised we don't see more games using SVOGI as it seems like a nice middle ground in terms of frame time cost and quality.

It’s a decent solution for diffuse GI but doesn’t help with detailed shadows or reflections.

 

[davis.anthony]

It’s a decent solution for diffuse GI but doesn’t help with detailed shadows or reflections.

Shadows I can live without and reflections, AC Shadows has some pretty terrible looking RT reflections so I personally don't consider that an issue.

 

[trinibwoy]

Shadows I can live without and reflections, AC Shadows has some pretty terrible looking RT reflections so I personally don't consider that an issue.

Yes that’s your personal choice. Now look at it from the perspective of a developer who has to choose between SVOGI and RT. One of which gives you a total lighting solution for your engine and another that’s just low resolution GI. It’s an easy choice if you’re thinking long term so not that surprising that SVOGI didn’t take off.

 

[davis.anthony]

Yes that’s your personal choice. Now look at it from the perspective of a developer who has to choose between SVOGI and RT. One of which gives you a total lighting solution for your engine and another that’s just low resolution GI. It’s an easy choice if you’re thinking long term so not that surprising that SVOGI didn’t take off.

And to think of it another way.

One solution works on more hardware, so will work on more consumers gaming PC's, which increases potential sales.

The other doesn't.

You talk about SVOGI not taking off while forgetting that we've had GPU's with hardware RT for 7 years now, and the adoption rate, has been piss poor at best.

 

[Shifty Geezer]

You talk about SVOGI not taking off while forgetting that we've had GPU's with hardware RT for 7 years now, and the adoption rate, has been piss poor at best.

SVOGI has been around 13+ years and seen far, far less adoption. If RTRT is 'piss poor', how do you grade SVOGI's rate of adoption?

 

[davis.anthony]

SVOGI has been around 13+ years and seen far, far less adoption. If RTRT is 'piss poor', how do you grade SVOGI's rate of adoption?

Whatever you describe as being worse than piss poor.

But I would argue that 13 years ago, there was still a lot to do with normal raster rendering so it took the back seat, and wasn't much of a priory as lighting is with modern games and their assets.

 

[trinibwoy]

And to think of it another way.

One solution works on more hardware, so will work on more consumers gaming PC's, which increases potential sales.

SVOGI will not juice sales. There’s zero marketing around it. It definitely won’t move product more than RT given the hype around that.

You talk about SVOGI not taking off while forgetting that we've had GPU's with hardware RT for 7 years now, and the adoption rate, has been piss poor at best.

What’s your reference point for a technology that was quickly adopted by game developers? RT requires entirely new data structures and art setup. Its adoption has been nothing short of miraculous given the high barrier to entry.

 

[davis.anthony]

SVOGI will not juice sales. There’s zero marketing around it. It definitely won’t move product more than RT given the hype around that.

I never said SVOGI would, I said offering SVOGI over RT would mean potentially more sales (and thus more money) as it would simply run a larger range of hardware, and thus would offer a larger consumer base to sell too.

What’s your reference point for a technology that was quickly adopted by game developers? RT requires entirely new data structures and art setup. Its adoption has been nothing short of miraculous given the high barrier to entry.

An HDR pipeline didn't take 7 years.

Pixel shaders didn't take 7 years.

 

[arandomguy]

An HDR pipeline didn't take 7 years.

In the context of this discussion though -

HDR settings uniformity is much needed

HDR experience for consumers continues to be a struggle and part of that is due to how it's implemented, naming conventions use for the options (or lack there of) and varying descriptions . Without going into HGIG vs Dynamic Tone Mapping etc, Just having a basic settings menu that's unified...

forum.beyond3d.com

It's a shit show on PC as most games won't even let you use HDR in a game unless you have it activated on Windows itself.

I can't imagine it's much trouble to give the game the ability to use HDR without it needing to be on in Windows itself first as some games do it.

 

[davis.anthony]

I'm not talking about HDR displays so that adds nothing to the discussion or to what I said.

 

[trinibwoy]

I never said SVOGI would, I said offering SVOGI over RT would mean potentially more sales (and thus more money) as it would simply run a larger range of hardware, and thus would offer a larger consumer base to sell too.

Or it will potentially get fewer sales due to less media coverage and exposure. This doesn’t seem like a compelling argument to lock your engine out of next gen shadows and reflections.

An HDR pipeline didn't take 7 years.

That’s just a framebuffer format and didn’t change the rendering pipeline much at all.

Pixel shaders didn't take 7 years.

The GeForce 3 launched in 2001. HL2, Doom 3 and Far Cry released in 2004 and were PC exclusives. The PS3 launched in 2006. Pixel shaders took a very long time to get to market. They would take even longer in today's gaming ecosystem.

When people criticize RT they often compare it to some imaginary tech that's adopted overnight. That's not how game development works and never has been.

 

[davis.anthony]

Or it will potentially get fewer sales due to less media coverage and exposure. This doesn’t seem like a compelling argument to lock your engine out of next gen shadows and reflections.

But it is, and so is the lack of performance on the low end.

That’s just a framebuffer format and didn’t change the rendering pipeline much at all.

It was new way of doing lighting so it wasn't as 'flat' and was adopted very quickly.

The GeForce 3 launched in 2001. HL2, Doom 3 and Far Cry released in 2004 and were PC exclusives. The PS3 launched in 2006. Pixel shaders took a very long time to get to market. They would take even longer in today's gaming ecosystem.

That's still much faster than RT and less than 7 years, and even some of the early attempts to use pixel shaders were more widespread than RT use was at the start.

Normal maps and SSAO had a hugely fast adoption rate in games.

When people criticize RT they often compare it to some imaginary tech that's adopted overnight. That's not how game development works and never has been.

I never said it anything has been adopted overnight, I merely highlighted that RT's adoption rate has been and continues to be piss poor and likely will be for the next 7-10 years.

 

[troyan]

Battlefield 5 with Raytracing was released two months after Turing. I cant remember which was the second game with pixelshader support in 2001. Most games had been released after the Xbox 1.

 

[davis.anthony]

And I remember it performing like ass until it was patched.

 

[DavidGraham]

I merely highlighted that RT's adoption rate has been and continues to be piss poor

RT is just DXR, since the introduction of DXR we have seen adoption of more than 215 titles (with dozens more being planned), not to mention numerous mods, remixes and apps. We have moved from ray tracing a single element to multiple elements to path tracing .. objectively speaking, that's more adoption and progress than any recent version of DirectX (DX10/DX11/DX12) .. DX12 especially was a laughing stock in adoption rate, but DXR was and still is the opposite.

and likely will be poor for the next 7-10 years

With Sony and AMD now fully on board, the pace of adoption should be massively accelerated.