The GoWR, TLOU2, and GoT, megathread! The soup of PS 1st Party goodness

Status
Not open for further replies.

Aidendelaney95

Well-known member
9 Jul 2022
348
449
29
I think we'll get another short trailer for Ragnarok this week to remind everyone about pre-orders going live (not that I think people will have forgotten :ROFLMAO:) Perhaps something similar to the one we got for Horizon Forbidden West, with a couple of new gameplay segments thrown in:

 

Yurinka

Veteran
VIP
21 Jun 2022
7,778
6,665
While that may be true, I don't think the game engine needs to be rewritten for the PS5. They got RT working in the engine and did some improvements with lighting. Also, it's sort of weird to think that they will port the PS4 game engine implementation for the PC and start from scratch with a PS5 engine that isn't portable to the PC.

With regards to the features you mentioned, we all need to be careful that we don't assume just because Epic came up with Nanite/Lumen that other studios will be just as capable. I'm not saying Insomniac isn't capable, but I'm saying that we shouldn't think all other games coming out (from various developers) that are just for PS5/XSX/PC will have an equivalent implementation just because we are in a new generation.
To add RT or the DualSense stuff to a past gen engine is basically nothing for game development compared to the paradign shift changes of things like Natite and Lumen or the stuff Cerny talked about.

We're talking about engines that don't need LODs anymore, not needing tricks like normal maps, ambinet occlusion, baked lighting, shadowing, etc because devs will be able to throw there super detailed models and textures and proper real time illumination. And all this done streaming a super fast speed allowing to stream the content of a room in the time that the door is opened, or even stream what it's behind the player in his same room as he turns to see it.

There are tons of huge paradigm changes involved for engine programmers, game programmers, artists or game designers or level designers. Phil Spencer was right when said that the technology of this generation will mean the biggest change for game development since moving from 2D to 3D games. This time we aren't talking only about being able to have slightly more polys/tris, texture sizes, resolution, framerate and some small new trick to fake lighting, shadowing, reflections or stuff like that.

As Cerny said, in the old system with HDDs and extra bottlenecks there were maze-like level design patterns or using tricks like elevators or slim corridors to hide the streaming and popping, or with stuff with walls, rocks, trees, buildings or big terrain changes in height or direction etc. to keep you a certain amount of time in certain area before you were able to move to (or see) other one that still wasn't streamed. And quite often the "door" to go back to a previous part of the level got stuck/broken/closed/etc and other tropes to don't allow you to go back (to free that part of the level from memory), or you went down through a cliff than later you can't go up.

In the same way, there's the part of damaging or destroying the environment. With all the tricks to fake detail, lighting, shadowing, reflections etc that were used in the past, to destroy let's say a wall of a dungeon would be a mess because many of these baked things would 'break' and their baked tricks would be visible. Now if objects are really super detailed and there is a great lighting, shadows, reflections and so on everything calculated in real time things are very different and there is no problem of breaking the wall of a dungeon in a random point and see proper light from the sun entering and properly changing the lighting, shadows etc. Or to allow you to move or interact with props that in the past where static not for level design reasons but to don't break any trick.

Unreal Engine 5 has been the first one to show a first iteration of this with Nanite and Lumen running in consoles, the other next gen engines are still working on it. The game teams are still learning, testing, prototyping, adapting everything to this. We're still a few years away from seeing the first games using this stuff, and a few years more away from them mastering these things and bringing the level design and game design to take advantage of it in the games where it makes sense.

Until all these things are ready, we're getting released AAA games made with the engine designed for last gen consoles, made using the same workflow than in previous gen consoles, that were started to be developed in 2017 or 2018.
 
D

Deleted member 13

Guest
To add RT or the DualSense stuff to a past gen engine is basically nothing for game development compared to the paradign shift changes of things like Natite and Lumen or the stuff Cerny talked about.
That depends on what you want to do with the RT. Nanite is of course a paradigm change. No argument there.

RT lighting and shading can absolutely be a paradigm change. Notice that these games putting out RT reflections is very easy to implement and a first step into getting something on the screen as that's a completely separate process from the light loop. It can be "added in" on top of the BRDF they have already done using the old ways. I'm not talking about that though. I'm talking about incorporating RT into the entire lighting process. That would be a paradigm shift - especially if you start using importance sampling to get all your BRDF data calculated for the final pixel color. As an example, all GI light probe and environment lookups would have to be all removed as that approach is completely different than RT GI and RT environment lighting. Shadow map usage would also be thrown away which would change your early shadow map generation process. All the tools to lit a scene would therefore need to be rewritten.
 
21 Jun 2022
419
390
First let's see if they release it on PC or not.

Some people had the theory that they would port their GaaS games faster on PC (something I'd understand). But Dreams, MLB, Destruction All Stars and Gran Turismo 7 are not announced for PC.


I think PS5 only games like Returnal, Ratchet Rift Apart or Demon's Sous will take way longer to be ported, I think they'll prefer to keep them as PS5 unique selling points for a while, and until then to release ports of PS4 games (some of them remade/remastered like the Uncharteds, TLOU or Spider-Man) before they get too old.

I think the Ghost of Tsushima port will be released months before its movie or sequel releases as a PS5 only game.

I thought a PS5 version of TLOU2 was going to be released bundled with TLOU1 remake or their TLOU online game to promote the tv show, but seems it won't be the case. So probably they'll wait for a PS5+PC remaster/"Director's Cut" of TLOU2 with maybe TLOU1 bundled as extra content a few months before the season 2 of the tv series goes live.

I think that for the next fiscal year we may see as the next PC ports Bloodborne (with a PS5 version half a year or so released with extra content), MLB23 -they only day one and published by MLB, an exception-, Uncharted Nathan Drake Collection (I feel weird they didn't release it before the current collection), Dreams (a few monts after it gets a PS5/PSVR2 version) and if lucky Sackboy. After them at the end of the FY but more probably in the next FY there would be TLOU2.


I think Elden Ring already won the goty, doesn't matter if GoWR is better than GoW 2018 or Elden Ring. It's multi and the MS and PC fans won't have any goty candidate exclusive to vote. Nintendo fans will get instead as always their "inclusivity representation" as one of their final nominees (games that wouldn't be there if weren't Nintendo exclusives) but if not their vote will be split between Bayonetta, Pokemon and Splatoon 3. Sony's fans instead will have their vote mostly on GoWR but split with Horizon 2 and GT7.

So everything combined, I think will win Elden Ring. I also think that even if both GoWR and Elden Ring end being as good equallly, and same amount of people would have played them, I think people would also favor Elden Ring for being a new IP and having added open world to their genre while GoWR is going to be pretty likely an awesome game but a direct, coninuist sequel that won't add such big chang to its series and genre.
If you follow the game of the year tracker blog you would know that it takes two last year was the first game in a long time to not be exclusive to win the bulk of the game of the Year awards.
 
  • brain
Reactions: Deleted member 140

Yurinka

Veteran
VIP
21 Jun 2022
7,778
6,665
If you follow the game of the year tracker blog you would know that it takes two last year was the first game in a long time to not be exclusive to win the bulk of the game of the Year awards.
True, but that year is a rare exception because covid caused a ton of top 2021 games from many companies being delayed to 2022. In the case of Sony, Horizon 2, GT7 or God of War Ragnarok.
 
D

Deleted member 140

Guest
If God of War Ragnarok is as good as 2018 it'll still lose to Elden Ring, tbh

fawn-cartoon.gif
 

Yurinka

Veteran
VIP
21 Jun 2022
7,778
6,665
That depends on what you want to do with the RT. Nanite is of course a paradigm change. No argument there.

RT lighting and shading can absolutely be a paradigm change. Notice that these games putting out RT reflections is very easy to implement and a first step into getting something on the screen as that's a completely separate process from the light loop. It can be "added in" on top of the BRDF they have already done using the old ways. I'm not talking about that though. I'm talking about incorporating RT into the entire lighting process. That would be a paradigm shift - especially if you start using importance sampling to get all your BRDF data calculated for the final pixel color. As an example, all GI light probe and environment lookups would have to be all removed as that approach is completely different than RT GI and RT environment lighting. Shadow map usage would also be thrown away which would change your early shadow map generation process. All the tools to lit a scene would therefore need to be rewritten.
Yes. We are now with the first steps and with hardware that may not be capable of offering proper full realtime illumination using RT for everyting, or at least not enough RT rays or bounces. But the idea I think is to get rid of all the methods and rendering steps to fake lighting, shadowing and model or texture detail but instead throw there super detailed meshes, each one with a single very detailed texture (no normals/displacement/etc) and proper materials and then put some real light sources.

Then to simplify the work the engine -using hardware very good at this- on the fly drops anything not seen (like off camera stuff, sides of the objects not visible by the camera, sub pixel detail, etc and getting rid of all the tricks to fake detail, lighting, shadows, etc. calculate the lighting of the visible stuff on real time.

I assume the goal -maybe not possible with current hardware is to throw rays once and use them for multiple things at the same time: lighting, shadowing, reflections, translucents stuff, 3D audio etc. Maybe now they do a low resolution version using a limited amount of rays and bounces and have some super optimized way to extrapolate it into a the needed resolution, probably in a similar way they scale up resolution with modern methods.

But I think the main idea is to avoid multiple passes and tricks and instead use a single global solution for big ass meshes and materials for the detail and a single global solution for lighting. Instead of spending rendering time on the multiple passes and tricks they'd spent all that on raw computing power to manage these meshes, materials and lighting. Modern GPUs are capable of that, unlike the ones of generations ago, specially combined with current super fast SSD/IO/memory management that allows streaming big ass stuff to the memory. Memory which btw now has a lot more space so don't need tricks also to fake certain stuff to save memory as happened back in the PS1.
 
  • Like
Reactions: Bo_Hazem

ethomaz

Rebolation!
21 Jun 2022
12,088
9,806
Brasil 🇧🇷
PSN ID
ethomaz
That's personal opinion I'm saying when it comes to critical score and sales even if the new God of War does as well as the first it won't do as well as Elden Ring
Do you mean the 94 vs 96 meta?]
And sales what do you mean? I don't think Elden Ring will sell more than GoW even being in several platforms.... Elden Ring is at 13.4m with pretty recent data (May I believe) while GoW is over 20m with October 2021 data (probably it is a big over that already... maybe even hit 25m after PC release).
 
Last edited:
  • Like
Reactions: Bo_Hazem
D

Deleted member 13

Guest
I assume the goal -maybe not possible with current hardware is to throw rays once and use them for multiple things at the same time: lighting, shadowing, reflections, translucents stuff, 3D audio etc. Maybe now they do a low resolution version using a limited amount of rays and bounces and have some super optimized way to extrapolate it into a the needed resolution, probably in a similar way they scale up resolution with modern methods.
This is exactly what I'm thinking of. Even CG industry cuts down samples and introduce noise into the scene just for it to be denoised by the hardware (or software). I believe this is the best way forward and artist will have a field day saving on setup time for scenes.
 
  • Like
Reactions: Bo_Hazem
21 Jun 2022
419
390
Do you mean the 84 vs 86 meta?]
And sales what do you mean? I don't think Elden Ring will sell more than GoW even being in several platforms.... Elden Ring is at 13.4m with pretty recent data (May I believe) while GoW is over 20m with October 2021 data (probably it is a big over that already... maybe even hit 25m after PC release).

It has 96 on Metacritic vs the first GoW's 94, so the new one has to be better than the original to even match Elden Ring.

The 13.4million was from one month of sales, it has consistently been #2 or #1 on NPD in the months since, also...


God of War hit 19.5 million three years after release

 
  • Like
Reactions: Bo_Hazem

ethomaz

Rebolation!
21 Jun 2022
12,088
9,806
Brasil 🇧🇷
PSN ID
ethomaz
It has 96 on Metacritic vs the first GoW's 94, so the new one has to be better than the original to even match Elden Ring.

The 13.4million was from one month of sales, it has consistently been #2 or #1 on NPD in the months since, also...


God of War hit 19.5 million three years after release

Ops… I just typed wrong :D… 96 vs 94 indeed.

And I don’t believe Elden Ring sell more than GoW no matter how much time it takes.
 

Alabtrosmyster

Veteran
26 Jun 2022
3,399
2,944
As Cerny said, in the old system with HDDs and extra bottlenecks there were maze-like level design patterns or using tricks like elevators or slim corridors to hide the streaming and popping, or with stuff with walls, rocks, trees, buildings or big terrain changes in height or direction etc. to keep you a certain amount of time in certain area before you were able to move to (or see) other one that still wasn't streamed. And quite often the "door" to go back to a previous part of the level got stuck/broken/closed/etc and other tropes to don't allow you to go back (to free that part of the level from memory), or you went down through a cliff than later you can't go up.
You forgot the Naughty Dog ladder!
 

Yurinka

Veteran
VIP
21 Jun 2022
7,778
6,665
It has 96 on Metacritic vs the first GoW's 94, so the new one has to be better than the original to even match Elden Ring.

The 13.4million was from one month of sales, it has consistently been #2 or #1 on NPD in the months since, also...


God of War hit 19.5 million three years after release

Elden Ring debuted in 3 platforms, GoW in one. GoW may end selling maybe some million more in PS now that GoWR is released, and in PC its debut had potential to end selling around 5M copies on PC alone.

Once it's said and done, GoW will have sold pretty likely over 25M copies on both platforms combined. Elden Ring had a super strong start, but let's see if it has legs.
 
  • Like
Reactions: Alabtrosmyster
D

Deleted member 13

Guest
Elden Ring debuted in 3 platforms, GoW in one. GoW may end selling maybe some million more in PS now that GoWR is released, and in PC its debut had potential to end selling around 5M copies on PC alone.

Once it's said and done, GoW will have sold pretty likely over 25M copies on both platforms combined. Elden Ring had a super strong start, but let's see if it has legs.
Yea. Even if GoW outsells ER, IMO it still won't be in the same league as ER as a game. ER just has way more components to it and therefore will stand the test of time like Zelda does.
 
  • Like
Reactions: Katajx

Yurinka

Veteran
VIP
21 Jun 2022
7,778
6,665
Yea. Even if GoW outsells ER, IMO it still won't be in the same league as ER as a game. ER just has way more components to it and therefore will stand the test of time like Zelda does.
Yes, GoW is in the league of a clear GOTY of its year and one of the best games of the generation. Plus also a game that boldly rebooted of the most important IPs in gaming and highly improved its series and genre in all areas: from gameplay and mechanics to narrative and visuals.

ER as of now is only in the GOTY candidate (and I bet winner) league that added open world to its formula. Let's see it after some years people remember ER and as one of the best of the generation.

I am pretty sure that if ER wouldn't have been released in a time where Sony had several big exclusives while Nintendo and specially MS not, or if ER would have been a Sony exclusive, its metacritic would have been pretty lower.
 
  • Like
Reactions: Alabtrosmyster
Status
Not open for further replies.