• Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.

NXGamer: Shadow of the Tomb Raider : Now Gen Update Analysis - PS5 | SX | SS | PRO

Arioco

Member
Didn't know it had a good CB method. Most of the time CBR looks pretty bad but it looks fine in this one.


There are very good CB implementations, Horizon, God of War, Detroit, Sekiro, any game using RE Engine...

And there are some bad implementations like FFXV or Red Dead Redemption.

Un general the results are good and I thinks CB is a good reconstruction technique that allows to save on resources that devs can use somewhere else.
 

01011001

Banned
The cb was visible for me. Series x looks like the way to go.

they still should have patched in an option on Series X to use the same CB solution.

the way it stands now, this patch was completely unnecessary on Series X, they literally changed nothing that the FPS boosted/pre patch version didn't do.

so this makes ot a bit weird to me that they even bothered to patch it.
 
Last edited:

AnotherOne

Member
Season 2 Fergie GIF by The Four
 

GametimeUK

Member
Checkerboard rendering obviously doesn't look as good as 4K, but the gains in performance make it worthwhile. I'll gladly cheat 4K for extra frames and I wish the option was more popular on PC too. I know we have DLSS, but it's still not pResent in most games I play.

If I had to play this game I'd go for the PS5. VRR is only going to help so much and some of those drops on Xbox are nasty. If Xbox had CBR and VRR then it would be the way to play.
 
For IGN he just did an amazing Flight Sim through the years on actual hardware of the time. Man, that must take so much time and dedication.

IGN is lucky to have his talent on board.
I'll go watch that! I always liked flight simulators (in general).

I was wondering about the IGN thing because he insisted on the fact that he is independent and wanted support (which he 100% deserve).
 
Last edited:
My problem with his analysis pops up within the first 1 minute and 29 seconds of the video. What's this nonsense about the Series S not having the GPU horsepower to achieve Xbox One X levels of performance, when in fact it can easily achieve, even exceed, that if it wasn't a BC game locked to the Xbox One S code and was instead properly made for the Series S?

Series S is not only capable of what an Xbox One S can do. So if it's locked to Xbox One S code, no amount of GPU power is going to get you Xbox One X levels of performance or visual quality. You listen a little longer, and he literally calls the Series S a console only capable of last gen performance. This is why I don't take this guy's work seriously.
 
Last edited:
My problem with his analysis pops up within the first 1 minute and 29 seconds of the video. What's this nonsense about the Series S not having the GPU horsepower to achieve Xbox One X levels of performance, when in fact it can easily achieve, even exceed, that if it wasn't a BC game locked to the Xbox One S code and was instead properly made for the Series S?

Series S is not only capable of what an Xbox One S can do. So if it's locked to Xbox One S code, no amount of GPU power is going to get you Xbox One X levels of performance or visual quality. You listen a little longer, and he literally calls the Series S a console only capable of last gen performance. This is why I don't take this guy's work seriously.
One x is more capable in its gpu and ram capacity. One x for graphics, and series s is better in game logic (cpu) and loading.
 

GametimeUK

Member
Other than the hair in this case, it’s hardly call it obvious.
Are you implying I suggested the visible difference is obvious? I said obviously native 4K looks better than checkerboard rendering because this is always the case (no matter how big or small the difference is). I did not say the visible difference is obvious. Just to clarify.
 
Last edited:

Shmunter

Member
My problem with his analysis pops up within the first 1 minute and 29 seconds of the video. What's this nonsense about the Series S not having the GPU horsepower to achieve Xbox One X levels of performance, when in fact it can easily achieve, even exceed, that if it wasn't a BC game locked to the Xbox One S code and was instead properly made for the Series S?

Series S is not only capable of what an Xbox One S can do. So if it's locked to Xbox One S code, no amount of GPU power is going to get you Xbox One X levels of performance or visual quality. You listen a little longer, and he literally calls the Series S a console only capable of last gen performance. This is why I don't take this guy's work seriously.
Wouldn’t it beg to reason if series S could run One X Fidelity at double the frame rate, it would be enabled?

Im not quite sure the issue here?
 
One x is more capable in its gpu and ram capacity. One x for graphics, and series s is better in game logic (cpu) and loading.

One X GPU is by no means more capable than what's inside the Series S. Series S GPU is a much more capable/advanced piece of hardware. Games properly designed around it despite having less RAM, less memory bandwidth and less TFLOPS will easily outclass what can be done on Xbox One X on both a performance and visual level.
 
Going by the comments here (haven't watched the video yet), shouldn't the Xbox version be a "considerably better with an unnoticeable minor trade-off" option because of native 4K already on the One X, with a framerate boost to 60 but maintaining an average of just under 58.8 vs a really good implementation of CB4K on PS5 (running via BC mode of PS4Pro) and unlocked framerate but maintaining an unnoticeable higher framerate of 59.6?

CB4K vs Native 4K (native always superior, but I want to know if the native version in the Xbox version is vastly superior to this one on PS5 and whether PS5 has visible artefacting) and a 0.8 fps avg difference in perf, which is unnoticeable and on top of that Xbox has VRR (PS5 still doesn't have it) which makes it even better)?

That said if you have a PS4 physical/digital copy or a PS+ version that you recently redeemed, you shouldn't fret (don't know if this is on gamepads, but considering how old this is, it should be?). It's not worthy of going and buying a new nextgen console over (not like you can anyway thanks to scalpers lmao) just to play this in higher fidelity.

This comparison is only for people who already have both the consoles, haven't already played and don't have Shadow of the Tomb Raider at all, and most importantly have a 4K native TV or monitor, which has VRR, trying to choose which version to buy, and for this particular set of people who check all the boxes in this sentence, get the Xbox One version for cheap and run it on the XSX via BC.

For people who have either of the new consoles, get it on that or play the version you have already. It looks great on both.

All things done, Sony should let go of this "PS4PRo mode" in their SDK and update it with an efficient path of completely utilizing the PS5's power, for the developers (especially 3rd party) for their PS4 BC solution going forward, with minimal required resources from them, like what Microsoft does for their XB1 BC.

There's no reason why both the new consoles shouldn't play this game at 4K Ultra of the PC version, with a constant locked 60 fps with no dips (the Xbox Series fps dip must be a bug). It isn't that graphically intensive (without ray traced shadows and lighting), their engines scales really well with hardware and these new machines should breeze through with little to no issues.

Oh and ffs, add VRR to PS5 already Sony you lazy clowns.
 
Wouldn’t it beg to reason if series S could run One X Fidelity at double the frame rate, it would be enabled?

Im not quite sure the issue here?

It's a RAM limitation, nothing more, nothing less. Xbox One games have access to a larger pool of RAM than Series S does, so since Series S easily exceeds the RAM allocation of Xbox One S that's why the Series S gets the backwards compatible versions of Xbox One S games, and not the Xbox One X enhanced games. For BC to just work, they need the game to fit into the RAM pool, and Xbox One X enhanced titles simply don't fit into the available RAM for Series S unless the dev redesigns something.

But Series S could easily run the Xbox One X version and better if it were designed to.
 
One X GPU is by no means more capable than what's inside the Series S. Series S GPU is a much more capable/advanced piece of hardware. Games properly designed around it despite having less RAM, less memory bandwidth and less TFLOPS will easily outclass what can be done on Xbox One X on both a performance and visual level.
Not sure what to tell you if you think less is more. New architecture can help with new techniques but it's not going to help old games much. 8gb @ 224gb/s, 2gb @56gb/s vs 12gb @ 336gb/s. 6 vs. 4tf. Big difference in favor of one x. Go read up on the stuff. Series S i'm betting would be incapable of running forza horizon 4 with 4k one x settings.
 
Not sure what to tell you if you think less is more. New architecture can help with new techniques but it's not going to help old games much. 8gb @ 224gb/s, 2gb @56gb/s vs 12gb @ 336gb/s. 6 vs. 4tf. Big difference in favor of one x. Go read up on the stuff. Series S i'm betting would be incapable of running forza horizon 4 with 4k one x settings.

Series S can run Psychonauts 2 at 2880 x 1620 at 60fps. Or do 1080p at 120fps. In other words, there's plenty of performance headroom to outperform Xbox One X.
 

Shmunter

Member
One X GPU is by no means more capable than what's inside the Series S. Series S GPU is a much more capable/advanced piece of hardware. Games properly designed around it despite having less RAM, less memory bandwidth and less TFLOPS will easily outclass what can be done on Xbox One X on both a performance and visual level.
Despite having enhanced features, are we sure the computational TF comparison between the 2 is not close? How many % TF efficiency does RDNA2 add over the past gen? It would have to be 30% at the minimum to match the computational capability.
 
Last edited:

01011001

Banned
My problem with his analysis pops up within the first 1 minute and 29 seconds of the video. What's this nonsense about the Series S not having the GPU horsepower to achieve Xbox One X levels of performance, when in fact it can easily achieve, even exceed, that if it wasn't a BC game locked to the Xbox One S code and was instead properly made for the Series S?

Series S is not only capable of what an Xbox One S can do. So if it's locked to Xbox One S code, no amount of GPU power is going to get you Xbox One X levels of performance or visual quality. You listen a little longer, and he literally calls the Series S a console only capable of last gen performance. This is why I don't take this guy's work seriously.

well One X is 1080p60, Series S is 900p60, I bet they could have gone with 1080p60 but with roughly the same amount of drops the One X had, because the One X version, while relatively stable, ran way worse than the Series S version does

this means the Series S is almost at One X level here but is more stable than One X

 
Last edited:
Despite having enhanced features, are we sure the computational TF comparison between the 2 is not close? How many % TF efficiency does RDNA2 add over the past gen? It would have to be 30% at the minimum to match the computational capability.
Series S GPU is two entire generations newer than what’s in the One X, and it’s mated to a significantly more capable CPU while targeting 1080p and 1440p. It’s not really comparable.
 

Kagey K

Banned
PS5 CB is very good on this game, but XSX 2016p native resolution sounds a bit better.

But PS5 has smoother FPS on stress points. Average performance is 59.6 on PS5 and 58.8 on XSX.

For me it's a trade-off, both have bit advantagens in different points.
Not to be a meme here, but VRR offsets that advantage and makes the .8 frame performance invalid.

I wish Sony would implement it already so everyone could stop crying about frame drops, and more devs would be encouraged to use it to their advantage.

If anything this video shows how little advantage there was to the "next gen" patch, similar to Fallen Order. Xbox was already running these framerates and resolutions
 
Last edited:

Shmunter

Member
59.6 to 58.8 on average omfg such huge difference. Omg omg.


Native 4k with minor 2 fps drop ftw.
I think NX dropped the ball here doing an avg over a too long a session.

Traversing the forest without anything happening hits 60 easily skewing the result.

During battles the framerate is almost 15-20fps higher on PS5, these are the stress points that affect the gameplay being combat heavy.
 

Kagey K

Banned
I think NX dropped the ball here doing an avg over a too long a session.

Traversing the forest without anything happening hits 60 easily skewing the result.

During battles the framerate is almost 15-20fps higher on PS5, these are the stress points that affect the gameplay being combat heavy.
So pkaying the game as intended skews the result now?

OK I think I've heard just about everything now.
 
Despite having enhanced features, are we sure the computational TF comparison between the 2 is not close? How many % TF efficiency does RDNA2 add over the past gen? It would have to be 30% at the minimum to match the computational capability.

Series S's advantages go well beyond computational capability. It also has a features like Mesh Shaders and Sampler Feedback Streaming that will absolutely blow Xbox One X out of the water. Sampler Feedback Streaming eliminates entirely the one real advantage Xbox One X had, which was the RAM capacity advantage for games.
 

Shmunter

Member
Series S's advantages go well beyond computational capability. It also has a features like Mesh Shaders and Sampler Feedback Streaming that will absolutely blow Xbox One X out of the water. Sampler Feedback Streaming eliminates entirely the one real advantage Xbox One X had, which was the RAM capacity advantage for games.
Yes, I recognise that. But this or other games are not built around those performance saving features. A game that does all that is unlikely to see the light of day on last gen.

So it’s not just a ram limitation affecting the resolution, it’s the delta in raw computation.
 
Last edited:

FritzJ92

Member
My problem with his analysis pops up within the first 1 minute and 29 seconds of the video. What's this nonsense about the Series S not having the GPU horsepower to achieve Xbox One X levels of performance, when in fact it can easily achieve, even exceed, that if it wasn't a BC game locked to the Xbox One S code and was instead properly made for the Series S?

Series S is not only capable of what an Xbox One S can do. So if it's locked to Xbox One S code, no amount of GPU power is going to get you Xbox One X levels of performance or visual quality. You listen a little longer, and he literally calls the Series S a console only capable of last gen performance. This is why I don't take this guy's work seriously.
Thank you. His reviews always derail from pure technological objective information to some unnecessary fanboy or bias crap that throws it off or me. DF remains the supreme benchmark of benchmarking console performance. 🤷🏾‍♂️
 

Shmunter

Member
Yeah. During most of the combat the fps is only slightly below 60. During one alpha heavy segment it drops much lower, but the comparison is not like for like at all. So saying "in combat PS5 has 20 more fps" is nonsense.
NX gamer deliberately went and re did the section to be as like for like as reasonably possible just for that reason, also in the video.

Anyway if one doesn’t care then who am into tell you to care. Even if the thread is about that very thing.
 
Top Bottom