What do gamers actually think about Ray-Tracing?

[Zarax]

Precisely. While I appreciate improved looks and I can see the potential savings in artwork time, we have to remember it's a hobby we're talking about and it has to be balanced against what people can actually afford.
I've seen the same phenomena in smartphones, you used to get a decent lower mid range model for 200ish euros, now even 300€ won't buy a significant improvement from 2 years ago.

 

[RBImGuy]

Well that is exactly the point, but I was showing how the HUB survey results supports this concept.

In very simple terms, game developers see that 85% of potential PC game customers cannot run RT at significant levels. Add the significant majority of gamers on last gen let alone current gen consoles and the return of investment into high levels of RT becomes minuscule.

As a result developers currently see higher levels of RT as not worth the investment. So we end up with the current reality of where RT is.

can check path of exile 2 that use global illumination and dont need RT for looks
and runs flawlessly great also
(gameplay isnt good but game looks great)

 

[ICDP]

It’s the same with Stalker 2. It has its problems but it can be very impressive looking and no hardware RT.

RT when done well does look great, it’s just not that often we see this level implemented without Nvidia throwing money at the developers. No subsidies = no heavy RT in the majority of games.

 

[gpuerrilla]

I dont know why the usuals arent all over the 50 series launch lapping it up. It seems nvidia have doubled down on the RT cores, tensor etc. to push it harder (referencing robbie's shared video here).

Eager to see if the 5070 is on par with a 4090 as showcased!

 

[mrk]

Spoiler alert, it isn't As other outlets have shown already. The 5070 numbers are at 1440p, the 4090 is at 4K lol, as per the nvidia slide smallprint

 

[gpuerrilla]

I know that , I wasn't exactly impressed with the jump from 30 > 40 series but that was more down to the price increases. You need to give old Jensen a panning for the deception!

I'm over halfway of that tuber you linked, the TOPS is the only thing that seems to have accelerated this gen and whatever GDDR7 brings.

 

[D3K]

I have discussed them at face value, you even quoted my post where I outlined a brief explanation of how the stats would be interpreted by game developers. The fact you can’t see that is not my problem. I tried telling you why your analysis of the survey was simplistic. I then went in to some details as to why you were wrong and even used a deliberately ridiculous analogy (all people who drink are alcoholics) to demonstrate this fallacy.

Are you having a seizure? You have thrown in some woolly logic and failed to disprove that 48% of people in HUBs survey use RT and therefore see it as worthwhile on some level. Do you think they are all hate-playing and self-flagellating by turning RT on despite thinking it's not worth it?

 

[Bill Turnip]

You have thrown in some woolly logic and failed to disprove that 48% of people in HUBs survey use RT and therefore see it as worthwhile on some level. Do you think they are all hate-playing and self-flagellating by turning RT on despite thinking it's not worth it?

26% of those 48% don't turn it on in some games, and therefore for those games would fall into the "no" categories as presumably the reason is that performance hit/visual uplift is not sufficient.

So over half of the people who can be said to consider it worthwhile sometimes don't bother.

 

[D3K]

26% of those 48% don't turn it on in some games, and therefore for those games would fall into the "no" categories as presumably the reason is that performance hit/visual uplift is not sufficient.

So over half of the people who can be said to consider it worthwhile sometimes don't bother.

This is a better argument, and I agree that the numbers can accommodate that as a probability.
Equally, as I said previously, the 36% of people who say the perf hit is too significant can be broken into 2 camps - those who simply cant, and those who refuse to because of the perf hit. The former may think RT is worth it on the whole but cannot afford it, yet.

So losing 13% (26/2) from your claim, but gaining 18% (36/2) from my point, then we end up with more than half of people who think it's worth it

 

[ICDP]

You are literally making up stats within stats. Stop thinking like a consumer who happens to like RT. It is clouding your narrative.

 

[D3K]

You are literally making up stats within stats. Stop thinking like a consumer who happens to like RT. It is clouding your narrative.

No, we're using the concept of probability - you know all about stats because you work in IT so I'm sure you get it - to posit permutations on the stats as none of us can prove one way or the other how those stats are sub-divided. Not without interviewing every person who voted. We seem to get that concept, but you seem to default to a poorly defined deterministic viewpoint and refuse to quantify or show your working.

 

[Calin Banc]

Gamers are happy to use ray tracing to various degrees, all of them. *As long as the visual improvement is significant and the GPU can run it. ) That poll is worded so wrong!

 

[Joxeon]

RT is fine, I like RT and it makes a game look better but the problem is that to run it well it especially heavy RT/PT is that it requires frame gen which makes the game look much worse so it’s currently a double edge sword.

 

[Bill Turnip]

It's all subjective. Do I like RT? I mean, sure, it can work. Do I want it in preference to higher frame rates? Absolutely not.

Frame Generation seems to be an answer of sorts, except it only really works when your frame rate is already at a decent level or latency is horrible. That doesn't sound like a solution.

DLSS and FSR are excellent tech, and to my mind the most welcome of the three. Are they perfect? No, I wouldn't claim they are, but for me their drawbacks are more than worth the result.

YMMV etc. etc.

 

[ICDP]

No, we're using the concept of probability - you know all about stats because you work in IT so I'm sure you get it - to posit permutations on the stats as none of us can prove one way or the other how those stats are sub-divided. Not without interviewing every person who voted. We seem to get that concept, but you seem to default to a poorly defined deterministic viewpoint and refuse to quantify or show your working.

I’m not going on about HUB but the overall RT reality as we are right now. It is not viable in mainstream hardware, as such it is still a few years from its potential being fully realised in games. Developers currently still don’t prioritise heavy RT due to these basic realities.

This isn’t an attack on RT, it’s saying the best is potentially yet to come.

 

[Pottsey]

I’m not going on about HUB but the overall RT reality as we are right now. It is not viable in mainstream hardware, as such it is still a few years from its potential being fully realised in games. Developers currently still don’t prioritise heavy RT due to these basic realities.

This isn’t an attack on RT, it’s saying the best is potentially yet to come.

Between the Steam Survey and current consoles hardware I would say RT is now viable in mainstream hardware at mainstream resolutions (1080P). Now that RT hardware is mainstream its only going to cause an increase in RT titles. Plus with consoles games now being made with RT those are just going to be transferred over to the PC with RT. Many of tomorrows games being developed now very much will be prioritised for RT.

 

[Tinek]

It's not an opinion, the numbers are plain to see. 48% use it, 52% don't. What you can disagree on is the rationale for the 36% of the stats not using it, but I already gave the whole number to you...

Yes, the numbers are plain too see: 26% Were in between with enabling it only in some games. That's not excluding the option that they simply don't find it worth it in the games they don't enable it in. Because they are on the fence and we don't have enough data to judge which way they lean more, we can exclude that from both sides. We are left with 52% don't find it worthy, 22% find it worthy and at best 26% undecided. Out of all the numbers, only 15% were able to actually run it on full settings. That's the proper interpretation of these numbers imho.

 

[CAT-THE-FIFTH]

Yes, the numbers are plain too see: 26% Were in between with enabling it only in some games. That's not excluding the option that they simply don't find it worth it in the games they don't enable it in. Because they are on the fence and we don't have enough data to judge which way they lean more, we can exclude that from both sides. We are left with 52% don't find it worthy, 22% find it worthy and at best 26% undecided. Out of all the numbers, only 15% were able to actually run it on full settings. That's the proper interpretation of these numbers imho.

Also this is a survey of viewers of an enthusiast's PC hardware channel. People who care enough about watching hardware reviews are more likely to have better hardware and are more invested in caring about visuals.

The issue is that enthusiasts on forums are only a miniscule amount of most gamers. This very forum,is one of the biggest tech forums in the world but is lucky to have 5000 concurrent users online and many of them are not talking about games or graphics. Even if you add all the tech forums,Reddit,etc you are talking maybe 10s of 1000s of gamers. It's an echo chamber of people interested in hardware,talking about hardware and wanting to justify an upgrade. Yet we have over 100 million PC gamers on Steam alone.

Look at games such as LoL,GTA:Online,etc. Games which make billions and hardly seem to care about graphics. Yet,how many in the graphics card sub-forum are waxing lyrical about playing such games? Probably none.

Even Fortnite,despite having an RT option is made to run on a potato. Lots of these companies don't want to exclude people on lesser hardware.

 

[Tinek]

There is no need of statistics.
How many sub $300 GPUs can run RT without the monitor becoming a blurry or pixelated mess due to excessive upscaling needed?

Answer that and you will get the realistic state or RT adoption.

It's even simpler and I asked the question before - Howe many AAA games that limit themselves to RT and expensive hardware to run well were financially successful? The answer is almost none. Huge majority are financial flops with steam played statistics confirming it. It's not that it was just RT fault - a lot of them are just really bad games. But publishers got a wake-up call, one could hope.

 

[D3K]

Yes, the numbers are plain too see: 26% Were in between with enabling it only in some games. That's not excluding the option that they simply don't find it worth it in the games they don't enable it in. Because they are on the fence and we don't have enough data to judge which way they lean more, we can exclude that from both sides. We are left with 52% don't find it worthy, 22% find it worthy and at best 26% undecided. Out of all the numbers, only 15% were able to actually run it on full settings. That's the proper interpretation of these numbers imho.

So let's take someone who doesn't bother turning it on for SOTTR because whats the point really, but turns it on for Metro Exodus since it is very much worth it. Where does that person lie in your "undecided" blanket rule? They don't. They are decided, they just turn it on where it truly is worth it. To speak with the conviction like you do comes across as naive.

It's not that it was just RT fault

Not even... find a source for a dev blaming RT please.