41 minutes ago, TravelRunner404 said:

The card is great. I use VR in my work so I am fortunate to grab these things and get to use them on the side. In the Varjo Aero with 39 PPD and DLSS I am getting 50-60 FPS with settings that I got 31 FPS with the 3090. The Varjo Aero experience really shines with this thing.

Hi. To clarify, you were getting 31 FPS with the 3090 using DLSS 2, right?  And with the 4090, you are getting 50 to 60 FPS using DLSS 3, with all the same settings, add-ons, configuration, hardware, etc?

If I am understanding what you wrote here correctly, you are getting almost double the FPS (50 to 60 FPS) with DLSS 3 using the 4090 vs DLSS 2 with the 3090 (31 FPS). Please correct me if I misunderstood what you wrote.

Edited October 14, 20223 yr by abrams_tank

I use EBBR and the Fenix for testing at ultra and on a 10900 @4.9 and a 3080 I get around 35 FPS, so seeing silky smooth frames at 70-75 is huge.

"Nothing that a good 2.XXX or 3.XXX series can do" Really !

Edited October 14, 20223 yr by Nyxx

I’ve got a 3080ti on 12th gen

if I can get 35fps smooth in New York in the Fenix I’m a happy camper. 

Does anybody else get the following blurries at a certain distance on panels with DLSS2?:

https://photos.app.goo.gl/5g4a5Wr5FRz7rwGf7

22 minutes ago, vincentrouleau said:

Does anybody else get the following blurries at a certain distance on panels with DLSS2?:

That's normal. Whenever you use any form of Temporal Antialiasing ( TAA and DLSS are all temporal-based ), anything that moves fast will have these issues on fast moving displays with text.

I'll be interested in feedback on this frame generation with people using head tracking and similar. 

Quote

9 minutes ago, virtuali said:

That's normal. Whenever you use any form of Temporal Antialiasing ( TAA and DLSS are all temporal-based ), anything that moves fast will have these issues on fast moving displays with text.

 

Thing is, I don't get this in TAA. only in DLSS2. Just wondering, As I would not use the DLSS3 feature with such ghosting. I can manager now not to saturate my 3080 12Gig in TAA.

23 minutes ago, Donka said:

I'll be interested in feedback on this frame generation with people using head tracking and similar. 

Likewise. 

4 hours ago, Claudius_ said:

Wrong: my comments are a lot salty to who choose to claim miraculous properties from a new (and expensive), graphic card.

You sound like a real sad and angry person.  What is it about other peoples purchasing decisions that you cannot tolerate?

4 hours ago, abrams_tank said:

If I am understanding what you wrote here correctly, you are getting almost double the FPS (50 to 60 FPS) with DLSS 3 using the 4090 vs DLSS 2 with the 3090 (31 FPS). Please correct me if I misunderstood what you wrote.

In VR at the max resolution of the Varjo Aero yes. In fact I increased shadows and clouds to high and still got over double. So increased settings.  But the Varjo Aero is almost a sure bet you were definitely GPU limited.

1 hour ago, DesLynam said:

You sound like a real sad and angry person.  What is it about other peoples purchasing decisions that you cannot tolerate?

To be fair, @Claudius_ wrote a completely fair and reasonable reply to my post. 

Can someone who went from a 3080/3090 to a 4090 report on the FPS increase using just TAA?

6 hours ago, hanhamreds said:

They did indeed, which is why I posted the link for this yesterday on the main DLSS 3 thread, you will need 100+ fps to start with basically, otherwise you will notice these artifacts

Which makes me wonder what the point of it is. So far I've only seen examples where the base FPS was very good already, sometimes better than what you even need. Achieving double that in that scenario seems like wasting your money on an overpriced GPU. The question is does it help you so much and produce double the FPS (say 25 base vs. 50 with DLSS3) in a demanding scenario like JFK stuffed with AI traffic, thick clouds lingering over the airport while on approach in the Fenix A320. I haven't seen any video like this yet. If it indeed causes artifacts at low FPS that's just making it a lot worse.

As an example, I'm getting around 80 FPS in Call of Duty on maxed out settings. If I were to use a 4090 with DLSS3 I would end up at around 160 FPS, maybe a little less. But what is this supposed to give me?

I haven't drowned myself in research on DLSS3 (and therefore the 4090) but what I have seen seems to suggest this is a lot of money for nothing (that is, if you're in it just to be able to use DLSS3 to get 140 FPS - the 4090 itself probably is a nice improvement in performance, especially in VR).

Edited October 14, 20223 yr by threegreen

1 hour ago, TravelRunner404 said:

In VR at the max resolution of the Varjo Aero yes. In fact I increased shadows and clouds to high and still got over double. So increased settings.  But the Varjo Aero is almost a sure bet you were definitely GPU limited.

Wow, that's a huge FPS win for people that don't mind the blurry glass panels, if you are getting double FPS with the 4090 using DLSS 3, versus the 3090 using DLSS 2.

9 hours ago, steve310002 said:

I don't believe for a second that 2000 or 3000 series could get quite as smooth frame rates as seen in the Fenix video above which was hitting 70+fps with DLSS 3. That would mean with the same scenario and DLSS 3 turned off, it would only be achieving 35 to 40fps!! A 2000 or 3000 series card would likely be struggling to hit 30 fps in the same situation (especially a 2000 series card) and wouldn't be as smooth.

Frankly, I can't wait to get myself a 4090 once the Intel 13th gen CPUs are released. I agree they are over priced and it is greedy of Nvidia, but anyone saying that a 2000 or 3000 series can achieve the same thing is deluding themselves. I think a lot of people are angry at the prices and trying to convince themselves that the 4090 and DLSS 3 isn't as good as it actually is.

I think what he's trying to say is that for the money, what is 50 fps vs. 90 fps in the sim?    Are you really going to notice it in the sim?

I can definitely tell a difference between 30 vs. 40 vs. 50 fps in MSFS.  But above 50, I really have a hard time telling any difference.