Jump to content

DLSS is now free


Benchboy

Do you want DLSS?  

329 members have voted

  1. 1. Do you want DLSS?

    • Yes
      293
    • No
      36


Recommended Posts

  • ED Team

I have already mentioned it to the team, but I have no news to share with you. 

 

thanks

  • Like 7
  • Thanks 1

smallCATPILOT.PNG.04bbece1b27ff1b2c193b174ec410fc0.PNG

Forum rules - DCS Crashing? Try this first - Cleanup and Repair - Discord BIGNEWY#8703 - Youtube - Patch Status

Windows 11, NVIDIA MSI RTX 3090, Intel® i9-10900K 3.70GHz, 5.30GHz Turbo, Corsair Hydro Series H150i Pro, 64GB DDR @3200, ASUS ROG Strix Z490-F Gaming, HP Reverb G2

Link to comment
Share on other sites

@BIGNEWY it would be great to know what ED think about DLSS/FSR integration on DCS engine, as long as some news about how is going multithreading and Vulkan work

  • Like 3

NZXT H9 Flow Black | Intel Core i5 13600KF OCed P5.6 E4.4 | Gigabyte Z790 Aorus Elite AX | G.Skill Trident Z5 Neo DDR5-6000 32GB C30 OCed 6600 C32 | nVidia GeForce RTX 4090 Founders Edition |  Western Digital SN770 2TB | Gigabyte GP-UD1000GM PG5 ATX 3.0 1000W | SteelSeries Apex 7 | Razer Viper Mini | SteelSeries Artics Nova 7 | LG OLED42C2 | Xiaomi P1 55"

Virpil T-50 CM2 Base + Thrustmaster Warthog Stick | WinWing Orion 2 F16EX Viper Throttle  | WinWing ICP | 3 x Thrustmaster MFD | Saitek Combat Rudder Pedals | Oculus Quest 2

DCS World | Persian Gulf | Syria | Flaming Cliff 3 | P-51D Mustang | Spitfire LF Mk. IX | Fw-109 A-8 | A-10C II Tank Killer | F/A-18C Hornet | F-14B Tomcat | F-16C Viper | F-15E Strike Eagle | M2000C | Ka-50 BlackShark III | Mi-24P Hind | AH-64D Apache | SuperCarrier

Link to comment
Share on other sites

i think the real potential of dlss lies in DLSS for VR.

i assume that both dlss and that AMD equivalent will run into problems with distance spotting / rendering of very small details (aircrafts that are only one or few pixels in size).

DLSS VR will maybe cope better with that, as it has double the amount of data to work with, because of the stereoscopic rendering.

@BIGNEWY have you already looked at dlss VR? it will maybe change the team's view on the technology, since - on paper - it should have much greater benefits than the pancake variant.

  • Like 1
Link to comment
Share on other sites

If not DLSS, I wish ED would integrate at least something (I guess that would mean FSR then). 

 

P.S. I know there is an FSR VR mod. 

  • Like 2

НЕТ ВОЙНЕ!

Gib full-fi Su-27 or MiG-29 plz!

AMD R7 3700X|32GB DDR4 RAM|Gigabyte RTX2070S Gaming OC|2TB NVMe SDD + 1TB SSD + 2TBB + 1TB HDD|Dell P3421W|Windows 10 Pro x64

TM Warthog|MFG Crosswind|Samsung Odyssey+|TrackIR 5

Modules: Mirage F1|Mi-24P|JF-17|F/A-18C|F-14A/B|F-5E|M-2000C|MiG-21bis|L-39|Yak-52|FC3|Supercarrier || Terrains: Persian Gulf|NTTR|Normandy|Syria

Link to comment
Share on other sites

3 minutes ago, killkenny1 said:

If not DLSS, I wish ED would integrate at least something (I guess that would mean FSR then). 

 

P.S. I know there is an FSR VR mod. 

or postFX AA. SMAA, TAA. These would help so much already.

  • Like 1
Link to comment
Share on other sites

3 hours ago, Benchboy said:

which may increase performance

… at the expense of quality. 
 

Fixed that for you. DLSS is not a magic performance boost. Like any other graphic setting it’s a compromise between quality and performance. 

  • Like 4
  • Thanks 1

i9-13900K @ 6.2GHz oc | ASUS ROG MAXIMUS Z790 HERO | 64GB DDR5 5600MHz | iCUE H150i Liquid CPU Cooler | 24GB GeForce RTX 4090 | Windows 11 Home | 2TB Samsung 980 PRO NVMe | Corsair RM1000x | LG 48GQ900-B 4K OLED Monitor | CH Fighterstick | Ch Pro Throttle | CH Pro Pedals | TrackIR 5

Link to comment
Share on other sites

9 minutes ago, SharpeXB said:

… at the expense of quality.

…which is a meaningless statement where no-one has actually stated otherwise, and which requires context to have any value or worth whatsoever.

 

The question you should be asking, but aren't for some unknowable reason, is how much less quality you get. Is the reduction less than you'd get for any other option that would increase performance by the same amount? Because if so, then yes, DLSS is indeed a magic performance boost: it does something no other graphics option can do.

 

Until you have that data, saying that it comes at the cost of quality is just a vapid and worthless platitude that challenges nothing.

  • Like 5
  • Thanks 3

❧ ❧ Inside you are two wolves. One cannot land; the other shoots friendlies. You are a Goon. ❧ ❧

Link to comment
Share on other sites

Irrespective of your positions or arguments  , your obvious jihad against SharpXB is seriously damaging your credibility .

38 minutes ago, Tippis said:

…which is a meaningless statement where no-one has actually stated otherwise, and which requires context to have any value or worth whatsoever.

 

The question you should be asking, but aren't for some unknowable reason, is how much less quality you get. Is the reduction less than you'd get for any other option that would increase performance by the same amount? Because if so, then yes, DLSS is indeed a magic performance boost: it does something no other graphics option can do.

 

Until you have that data, saying that it comes at the cost of quality is just a vapid and worthless platitude that challenges nothing.

 


Edited by Svsmokey
  • Like 3
  • Thanks 1

9700k @ stock , Aorus Pro Z390 wifi , 32gb 3200 mhz CL16 , 1tb EVO 970 , MSI RX 6800XT Gaming X TRIO , Seasonic Prime 850w Gold , Coolermaster H500m , Noctua NH-D15S , CH Pro throttle and T50CM2/WarBrD base on Foxxmounts , CH pedals , Reverb G2v2

Link to comment
Share on other sites

25 minutes ago, Svsmokey said:

Irrespective of your positions or arguments  , your obvious jihad against SharpXB is seriously damaging your credibility .

😂 Not really no.

Mainly because that would be an ad hominem — making an assumptions about the statement based on the person rather than the merits of the statement itself — but also because it doesn't really damage anyone's credibility to point out that a contextless vapid statement is vapid unless put into context. Well, except maybe the person who tried to offer said vapid statement as some kind of counter-argument against a position no-one has particularly expressed.


Edited by Tippis
  • Like 2

❧ ❧ Inside you are two wolves. One cannot land; the other shoots friendlies. You are a Goon. ❧ ❧

Link to comment
Share on other sites

What I still don't get in VR: I want my scene rendered in 3000x3000 pixels to look better and more detailed on a e.g. Reverb G2. How can DLSS help me when it renders all with less resolution and upscales it afterwards to the resolution of that reverb? Am I outside of the target group?


Edited by Tom Kazansky
  • Like 1
Link to comment
Share on other sites

5 minutes ago, Tom Kazansky said:

What I still don't get in VR: I want my scene rendered in 3000x3000 pixels to look better and more detailed on a e.g. Reverb G2. How can DLSS help me when it renders all with less resolution and upscales it afterwards to the resolution of that reverb? Am I outside of the target group?

It won't be more detailed, pretty much by definition.

It may look better, but that depends on what aspects you put into your judgement of “look good” and what you're comparing against.

 

For the scenario you describe, what it could bring you is better and faster upscaling than if you relied on the headset's (or its drivers') own upscaling algorithms. However, this gets into very muddy waters of what the benefits are to making the headset do as little work as possible. Do you get something for just feeding it its final resolution directly, compared to letting it do its thing? You're meant to get better quality and higher performance assuming you're upscaling from and to the same baseline and target resolutions. And also possibly more tweakability and continuous improvements as the graphic card drivers improve, which usually happens at a faster pace than the headset drivers. You can then make a choice to perhaps sacrifice some of that increased performance for having a more detailed base image that gets upscaled.

 

If you're not upscaling to begin with, you get a very different proposition. The question instead becomes one of, maybe you should start? What reductions in quality would you be willing to accept in exchange for increased performance. With good resolution compensation (which is basically what this is), at what point do you start noticing a difference, and at what point is that difference large enough that it no longer looks acceptable compared to the raw-render alternative? And how much, if any, performance do you gain before you hit that limit? The presumption (or sales pitch, rather) of these techniques is that the ∆performance grows faster than the ∆quality, so you end up at a new equilibrium between the two where you get more speed for a largely imperceptible reduction in image quality. And then the pixel peeping begins… 😄 

  • Like 1

❧ ❧ Inside you are two wolves. One cannot land; the other shoots friendlies. You are a Goon. ❧ ❧

Link to comment
Share on other sites

18 minutes ago, Tippis said:

😂 Not really no.

Mainly because that would be an ad hominem — making an assumptions about the statement based on the person rather than the merits of the statement itself — but also because it doesn't really damage anyone's credibility to point out that a contextless vapid statement is vapid unless put into context. Well, except maybe the person who tried to offer said vapid statement as some kind of counter-argument against a position no-one has particularly expressed.

 

I respectfully suggest that you revisit your own arguments , applying the same criteria , as well as the trendline of to whom you are responding .

 

  • Like 3
  • Thanks 1

9700k @ stock , Aorus Pro Z390 wifi , 32gb 3200 mhz CL16 , 1tb EVO 970 , MSI RX 6800XT Gaming X TRIO , Seasonic Prime 850w Gold , Coolermaster H500m , Noctua NH-D15S , CH Pro throttle and T50CM2/WarBrD base on Foxxmounts , CH pedals , Reverb G2v2

Link to comment
Share on other sites

Why bother implementing DLSS, which only works on Nvidia hardware, when AMD's offering (FSR) works on both AMD and Nvidia hardware?

 

 

Either way, for those that are interested and need a tl;dr version of the tech: DLSS+FSR is just a clever trick. Some clever engineer worked out it's quicker and less resource intensive to render at lower resolution and upscale, than it is to just render at the higher resolution. That's it. Nvidia do some very clever AI based upscaling, but at a high level, that's all it does.

  • Like 1
Link to comment
Share on other sites

7 minutes ago, Tippis said:

It won't be more detailed, pretty much by definition.

It may look better, but that depends on what aspects you put into your judgement of “look good” and what you're comparing against.

 

For the scenario you describe, what it could bring you is better and faster upscaling than if you relied on the headset's (or its drivers') own upscaling algorithms. However, this gets into very muddy waters of what the benefits are to making the headset do as little work as possible. Do you get something for just feeding it its final resolution directly, compared to letting it do its thing? You're meant to get better quality and higher performance assuming you're upscaling from and to the same baseline and target resolutions. And also possibly more tweakability and continuous improvements as the graphic card drivers improve, which usually happens at a faster pace than the headset drivers. You can then make a choice to perhaps sacrifice some of that increased performance for having a more detailed base image that gets upscaled.

 

If you're not upscaling to begin with, you get a very different proposition. The question instead becomes one of, maybe you should start? What reductions in quality would you be willing to accept in exchange for increased performance. With good resolution compensation (which is basically what this is), at what point do you start noticing a difference, and at what point is that difference large enough that it no longer looks acceptable compared to the raw-render alternative? And how much, if any, performance do you gain before you hit that limit? The presumption (or sales pitch, rather) of these techniques is that the ∆performance grows faster than the ∆quality, so you end up at a new equilibrium between the two where you get more speed for a largely imperceptible reduction in image quality. And then the pixel peeping begins… 😄 

Still don't get it. My G2 is not upscaling it is downscaling from a 3000×3000 source. For me better = more details in the cockpit. DLSS does the opposite. I saw screenshots of other games where DLSS reduces the details by renderung a lower resolution (and scales it up) to gain frame rate.


Edited by Tom Kazansky
Link to comment
Share on other sites

3 minutes ago, Svsmokey said:

I respectfully suggest that you revisit your own arguments , applying the same criteria , as well as the trendline of to whom you are responding .

Ok.

 

No ad hominems. Context provided. Direct response to something someone said rather than some random truism that no-one was denying to begin with. As for trendlines, the fact that he does that kind of thing a lot only reflects on him, not me. Looks like I'm in good shape. 🙂 

 

2 minutes ago, Buzzles said:

Why bother implementing DLSS, which only works on Nvidia hardware, when AMD's offering (FSR) works on both AMD and Nvidia hardware?

The standard reason is that cross-vendor support is usually quite iffy at the start, and before the… well… standard is set, both sides have an irritating fondness for not doing “the other guy's” solution well. It's the problem with all new tech, really. If you want to jump on the bandwagon early on, buying into vendor-specific solutions is an unfortunate part of the package. In time, if the tech offers a solid benefit, it'll converge towards a single solution that will work for everyone but that's a card generation or three down the line… 😕

 

5 minutes ago, Tom Kazansky said:

Still don't get it. My G2 is not upscaling it is downscaling from a 3000×3000 source. For me better = more details in the cockpit. DLSS does the opposite. I saw screenshots of other games where DLSS reduces the details by renderung a lower resolution (and scales it up) to gain frame rate.

Yes, that's sort of the whole issue: if all you end up doing is super-sampling an upscaled image where the source is already at the target resolution, why bother? Why not just feed that image directly to the headset and skip all the seemingly self-defeating up-and-down-scaling? It really comes down to, how much detail do you gain in the supersampling? How much detail do you lose (or, more accurately, don't you get to begin with) in the upscaling? If it's just 1:1, all you end up with is a hideously overcomplicated anti-aliasing, which may if anything just wash out details that are actually important to preserve.

 

Where you could conceivably gain something is in the margin between the two. Maybe you get sufficient detail by rendering at, say, 2560 rather than the full 3000, and letting the fancy algorithms bounce that image back and forth until it arrives at the final 2160… but that all depends on what the headset requires as an output resolution from the card. If you can set the final render resolution to anything, it's still pretty pointless to add the upscaling to that — just feed it the lower-than-maximum-but-still-higher-than-headset image and let it do the downscaling like it would anyway.

 

So there could still be some room for the same consideration of lowering the actual rendering resolution to the point where you just start losing details, but it's a thin margin and it only works on the assumption that the display hardware has a fixed rendering resolution that it absolutely must have. And while I'm not familiar with the G2, that seems quite unlikely.

❧ ❧ Inside you are two wolves. One cannot land; the other shoots friendlies. You are a Goon. ❧ ❧

Link to comment
Share on other sites

@Tom KazanskyDLSS its not a simply upscaler, its a AI assisted one, wich make a difference.

I dont know what titles you saw where theres a (noticiable) loss of detail, but since DLSS 2.0 thats not a problem, theres games like Marvel Avengers or Death Stranding that they not only dont loss detail, they add detail.

And about why not FSR instead of DLSS, well, FSR works fine but DLSS is better and AI assisted upscaling is the way to go for future, even Microsoft has stated that the upscaler that they will release for DirectX will be AI assisted, but still, better FSR than nothing


Edited by 5ephir0th

NZXT H9 Flow Black | Intel Core i5 13600KF OCed P5.6 E4.4 | Gigabyte Z790 Aorus Elite AX | G.Skill Trident Z5 Neo DDR5-6000 32GB C30 OCed 6600 C32 | nVidia GeForce RTX 4090 Founders Edition |  Western Digital SN770 2TB | Gigabyte GP-UD1000GM PG5 ATX 3.0 1000W | SteelSeries Apex 7 | Razer Viper Mini | SteelSeries Artics Nova 7 | LG OLED42C2 | Xiaomi P1 55"

Virpil T-50 CM2 Base + Thrustmaster Warthog Stick | WinWing Orion 2 F16EX Viper Throttle  | WinWing ICP | 3 x Thrustmaster MFD | Saitek Combat Rudder Pedals | Oculus Quest 2

DCS World | Persian Gulf | Syria | Flaming Cliff 3 | P-51D Mustang | Spitfire LF Mk. IX | Fw-109 A-8 | A-10C II Tank Killer | F/A-18C Hornet | F-14B Tomcat | F-16C Viper | F-15E Strike Eagle | M2000C | Ka-50 BlackShark III | Mi-24P Hind | AH-64D Apache | SuperCarrier

Link to comment
Share on other sites

it is indeed the case that DLSS can add detail, mainly because it's temporal and uses information from already rendered frames.

in VR mode it can also use information from both eyes to reconstruct an image.

i think the following video gives a good udnerstanding of how DLSS works by letting it upscale from ridiculously low resolution.

 

the main question will be, how well it can handle the specifics of DCS. i'm very optimistic that it will shine in VR because of the extra raw data the stereoscopic rendering produces.

@Tom Kazansky you could theoretically simply view it as an anti aliasing techique, because you could use it to output even higher resolution to downsample again, thereby reducing aliasing. have a look at the video, i think your questions are covered in it quite well. in the end it comes down to marketing, but that does not change the fact that it is a very interesting technology.


Edited by twistking
  • Like 1
  • Thanks 1
Link to comment
Share on other sites

11 minutes ago, twistking said:

it is indeed the case that DLSS can add detail, mainly because it's temporal and uses information from already rendered frames.

in VR mode it can also use information from both eyes to reconstruct an image.

i think the following video gives a good udnerstanding of how DLSS works by letting it upscale from ridiculously low resolution.

 

the main question will be, how well it can handle the specifics of DCS. i'm very optimistic that it will shine in VR because of the extra raw data the stereoscopic rendering produces.

@Tom Kazansky you could theoretically simply view it as an anti aliasing techique, because you could use it to output even higher resolution to downsample again, thereby reducing aliasing. have a look at the video, i think your questions are covered in it quite well. in the end it comes down to marketing, but that does not change the fact that it is a very interesting technology.

 

Thanks for that video. Now I start to get it. Very interesting technology that might help even high resolutions look better and faster than those with non-AI-anti-aliasing-technologies. Here is my YES.

  • Like 1
Link to comment
Share on other sites

3 hours ago, Tippis said:

[snip]

 

The standard reason is that cross-vendor support is usually quite iffy at the start, and before the… well… standard is set, both sides have an irritating fondness for not doing “the other guy's” solution well. It's the problem with all new tech, really. If you want to jump on the bandwagon early on, buying into vendor-specific solutions is an unfortunate part of the package. In time, if the tech offers a solid benefit, it'll converge towards a single solution that will work for everyone but that's a card generation or three down the line… 😕

 

[snip]

I get where you're coming from, but DLSS requires Nvidia's Tensor technology, it's hardware based. Ergo, it's a non starter unless they offer that out to AMD/Amd vendors.
FSR already works across cards and older stuff too.

Harks back the days of Physx hardware acceleration, where we had the hardware version on a specific card (Nvidia) vs the non accelerated software version running on cpu's. Guess which one won out in the end...

  • Thanks 1
Link to comment
Share on other sites

5 hours ago, Ikaros said:

I don't understand why anyone would say no. I just used it for the first time in another game and the performance improvement is jaw dropping. 

 

I believe most who say 'no' are saying 'please not vendor-specific'. I'd love to have DLSS-type tech for me (as I, too, am flying VR, and the temporal AA 'no shimmer' feature alone is to kill for). But as someone who got burned one time too many with vendor-bound features I now advocate waiting until an independent standard emerges. If you don't agree - well, I still have some 3dfx Voodoo-bound titles lying around in my discard pile for you (those that did not provide an OpenGL version as fallback). 

 

  • Like 1
Link to comment
Share on other sites

i think one important thing to understand is, that DLSS works fundamentally different than AMD FSR.

AMD FSR is indeed nothing more than a fancy upscaler. It does some combination of vector scaling, smoothing, sharpening etc. It has no "concept" of what is in the scene, it just applies a combination of effects to create clarity in an upscaled image, that would otherwise look more blurry and/or aliased.

When it was first shown off by AMD i did expect nothing to be honest, because this type of non-aware upscaling has been around for ages and never looked good. However, most reviewers found it to work surprisingly good, definitely better than maybe expected and at least comparable to Nvidia DLSS, which is quite an archivment considering that DLSS is technically way more sophisticated.

 

so from that you could definitely argue that AMD FSR would be the sensible choice for DCS, because it is hardware agnostic and at least comparable to DLSS in quality, HOWEVER i fear that FSR would not work particularly well in DCS, while i believe that DLSS would.

The reason is, that in DCS it's not just about getting rid of jagged edges and creating a smooth picture with great clarity. In DCS you need detail (detail is not the same as clarity).

Since FSR has no concept of the scene and no temporal element, it is less capable of maintaining precise detail like small text in cockpit displays or aircrafts in the distance. There is a high chance that those details would get lost. The scene would still look nice, with smooth edges, but you'd definitely loose pixel detail, which would be more of a critical issue in DCS than it would be in other titles.

 

DLSS on the other hand is equipped to maintain Pixel level details and even create new pixel level detail (hyper-resolution) through the temporal element and it's ability to get clues from both cameras of a stereoscopic renderer. I don't know how well that would work within DCS, but i'm very optimistic after all what i've learned about it.

 

I want to say again, that AMD FSR might generally be better - if only for the fact that it's vendor agnostic - i just think that because of how it operates, it won't work too well with DCS and it's unique properties.

I hope the ED devs have a sincere look at both technologies and make a more educated decision about it than we can.


Edited by twistking
  • Like 8
Link to comment
Share on other sites

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...