-
Ever wanted an RSS feed of all your favorite gaming news sites? Go check out our new Gaming Headlines feed! Read more about it here.
Digital Foundry - Red Dead Redemption II PC: Every Setting Explained (Dictator's Magnum Opus)
- Thread starter RadioactiveLobster
- Start date
I am baffled, raw specs are meaningless when comparing 2 different architectures. The 2080 is faster despite having only 2560 cuda cores against the 1080 Ti's 3584. Hence, its more efficient. It also has lower TDP.
No one buys monster GPU's specifically to play older titles in which you get insane amount of performance anyway regarldess of what you choose. In relevant, modern titles, the 2080 is generally faster, mainly because Turing plays better with low level API's like DX12 and vulkan. This has been historically a weak spot for Pascal.
VRAM has not been an issue for the 2080 so far, even in 4k. So not sure what to say there. Maybe in the future it will matter, but there's a much higher chance that, by the time it does matter, neither card will be able to run acceptable framerates at resolutions and graphical settings where the VRAM difference would matter.
I've literally linked 2 articles, one showing a 11 game average where the 2080 faster, the other showing the average of over 20 games.
Honestly, there's a bazillion benchmarks to prove that the 2080 is faster in general. Why are you denying this so hard? It makes zero sense, especially in a thread about a game where the 1080 Ti gets absolutely spanked by Turing and AMD cards that it usually keeps pace with.
And in 99% of games the RTX 2080 is either tied with the GTX 1080 Ti, slower than the 1080 Ti, or faster than it by 5-10% which is margin of error.
Not mistaking anything.
I know the difference between VRAM usage and VRAM allocation. Doesn't change the facts
The majority of people saying there's no difference are from just looking at screenshots/videos of the game. It goes without saying if you're on a low/mid PC running medium settings it's not going to be a massive difference from the 1X version.
Everyone rebutting you has been posting actual 'facts' with actual benchmarks, you're just pulling out "99%!" and "the extra VRAM matters!" with nothing to back it up.
99% of monitors don't support 1000 nit HDR10/Dolby Vision support. Not hard to prove. The number of monitors that support that spec are less than 5 monitors in the world out of thousands.
I literally posted an official article by AMD proving that more than 8GB frame buffer is useful. Guess you must have missed it.
Under display options. Maybe check if you have hdr enabled in windows 10? I did not have issues with that, but the game restricted audio options for me until I had enabled surround in windows 10, so the options seem to change somewhat based on your windows 10 settings
So the most obvious settings that actually differentiate graphics (without zooming 400%) between X1X and PC is the Lighting Quality set to Ultra and Water Physics set to Medium and up.
Running everything at X1X settings on a system which costs more than twice as much and runs every other game at almost double the framerate of the X1X is still bad optimization.
Running everything at X1X settings on a system which costs more than twice as much and runs every other game at almost double the framerate of the X1X is still bad optimization.
Water physics slider actually makes a difference?
I've heard it makes zero difference and the slider does nothing but cut performance in half
You don't even need to look at the visual settings to claim bad optimisation. Just look at the CPU behaviour on i5s and how awful the stuttering is in every town in the game.
Water ripples from shots, I found 50% to not cost that much. Only the highest setting tanks performance.
The comparison was GTX 1080 Ti vs RTX 2080, where both are functionally identical where 1080 Ti is faster or slower by 5-10%, or even as much as 40+% faster in VRAM limited scenarios, but ok.
My "babble" was off topic but I mean, remove the screen and PC gaming is still relatively expensive, is it not? I don't understand where the defense is coming from. Personally I still think it's worth every penny and I've already planned my next upgrade, but if people think it's not worth it then I have no issues with that. For me it's 30fps vs 60+ fps with mods and triple screen capabilities, I could never go back.
Aren't all desktop i5s missing hyperthreading and are either 4 or 6 thread cpus? I wonder if Rockstar just built the game to work best at minimum with the number of threads available on the consoles (8 threads).
I guess the minimum requirements asking for a 2500k doesn't help though
Why are you comparing specs between two different architectures??? (Pascal ≠Turing)
And seriously where do you get your info to think the Ti is faster than a 2080?
Grafikkarten-Rangliste 2024: 31 GPUs im Benchmark
Grafikkarten-Rangliste 2024 mit Nvidia-, AMD- und Intel-Grafikkarten: Benchmark-Ăśbersicht mit allen wichtigen Grafikchips von Nvidia, AMD und Intel.
Grafikkarten-Rangliste 2020: GPU-Vergleich
Welche Grafikkarte kaufen? ComputerBase liefert Kaufberatung mit aktuellen Tests, Benchmarks, Bestenlisten und Vergleiche fĂĽr den Dezember.
www.computerbase.de
The 2080 is around 8/9% faster on average
It's the close to X1X setting, pretty simple. Imo, the higher settings are not worth the performance cost at all.
TBH I'm pretty disappointed they haven't made higher quality textures for PC. Some of them are pretty muddy
btw, forgot to say: Thanks Alex for this awesome video. Enjoyed every minute of it!
Last edited:
I didn't try that yet, thank you! I normally don't have to have it enabled for games that support HDR, but that must be it.
Why are you comparing specs between two different architectures??? (Pascal ≠Turing)
And seriously where do you get your info to think the Ti is faster than a 2080?
Grafikkarten-Rangliste 2024: 31 GPUs im Benchmark
Grafikkarten-Rangliste 2024 mit Nvidia-, AMD- und Intel-Grafikkarten: Benchmark-Ăśbersicht mit allen wichtigen Grafikchips von Nvidia, AMD und Intel.www.pcgameshardware.de
Grafikkarten-Rangliste 2020: GPU-Vergleich
Welche Grafikkarte kaufen? ComputerBase liefert Kaufberatung mit aktuellen Tests, Benchmarks, Bestenlisten und Vergleiche fĂĽr den Dezember.
The 2080 is around 8/9% faster on average
We are only supposed to be testing old games apparently.
I Show in the Video how it runs at more than double the framerate of x1x
Btw everyone, thanks for such kind words. Taking a day of tomorrow, but will reply to everyone here very soon.
I'd be curious as well. If you play on lower res there's still quite some headroom there.
I can't imagine how much work this must have been, you've definitely deserved your day off. Fantastic video!
Great article, that must have been a lot of work.
I haven't seen a single review that shows 1080ti being on avg better than the 2080.
Here is a benchmark from techpowerup with more than 20 games.
Is xbox one x good value? yes of course it is but comparing it with the highest end components is meaningless, high end parts are always the worst value, that is especially true for the current gpu line up. A comparison that would make sense would be with similar performing pc parts. But in this particular case we couldn't even compare them accurately even if we wanted to with several settings lower than low on the console.
Gpu prices have gone up unfortunately but you don't need a very expensive pc for a decent experience. You can get a good value pc that will last you a while for 750$ - 800$ us. Example article
I haven't seen a single review that shows 1080ti being on avg better than the 2080.
Here is a benchmark from techpowerup with more than 20 games.
Is xbox one x good value? yes of course it is but comparing it with the highest end components is meaningless, high end parts are always the worst value, that is especially true for the current gpu line up. A comparison that would make sense would be with similar performing pc parts. But in this particular case we couldn't even compare them accurately even if we wanted to with several settings lower than low on the console.
Gpu prices have gone up unfortunately but you don't need a very expensive pc for a decent experience. You can get a good value pc that will last you a while for 750$ - 800$ us. Example article
I feel bad about calling this thread predictable before, didn't expect to see someone going on a rant about the 1080ti here.
If you are happy with 1080p/60fps you can basically max out the visuals on a 2070 Super and it looks incredible.
If you are happy with 1080p/60fps you can basically max out the visuals on a 2070 Super and it looks incredible.
Sadly the threads are predictable, they always devolve into arguments over petty shit :(
Ok, so I quickly tested the optimized XBX-equivalent settings and got the following rough numbers at 4k with the benchmark (using DSR on my 1080p monitor):
1st scene: 46 fps
2nd scene: 52-58 fps
3rd scene: 52 fps
4th scene: 53 fps
5th scene: 42-55 fps (drops to 42+ on horseback scene. Lowest drop is 32 during final shootout. Interestingly, this final scene had snow in it so not sure if that affected performance)
Min: 33.76 fps
Avg: 45.24 fps
This is on a 1080 Ti (oc'ed a bit), stock 3700x and 32GB ram.
To be honest, I was expecting much higher numbers than that with my PC. My initial expectation was hitting 60 fps with similar console settings at 4k, which I am so far able to achieve with other games (usually with even better fidelity too)
1st scene: 46 fps
2nd scene: 52-58 fps
3rd scene: 52 fps
4th scene: 53 fps
5th scene: 42-55 fps (drops to 42+ on horseback scene. Lowest drop is 32 during final shootout. Interestingly, this final scene had snow in it so not sure if that affected performance)
Min: 33.76 fps
Avg: 45.24 fps
This is on a 1080 Ti (oc'ed a bit), stock 3700x and 32GB ram.
To be honest, I was expecting much higher numbers than that with my PC. My initial expectation was hitting 60 fps with similar console settings at 4k, which I am so far able to achieve with other games (usually with even better fidelity too)
Last edited:
I have a 2060 and got like ~80 FPS@1080p with the X1X settings and having such a hard time balancing what to bump up because alot of settings just look amazing.
Nothing to detract from your video that's some thorough work!
Your GFX card alone is more than double the price of a X1X, I was comparing it to my system which is a GTX 2070 with a 9700k :)
Can you imagine spending hours testing a bunch of individual graphics settings, explaining what they change and how they affect performance, and then coming back to a thread full of "but you need to buy a monitor!" and weird currency-mixing value arguments?
As always, Dictator, thanks for all the hard work. When I finally try out RDR2, it'll really help to refer to your stuff to get the most out of it!
As always, Dictator, thanks for all the hard work. When I finally try out RDR2, it'll really help to refer to your stuff to get the most out of it!
Other than water physics what are the most performance intensive features?
I'm trying to push up performance a tad and can't decide what to cut down.
I'm trying to push up performance a tad and can't decide what to cut down.
The video is well done but I don't think it says much about optimization specifically.
To know about optimization you would have to run on PC on similar hardware, and see how close they are. I'm pretty sure the results would be bad since from all benchmarks I've seen the game doesn't seem to appreciate much slightly older hardware. (we used to match PS4 performance with a 750 Ti, I seriously, seriously doubt it will work here)
It just seems to say the game's optimization was done on top of the line hardware.
The bottom line also seems to be that to get small benefits to visual quality you have to pay a much higher hardware cost. So the game does scale up, but quite poorly.
An actual request to DF: could you please make a video where you do this? Use PS4 basic quality settings and try to match with the lowest PC hardware possible. That could be a serious final answer on optimization.
(but I do understand these videos are especially made for enthusiasts with super bleeding edge hardware, and that look... flattering to both software and hardware companies, and so it's understandable there's an incentive going in that direction)
To know about optimization you would have to run on PC on similar hardware, and see how close they are. I'm pretty sure the results would be bad since from all benchmarks I've seen the game doesn't seem to appreciate much slightly older hardware. (we used to match PS4 performance with a 750 Ti, I seriously, seriously doubt it will work here)
It just seems to say the game's optimization was done on top of the line hardware.
The bottom line also seems to be that to get small benefits to visual quality you have to pay a much higher hardware cost. So the game does scale up, but quite poorly.
An actual request to DF: could you please make a video where you do this? Use PS4 basic quality settings and try to match with the lowest PC hardware possible. That could be a serious final answer on optimization.
(but I do understand these videos are especially made for enthusiasts with super bleeding edge hardware, and that look... flattering to both software and hardware companies, and so it's understandable there's an incentive going in that direction)
Last edited:
Md Ray the absolute mad man.
Just drop the resolution to 1800p and call it a day? You'l most likely be able to maintain 60fps and even bump some settings here and there. Not sure why people are so offended by their 1080ti not being on par with newer hardware. Older architecture and more RAM won't save you from the inevitable. Similar results will only happen more often as drivers and newer games get more optimized for more modern architecture. Pascal is bad at Async.
This is coming from a 1080ti owner btw.
Just drop the resolution to 1800p and call it a day? You'l most likely be able to maintain 60fps and even bump some settings here and there. Not sure why people are so offended by their 1080ti not being on par with newer hardware. Older architecture and more RAM won't save you from the inevitable. Similar results will only happen more often as drivers and newer games get more optimized for more modern architecture. Pascal is bad at Async.
This is coming from a 1080ti owner btw.
MSAA is costly. 2x takes me from 75fps to 60fps or there abouts. Close Volumetrics can hit hard. I have that at medium and Far Volumetrics at High.
Textures, Trees and lighting ar at Ultra and everything else at High except water at medium. Now I have a 2080ti but also an i5 [email protected] and 32GB 3200RAM. Usually get over 80fps (sometimes 90fps if free of trees and animals/people) in the wilds but in towns varies from as low as 60fps to 75fps. I'd say my average is around 80. No stuttering on DX12 or frame dipping but I would sometimes get serious frame drops down to 5fps on Vulkan. Not often though, once an hour maybe.
I know I have a serious GPU but only one crash (when I had the game minimized) and great performance in general. And it looks fantastic.
Not offended at all. Just not used to seeing the 1080 Ti struggling this much. My native res is 3440x1440 by the way, and I get similar or higher framerates than the above at higher settings too so it isn't too bad so far.
The 2080 is consistently 10% faster on average. Average of 15 games in the first link and 23 games in the second link:
Grafikkarten-Rangliste 2024: GPUs im Vergleich
Welche Grafikkarte kaufen? Für den März gibt es Empfehlungen mit Nvidia GeForce RTX 4000, AMD Radeon RX 7000 und Intel Arc.
www.computerbase.de
Last edited:
The 2080 is consistently 10-15% faster on average. Average of 15 games in the first link and 23 games in the second link:
Grafikkarten-Rangliste 2024: GPUs im Vergleich
Welche Grafikkarte kaufen? Für den März gibt es Empfehlungen mit Nvidia GeForce RTX 4000, AMD Radeon RX 7000 und Intel Arc.
But which one runs Crysis faster mate.
Yeah I'm on 2080ti too. Just pumped POM and Tesselation up to ultra and POM on ultra in particular looks really noticeably nice. Might pop tesselation back down to high tho.
Can we get back onto the topic of this video, RDR2 on PC and its settings please? 99% of games are not in this thread. Thank you.
I would like so much a new Crysis that plays more like the first one and that is also the graphical leap that the first one was.
It would also probably sell decent enough now that piracy is not such a concern as it was those days.
I don't think that's really a viable usage of DF's limited time. I definitely want to see more comparisons done with say, 580/1660 and the like as that performance class is still very popular and is competitive with the console refreshes in most specs/games, but going lower than that, in late 2019 no less, I just fail to see the point. The 750ti hasn't been available to purchase new for a long time for a reason. Any recently modern titles are going to struggle on that kind of hardware, this is not something that really needs to be analyzed in-depth.