Yea, I thought the same, but I did several test low/med/high/ultra, even killed the game and relaunched.
But Im still leaning towards it being my system somehow, lol.
Thanks for the effort!
Advanced Questions on Graphic Settings
- Thread starter Exigeous
- Start date
Yea, I thought the same, but I did several test low/med/high/ultra, even killed the game and relaunched.
But Im still leaning towards it being my system somehow, lol.
Thanks for the effort!
Good to know - I didn't take time for this today as I wanted to finish the first draft with everything I've done so far. I'm honestly not great with Premier Pro so I've been wanting an excuse to learn a few things (like fly-out titles, etc) so I spent 4-5 hours this morning on that.
Okay so I don't usually share drafts of videos but here is the first 20 minutes of what I have. I'm still missing volumetric and the other planetary stuff so I'll work on that tonight. There are certainly a few bugs (missing titles, video needs to shift down as I added the timeline last minute, etc) but if you wanna see what I'm thinking here goes. Oh and of course feedback welcome
Source: https://youtu.be/DqZNAJAqUJg
Thanks, it's by far the most work I've ever put into a video - I've had it on my list since literally the first day I started my channel 2.5 years ago so I'm glad to finally get around to it but glad I waited so I had the skills to understand all of it.
Now, if I can just find volumetric I'd be so happy. I thought a few times I had, when I screamed into Discord to only then be disappointed I got lots of laughs...
For the AA tests you appear to only be looking at the degree to which jaggies are smoothed when assessing the desirability of the settings.
I agree that FXAA has a small edge in this regard, but it comes at the cost of blurring the the rest of the view much more noticeably than the other modes. Overall, I've always felt SMAA was the best option...performance difference is marginal, it does a pretty good job at removing jaggies, but preserves details elsewhere much better.
Also, all of ED's shadow presets are horrible and while I agree with testing what the game comes with as a baseline, I'd highly recommend anyone who cares about shadow quality to look into manually editing the tables themselves (or using the work I've done elsewhere as a starting point). Unfortunately, with the shadow system the game uses, there will always be compromises, but the effect can still be significantly improved.
I'll see if I can find an example of the discrepancy between the volumetric settings, if indeed it hasn't been depreciated.
Finally got around to taking another look at the volumetric effects and I couldn't see any visual or performance difference of any kind between the settings.
I even tried to induce a greater discrepancy by softening the quality of 'low' and jacking up the quality of 'ultra' with the following addition to my GraphicsConfigurationOverride.xml:
I checked light cones and the light scattering through ice canyons on Europa, zero difference, as far as was discernible, confirming Exigeous' account. If the effect was present and controlled by the listed setting, I should have seen some sort of performance hit with twice the blur samples and four times the resolution. Nothing.
Frontier seems to have depreciated the setting at some point and not only neglected to pull it from the configuration files (which has been done in the past), but also failed to remove it from the UI options, which is odd.
I even tried to induce a greater discrepancy by softening the quality of 'low' and jacking up the quality of 'ultra' with the following addition to my GraphicsConfigurationOverride.xml:
XML:
<?xml version="1.0" encoding="UTF-8" ?>
<GraphicsConfig>
<Volumetrics>
<Low>
<LocalisationName>$QUALITY_LOW;</LocalisationName>
<StepsPerPixel>32</StepsPerPixel>
<DownscalingFactor>4</DownscalingFactor>
<BlurSamples>2</BlurSamples>
<TwoPassBlur>false</TwoPassBlur>
<InScatterSteps>1</InScatterSteps>
<StepMultiplier>8.0</StepMultiplier>
<RingQuality>0</RingQuality>
<FogMotesEnabled>false</FogMotesEnabled>
</Low>
<Ultra>
<LocalisationName>$QUALITY_ULTRA;</LocalisationName>
<StepsPerPixel>32</StepsPerPixel>
<DownscalingFactor>1</DownscalingFactor>
<BlurSamples>4</BlurSamples>
<TwoPassBlur>true</TwoPassBlur>
<InScatterSteps>2</InScatterSteps>
<StepMultiplier>4.0</StepMultiplier>
<RingQuality>3</RingQuality>
<FogMotesEnabled>true</FogMotesEnabled>
</Ultra>
</Volumetrics>
</GraphicsConfig>I checked light cones and the light scattering through ice canyons on Europa, zero difference, as far as was discernible, confirming Exigeous' account. If the effect was present and controlled by the listed setting, I should have seen some sort of performance hit with twice the blur samples and four times the resolution. Nothing.
Frontier seems to have depreciated the setting at some point and not only neglected to pull it from the configuration files (which has been done in the past), but also failed to remove it from the UI options, which is odd.
Informative video - one more thing about Shadows is that they render farther as the setting increases - I use the Ultra level so that I don’t get a shadow “cut-off line” when flying towards the big Stations.
Edit: several screenshots to illustrate -
Shadows Low (no projected shadows visible at this distance):
Shadows Medium (projected shadows visible to the habitation ring):
Shadows High (shadows visible all the way to the dock...):
(but if I turn my head, the shadow cut-off point becomes visible):
Ultra Shadows (basically the same as high but without the "cut-off" when looking around):
Shadows Medium (projected shadows visible to the habitation ring):
Shadows High (shadows visible all the way to the dock...):
(but if I turn my head, the shadow cut-off point becomes visible):
Ultra Shadows (basically the same as high but without the "cut-off" when looking around):
Last edited:
First, this was a great video, i actually learned alot from seeing the visual quality comparisons between the effects. Im actually going to change the shadows and bloom settings i use from it, so nice one.
There's might be of a logical problem with the core comparison though. I suspect that the percentage impacts of each of the settings might change depending on the card you're using. You've got the best card you can have right now, which is good, but at the same time, the majority of people who who are going to need to even bother with any of that are people running lower hardware... so the percentage differences might not apply out of the gate. But absolutely, i did find it super useful to see the actual differences themselves, which has saved me those many hours doing the comparisons, so there's great content in that video anyway.
Also just for the bloom setting, i think the glowing halo around stars might be the other main use of this effect, not sure if you want to check out the visual differences after dropping out of supercruise say to see if its makes a difference.
Yep, I've added a new section showing the detail differences in a station which I agree is a bit more useful. That said to my eyes I still find FXAA to look the best all around but it does give a less one dimensional view that I had there, thanks for the feedback.
Exactly my findings and basically what I've added to the video. The crazy thing is in certain places, like the Guardian Beacon or Thargoid structures there is a HUGE performance hit on average for Ultra - like 22% on average over all other settings. I tested this multiple times, averaging about 18 different tests and that was consistent.
Thanks, that's the whole idea so glad it came through.
Totally agree and I'm addressing this in the intro to the video (which wasn't in the draft). Unfortunately I only have the one PC to test with so it's all I can do. Plus the 60 hours I have in this is enough already!
Almost positive that's based on Environment quality or FX and not bloom. There are several things that you'd think would be one setting to only be another.
And now I still can't f'ing find Material quality. As a final test I'm going to run the SRV training missions with the different settings to see if I can at least see a FPS difference in them, it's all I can do at this point. If I do get a reply from Frontier I'll put out an updated video covering that and Volumetric, depending on what they say
Thanks for all the input everyone - hope to have this done in the next 8 hours. Looking forward to everyone's response.
Oh - last question - title ideas?
"Elite Dangerous Graphical Deep Dive"
"Elite Graphic Settings FULLY Explained"
"Every Elite Graphic Setting FULLY Analyized"
"Understanding ALL Elite Dangerous Graphic Settings"
Thoughts???
~X
I'd vote for that one.
HAHAHAHA - hysterical!
GOOD NEWS - I FOUND WHAT MATERIAL QUALITY AFFECTS!!!!!!!!!!
It's the atmospheres of planets. Tested at a few earth likes and gas giants and while the visual differences are slight (especially in gas giants) there is absolutely a difference visually and in terms of frame rate!! FINALLY!!!
Whew. Just hit me on a whim to go check that for some reason and boy am I glad I did!!!
Finally video soon.
~X
If there is a performance hit, the setting has to be doing something.
I didn't evaluate fog/mist, or any of the alien structures, in the tests I did yesterday, but it looks like I'll have to take another look. My default personal override settings include tweaks to the ultra volumetric preset, and I distinctly recall seeing both a visual and performance difference at some point. Maybe I'll visit one of these areas you've noted a large performance hit and try to disable the volumetric effects entirely to see if there is a visual difference then.
Agreed - and I say as such in the video. My strong assumption is there is a bug somewhere and that what is supposed to be displayed isn't as I still after dozens and dozens of tests can see no difference visually. UGH this one section has taken me over 20% of the total testing work!!
The good news is i'm done, rendering now.
And here she is - running it by a few other friends before I share it with everyone but given the help you guys have been I figured I'd share it here first as well. Looking forward to your thoughts as given the amount of work I put into this one I hope people find it useful.
Source: https://youtu.be/G7efYzpquIs
EDIT: Updated URL
~X
EDIT: Updated URL
~X
Last edited:
