Jump to content

Low image quality on newer GPUs

Demiqas

Awesome. Thanks. :)

 

Ok, now we can start to draw some conclusions. Tic-Tac is using 2xMSAA with TXAA off. In my screenshots with 4xMSAA, even with TXAA off, there is noticeably less aliasing throughout the image.

 

With the settings you're using, which is a lower level of AA filtering, I would say; yes that is "normal" to see that amount of aliasing.  

 

I would hypothesize and be willing to bet that at 8xMSAA the image quality would be much more like @Ryan_Vickers Photoshop'd images. ;)

There will be literally no noticable difference between 4xmsaa and 8xmsaa. I've tried this countless amount of times. If something like 8xmsaa would solve these issues I wouldn't have made this thread in the first place.

Link to comment
Share on other sites

Link to post
Share on other sites

 

 

Except now I run 4xMSAA and HiRes Shadows OFF . Maybe some settings in NCP could help? Besides editing Settings.xml and changing <MSAAQuality value="0" />

 

 

Cosmic Council Department of Defense ; Interplanetary Class 3 Relations & Diplomatic Affairs - OFFICE 117

Link to comment
Share on other sites

Link to post
Share on other sites

There will be literally no noticable difference between 4xmsaa and 8xmsaa. I've tried this countless amount of times. If something like 8xmsaa would solve these issues I wouldn't have made this thread in the first place.

 

Can you please post up a (full size) screenshot of the same scene with both settings (4x and 8x MSAA) so we can compare?  

My Systems:

Main - Work + Gaming:

Spoiler

Woodland Raven: Ryzen 2700X // AMD Wraith RGB // Asus Prime X570-P // G.Skill 2x 8GB 3600MHz DDR4 // Radeon RX Vega 56 // Crucial P1 NVMe 1TB M.2 SSD // Deepcool DQ650-M // chassis build in progress // Windows 10 // Thrustmaster TMX + G27 pedals & shifter

F@H Rig:

Spoiler

FX-8350 // Deepcool Neptwin // MSI 970 Gaming // AData 2x 4GB 1600 DDR3 // 2x Gigabyte RX-570 4G's // Samsung 840 120GB SSD // Cooler Master V650 // Windows 10

 

HTPC:

Spoiler

SNES PC (HTPC): i3-4150 @3.5 // Gigabyte GA-H87N-Wifi // G.Skill 2x 4GB DDR3 1600 // Asus Dual GTX 1050Ti 4GB OC // AData SP600 128GB SSD // Pico 160XT PSU // Custom SNES Enclosure // 55" LG LED 1080p TV  // Logitech wireless touchpad-keyboard // Windows 10 // Build Log

Laptops:

Spoiler

MY DAILY: Lenovo ThinkPad T410 // 14" 1440x900 // i5-540M 2.5GHz Dual-Core HT // Intel HD iGPU + Quadro NVS 3100M 512MB dGPU // 2x4GB DDR3L 1066 // Mushkin Triactor 480GB SSD // Windows 10

 

WIFE'S: Dell Latitude E5450 // 14" 1366x768 // i5-5300U 2.3GHz Dual-Core HT // Intel HD5500 // 2x4GB RAM DDR3L 1600 // 500GB 7200 HDD // Linux Mint 19.3 Cinnamon

 

EXPERIMENTAL: Pinebook // 11.6" 1080p // Manjaro KDE (ARM)

NAS:

Spoiler

Home NAS: Pentium G4400 @3.3 // Gigabyte GA-Z170-HD3 // 2x 4GB DDR4 2400 // Intel HD Graphics // Kingston A400 120GB SSD // 3x Seagate Barracuda 2TB 7200 HDDs in RAID-Z // Cooler Master Silent Pro M 1000w PSU // Antec Performance Plus 1080AMG // FreeNAS OS

 

Link to comment
Share on other sites

Link to post
Share on other sites

Can you please post up a (full size) screenshot of the same scene with both settings (4x and 8x MSAA) so we can compare?  

Sorry, as I mentioned I already got rid of my PC. If you were to take a screenshot of 4x vs 8x you can zoom in and compare pixels to see a difference, but good luck seeing any when moving around in a game. 

Link to comment
Share on other sites

Link to post
Share on other sites

Sorry, as I mentioned I already got rid of my PC. If you were to take a screenshot of 4x vs 8x you can zoom in and compare pixels to see a difference, but good luck seeing any when moving around in a game. 

So now we have no way to get proof from you further which means your thread is ultimately useless, at least to yourself.

CONSOLE KILLER: Pentium III 700mhz . 512MB RAM . 3DFX VOODOO 3 SLi

 

 

Link to comment
Share on other sites

Link to post
Share on other sites

Except now I run 4xMSAA and HiRes Shadows OFF . Maybe some settings in NCP could help? Besides editing Settings.xml and changing <MSAAQuality value="0" />

And does it look any better? Going between 2x and 4x MSAA I can see the difference in-game.

 

I would recommend not touching any of the settings in NVCP except antistropic filtering and leaving everything else set to "let the application decide". I've always had better results tweaking the in-game settings vs the NVCP or AMD catalyst settings.

 

With the settings I'm running in GTAV, the game honestly looks great, IMO. Could the AA be improved? Possibly. But here's the thing; my GPU can't run the max level of AA and there's a ton of detail in the game. So I [have to] except the level of aliasing I see in the game at the settings I run. The aliasing we see is "normal", relative to the level of AA we're using in each game.  

My Systems:

Main - Work + Gaming:

Spoiler

Woodland Raven: Ryzen 2700X // AMD Wraith RGB // Asus Prime X570-P // G.Skill 2x 8GB 3600MHz DDR4 // Radeon RX Vega 56 // Crucial P1 NVMe 1TB M.2 SSD // Deepcool DQ650-M // chassis build in progress // Windows 10 // Thrustmaster TMX + G27 pedals & shifter

F@H Rig:

Spoiler

FX-8350 // Deepcool Neptwin // MSI 970 Gaming // AData 2x 4GB 1600 DDR3 // 2x Gigabyte RX-570 4G's // Samsung 840 120GB SSD // Cooler Master V650 // Windows 10

 

HTPC:

Spoiler

SNES PC (HTPC): i3-4150 @3.5 // Gigabyte GA-H87N-Wifi // G.Skill 2x 4GB DDR3 1600 // Asus Dual GTX 1050Ti 4GB OC // AData SP600 128GB SSD // Pico 160XT PSU // Custom SNES Enclosure // 55" LG LED 1080p TV  // Logitech wireless touchpad-keyboard // Windows 10 // Build Log

Laptops:

Spoiler

MY DAILY: Lenovo ThinkPad T410 // 14" 1440x900 // i5-540M 2.5GHz Dual-Core HT // Intel HD iGPU + Quadro NVS 3100M 512MB dGPU // 2x4GB DDR3L 1066 // Mushkin Triactor 480GB SSD // Windows 10

 

WIFE'S: Dell Latitude E5450 // 14" 1366x768 // i5-5300U 2.3GHz Dual-Core HT // Intel HD5500 // 2x4GB RAM DDR3L 1600 // 500GB 7200 HDD // Linux Mint 19.3 Cinnamon

 

EXPERIMENTAL: Pinebook // 11.6" 1080p // Manjaro KDE (ARM)

NAS:

Spoiler

Home NAS: Pentium G4400 @3.3 // Gigabyte GA-Z170-HD3 // 2x 4GB DDR4 2400 // Intel HD Graphics // Kingston A400 120GB SSD // 3x Seagate Barracuda 2TB 7200 HDDs in RAID-Z // Cooler Master Silent Pro M 1000w PSU // Antec Performance Plus 1080AMG // FreeNAS OS

 

Link to comment
Share on other sites

Link to post
Share on other sites

Sorry, as I mentioned I already got rid of my PC. If you were to take a screenshot of 4x vs 8x you can zoom in and compare pixels to see a difference, but good luck seeing any when moving around in a game.

*slow clap*

All you've provided to your own thread is heresay.

 

G3258 V 860k (Spoiler: G3258 wins)

 

 

Spoiler

i7-4790K | MSI R9 390x | Cryorig H5 | MSI Z97 Gaming 7 Motherboard | G.Skill Sniper 8gbx2 1600mhz DDR3 | Corsair 300R | WD Green 2TB 2.5" 5400RPM drive | <p>Corsair RM750 | Logitech G602 | Corsair K95 RGB | Logitech Z313

Link to comment
Share on other sites

Link to post
Share on other sites

So now we have no way to get proof from you further which means your thread is ultimately useless, at least to yourself.

Knew it'd happen from the start.

Also, had to be the 666th reply to this topic.

 

G3258 V 860k (Spoiler: G3258 wins)

 

 

Spoiler

i7-4790K | MSI R9 390x | Cryorig H5 | MSI Z97 Gaming 7 Motherboard | G.Skill Sniper 8gbx2 1600mhz DDR3 | Corsair 300R | WD Green 2TB 2.5" 5400RPM drive | <p>Corsair RM750 | Logitech G602 | Corsair K95 RGB | Logitech Z313

Link to comment
Share on other sites

Link to post
Share on other sites

Ryan_Vickers made a great contribution to the topic with his work on Photoshop! That's precisely what I was missing, some form of a more precise measurement.

I guess we only need one thing now to end the debate: make the same analysis using Ultra Settings. I remember using 8xMSAA and DSR and still being bothered with the quality, but we need to measure that in detail to confirm if there is a problem.

Link to comment
Share on other sites

Link to post
Share on other sites

Sorry, as I mentioned I already got rid of my PC. If you were to take a screenshot of 4x vs 8x you can zoom in and compare pixels to see a difference, but good luck seeing any when moving around in a game.

 

I will attempt running 8x MSAA again and hopefully the game doesn't crash before I can get a screenshot taken. If I can't get it, I can have my friend take comparison shots between 4x and 8x as his 980Ti should be able to handle it with the extra Vram. 

 

Going back to the article you posted on the first page, we find this quote: 

Today’s graphics cards are usually powerful enough to use 16x anisotropic texture filtering at all times, plus they are using better texture filters inherently, which means that textures are actually rendered “too sharp” which results in aliased textures. Nvidia has a setting in their driver though which is called “Texture filtering – Negative LOD bias” which has two options: “Allow” and “Clamp”. Setting it to “Clamp” would simply ignore all negative Mip LOD bias from a game engine and use the neutral 0.0 setting instead.

 

As well as this:

The “Texture filtering – Negative LOD bias” is still available in Nvidia’s current drivers, but it doesn’t work anymore, and apparently it had been like this for quite a while allready. (if you are using a more recent graphics card, that is.)

 

Which ties into this:

Using a GTX400 series card or newer, the setting doesn’t work anymore (at least not in DirectX applications)

 

So, we have an interesting combination of things going on here:

 

-Higher levels of detail in modern games, WAY more than ever before. 

-Textures being rendered excessively sharp (on newer GPUs).

-Lack of or broken LOD bias option(s) (apparently since the GTX 200 series).

-1080p resolution with limited number of pixels and somewhat low pixel density (by today's standards).  

 

Considering the above, I can concede that there has been a change in image quality over time. However, the image quality still remains "normal" in relation to the above combination of factors. The LOD bias issue, specifically, would seem to be a somewhat significant change, which brings be to ask the following questions:

 

-If LOD bias was locked out starting with the GTX 400 series (which launched back in 2010), then why haven't we heard more about this over the last 5-6 years? 

-Is it possible that not many people knew about or even used the LOD bias option(s) in games back when it was functioning and thus; to this day, didn't know such thing even existed? (I am one of those people). If the answer to this question is yes, then that would explain why the majority has found this to be a non-issue. 

-If LOD bias has been locked out starting with the GTX 400 series and is non-functional in all GTX 900 series cards, then would it not be impossible for the image quality to be "propped up" (using your words) as if the LOD bias is functioning, then degrade/become non-functioning over time (a few weeks as you claim)?

My Systems:

Main - Work + Gaming:

Spoiler

Woodland Raven: Ryzen 2700X // AMD Wraith RGB // Asus Prime X570-P // G.Skill 2x 8GB 3600MHz DDR4 // Radeon RX Vega 56 // Crucial P1 NVMe 1TB M.2 SSD // Deepcool DQ650-M // chassis build in progress // Windows 10 // Thrustmaster TMX + G27 pedals & shifter

F@H Rig:

Spoiler

FX-8350 // Deepcool Neptwin // MSI 970 Gaming // AData 2x 4GB 1600 DDR3 // 2x Gigabyte RX-570 4G's // Samsung 840 120GB SSD // Cooler Master V650 // Windows 10

 

HTPC:

Spoiler

SNES PC (HTPC): i3-4150 @3.5 // Gigabyte GA-H87N-Wifi // G.Skill 2x 4GB DDR3 1600 // Asus Dual GTX 1050Ti 4GB OC // AData SP600 128GB SSD // Pico 160XT PSU // Custom SNES Enclosure // 55" LG LED 1080p TV  // Logitech wireless touchpad-keyboard // Windows 10 // Build Log

Laptops:

Spoiler

MY DAILY: Lenovo ThinkPad T410 // 14" 1440x900 // i5-540M 2.5GHz Dual-Core HT // Intel HD iGPU + Quadro NVS 3100M 512MB dGPU // 2x4GB DDR3L 1066 // Mushkin Triactor 480GB SSD // Windows 10

 

WIFE'S: Dell Latitude E5450 // 14" 1366x768 // i5-5300U 2.3GHz Dual-Core HT // Intel HD5500 // 2x4GB RAM DDR3L 1600 // 500GB 7200 HDD // Linux Mint 19.3 Cinnamon

 

EXPERIMENTAL: Pinebook // 11.6" 1080p // Manjaro KDE (ARM)

NAS:

Spoiler

Home NAS: Pentium G4400 @3.3 // Gigabyte GA-Z170-HD3 // 2x 4GB DDR4 2400 // Intel HD Graphics // Kingston A400 120GB SSD // 3x Seagate Barracuda 2TB 7200 HDDs in RAID-Z // Cooler Master Silent Pro M 1000w PSU // Antec Performance Plus 1080AMG // FreeNAS OS

 

Link to comment
Share on other sites

Link to post
Share on other sites

A contributing factor to the GTA screenshots is that msaa does not fully apply. The internal to the car is not acted on by msaa and will only be impacted by post processing aa, like fxaa or smaa. This is one of those problems that impacts all lighting in game and with no good ppaa implementations its kind of inevitable it will look rubbish.

This difference in geometry based aa and post processing effect impacts image quality in a lot of games.

Link to comment
Share on other sites

Link to post
Share on other sites

So now we have no way to get proof from you further which means your thread is ultimately useless, at least to yourself.

 

*slow clap*

All you've provided to your own thread is heresay.

 

To be fair, @Demiqas did provide example images on several occasions showing a "before/after" comparison and linked to an article that addressed the LOD bias issue (which is a real thing).

 

There isn't anything we can do about the LOD bias and we (apparently) haven't even had the option for 5 years now (with Nvidia GPUs anyways). Now, if the option was never there to begin with, then it's not possible for it to have been functioning (or for the image quality appear as such) when the card is first used, and then degrade over a period of several weeks. That is the claim that needs to be proven at this point. 

My Systems:

Main - Work + Gaming:

Spoiler

Woodland Raven: Ryzen 2700X // AMD Wraith RGB // Asus Prime X570-P // G.Skill 2x 8GB 3600MHz DDR4 // Radeon RX Vega 56 // Crucial P1 NVMe 1TB M.2 SSD // Deepcool DQ650-M // chassis build in progress // Windows 10 // Thrustmaster TMX + G27 pedals & shifter

F@H Rig:

Spoiler

FX-8350 // Deepcool Neptwin // MSI 970 Gaming // AData 2x 4GB 1600 DDR3 // 2x Gigabyte RX-570 4G's // Samsung 840 120GB SSD // Cooler Master V650 // Windows 10

 

HTPC:

Spoiler

SNES PC (HTPC): i3-4150 @3.5 // Gigabyte GA-H87N-Wifi // G.Skill 2x 4GB DDR3 1600 // Asus Dual GTX 1050Ti 4GB OC // AData SP600 128GB SSD // Pico 160XT PSU // Custom SNES Enclosure // 55" LG LED 1080p TV  // Logitech wireless touchpad-keyboard // Windows 10 // Build Log

Laptops:

Spoiler

MY DAILY: Lenovo ThinkPad T410 // 14" 1440x900 // i5-540M 2.5GHz Dual-Core HT // Intel HD iGPU + Quadro NVS 3100M 512MB dGPU // 2x4GB DDR3L 1066 // Mushkin Triactor 480GB SSD // Windows 10

 

WIFE'S: Dell Latitude E5450 // 14" 1366x768 // i5-5300U 2.3GHz Dual-Core HT // Intel HD5500 // 2x4GB RAM DDR3L 1600 // 500GB 7200 HDD // Linux Mint 19.3 Cinnamon

 

EXPERIMENTAL: Pinebook // 11.6" 1080p // Manjaro KDE (ARM)

NAS:

Spoiler

Home NAS: Pentium G4400 @3.3 // Gigabyte GA-Z170-HD3 // 2x 4GB DDR4 2400 // Intel HD Graphics // Kingston A400 120GB SSD // 3x Seagate Barracuda 2TB 7200 HDDs in RAID-Z // Cooler Master Silent Pro M 1000w PSU // Antec Performance Plus 1080AMG // FreeNAS OS

 

Link to comment
Share on other sites

Link to post
Share on other sites

A contributing factor to the GTA screenshots is that msaa does not fully apply. The internal to the car is not acted on by msaa and will only be impacted by post processing aa, like fxaa or smaa. This is one of those problems that impacts all lighting in game and with no good ppaa implementations its kind of inevitable it will look rubbish.

This difference in geometry based aa and post processing effect impacts image quality in a lot of games.

 

In other words; what we are "seeing" in-game is "normal", all factors considered. Would you agree? 

 

All factors being: Resolution, in-game settings, game graphics engine, lighting effects, etc.

My Systems:

Main - Work + Gaming:

Spoiler

Woodland Raven: Ryzen 2700X // AMD Wraith RGB // Asus Prime X570-P // G.Skill 2x 8GB 3600MHz DDR4 // Radeon RX Vega 56 // Crucial P1 NVMe 1TB M.2 SSD // Deepcool DQ650-M // chassis build in progress // Windows 10 // Thrustmaster TMX + G27 pedals & shifter

F@H Rig:

Spoiler

FX-8350 // Deepcool Neptwin // MSI 970 Gaming // AData 2x 4GB 1600 DDR3 // 2x Gigabyte RX-570 4G's // Samsung 840 120GB SSD // Cooler Master V650 // Windows 10

 

HTPC:

Spoiler

SNES PC (HTPC): i3-4150 @3.5 // Gigabyte GA-H87N-Wifi // G.Skill 2x 4GB DDR3 1600 // Asus Dual GTX 1050Ti 4GB OC // AData SP600 128GB SSD // Pico 160XT PSU // Custom SNES Enclosure // 55" LG LED 1080p TV  // Logitech wireless touchpad-keyboard // Windows 10 // Build Log

Laptops:

Spoiler

MY DAILY: Lenovo ThinkPad T410 // 14" 1440x900 // i5-540M 2.5GHz Dual-Core HT // Intel HD iGPU + Quadro NVS 3100M 512MB dGPU // 2x4GB DDR3L 1066 // Mushkin Triactor 480GB SSD // Windows 10

 

WIFE'S: Dell Latitude E5450 // 14" 1366x768 // i5-5300U 2.3GHz Dual-Core HT // Intel HD5500 // 2x4GB RAM DDR3L 1600 // 500GB 7200 HDD // Linux Mint 19.3 Cinnamon

 

EXPERIMENTAL: Pinebook // 11.6" 1080p // Manjaro KDE (ARM)

NAS:

Spoiler

Home NAS: Pentium G4400 @3.3 // Gigabyte GA-Z170-HD3 // 2x 4GB DDR4 2400 // Intel HD Graphics // Kingston A400 120GB SSD // 3x Seagate Barracuda 2TB 7200 HDDs in RAID-Z // Cooler Master Silent Pro M 1000w PSU // Antec Performance Plus 1080AMG // FreeNAS OS

 

Link to comment
Share on other sites

Link to post
Share on other sites

@Ryan_Vickers @Tic-Tac @Demiqas

 

Not sure why the game crashed last time I tried 8xMSAA, but it's working fine now (except for lower fps). Here's GTAV comparing 4x and 8x MSAA with TXAA off (because it doesn't let you use TXAA with 8xMSAA). 

 

4xMSAA:

0OTvnV6.png

 

8xMSAA:

PQOvNXq.png

 

Don't know about you, but I definitely see a difference in improvement. Can see throughout the image where it clearly looks better at 8x. 

 

I think this pretty much speaks for itself. ;)

My Systems:

Main - Work + Gaming:

Spoiler

Woodland Raven: Ryzen 2700X // AMD Wraith RGB // Asus Prime X570-P // G.Skill 2x 8GB 3600MHz DDR4 // Radeon RX Vega 56 // Crucial P1 NVMe 1TB M.2 SSD // Deepcool DQ650-M // chassis build in progress // Windows 10 // Thrustmaster TMX + G27 pedals & shifter

F@H Rig:

Spoiler

FX-8350 // Deepcool Neptwin // MSI 970 Gaming // AData 2x 4GB 1600 DDR3 // 2x Gigabyte RX-570 4G's // Samsung 840 120GB SSD // Cooler Master V650 // Windows 10

 

HTPC:

Spoiler

SNES PC (HTPC): i3-4150 @3.5 // Gigabyte GA-H87N-Wifi // G.Skill 2x 4GB DDR3 1600 // Asus Dual GTX 1050Ti 4GB OC // AData SP600 128GB SSD // Pico 160XT PSU // Custom SNES Enclosure // 55" LG LED 1080p TV  // Logitech wireless touchpad-keyboard // Windows 10 // Build Log

Laptops:

Spoiler

MY DAILY: Lenovo ThinkPad T410 // 14" 1440x900 // i5-540M 2.5GHz Dual-Core HT // Intel HD iGPU + Quadro NVS 3100M 512MB dGPU // 2x4GB DDR3L 1066 // Mushkin Triactor 480GB SSD // Windows 10

 

WIFE'S: Dell Latitude E5450 // 14" 1366x768 // i5-5300U 2.3GHz Dual-Core HT // Intel HD5500 // 2x4GB RAM DDR3L 1600 // 500GB 7200 HDD // Linux Mint 19.3 Cinnamon

 

EXPERIMENTAL: Pinebook // 11.6" 1080p // Manjaro KDE (ARM)

NAS:

Spoiler

Home NAS: Pentium G4400 @3.3 // Gigabyte GA-Z170-HD3 // 2x 4GB DDR4 2400 // Intel HD Graphics // Kingston A400 120GB SSD // 3x Seagate Barracuda 2TB 7200 HDDs in RAID-Z // Cooler Master Silent Pro M 1000w PSU // Antec Performance Plus 1080AMG // FreeNAS OS

 

Link to comment
Share on other sites

Link to post
Share on other sites

@Ryan_Vickers @Tic-Tac @Demiqas

 

Not sure why the game crashed last time I tried 8xMSAA, but it's working fine now (except for lower fps). Here's GTAV comparing 4x and 8x MSAA with TXAA off (because it doesn't let you use TXAA with 8xMSAA). 

 

4xMSAA:

 

 

8xMSAA:

 

 

Don't know about you, but I definitely see a difference in improvement. Can see throughout the image where it clearly looks better at 8x. 

 

I think this pretty much speaks for itself. ;)

 

There are minor differences that are barely noticeable, but they are there (i.e. where the windows meet the top of the doors, definitely more jaggys in 4x). To be fair, its only 1080p so at higher resolutions there might be no difference because of texture scaling.

R9 3900XT | Tomahawk B550 | Ventus OC RTX 3090 | Photon 1050W | 32GB DDR4 | TUF GT501 Case | Vizio 4K 50'' HDR

 

Link to comment
Share on other sites

Link to post
Share on other sites

@Ryan_Vickers @Tic-Tac @Demiqas

 

Not sure why the game crashed last time I tried 8xMSAA, but it's working fine now (except for lower fps). Here's GTAV comparing 4x and 8x MSAA with TXAA off (because it doesn't let you use TXAA with 8xMSAA). 

 

4xMSAA:

 

 

8xMSAA:

 

 

Don't know about you, but I definitely see a difference in improvement. Can see throughout the image where it clearly looks better at 8x. 

 

I think this pretty much speaks for itself. ;)

Oh look, exactly what I told you. You can barely see differences in pixels while zooming in and focussing on those spots. Now I wonder how much, if any, the difference is while actually playing the game.

Link to comment
Share on other sites

Link to post
Share on other sites

There are minor differences that are barely noticeable, but they are there (i.e. where the windows meet the top of the doors, definitely more jaggys in 4x). To be fair, its only 1080p so at higher resolutions there might be no difference because of texture scaling.

Indeed. It's not night and day, but it's there. Similar to how the images looked after @Ryan_Vickers Photoshop'd them. Totally agree, there are limitations at 1080p where once you reach a certain point, no amount of AA will make a difference.

 

Oh look, exactly what I told you. You can barely see differences in pixels while zooming in and focussing on those spots. Now I wonder how much, if any, the difference is while actually playing the game.

 

But that's not what you said. You said this:

 

There will be literally no noticable difference between 4xmsaa and 8xmsaa. I've tried this countless amount of times. If something like 8xmsaa would solve these issues I wouldn't have made this thread in the first place.

Yet there is some noticeable difference. It's not enough to make it worth the performance hit, but it's there.

 

Now, if we compare with Tic-Tac's earlier screenshot with the red car at only 2xMSAA, there is most definitely quite a difference between 2x and 8x (and even 4x, IMO). Tic-Tac was complaining there was too much aliasing asking, "is this normal?" Yes it is! When you're playing the game with only 2xMSAA, what do you expect? lol. Furthermore, I've just shown the image quality can be improved by turning up the AA, plain and simple.

 

But like I said a few posts earlier; the burden of proof is on you now, to prove the image quality was "propped up" when you first used your new GPU and somehow degraded over the course of a few weeks.

 

Please go back and read this post: http://linustechtips.com/main/topic/499767-low-image-quality-on-newer-gpus/page-34#entry6996495

 

    

 

 

 

My Systems:

Main - Work + Gaming:

Spoiler

Woodland Raven: Ryzen 2700X // AMD Wraith RGB // Asus Prime X570-P // G.Skill 2x 8GB 3600MHz DDR4 // Radeon RX Vega 56 // Crucial P1 NVMe 1TB M.2 SSD // Deepcool DQ650-M // chassis build in progress // Windows 10 // Thrustmaster TMX + G27 pedals & shifter

F@H Rig:

Spoiler

FX-8350 // Deepcool Neptwin // MSI 970 Gaming // AData 2x 4GB 1600 DDR3 // 2x Gigabyte RX-570 4G's // Samsung 840 120GB SSD // Cooler Master V650 // Windows 10

 

HTPC:

Spoiler

SNES PC (HTPC): i3-4150 @3.5 // Gigabyte GA-H87N-Wifi // G.Skill 2x 4GB DDR3 1600 // Asus Dual GTX 1050Ti 4GB OC // AData SP600 128GB SSD // Pico 160XT PSU // Custom SNES Enclosure // 55" LG LED 1080p TV  // Logitech wireless touchpad-keyboard // Windows 10 // Build Log

Laptops:

Spoiler

MY DAILY: Lenovo ThinkPad T410 // 14" 1440x900 // i5-540M 2.5GHz Dual-Core HT // Intel HD iGPU + Quadro NVS 3100M 512MB dGPU // 2x4GB DDR3L 1066 // Mushkin Triactor 480GB SSD // Windows 10

 

WIFE'S: Dell Latitude E5450 // 14" 1366x768 // i5-5300U 2.3GHz Dual-Core HT // Intel HD5500 // 2x4GB RAM DDR3L 1600 // 500GB 7200 HDD // Linux Mint 19.3 Cinnamon

 

EXPERIMENTAL: Pinebook // 11.6" 1080p // Manjaro KDE (ARM)

NAS:

Spoiler

Home NAS: Pentium G4400 @3.3 // Gigabyte GA-Z170-HD3 // 2x 4GB DDR4 2400 // Intel HD Graphics // Kingston A400 120GB SSD // 3x Seagate Barracuda 2TB 7200 HDDs in RAID-Z // Cooler Master Silent Pro M 1000w PSU // Antec Performance Plus 1080AMG // FreeNAS OS

 

Link to comment
Share on other sites

Link to post
Share on other sites

You must consider that I've used 4xMSAA while that "red car" screenshot was made and that's why I've written that doesn't seem normal to me...

Cosmic Council Department of Defense ; Interplanetary Class 3 Relations & Diplomatic Affairs - OFFICE 117

Link to comment
Share on other sites

Link to post
Share on other sites

You must consider that I've used 4xMSAA while that "red car" screenshot was made and that's why I've written that doesn't seem normal to me...

 

Wait, now you're changing your story? I asked you what settings you used when you took that screenshot and you linked me to where you posted your settings showing 2xMSAA.   

My Systems:

Main - Work + Gaming:

Spoiler

Woodland Raven: Ryzen 2700X // AMD Wraith RGB // Asus Prime X570-P // G.Skill 2x 8GB 3600MHz DDR4 // Radeon RX Vega 56 // Crucial P1 NVMe 1TB M.2 SSD // Deepcool DQ650-M // chassis build in progress // Windows 10 // Thrustmaster TMX + G27 pedals & shifter

F@H Rig:

Spoiler

FX-8350 // Deepcool Neptwin // MSI 970 Gaming // AData 2x 4GB 1600 DDR3 // 2x Gigabyte RX-570 4G's // Samsung 840 120GB SSD // Cooler Master V650 // Windows 10

 

HTPC:

Spoiler

SNES PC (HTPC): i3-4150 @3.5 // Gigabyte GA-H87N-Wifi // G.Skill 2x 4GB DDR3 1600 // Asus Dual GTX 1050Ti 4GB OC // AData SP600 128GB SSD // Pico 160XT PSU // Custom SNES Enclosure // 55" LG LED 1080p TV  // Logitech wireless touchpad-keyboard // Windows 10 // Build Log

Laptops:

Spoiler

MY DAILY: Lenovo ThinkPad T410 // 14" 1440x900 // i5-540M 2.5GHz Dual-Core HT // Intel HD iGPU + Quadro NVS 3100M 512MB dGPU // 2x4GB DDR3L 1066 // Mushkin Triactor 480GB SSD // Windows 10

 

WIFE'S: Dell Latitude E5450 // 14" 1366x768 // i5-5300U 2.3GHz Dual-Core HT // Intel HD5500 // 2x4GB RAM DDR3L 1600 // 500GB 7200 HDD // Linux Mint 19.3 Cinnamon

 

EXPERIMENTAL: Pinebook // 11.6" 1080p // Manjaro KDE (ARM)

NAS:

Spoiler

Home NAS: Pentium G4400 @3.3 // Gigabyte GA-Z170-HD3 // 2x 4GB DDR4 2400 // Intel HD Graphics // Kingston A400 120GB SSD // 3x Seagate Barracuda 2TB 7200 HDDs in RAID-Z // Cooler Master Silent Pro M 1000w PSU // Antec Performance Plus 1080AMG // FreeNAS OS

 

Link to comment
Share on other sites

Link to post
Share on other sites

Wait, now you're changing your story? I asked you what settings you used when you took that screenshot and you linked me to where you posted your settings showing 2xMSAA.   

 

 

Those are my current settings : 

 

Except now I run 4xMSAA and HiRes Shadows OFF . Maybe some settings in NCP could help? Besides editing Settings.xml and changing <MSAAQuality value="0" />

 

Sett.pngSett1.pngSett2.png

Cosmic Council Department of Defense ; Interplanetary Class 3 Relations & Diplomatic Affairs - OFFICE 117

Link to comment
Share on other sites

Link to post
Share on other sites

look at the windows and big doors on the balcony.(played on amd card)
Link to comment
Share on other sites

Link to post
Share on other sites

I have nvidia physx downloaded (dont know what it is) but i have an amd card. Should I get rid of it to help with this problem?

Link to comment
Share on other sites

Link to post
Share on other sites

I have nvidia physx downloaded (dont know what it is) but i have an amd card. Should I get rid of it to help with this problem?

yes get rid of it, it will not install.  It will also not affect this problem though.

Solve your own audio issues  |  First Steps with RPi 3  |  Humidity & Condensation  |  Sleep & Hibernation  |  Overclocking RAM  |  Making Backups  |  Displays  |  4K / 8K / 16K / etc.  |  Do I need 80+ Platinum?

If you can read this you're using the wrong theme.  You can change it at the bottom.

Link to comment
Share on other sites

Link to post
Share on other sites

look at the windows and big doors on the balcony.(should get better quality after a few moments)

 

Thanks for posting up your footage. Try turning the "Sharpness" all the way down and see if that makes any difference. 

 

Otherwise, yeah, that is typical aliasing as a result of the level of detail vs the limitations of the resolution. The light is reflecting on the edges of the window sill, for example, which is supposed to show as a very thin white line, but the pixels are too large to properly represent this detail properly and realistically. It's typically more noticeable with thin lines that are slightly off vertical and horizontal angles.   

 

Unfortunately for this game there are no various levels of AA to choose from. It is just simply "AA on or off", so we don't know what type or how much it is applying (2x, 4x, 8x?). That being said, this is "normal" for this game at this resolution. LOD vs limitations of the resolution.

 

I have this game as well, so maybe I'll jump in and do some of my own testing in that scene to see if any improvement can be made by tweaking the sharpness or by using DSR, but I don't think much can be done about it. Just the way it is.

My Systems:

Main - Work + Gaming:

Spoiler

Woodland Raven: Ryzen 2700X // AMD Wraith RGB // Asus Prime X570-P // G.Skill 2x 8GB 3600MHz DDR4 // Radeon RX Vega 56 // Crucial P1 NVMe 1TB M.2 SSD // Deepcool DQ650-M // chassis build in progress // Windows 10 // Thrustmaster TMX + G27 pedals & shifter

F@H Rig:

Spoiler

FX-8350 // Deepcool Neptwin // MSI 970 Gaming // AData 2x 4GB 1600 DDR3 // 2x Gigabyte RX-570 4G's // Samsung 840 120GB SSD // Cooler Master V650 // Windows 10

 

HTPC:

Spoiler

SNES PC (HTPC): i3-4150 @3.5 // Gigabyte GA-H87N-Wifi // G.Skill 2x 4GB DDR3 1600 // Asus Dual GTX 1050Ti 4GB OC // AData SP600 128GB SSD // Pico 160XT PSU // Custom SNES Enclosure // 55" LG LED 1080p TV  // Logitech wireless touchpad-keyboard // Windows 10 // Build Log

Laptops:

Spoiler

MY DAILY: Lenovo ThinkPad T410 // 14" 1440x900 // i5-540M 2.5GHz Dual-Core HT // Intel HD iGPU + Quadro NVS 3100M 512MB dGPU // 2x4GB DDR3L 1066 // Mushkin Triactor 480GB SSD // Windows 10

 

WIFE'S: Dell Latitude E5450 // 14" 1366x768 // i5-5300U 2.3GHz Dual-Core HT // Intel HD5500 // 2x4GB RAM DDR3L 1600 // 500GB 7200 HDD // Linux Mint 19.3 Cinnamon

 

EXPERIMENTAL: Pinebook // 11.6" 1080p // Manjaro KDE (ARM)

NAS:

Spoiler

Home NAS: Pentium G4400 @3.3 // Gigabyte GA-Z170-HD3 // 2x 4GB DDR4 2400 // Intel HD Graphics // Kingston A400 120GB SSD // 3x Seagate Barracuda 2TB 7200 HDDs in RAID-Z // Cooler Master Silent Pro M 1000w PSU // Antec Performance Plus 1080AMG // FreeNAS OS

 

Link to comment
Share on other sites

Link to post
Share on other sites

Guest
This topic is now closed to further replies.


×