Graphic downgrade

[GuyNwah]

Agreed $250 is too much for PC game, just can't agree with claim overall visuals would look similar on both platforms.

Nobody who is not deliberately misrepresenting what developers said, is saying anything like that.

Nobody has said overall visuals would look "similar". The PC version has been seen to be clearly superior on PCs capable of rendering it in superior quality.

The use of common textures is not a downgrade of the PC version, nor does it cause the PC version to display inferior quality.

 

[erxv]

Just out of curiosity, because this part of the subject seems to never get enough attention.

What do you guys, from both sides of beliefs and arguments, think about screenshots?

I think they are not only more modern in terms of the build they are taken from, cause it's easier than doing videos, but also they look really really good.
Foliage for example, in recent screenshots, I'd say is not downgraded/of-worse-quality at all compared to VGX, SOD, and old screens. There's still the art direction change of course, which makes everything look different, but I dont see less detail.

They are heavily "prepared" and taken with care of course, but so were SOD and VGX as we know, VGX especially has such short scenes that they can be considered not much more than moving screenshots except for the combat part. I also think these might be the closes representation we have from Ultra footage, or something similar to it. However there are a few older ones, that look more like the recent gameplay videos, as if its another graphical setting altogether, and the grass difference is present there.

But point is, out of modern footage, its by far the best looking thing we have, so its rather interesting...

screenshots are pretty much the most useless way to look at a game, (unless you just want pretty pictures) along with prerendered CGI trailers...

screenshots are 95% of the time photoshopped to oblivion.... the only REAL representation of how a game will look is ACTUAL gameplay.....

 

[GuyNwah]

screenshots are pretty much the most useless way to look at a game, (unless you just want pretty pictures) along with prerendered CGI trailers...

screenshots are 95% of the time photoshopped to oblivion.... the only REAL representation of how a game will look is ACTUAL gameplay.....

Agreed. Screenshots are a great way to represent a game when the gameplay is a tour of an art gallery. They are misleading when taken out of a sequence with objects in motion, and they are doubly misleading when they are then photoshopped.

Gameplay footage as it's commonly distributed is not much better. You can't base any conclusions on footage that's been through lossy compression.

You can use screenshots to illustrate patent flaws, like the defective grass clipping. But you can't use a screenshot to judge the quality of a moving object.

 

[sidspyker]

Depends on the context, I would watch the gameplay to see... how the game plays and generally looks like.

HOWEVER, the most common video compression is in the 4:2:0 ratio of luma and chroma, that right there means 50% of the colour data is discarded and not visible in the video. This works fine for live action footage but animated/rendered content suffers a lot due to it. So if I want an accurate idea of what the game really 'looks' like then screenshots are the only way.

Yes bullshots exist but you can usually tell what's a shopped picture and what isn't, so far Witcher hasn't had any bullshots that I can think of. Ubersampled yes but not shopped.

 

[SofaJockey]

.... the only REAL representation of how a game will look is ACTUAL gameplay.....

For me, this very insightful statement would have happily substituted for 137 pages of fairly pointless argument.
I'm looking forward to May when I expect the game will be CDPR's best blend of graphics and performance, because why wouldn't it be?

 

[GuyNwah]

4:2:0 is actually devastating to any kind of animated video. (To be strict, it's "subsampling", not "compression", but its purpose is the same: reduce the bandwidth.)

50 to 75 percent of the chroma content is lost to subsampling (50% at 4:2:2, 75% at 4:2:0)

Here's an example of the impact, using stills because the impact is easier to see that way.

 

[sidspyker]

Yeah I'm aware it's chroma subsampling, I just didn't want to make the post jargon dense for people unfamiliar with the terms

I have 'fought' against it too much because I kept getting the result that made me upset, eventually realized it's not a battle I can win so I started doing 4:4:4 encodes for myself and 4:2:0 for other purposes

 

[GuyNwah]

Yeah I'm aware it's chroma subsampling, I just didn't want to make the post jargon dense for people unfamiliar with the terms

I have 'fought' against it too much because I kept getting the result that made me upset, eventually realized it's not a battle I can win so I started doing 4:4:4 encodes for myself and 4:2:0 for other purposes

You'd never catch me doing jargon subsampling Seriously, live-action content doesn't suffer much at 4:2:0. So it's used for 4K/60Hz video in HDMI 2.0. But animated content (and desktop displays) tolerate 4:2:2 but get hammered at 4:2:0, like somebody smeared Vaseline on the screen.

 

[dasega]

Nobody who is not deliberately misrepresenting what developers said, is saying anything like that.

Nobody has said overall visuals would look "similar". The PC version has been seen to be clearly superior on PCs capable of rendering it in superior quality.

The use of common textures is not a downgrade of the PC version, nor does it cause the PC version to display inferior quality.

If thats the case how to explain difference in textures quality in posted here screenshots comperations?

 

[TheDeathRun]

If thats the case how to explain difference in textures quality in posted here screenshots comperations?

Most of that difference was attributed to a sharpening filter pages and pages ago. The comparisons used were pretty good, showing quite a difference when the picture was sharpened and when it was not (though which was "better-looking" was subjective and depended on the viewer). Furthermore, a developer (Damien Monnier) confirmed on his twitter page (in response to a tweet) that the difference between 2013 footage and late 2014 footage was a sharpening filter. While this by no means absolutely proves CDPR only got rid of the sharpening filter, it is currently the most likely explanation. Does that answer your question?

 

[sidspyker]

Sharpening is the difference when it comes to textures. I and @C0bR talked about this and came to that conclusion many months ago. Try playing Witcher 2 with sharpening off, it's quite the blurry mess, I just can't bear it. It's not as severe a case with Witcher 3 so far for me.

This really applies to almost all games. I did the same with AC3 2-3 years ago, used lumasharpen and voila the textures looked quite alot sharper(as they should, obviously)

Spoiler: Example 'comparision' shot

 

[dasega]

Most of that difference was attributed to a sharpening filter pages and pages ago. The comparisons used were pretty good, showing quite a difference when the picture was sharpened and when it was not (though which was "better-looking" was subjective and depended on the viewer). Furthermore, a developer (Damien Monnier) confirmed on his twitter page (in response to a tweet) that the difference between 2013 footage and late 2014 footage was a sharpening filter. While this by no means absolutely proves CDPR only got rid of the sharpening filter, it is currently the most likely explanation. Does that answer your question?

No it doesn't. I can't paste images as I am using my mobile but if you go back to page 4 of this thread you will see comparation of the flying creature Geralt is fighting with, look closer at the wings , you are looking at 512 vs 1-2K textures and its nothing to do with sharpness filter.

 

[sb87]

So yeah maybe it has been downgraded, still looks better than most games out today. The Witcher is no Crysis, you buy it for the story and gameplay not for pretty graphics. The Witcher is like a woman you grow to like, you like her for who she is, she doesn't have to be gorgeous, but if she is, it's just a really nice bonus.

 

[Zabanzo]

comparation of the flying creature Geralt is fighting with, look closer at the wings , you are looking at 512 vs 1-2K textures

You are talking about Geralt fighting Sirens in E3 2013 trailer right? The creature from january gameplay footage is Echidna, a different kind of siren, so its possible that echidnas just have a different model, I could be wrong of course, but its a possibility.

 

[sidspyker]

Yeah those were Sirens and these are Echidna, might be different... even if they weren't monsters have different subspecies. If you didn't notice the Griffins we've been shown is a slightly different model a few times because of that.

 

[shinstyle]

obsolete

 

[sidspyker]

That is high end, you have nothing to worry about.

 

[TheDeathRun]

No it doesn't. I can't paste images as I am using my mobile but if you go back to page 4 of this thread you will see comparation of the flying creature Geralt is fighting with, look closer at the wings , you are looking at 512 vs 1-2K textures and its nothing to do with sharpness filter.

That would be an intriguing comparison, except that they appear to be two completely different creatures. Open the pictures in new tab or something to view in full size.

The 2013 footage features the following creature:

View attachment 11341

January 2015 gameplay featured the following creature:

View attachment 11340

Notice the giant red arrows. While the creatures appear to be similar, 2015 creature possesses ridiculously nice squiggles on its back. 2013 creature doesn't. This can mean one of two things:

1) They are different creatures, and so cannot be used as comparison because the 2015 creature might have had those wings since it was first created.
2) 2013 creature's design was changed to the 2015 one. Which can be good or bad, depending on your preference. If this is the case then Mr. Squiggles tickles my fancy more than Mr. Scales, because... I like those squiggles.

Either way, the point is you can't really compare these two creatures until we have more info.

 

[C0bR]

Not sure if this article has already been mentioned (probably) but here it goes anyway:

Badowski said "it will be possible to spot differences in, for example, Nvidia HairWorks, but a strong PC will be needed." "Platform unification exists," but it doesn't necessarily equate to dumbing down PC graphics or performance. According to Badowski, "The Witcher 3: Wild Hunt" graphics engine is designed to look spectacular on any platform. It's true that gameplay will be limited to 30 fps at launch, but that's a PC specific improvement of top priority.

taken from http://www.designntrend.com/articles/43816/20150313/witcher-3-wild-hunt-cd-projekt-red-day-one-patch-gameplay-pc-graphics-comparison.htm

What a way to completly twist what Badowski said (as if that original interview wasn't already catastrophic). The last sentence is just completly made up by whoever wrote this pile of garbage.

 

[shinstyle]

obsolete