• Hey Guest. Check out your NeoGAF Wrapped 2025 results here!

Digital Foundry :- Does resolution really matter?

Status
Not open for further replies.
When I owned a 1366x768 resolution TV, I ran VGA from 360 into it and chose that exact resolution on the 360 menu. Native resolution to my display is important to me.

Yea, but you couldn't do this with the PS3, so in the case of those ports you'd be referencing, the PS3 version would be doing what he describes.
 
When you're supposedly a journalist from a reputable tech site and you make ridiculous comments like "Next gen gameplay!" and now, "Resolution doesn't matter", something would be wrong if he's not infamous.
The point, if anyone actually read the article, isn't that resolution doesn't matter now but that it matters less than it used to thanks to increased use of post-processing.

Of course we can't actually discuss the article or the points it raises because too many would rather slam the source for "his" bias than dispute even a single point.
 
I want to reiterate what I said on page 1...



I'm operating under an assumption that the majority of HDTVs used for game consoles are not native 720p. The majority now are probably 1080p and have been for a while now but, prior to that, they were likely 1366x768 (or similar). 720p displays are a rare breed limited almost exclusively to earlier DLP rear projection sets. For all intents and purposes, they don't really exist and I'd argue that less than 1% of HDTVs in use throughout the world are actually native 1280x720.

So, in the majority of cases, people will be viewing games with double scaling. I feel that scaling resolutions like 576p and 600p to 1080p look dramatically worse than scaling 720p to 1080p.

The numbers of pixels present in the signal versus the number in the display is what is important here.

There's obviously no way to account for every situation but I do believe, in general, the difference between 900p and 1080p is less significant than sub-720p resolution games compared to those rendering at 720p due, mostly, to the types of displays people are using.

Now, if you were to perform this same comparison using a native 1280x720 32" display versus a 50" 1080p display things would be different. A 600p game on that 720p display would likely appear rather similar in a lot of ways to a 900p game on a 1080p display if those sizes were used and the viewing distances matched up. That's not a realistic scenario, though.

I also think times have simply changed which is why one can discuss that resolution is less important NOW than it was in the past. This is actually a key point that people are missing when they say "BUT WHY WAS RESOLUTION A BIG DEAL IN THE PAST BUT NOT NOW BIASSSSS?!?" - anti-aliasing techniques have improved dramatically. The type of AA used in last generation games was often rather poor and post-process AA didn't really even make an appearance until halfway through. Combining modern AA techniques with higher number of pixels does reduce the impact of running at a non-native resolution.

Another thing to consider - those PS3 ports that were often bitched about didn't JUST run at lower resolutions they often also ran at lower frame-rates and featured other compromises. Since I dislike buying games on Microsoft platforms for the most part I was pretty much forced to abandon buying console games (outside of exclusives) during the second half of last gen as PS3 iterations just couldn't cut it. There's a point where my console library just stops and switches to the PC entirely as a result. I still think exclusive games did some amazing stuff on PS3, though, but ports were often pretty bad.

Now, for my own tastes, I actually don't really MIND lower resolutions provided I can choose the scaling method. With the XRGB I always tend to use the 2:1 pixel mode for 480i/480p content with nearest neighbor filtering. This produces a centered 1280x960 image with sharp pixels. 240p looks fantastic as well when scaled properly. When scaled IMPROPERLY, however, with lots of filtering 240p and 480p look absolutely atrocious on modern displays.


I hope not too many people feel that way. I've been trying my goddamned best to put out good stuff. I'm no master of this by any means but I want to present interesting findings for people and I do it without letting bias enter into the equation. I dislike what Microsoft has wrought on the industry but I don't let that personal distaste influence the results.

I'm assuming that your comparisons were on equipment that was set up correctly therefore the double scaling thing would have been irrelevant for DF articles?

Can you explain the point of the articles? Are you trying to get devs to have parity in resolution and use the additional ps4 power for additional effects, use 720p, or what? They aren't clear on what the point is and the reader is expected to read between the lines.
 
I'm assuming that your comparisons were on equipment that was set up correctly therefore the double scaling thing would have been irrelevant for DF articles.

Why would it be irrelevant?

The articles aren't there just for number purposes. If nobody at home is likely to have a 720p native screen, what good would making the recommendations based on that scenario be?
 
The point, if anyone actually read the article, isn't that resolution doesn't matter now but that it matters less than it used to thanks to increased use of post-processing.

Of course we can't actually discuss the article or the points it raises because too many would rather slam the source for "his" bias than dispute even a single point.

The framing of the article matters. He came across as baffled that people would choose PS4 for resolution and then immediately posed questions about Nielson's methods because he was skeptical about the results. This was in the first paragraph. It becomes difficult to read an article with complete objectivity when the author started it out in such a way.
 
I'm assuming that your comparisons were on equipment that was set up correctly therefore the double scaling thing would have been irrelevant for DF articles.
I wasn't present for those pieces, but, like I said, the reason it's less critical now is due to more overall pixels combined with MUCH improved AA and additional post fx. The type of stuff you see in The Order and Ryse. That stuff is amazing but couldn't have been achieved last gen. If it had, those resolution differences would have been less significant.

It's still irrelevant due to lack of actual 720p screens.
 
So I guess if you don't agree with the article or the reasoning for it you are just a blowhard that has bad reading comprehension. You might even get discredited with a selective quote. Glad this thread played out like every other DF thread.
 
The point, if anyone actually read the article, isn't that resolution doesn't matter now but that it matters less than it used to thanks to increased use of post-processing.

Of course we can't actually discuss the article or the points it raises because too many would rather slam the source for "his" bias than dispute even a single point.

Resolution is a basically catch-all phrase that encompasses the overall visual quality of a game for more uninformed consumers. They know PS4 games look better and resolution is the bullet point for that. The guy who wrote this article knows this too. He is still bias and doing damage control.
 
I don't think people are calling YOU out for being biased. Unfortunately, Leadbetter is (has) ruining the site's reputation for some, and it's easier to blast DF than look at each writer separately. For what it's worth, I don't feel any general ill will towards you or the site as a whole.

The only thing recently that's bugged me (aside from the entire catalog of Leadbetter articles) is the sudden, selective focus on gameplay or secret sauce (infrastructure/online community) in what used to be technical write-ups. If this is going to be an evolution of the way DF writes their articles, shouldn't it happen to every game? The games it's been used on thus far put a bit of doubt in my mind as to the site's neutrality, but I'm willing to wait and see if it's fixed.
 
What the fuxk is going on over at DF? Since the PS4 has been dropping games left and right at full HD folks have gone nutty. So where was this article last gen? Give me a fucking break.
That's weird, it used to matter.....
ix5hNAArIgXiW.png
Exactly. Why the back pedaling now?
 
Why would it be irrelevant?

The articles aren't there just for number purposes. If nobody at home is likely to have a 720p native screen, what good would making the recommendations based on that scenario be?

Double scaling wouldn't occur on correctly set up equipment.
 
Double scaling wouldn't occur on correctly set up equipment.

It'd happen on any sub-720p game on PS3 running on a non-720p screen (pretty much every screen in existence). The PS3 doesn't do what the 360/XB1/PS4 does by scaling internally to match the display's native resolution.

It's not simply a case of people having their shit set up incorrectly.
 
Let's set aside xbox vs ps4 for the moment.

Which would provide a better experience
- 1080p/60 with no AA
- 900p/60 with 2xMSAA
- 720p/60 with 4xMSAA

Before seeing the likes of Ryse, I'd have said native resolution is the best, but that game has shown me that high image quality at 900p, combined with a decent scaler can clearly provide better image quality than some native 1080p games. So personally I think at least some level of AA is more important than resolution for the sake of resolution. (Obviously higher is better if you can have both)

Framerate I'm ok with 30fps as long as it is locked. Solid 30fps > unlocked 60fps with lots of drops.
 
I don't think people are calling YOU out for being biased. Unfortunately, Leadbetter is (has) ruining the site's reputation for some
Except he has been called out on it. Several times for several different articles where he was accused of doing the same thing that Leadbetter gets accused of - downplaying the XB1 version's disadvantages or not hyping up the PS4's advantages, however you want to phrase it.
 
Let's set aside xbox vs ps4 for the moment.

Which would provide a better experience
- 1080p/60 with no AA
- 900p/60 with 2xMSAA
- 720p/60 with 4xMSAA

Before seeing the likes of Ryse, I'd have said native resolution is the best, but that game has shown me that high image quality at 900p, combined with a decent scaler can clearly provide better image quality than some native 1080p games. So personally I think at least some level of AA is more important than resolution for the sake of resolution. (Obviously higher is better if you can have both)

Framerate I'm ok with 30fps as long as it is locked. Solid 30fps > unlocked 60fps with lots of drops.

I'd go with 900p with 2xMSAA.

Based on Forza Motorsport 5 and Dead or Alive 5: Last Round... 1080p with no AA is pretty nasty.

Except he has been called out on it. Several times for several different articles where he was accused of doing the same thing that Leadbetter gets accused of - downplaying the XB1 version's disadvantages or not hyping up the PS4's advantages, however you want to phrase it.

Yup, that Dragon Age Inquisition thread was ridiculous.
 
No it wouldn't but what does that matter when nobody will actually be playing on actual 1280x720 displays?

I'm kind of disturbed by the number of people who don't realize this. 720p content will not be 1:1 on nearly any consumer HDTV released in the better part of the last decade.
 
Let's set aside xbox vs ps4 for the moment.

Which would provide a better experience
- 1080p/60 with no AA
- 900p/60 with 2xMSAA
- 720p/60 with 4xMSAA

Before seeing the likes of Ryse, I'd have said native resolution is the best, but that game has shown me that high image quality at 900p, combined with a decent scaler can clearly provide better image quality than some native 1080p games. So personally I think at least some level of AA is more important than resolution for the sake of resolution. (Obviously higher is better if you can have both)

Framerate I'm ok with 30fps as long as it is locked. Solid 30fps > unlocked 60fps with lots of drops.

I'm fine with unlocked framerate as long as it is above 30. Killzone and inFAMOUS played great unlocked.
 
Let's set aside xbox vs ps4 for the moment.

Which would provide a better experience
- 1080p/60 with no AA
- 900p/60 with 2xMSAA
- 720p/60 with 4xMSAA

Before seeing the likes of Ryse, I'd have said native resolution is the best, but that game has shown me that high image quality at 900p, combined with a decent scaler can clearly provide better image quality than some native 1080p games. So personally I think at least some level of AA is more important than resolution for the sake of resolution. (Obviously higher is better if you can have both)

Framerate I'm ok with 30fps as long as it is locked. Solid 30fps > unlocked 60fps with lots of drops.
For the living room, I'd go with 900p and 2xMSAA.

For my desk, I'd probably go with 1080p and no AA. It wouldn't be pretty, but most PC/monitor setups don't handle scaling as well as console/TV setups. I'd probably want to see them side by side for comparison, though. Depends on the game, too.

Completely opposite on the performance, though. I'll take 45-60fps over 30fps most of the time.
 
Pc monitors and cards don't have scalers tho. It's a completely different argument. Yet we see it so much

The scalers on PC monitors generally aren't as good because there's an assumption that they'll be used for native resolution content, but the majority do have scalers. If they had no scalers, you wouldn't be able to display non-native-res material at all.
 
I'm kind of disturbed by the number of people who don't realize this. 720p content will not be 1:1 on nearly any consumer HDTV released in the better part of the last decade.


This is the whole point tho. People don't really know anything, but they hear something and stick too it. Form an opinion from that. The 360 had to scale to the guys 1366 monitor through a vga cable doing exactly what the xbox one is doing, I believe. Yet he believes it was native. Surely the game was still rendered at the res it was developed at.

It should be does the game look good.

I know cod aw is lower res on xbox one but it still looks good.
 
Pc monitors and cards don't have scalers tho. It's a completely different argument. Yet we see it so much

The vast majority of HD TV's do have built in scalers though don't they? That's my assumption anyway. Not very good ones usually but they certainly help.
 
Except he has been called out on it. Several times for several different articles where he was accused of doing the same thing that Leadbetter gets accused of - downplaying the XB1 version's disadvantages or not hyping up the PS4's advantages, however you want to phrase it.
Yeah, that particular thread was absolutely ridiculous and shows how warped people's views can be.

Pc monitors and cards don't have scalers tho. It's a completely different argument. Yet we see it so much
Of course they have scalers.

I must say, though, I use 1:1 pixel mode most often on my 1440p 32" monitor as the size is large enough to allow for smaller windows to remain playable. 1920x1080 at 1:1 is the size of a 24" 1080p monitor while even something like 1024x768 is still bigger than a 19" 4:3 display. It's quite usable that way.
 
Let's set aside xbox vs ps4 for the moment.

Which would provide a better experience
- 1080p/60 with no AA
- 900p/60 with 2xMSAA
- 720p/60 with 4xMSAA

Before seeing the likes of Ryse, I'd have said native resolution is the best, but that game has shown me that high image quality at 900p, combined with a decent scaler can clearly provide better image quality than some native 1080p games. So personally I think at least some level of AA is more important than resolution for the sake of resolution. (Obviously higher is better if you can have both)

Framerate I'm ok with 30fps as long as it is locked. Solid 30fps > unlocked 60fps with lots of drops.

It goes even deeper. What scaling algorithm do you use? What shaders/lighting are you using? AF? Even lighting models can cause certain visual artifacts to appear more or less. The question depends on a lot of factors.
 
Just checked, both of my two PC monitors do indeed have built in scaling. Probably not very good ones though, as non native resolutions still look pretty poor.
 
I disagree. Ryse looks very blurry just like every other 900p game. To me that looks much worse than a 1080p image without AA.


Does Ryse look blurry because it's 900p, or for other reasons? I've seen games that are 900p vs 1080p where overall image look just as sharp, texture quality wise of course, since the 900p image has a few more jagged edges, but it's not the Vaseline smear that we see in some of the Ryse captures. It could be their method of AA or some kind of diffuse lighting technique.

DF catpures:
jpg

jpg
 
I think if DF wrote individual articles for each platform with the same testing and results reporting we wouldn't see a lot of this hyperbolic and silly platform pandering. "My console does this better and I need to have it justified!" It doesn't matter to the individual consumer who takes in many more factors than just Resolution. The point of the article is that there is more to it than resolution, and there can be sacrifices that can make for a pleasurable experience.

However, they will not break down and not compare the different platforms, because the shit storm of the comparisons gives them the clicks they desire.
 
I'd go with 900p with 2xMSAA.

Based on Forza Motorsport 5 and Dead or Alive 5: Last Round... 1080p with no AA is pretty nasty.

What i find annoying is that in the same breath you have people like Ricahrd saying resolution doesn't matter for people who bought the PS4 for the power gap over the XB1, we have DF's own analysis of DOA5LR which has the results in direct comparison of 720p and 1080p both with no AA.

"In terms of the actual remastering, we're looking at a pleasant enough, if somewhat workman-like effort. As expected, the jump to native 1080p allows for smaller details and distant scenery to resolve with more clarity, with the game boasting a crisp appearance that complements the clinical anime art style the series is known for. There are minor visual improvements here and there, but by and large, the move to a fixed full HD resolution is the most obvious boost to overall image quality. However, a complete lack of anti-aliasing is unexpected and disappointing, leading to some shimmering around more detailed parts of the environment.


By comparison, the last-gen editions of Dead or Alive 5 Last Round use a dynamic framebuffer, aiming for native 720p but dropping into sub-HD territory when the engine is under load. While the upscale to 1080p looks OK - despite the lack of anti-aliasing - image quality takes a visible dive when the dynamic framebuffer kicks in, producing fuzzy detail that does no favours to some of the low resolution artwork.
"


 
Of course it matters. I've got a 65" TV.

As for mentioning the Gamespot 'research' blind testing if people could tell the difference between PC, Xbone and PS4?

He tricked everyone with Unity. However, the 'trick' aside, any test to see if people can see differences should not contain Unity in any case due to parity.
It was badly conceived nonsense.

Definitely the case. You can't ask people to see a difference when the attempted source material wasn't intended to have differences in the first place. It's self fullfilling.

@Seanspeed Just recently in Rev Evil 2 face off, Dark was called out for bias against PS4. So yeah, he's definitely been accused more than once. Same thing happened with the last face off Dark did.
 
Status
Not open for further replies.
Top Bottom