I know at one point it was 720p and 900p, but don't know what they will release as.
EDGE: "Power struggle: the real differences between PS4 and Xbox One performance"
- Thread starter Musiol
- Start date
I know at one point it was 720p and 900p, but don't know what they will release as.
soundscream
Member
That quote from the second link isnt in the OXM article.
It probably has dynamic because that's what PS3 and 360 do.
Even if he had said it (which I doubt he did) I'm sure it wouldn't be in the OXM article xD.
What resolution is that picture?
1920x1080, it's a grab from the Direct feed Gamersyde stuff.
Pumpkincutter
Banned
bad news for everyone.. who want's 900p?
qa_engineer
Member
1920x1080p it appears to be a frame buffer grab, but someone else van confirm.
It's direct feed footage provided by Microsoft to Gamersyde. It is the best quality footage you can get.
EDIT: the image itself is a 2.2mb PNG.
Netherscourge
Banned
http://ps4daily.com/2013/10/battlefield-4-developer-talks-playstation-4/
Different DICE developer, same idea.
From 10/14/2013
Nothing about the XB1 resolution though, just the PS4.
FULL INTERVIEW
Different DICE developer, same idea.
From 10/14/2013
Nothing about the XB1 resolution though, just the PS4.
FULL INTERVIEW
Pimpbaa
Member
A jpeg made from an almost already compressed video feed is not exactly best quality you can get.
edit: png? Then the heavy compression artifacts are from the video feed.
viveks86
Member
I second this.
Again, if you read the original interview, there is no mention of 900p. It seems like these blogs are extrapolating, based on videos that were released earlier. Dice had refused to confirm the final resolution.
Even in compressed images you can still see the lack of AA. If the game does use some sort of post-processing AA, it's not very effective. I don't know why they'd go to all the effort of having lovely reflective car bodies when there are masses of jaggies everywhere. IQ trumps effects every time.
No, they can't. They don't exactly have as much leverage in the gaming industry like many people seem to believe.
And even if they do shell out money to publishers to lower the resolutions of PS4's versions of games, it'll be impossible for them to, since they would have to pay every single publisher a lot of money to do so. Not to mention that it'll make their investors/shareholders very furious.
Liabe Brave
Member
I have a couple for you, Skeff. The analyses are over in my pixel counting thread, but here's the results for a few more One games:
Lococycle appears to be native 1080p.
Powerstar Golf appears to be native 720p. But, it seems like it might be using the display planes functionality to render the UI in 1080p on top of that.
Not bad news for everyone, there's a thing called PC.
Low IQ: 360/PS3
Medium IQ: Xbone
High IQ: PS4
Very (very) High IQ: PC
We've gone over this already. There's no mention of the resolution.
Thank you very much, will update. Also will add your info to some of the other games, thaks for the post, out of curiousity, which part of Lococycle did you do the analysis on? I couldn't see clearly which section it was, personal curiosity mainly.
viveks86
Member
I find this a convincing argument. Also, isn't it plain old bribery? I can imagine one offs, but something illegal at such a large scale is near impossible to pull off. Every dev will know it if such a deal actually happens, since they would be the ones asked to implement parity. If information leaks, MS will get sued for millions. It is an extremely unlikely conspiracy theory IMHO.
If games do end up with parity, it would have to do with devs rushing the product and not having time to optimize for each platform. Create some baseline and meet that baseline. That can really happen only with launch games, if at all.
viveks86
Member
Thanks for sharing! Just curious, did you attempt pixel counting on the original Ryse videos? Do they show up as 1080p or 900p?
viveks86
Member
Are you sure about that? I remember websites quoting 720p and then later correcting the article as 'unconfirmed'.
Here is a quote from dual shockers:
We made a list of confirmed resolutions, but Albert sure as shit wouldn't want to look at it:
Forza: 1080p - http://uk.ign.com/articles/2013/09/17/forza-5-runs-at-native-1080p-ryse-does-not
CoDoG 1080p - http://www.eurogamer.net/articles/2...s-runs-at-1080p-and-60fps-on-xbox-one-and-ps4
Fifa 1080p - http://www.eurogamer.net/articles/2...ference-between-current-gen-and-next-gen-fifa
LocoCycle 1080p - Pixel count by Liabe Brave - http://www.neogaf.com/forum/showpost.php?p=86349940&postcount=35
PowerstarGolf - 720p with 1080p UI - Pixel count by Liabe Brave - http://www.neogaf.com/forum/showpost.php?p=86349940&postcount=35
Ryse: 900p - http://uk.ign.com/articles/2013/09/17/forza-5-runs-at-native-1080p-ryse-does-not
KI: 720p - http://gamingbolt.com/killer-instin...s-removed-or-optimized-praises-xbox-one-cloud
DR3: Dynamic bullshit - actually unable to find a solid link on this, many available but no direct dev comments could be found easily. - Albert could clarify if he wanted to, even if he didn't know it would be 1 phonecall away.
Driveclub: 1080p - http://www.eurogamer.net/articles/digitalfoundry-hands-on-with-playstation-4
KZ:SF: 1080p - http://www.eurogamer.net/articles/digitalfoundry-hands-on-with-playstation-4
Knack: 1080p - http://www.eurogamer.net/articles/digitalfoundry-hands-on-with-playstation-4
Infamous: 1080p - http://www.eurogamer.net/articles/digitalfoundry-hands-on-with-playstation-4
the Order: 1920x800 (artistic- 2.40 cinema aspect ratio, retains 1:1 pixel mapping) - Seriously, the resolution of this one is everywhere, find it yourself
ACIV: 1080p - constantly referred to as 1080p, eurogamer comment here - http://www.eurogamer.net/articles/digitalfoundry-tech-analysis-assassins-creed-4
CoD
Fifa 1080p - http://www.eurogamer.net/articles/2...ference-between-current-gen-and-next-gen-fifa
Resogun: 1080p - http://www.eurogamer.net/articles/digitalfoundry-vs-resogun
Flower: 1080p - https://twitter.com/TDMoss/status/385907407475314688 - talks about 60fps in tweet but in comments confirms 1080p
DC:UO 1080p - http://massively.joystiq.com/2013/06/13/e3-2013-dcuo-plans-to-remain-completely-free-on-ps4/
FF14 1080p - http://www.dualshockers.com/2013/09...-fps-1080p-equivalent-to-highest-pc-settings/
Blacklight Retribution 1080p - http://www.thesixthaxis.com/2013/09/06/hands-on-with-blacklight-retribution-for-playstation-4/
Hohokum 1080p - http://blog.us.playstation.com/2013/05/07/hohokum-coming-to-ps4-ps3-ps-vita-in-2014/
Strider 1080p - http://blog.us.playstation.com/2013/10/10/strider-hands-on-remember-your-ninja-training/
They are just about the only confirmed resolutions out there.
EDIT: If anyone provides a valid link with a developer quote or a pixel analysis I'll add the game to the list.
EDIT2:
There is a thread here: www.neogaf.com/forum/showthread.php?p=86349940 which not only does some pixel counting it also explains how it is done if anyone is interested, it also confirms the figures above for Killzone, Knack, Infamous, Resogun, Driveclub and Infamous.
Also of note: Ryse was 1080p native at E3.
updated, and viveks, yes the original Ryse footage was analyzed in that thread and it was native 1080p, How very interesting!
SneakyStephan
Banned
Maybe not with ps4 (400 euros, if you ignore the paywall, lol) hardware wise, but a 500 euro steambox will shit on xbone.
Right now a pc with xbone equivalent cpu, a hd 7790, 500gb hdd and the other bits is really cheap to build, but noone would build such a pc for games right now, the cpu is much too low end and the gpu isn't powerful enough.
fake edit; ye just checked and to build one with parts from my local pc shop costs 409 euros.
(so no cashback, bundle deals or other things you'd look for to save money, just a quick build)
add 60 for a controller and that's 469 euros.
But again, noone in their right mind would pick parts like these for gaming:
1: because they're shitty and too low end, especially the cpu
2 you can get way better performance/dollar by spendin a marginal amount extra
- for 30 euros more you can have a 50 percent faster gpu (7850, similar to what's in ps4) - for 10 euros more you can have a 1TB hard drive that's also faster
There should definitely be low end steamboxes that compete with xbone price
MS really fucked themselves by including kinect and seting the price so high.
If they left out the kinect and sold it at 300 it would 've been similar to the 360 at release and be able to compete with ps4 (and steambox range pcs once those are out)
You can find some quotes here:
http://m.neogaf.com/showthread.php?t=687839
I read the quotes from the twitter post but I think the guy got in trouble or something and had to take them down. But originally he had said PS4 unconfirmed but higher than 720p and XB1 confirmed 720p for final game.
viveks86
Member
That is so strange! Can the upscaler fool pixel counting? It can't be that perfect. Was Crytek openly lying when they said it was 900p all along? How come there was not much of a fuss on Gaf about it?
No the upscaler can't fool pixel counting. I'm not going to say they lied, but it was 1080p and is now 900p.
To Clarify, Pixel counting cannot show supersampling or rendering at a higher resolution than the picture it is checking, so technically, Ryse was AT LEAST 1080p
viveks86
Member
Oh! That's the same article that I posted my quote from! Check the link on the OP. They had updated the post since, which doesn't reflect in the OP. There's been a lot of confusion and stuff lost in translation during that time because it came out of TGS.
True. I remember this as well. But since it was taken down, we don't have an official confirmation anymore. Though 720p is likely, I wouldn't be comfortable seeing it on Skeff's list till they reconfirm.
It can show supersampling.
Project Alice
Banned
Not bad news for everyone, there's a thing called PC.
Low IQ: 360/PS3
Medium IQ: Xbone
High IQ: PS4
Very (very) High IQ: PC
Do we have any proof of the Medium IQ/High IQ for the next gen consoles? Or is this your interpretation of what you think will happen?
Liabe Brave
Member
I used one of the stabilizer fins on the enemy robot.
Also, as another example I counted One's Zoo Tycoon. It seems to be native 1080.
No problem! As I've said before, I am not an expert at this, it's just that pixel counting is pretty easy once you understand the concept. I encourage you to check out the thread and try it yourself.
Regarding Ryse, as Skeff said the Combat Vidoc released by Crytek (the least-compressed media we have) counted at native 1080p. Though I haven't posted them anywhere, I've also counted several different promotional shots released by Microsoft, and they're all native 1080p (or supersampled) as well. To the best of my knowledge, the only other direct feed we have of Ryse is from E3, and it was released at 720p so can't be effectively counted. I don't believe any direct feed of the game running at its native res has been released anywhere.
viveks86
Member
But that's not what Crytek admitted. They categorically said it was 900p all along. Personally, I don't really care about that number itself, but this smells fishy. :/
This was the footage that has me confused as well. Didn't it have a watermark that said actual in-game footage?
EDIT: Correction. It says "This does not represent final game quality". LOL. Smart! Who would have thought it would be lesser!
I thought there was a download of the Ryse e3 video on Xbox 360 that was 60 fps as well.
Edit: can't confirm but will get back if it was 60fps.
They definitely lied, I just didn't want to sound mean to them as we know they're crunching hard right now.
Liabe Brave
Member
Can you clarify? Examination of pixels can certainly indicate AA being used, but I'm not savvy enough to know how you'd differentiate MSAA from SSAA (or other techniques). I also imagine it'd be tough to reliably calculate the intermediate size of a supersampled image.
This isn't to say that there's no indication at all. For example, in that Ryse Combat Vidoc, not only is it rendered at (at least) 1080p, the AA is extremely good, with blending pixels galore on every edge. Since the footage appears to come from a high-powered PC, I've strongly suspected supersampling since I first counted it. Any info you can give on how to qualify/quantify AA methods would be much appreciated.
My cynical take on it is that the game was always 900p on Xbox One but the native 1080p footage was taken from an internal PC build.
viveks86
Member
My heart goes out to the devs too. It must be total chaos in there right now. But stuff like this should be called out for what is is. Ugh! I hate this when it happens.
LOL.
Judging from comments on Twitter a lot of teams are crunching like mad right now, not just Crytek. I remember Yosp saying to some of the GG guys that he hoped they would get to go home at some point over the weekend.
I know the feeling. A comment like that from Yosp would be a nice gesture while your working your ass off.
viveks86
Member
Pretty sure they are. Which is what makes me hope driveclub isn't actually delayed. There might have been internal murmurs out of frustration, which somehow got out to our beloved insiders. There may have been no actual decision made at all. Just my wishful theory.
Developers generally forget what their out of work friends look like during the last couple of weeks before certification.
For basic supersampling where the internal resolution is simply higher than the actual image, you count the same way - except the number of steps will be higher than the pixels in an area (as opposed to the reverse if it was lower resolution).
Example:
Can't read the link, require registration, but wouldn't that be impossible, if it is only 10 pixels high you cannot count more than 10 steps, It's impossible...
EDIT: that's 12. Firstly there are only 15 dot's but the steps are the intervals between the dots which means 14 intervals and 14 steps, also 2 of them are incorrect and count the AA, the blue dots are not correct steps, they just appear to be with AA. It is not clear what AA is being used, although likely it is SS, it is not a confirmation. or am i wrong? - I see what you mean, theres definitely a pattern, but I think it could be difficult in real life examples.
Rehosted it. They're not really steps, but its a pattern to look for. If it wasn't supersampled there would be a proper step where the blue crosses are.
Liabe Brave
Member
To be fair, the 1080p video released by Crytek says right on it that it's not representative of the game. The only other misleading media has been native 1080p screens, but promo shots have been inaccurate this way for lots of games, for decades. There's a thread on GAF currently about a "PS4 screenshot" of NBA 2k14 that's rendered at far higher res than PS4 will attain. These things are unfortunately common.
viveks86
Member
Yeah I realize that now. I'm reminded of how all the burgers look on the menus versus what they actually look like (I'm looking at you Mc Donalds!) We get misled all the time. Sigh...
Liabe Brave
Member
Okay, I get it--patterns in the AA density. Makes sense, but I don't know how applicable it'll be in real life. I tried to apply the idea to one of the Ryse captures, and I just couldn't do it. The patterning is much less stark than the example. I don't know if that makes supersampling unlikely, or it means the supersampling used a much higher differential, or neither.
That's probably not the AA, but the effect of the resampling algorithm. You can observe the pattern yourself if you just create a near horizontal line in paint and downscale it with another program.
Yeah its very hard if you dont have clear captures and Ryse isn't a naturally clean game anyway with all the post processing.