Frame Rate Analysis Thread (Simple Rules Post #2)

[grandmaster]

not for upscale test, just for understand his strange Vsync

Ah right. I've got PAL retail Dead Rising promo disc here which doesn't work on my NTSC 360 Elite, but I do have debug code so will have to find that before doing the test. It's around here somewhere but I've not touched it in years...

i don't remember if there is tearing but they are slowdowns

I'm glad you said that because when I did look at the game I don't recall seeing v-lock problems.

They are very interesting results. I have to say i'm suprised the 360 is sticking to 60fps so closely.

When you mention the chopper/boat/rain intro sequence, is that were you are actually inside the cockpit flying around the boat? And is there any camera movement during the clip? I.e. are you looking around the cockpit andd over the boat throughout or does the camera just stay at its starting position?

Yes it's the sequence you refer to. I think you can move your head about to change the view, but I didn't as I wanted the test to be as close as possible to the 1080p clip I took subsequently.

I may measure Virtua Fighter 5 later if I can find some free time. It's a curious one as it's rescaling from 1024x1024 downwards and upwards depending on the selected resolution, and of course with the exhibition matches I should be able to get frame accurate comparisons. Unfortunately the exhibition matches are different on PS3. If I was going to do comparisons we could only do an overall average or else measure the pre-match character intros which would be like for like.

I'm not sure if this is relevant or not, but the new Digital Foundry TrueHD hardware can capture at 60fps at any HDTV or VESA resolution up to 1920x1080 so it will handle any of the VGA/DVI resolutions that 360 supports.

 

[goonergaz]

There was no screen tear on PS3. It has the traditional v-sync in this game.

The whole point of these tests is to compare the gameplay experience that you get from of different versions of a game or maybe even different games, not the GPUs. These comparisons can't be used to judge the difference in hardware speed because v-sync will always give slower results; moreover, the effect isn't consistent. Depending on the exact sequence of rendering times, it can either mask or accentuate the speed differences between platforms.

Sure, but you still don't have any reason to say that the variance is +/- 5%. The GTA4 tests were cut-scenes with only minor differences, the GRID tests have little variation, and for other in-game tests, a reasonable length (say 1-2 minutes) is enough to even out these variances for the most part.

Good points about the rain, as at least one member is saying there is some correlation between weather and tearing. The test 4 parity is interesting, but it should be noted that tests 4 and 5 (which also shows 360 with a smaller lead) are the shortest clips tested at ~15s.

thanks for your comments, BTW when I was talking about PS3 screen tear I meant maybe with vlock off it was so bad they had to enable it. Also on my +/- 5% I must have been having a bad day because I could have sworn there were bigger differences in some of the PS3 results. FTR I know a couple of guys who have mentioned bad screen tear in rain on X360 - also one guy stating a worse f/r on X360 when driving fast down a road.

anyway, thanks again

 

[grandmaster]

The driving fast down a road effect is there on both consoles and worse on PS3. Because you're travelling quickly the low frame rate combined with its uneven nature shows big jumps in road-side scenery.

Driving the truck at high speeds over the bridge during the "Rigged to Blow" mission produces these results:

Xbox 360: 28.546fps (6.8% screen tear)
PS3: 24.340fps

You're not the first to mention it which was why I did the test previously, but because it's not exactly like-for-like (but same scenery and very similar traffic levels), I didn't post it.

 

[pjbliverpool]

The driving fast down a road effect is there on both consoles and worse on PS3. Because you're travelling quickly the low frame rate combined with its uneven nature shows big jumps in road-side scenery.

Driving the truck at high speeds over the bridge during the "Rigged to Blow" mission produces these results:

Xbox 360: 28.546fps (6.8% screen tear)
PS3: 24.340fps

You're not the first to mention it which was why I did the test previously, but because it's not exactly like-for-like (but same scenery and very similar traffic levels), I didn't post it.

Cool. It actually feels a lot lower on screen IMO. I was expecting something in the teens on the PS3 version. Good to see its not that bad.

 

[Mr Deap]

Grandmaster is it possible to do some "physic crash test" in few selected multiplatform games with the framerate test. That would be interesting.

 

[grandmaster]

I'd need points where the video would be comparable; that would be quite tricky.

 

[RobertR1]

grandmaster,

Special request but would you please do GT5. Run the London 2 lap race S class and try to stay in the middle of a pack of cars. I'd love to see what that average frame rate and tearing frames is.

Thanks.

 

[Jesus2006]

The driving fast down a road effect is there on both consoles and worse on PS3. Because you're travelling quickly the low frame rate combined with its uneven nature shows big jumps in road-side scenery.

Driving the truck at high speeds over the bridge during the "Rigged to Blow" mission produces these results:

Xbox 360: 28.546fps (6.8% screen tear)
PS3: 24.340fps

You're not the first to mention it which was why I did the test previously, but because it's not exactly like-for-like (but same scenery and very similar traffic levels), I didn't post it.

Do we know if the circumstance/environments are the same (number of pedestrians, cars, effects) in that scene?

 

[grandmaster]

I agree it would be an interesting test.

Unfortunately, the Euro version of GT5 has no London circuit in the S class section... it only appears once during the whole event cycle. The London event has been replaced by the High Speed Ring. I did an arcade mode test but it's pretty useless in that the cars are evenly spaced. Best I could do was three cars on-screen once.

With three cars on-screen and close-up it dropped to about 57fps at 1080p but I only had the cars on-screen for around three seconds.

I can't go online because stupidly, PS3 TEST units will download updates but won't install them.

Do we know if the circumstance/environments are the same (number of pedestrians, cars, effects) in that scene?

It's on a bridge, there are no pedestrians. It's pointless to debate as the car positioning is different but curiously both clips feature 31 vehicles. I suspect that there is some element of scripting to some of the driving missions though. I've noticed patterns of traffic repeating cross-format in these circumstances.

Please no more GTA IV questions. It runs less smoothly on PS3 - time to face the truth! It doesn't make it a bad game!

 

[2real4tv]

What about Mass Effect?

 

[Squilliam]

That would be interesting, also Halo 3, RFOM and forza - for comparison to Grid and GT5P

 

[pjbliverpool]

I'd be quite interested in Oblivion actually. Thats one that was released on both systems and seamingly struggled.

 

[inpHilltr8r]

I can't go online because stupidly, PS3 TEST units will download updates but won't install them.

You need a special version of the firmware. You should talk to whoever gave you the TEST kit in the first place.

...also, reading this thread makes me feel slightly naked. You guys were never this clinical about back buffer resolutions, and exact fps before. What happened?

 

[grandmaster]

I have all the latest firmware updates, what I mean to say is that if a game requires an update - eg, GTA IV's recent update, the PS3 TEST will download the update but lacks the ability to decrypt the .pkg.

I'm working like crazy on a project at the moment, then off on assignment for two weeks on Sunday, so the thread will be on hiatus for a little while. Maybe I can get a couple of results in before I go though from archive footage.

 

[inpHilltr8r]

The game update itself? Weird. Worked on my devkit (for GTA and others), but then I haven't tried it on a TEST.

 

[manux]

This is somewhat offtopic but somewhat still related to capturing frames from games. How about actually analyzing the graphics output of consoles in different modes(rgb full, limited, y pB... and also dvd/blu-ray/games/viewing photos) and seeing if the consoles actually produce the full colour range.

There is for example some talk that ps3 full rgb would only scale the values from limited. Researching this should make for a nice article... Also outputting same picture file from ps3 and xbox360 could provide some light if there is something funky happening with the colour output´between the consoles.

 

[grandmaster]

The only problem with that is that the picture viewer on the Xbox 360 is completely awful. See here. Secondly, it is off-topic and probably better off in its own thread. If people want limited RGB vs full RGB shots, check out the RSX thread here.

 

[Fafalada]

...also, reading this thread makes me feel slightly naked. You guys were never this clinical about back buffer resolutions, and exact fps before.

Do remember this is B3D not your garden variety game forum.

What happened?

HD happened
Pixel counting was mostly impossible before it, and 99% of userbase used displays that 'postprocess' your image into something at least 2 times lower then real resolution (even CRT SDTVs that cost thousands of $ in Y2K had blurovision effect).
Resolution tricks last gen were perceptually equivalent on most displays(thus making resolution arguments mostly irrelevant) - this isn't the case for HDTVs this gen. In addition - this gen actually introduced resolution settings in the bloody machines themselves - which was unheard of in the world of consoles before. It's Sony and MS that put the spotlight on display-resolution variations, people are just responding with natural curiosity.

As for framerates - without uber-fancy capture equipment (which has not been available for too long either afaik, especially at accessible prices), it's completely impossible to get any quantification, so yea, I think it was a tech limitation as well.
Surely cross-platform fps comparisons were par for the course ever since console wars started - there just wasn't a way to quantify them before.

 

[grandmaster]

It is also a case of hardware developers making myriad claims about the capabilities of their hardware and game-makers attempting to pull the wool over our eyes in terms of the output of their software.

Microsoft made noises about a mandatory minimum of 720p, and yet their biggest-selling first party game is 640p. Sony heralded a 1080p revolution with PS3 and yet there's only two games (RR7 and VT3) that I enjoy playing at 1080p.

Software houses are continually pumping out screenshots that look nothing like the final game and - perhaps more pertinently for this thread - making claims about their cross-platform development skills that do not bear up to scrutiny.

 

[novcze]

Microsoft made noises about a mandatory minimum of 720p, and yet their biggest-selling first party game is 640p. Sony heralded a 1080p revolution with PS3 and yet there's only two games (RR7 and VT3) that I enjoy playing at 1080p.

1080p is also blu-ray resolution, so that nothing wrong with Sony PR talk.

Super Stardust HD and GT5 are fine examples of 1080p games (but not Full HD games )