PDA

View Full Version : Detail vs. Framerate



Roadster
28th January 2008, 11:18 AM
First of all, let me just start off by saying that Wipeout Pulse is an extremely beautiful game (as is Pure), with graphics easily amongst the best on the PSP from both a technical and artistic perspective.

However, I have to question the wisdom of putting so much graphical detail into a game that's pretty much about racing at incredibly high speeds. Especially when it affects the frame-rate so much. Not so bad on time trials, zone etc. but very noticeable on 8-ship races, especially at the higher speed classes. It's certainly an improvement over Pure, but still not ideal.

Personally, I'd rather have less detail, and a smoother, more consistent frame-rate - a bit like what F-Zero X did back on the N64, but obviously not as extreme. What do others think?

lunar
28th January 2008, 11:52 AM
I agree with you, really. I think the framerate on Pulse is worse than Pure on some tracks when racing AI ships. I never really noticed framerate drops in Pure at all, except on multiplayer, and they never bothered me offline. In single player, Pulse framerate is quite disturbingly low, for me, at some moments and on some tracks.

Ash-Omen
28th January 2008, 12:34 PM
i think the framerates fine its a very detailed game and ive noticed the framerate does drop allittle on 8 player phantom races but not by a large amount its still perfectly playable, lol it wouldent make much difference anyway cos 8 player phantom is alittle too hard too know wat the hells goin on anyway :g

GTAce
28th January 2008, 01:20 PM
Hmmm i dont have a PSP so i cant say how much the gameplay suffers from the FPS drops but i think on a Handheld console like the PSP you can get over it.
See, Sony must also show what the PSP can do and i think Pulse does this very well.
If you want a really flawless and fluid experience you should stay on the home consoles and wait for WipEout HD (1080p @60 frames iirc).
You cant expect that from a Handheld imo.

Frances_Penfold
28th January 2008, 01:28 PM
Interesting post :)

I agree that framerate, in general, is almost always more important than graphical detail.

On the other hand, the game modes in Wipeout Pure and Pulse that I cherish most-- namely, time trials and speed lap and free play-- hold up really well in the frame-rate department. So on balance I'm willing to give up some frames-per-second in multiplayer to maintain the beautiful and stunning track detail that I appreciate in single player time trialing.

Wipeout seems to have a long tradition of creating cool tracks with interesting backgrounds-- it has always differed from F-zero in this regard. Egg has made some interesting posts over the last year or two about tricks Studio Liverpool has to maintain the framerate while keeping as much detail as possible. (Things like inclusion of tunnel sections, that keep the game from having to show so much graphical detail at some points, and clouds as well.)

As an aside-- do folks think that the less detailed and "cel-shaded" look of tracks in Zone mode something that was implemented to maintain the framerate even when the game is racing at insane speeds?

Dogg Thang
28th January 2008, 01:56 PM
Pulse isn't flawless by any means but the frame rate seems pretty good to me. I was certainly impressed when I first loaded it up and going back to Pure reveals just how much better it is. Yes, things get choppier when it gets very busy but Pulse is one of the smoother PSP games to me.

In fact the only one I can think of that is silky smooth on the PSP is Ridge Racer, which is lovely.

Yeah, if it were up to me, I'd put smoothness over detail but I think Pulse runs much better and that counts for a lot.

ZenDJiNN
28th January 2008, 04:27 PM
I'm with you there Frances..... i often use freeplay just to wander about the tracks, looking at the buildings & the scenery..... even imagining what that it's a place i could quite easily visit and maybe get a place to live. I'd LOVE to go for a walk around Moa or Platinum for instance, but as the closest i can get (at the moment) is drifting slowly around in a ship, i'll take that...... detail means a lot to me personally, and i think SL have "Nailed it" given the limits of a machine that's quite old now spec wise....

Still love Pure as well though, and often do the same there. In fact, i have a nice little condo to rent in Iridia if anyone's interested! :g

phl0w
28th January 2008, 08:33 PM
I'd choose framerate over detail any day of the week. I can't play Forza or PGR to the day because they stutter so much. Now there are people out there that think there's no noticeable difference between 30 or 60fps, and maybe to some people there isn't. I, on the other hand, can't stand it and get headaches from playing @30fps over a longer period of time. First thing I noticed was the bad framerate of Pulse especially in Eliminator and when a Ghost is loaded. I think of it like this: You can get used to slightly less detailed tracks and ships, but a dropping framerate is something that you can't get used to because it's so random, and effectively alters gameplay because physics, thus your ships handling is dependent of framerate. I think that from all WOs 3 has the best graphics. Despite its obvious lower polygon count and less sophisticated tech than Pure/ Pulse it looks the most realistic and runs smoothest too.

CR4SH3D
28th January 2008, 08:39 PM
thats a good point, 3 does almost look as good and runs flawlessly on less powerful hardware, however i thought the human eye can only pick up 25fps? which is why films are recorded at that isnt it?

Dogg Thang
28th January 2008, 09:02 PM
No, the human eye can definitely tell the difference at higher frame rates. You only have to compare something running at a steady 60fps to something at 30fps to see it.

lunar
28th January 2008, 09:05 PM
I think 24 fps is the minimum number of frames per second needed to create the illusion of movement. You can still sense a greater smoothness with a higher framerate though.

I once read on here that the main reason the original Wipeout was a racer with walled in tracks, in the first place, was to avoid problems with draw-in, framerate etc. God knows where that was posted though. I`m not one of these people whose eyes start bleeding the moment the action drops below 60fps, I spent almost a year playing the original Wipeout so I can put up with most things, but it did seem to me on Outpost 7 and Talons Junction that the framerate was less than 24fps at times, and this was worse than Pure ever got, but everyone seems more or less happy, so maybe it was just my brain that was malfunctioning.

lochiebrad
28th January 2008, 09:36 PM
if it's a steady 30fps and NEVER drops i'd prefer that to a 60fps that often drops to 50 or lower, because it's the change in frame rate that puts you off. Low and steady wins the race, but I'd say 24 is the minimum.

Medusa
28th January 2008, 10:06 PM
I'll take the detail over framerate in this game because the detail makes it look more realistic, which it seemed like Pure was missing (to me anyways). Lots of people will stick with the slower classes anyways and the game can handle the detail at those speeds just fine.

phl0w
28th January 2008, 10:22 PM
the game can handle the detail at those speeds just fine.
Yet a steady framerate is much more crucial on Phantom...


, so maybe it was just my brain that was malfunctioning.
No, it's the game (or the PSP, your choice who's to blame ;) ), your brain works just fine ^^
Outpost7, Talon's and Moa have got some serious problems in SR and online, be it heavy drops in framerate or ugly screen tearing on the latter track.

Ash-Omen
29th January 2008, 06:16 AM
I think the games a technical achievement most detailed game on psp and its got some really nice lighting effects, frame rate for me has never been a problem, maybe ur psp's r dieing lol. anyway if u compare this too pure u can really see the difference 333mhz can make.

bloseth
29th January 2008, 06:24 AM
Just play 'Ridge Racer' for a few sec and go back to WO. You'll see the difference for sure. I think RR runs at about 60fps.

Ash-Omen
29th January 2008, 06:59 AM
lol yeah but look at the difference in detail, thats like comparing Pulse too LocoRoco :lol

bloseth
29th January 2008, 07:09 AM
I know, just an example to show difference in framerate :robot

... and to clear any misunderstandings: I prefer WO over RR _any_ day...

phl0w
29th January 2008, 07:22 AM
the difference 333mhz can makeAgain, the game does NOT run @333Mhz, Colin Berry and Egg stated this quite a few times now. If yours, however, is running at a higher CPU clock due to CFW used I can understand why you wouldn't notice the framerate, because it's super smooth @333.
I won't jump to conclusions, neither is it my intention to accuse anyone, but: If you read about the game on the net in general or on other boards there's no talk about the framerate, although it's usually a major factor when talking about new racing games. I am under the impression that for many (present company excepted) the framerate problem is a non existant one, because they thought Pulse was a game utilizing the PSP's full power, thus switched their CPUs to 333. Is that possible or am I completely paranoid? :sonar

Dogg Thang
29th January 2008, 07:24 AM
But would it be a bad thing if WO ran as smooth as RR but had to sacrifice the detail? For me, no. RR, in spite of its lack of detail, has some lovely looking sections mostly due to clever use of the layout to maximise showing off the good bits and great lighting. And the frame rate is sweet.

phl0w
29th January 2008, 07:28 AM
Although I can see the point in taking RR's framerate as comparison, I think there's more behind it than just the higher detail. Let's not forget that RR features fairly simple game mechanic (which does not mean they're bad!), while I can imagine that WO's physics have a greater impact on CPU usage.

lunar
29th January 2008, 08:26 AM
To answer the original post, which I never did, I`d go for framerate. At rapier and above I`m too busy trying to keep it all together to appreciate the scenery anyway. For me the problem is only siginificant in single player races, rapier and above. I really struggled in places on the grid 11 rapier tour because of changing framerates, and at time it was like racing with online lag. There`s one point on Basilico White, just after the jump as you turn left towards the funky blue lights that the framerate always dips. It just messes with my timing and causes me to turn too soon on the right turn that follows. I think the point that lochiebrad made, that the framerate needs to be consistent whatever it happens to be, is correct. God, I sound like a right moaner, but I can live with it all because once the single player game is cleared it`ll be online and TT only, and that seems all plenty smooth enough for me.

Ash-Omen
29th January 2008, 08:31 AM
Again, the game does NOT run @333Mhz, Colin Berry and Egg stated this quite a few times now. If yours, however, is running at a higher CPU clock due to CFW used I can understand why you wouldn't notice the framerate, because it's super smooth @333.
I won't jump to conclusions, neither is it my intention to accuse anyone, but: If you read about the game on the net in general or on other boards there's no talk about the framerate, although it's usually a major factor when talking about new racing games. I am under the impression that for many (present company excepted) the framerate problem is a non existant one, because they thought Pulse was a game utilizing the PSP's full power, thus switched their CPUs to 333. Is that possible or am I completely paranoid? :sonar

its doesent use 333mhz? thats kinda stupid lol and no im not using CFW im using official 3.80, well its sure running well for 222mhz or is it using alittle more than 222mhz?

then again i understand why it dont use 333mhz cos the batterie life would be hit pretty hard

stin
29th January 2008, 08:32 AM
I didn`t even noticed!:o I was just too busy finding the lines and avoiding the walls!, oh well I will just stick to golf so I can see the scenery!8)

stevie:sonar

phl0w
29th January 2008, 11:00 AM
PSPs are not running on 222mhz, only the first generation without FW update did. It's @266 since FW 2.something, and 333 since 3.53(was it?), So Pulse should run @266 or 300 (PSP being capable of 20,75,100,200,222,266,300,333)

Asayyeah
29th January 2008, 11:09 AM
Maybe Colin could answer us to that question : At what exact speed of CPU does Pulse runs : for me it's between 300 to 333 mghz in SR & roughly less in multiplayer.

Roadster
29th January 2008, 12:12 PM
Hmm, this has started an interesting discussion. Funny that people mention Ridge Racer, as that game looks beautiful and runs at a rock-steady 60fps. It has nowhere near the amount of effects and detail that Pulse has, but it still looks stunning on the handheld, and it was one of the launch titles (which lead to very high expectations of future PSP titles, for me anyway).

Also, Pulse does not run in 333mhz. I know this because I'm using CFW, and switching to a higher clock-speed makes a massive difference. Single player races are much more bearable, and Time Trial, Speed Lap and Zone modes run at 60fps. Interesting because I always assumed Pulse would be running at 333mhz (before I joined this forum and found out the truth), which the made the frame-rate even more disappointing when I first got my copy.

Someone mentioned that it's the inconsistency of the frame rate, rather than the frame rate itself, that's the problem - and I definitely agree. I'd rather that the whole game was locked down to 30fps, so that there wasn't such a noticeable difference between, say, Time Trial and Eliminator modes. Even at 333mhz the disparity is obvious, so I wonder what it must be like for users on original firmware.

phl0w
29th January 2008, 12:53 PM
so I wonder what it must be like for users on original firmware.
Well, since I know the difference too, I can tell you it's horrible. It's really hard to go back to default speeds with Pulse once you experienced the full glory of 333mhz. Why, you ask? FYI records achieved at other speeds than the game's default are not legitimate on the WOZ tables or in Challenges. It's a different thing on the official rankings though, because there's no such "code of honour" and God knows how many ranked times were actually clocked @333.

Frances_Penfold
29th January 2008, 02:31 PM
Do we have an estimate for fps in Time Trial / Speed Lap / Zone modes? Minus some screen tearing (why hello there, Moa Therma) the framerate seems pretty smooth to me in these modes.

I guess what I am getting at is this-- do we think that CFW users have an advantage in Wipeout Pulse for Time Trial modes? I sure hope not. My experience so far is that, for me, the framerate isn't really a problem for these modes.


if it's a steady 30fps and NEVER drops i'd prefer that to a 60fps that often drops to 50 or lower, because it's the change in frame rate that puts you off. Low and steady wins the race, but I'd say 24 is the minimum.

I agree with this. I'm not sure what exact fps I can see or cannot see, but what I REALLY notice is the drop in framerate.

NeXaR_QroN
29th January 2008, 02:48 PM
In default clock, whatever it is for pulse, frame rates are not a "problem" for TT/SL, BUT that doesn't mean you'll not notice the diference with 333mhz. It exists, but of course, less than in single race, let alone eliminator mode.

phl0w
29th January 2008, 06:29 PM
:| This is a very dicey case, what with all the talk about CFW. Maybe we can have infoxicated make a statement whether or not it's OK to have a discussion on how the game plays on different clockrates. Afterall it got tested for Pure too (it was Assayyeah who reported back then, IIRC), to see if there's an advantage.

infoxicated
29th January 2008, 07:05 PM
If the research done by Arnaud on Pulse is anything to go by then the game runs at a different speed on custom firmware than it does on official firmware. If you're racing online or in multiplayer with custom firmware you have an advantage, whether you're forcing up the clock speed or not.

NeXaR_QroN
29th January 2008, 07:13 PM
Can I have a link to some details on Arnaud's research? (if it's posted on the forum). From mine, at defaults I only noticed faster loading times, and only if booting from MS instead of UMD. But then again, I've never been too much sensitive at framerate changes (not only in wipeout), unless the changes are big.

Asayyeah
30th January 2008, 11:39 PM
That's right i did tests and it's not finished at all.
Now it's gonna be more precise cause i am cumulating datas.
Rendez-vous end of february for the results.

Without authorization of Rob i won't publish them live but PM to those who are asking maybe.

NeXaR_QroN
31st January 2008, 01:01 PM
Thanks. Then I'll be waiting for your results.

Vincent_VII
31st January 2008, 05:16 PM
I would say that anyone running Pulse at 333mhz is at a definite advantage.

Even if the game ran at a normal speed at 333mhz, the smoother frame rate and more responsive controls in hairy situations would give you a huge advantage.

It would be like someone trying to play Quake on a slower computer with a choppy frame rate on a network against someone with a much faster computer and higher frame rate. Both players will experience the same passing of time in the game world but the player with a faster system will be able to aim and judge movement more precisely.

It's just a shame that Pulse doesn't run at 333mhz natively.

Asayyeah
1st February 2008, 03:23 PM
Simple question :
If 'defaulted' mode allows a CFW to be like a original firmware PSP, the 2 psp should be completely identical when running a game, right?

NeXaR_QroN
1st February 2008, 03:31 PM
Mmm... In theory, yes. Only things I can think could make a difference are these

1. "IF" the system prx's on CFW are not encrypted (which I don't know), the PSP would not have to decrypt them, so it may speed up some tasks. Again, I don't know.

2. IF the CFW PSP is using plugins, that would take RAM and CPU cycles, probably doing the opposite effect to point 1.

There are probably more minor variables to take into account, but that two should be the main two. And anyways, I don't know how much could that affect the speed.

And now, a simple question from me:

How are you measuring framerates on official firmware?

Vincent_VII
1st February 2008, 03:35 PM
Simple question :
If 'defaulted' mode allows a CFW to be like a original firmware PSP, the 2 psp should be completely identical when running a game, right?

No. CFW does not always run at the same speed as OFW. CFW has been hacked and altered so speed cannot be relied on. Sometimes clocking the system to 333mhz is required so that CFW PSPs will play games at a near normal speed. Add: The team responsible for CFW has often said they must remove/add things to the FW when rebuilding it so that it will fit on the PSP's flash. That alone could drastically alter the speed at which the PSP itself runs.

I have played Pulse on a CFW PSP and then an OFW PSP. Even at default, Pulse hiccups badly on the CFW PSP when compared to the OFW one. When the CFW PSP was clocked at 333mhz, the game world seemed to run at the same speed as the OFW one. I didn't have a chance to run the game clocks against each other, however. The 3, 2, 1 count seemed the same.


And now, a simple question from me:

How are you measuring framerates on official firmware?

Maybe he has access to a development kit. ;)

NeXaR_QroN
1st February 2008, 03:37 PM
Then that CFW PSP definitely HAS a problem. From my experience, "OFW" and "defaulted CFW" run at same speed (or so I can see), and 333Mhz CFW runs WAY smoother. Unless that's caused by heavy CPU activity on the background (like that MP3 player in the background, don't remember the name)

Vincent_VII
1st February 2008, 03:47 PM
Possibly. Not every PSP is the same. Several hardware revisions have been made since it's release. But that is a "whole other can o' worms".

Asayyeah
1st February 2008, 03:58 PM
No dev kit, here : i can't measure fps

Thx for answering my questions NeXaR_QroN & Vincent_VII

Flashback Jack
1st February 2008, 04:07 PM
Even at default, Pulse hiccups badly on the CFW PSP when compared to the OFW one.

Not necessarily the firmware's fault. Convert your .CSO to .ISO and test again. Texture decompression requires time and CPU cycles.

- F

omega329
1st February 2008, 04:31 PM
hang on a minute, I thought CFW wasn't meant to be discussed on these boards:naughty. I think i can here someone coming...

NeXaR_QroN
1st February 2008, 04:42 PM
Well, it's being discussed here to actually try and settle wether it's the same or not in regards to actual playing, to determine if it should be considered "cheating" or not to use it to play pulse, regardless of clock speed. I think that's important.

But yes, we should be doing this via PM, MSN, IRC... If anyone's interested, I'm on iRC, and my msn tag is in the "PSN IDs and MSN Tags" thread.

Vincent_VII
1st February 2008, 05:14 PM
CSO vs ISO - this shouldn't matter. Only when loading the level. Once the data is loaded into the PSP memory it will run the same regardless. If the data is steaming from the source (ISO, CSO or UMD), that is a totally different story.

Hmm... I would love to discuss but I am at work.

NeXaR_QroN
1st February 2008, 06:45 PM
At least it would be streaming the music, if nothing else.

feisar rocket
28th January 2009, 03:41 PM
you no all yal shold be banned yo no not to talk about cfw all yal will be in troble

cnmth
28th January 2009, 04:12 PM
sighface..

whats with all those ****ty reply's ? :naughty

Kscorps
28th January 2009, 08:00 PM
No, the human eye can definitely tell the difference at higher frame rates. You only have to compare something running at a steady 60fps to something at 30fps to see it.

For example, after I was playing HAlo 3 religiously, I then got COD4 and though something was wrong with my Xbox lol!

stin
28th January 2009, 08:09 PM
Funnily enough!, as I was watching my son who plays Halo3 on my 42" TV`s screen, but I`d thought, it was a HD till COD4 came along which 100 times better than Halo3! go figure!.

And yes, my son who have got XBox360 and I have PS3 ;)

stevie:)

Seraph
5th February 2009, 09:53 PM
Don't really know about those tecnical things but if it DOES matter and someone can cheat through these methods I'd say it sucks... Back on the subject, I do prefer fps over detail but as many have said in pulse I think few are those who are going to play an 8 player match in phantom so it doesn't matter much about fps. I'm with fps all the way but when trying to avoid the track walls so as to race a perfect lap I don't have the time to look at the scenery. Also I think that is why in zone mode the whole track goes white or green or whatever with just a bit of shade so as to distinguish the track details, in order to have as high a framerate as possible. Otherwise your psp would like flare up in zone 60+ :P