Support NeoGAF

[Slayer-33]

the problem with gt5's sound is that while it sounds good in the cockpit view, it sounds absolutely rubbish in the replays; it's like a bunch of vacuum cleaners racing around the track.

lol, that's why they need to up the sound quality and use an audio system like Forza, shit will get more exciting.

You should be able to use your home theater to adjust the sound in your living room to your liking. Everyone has different setup and different ears. The game should try to mimic th "real" or raw sound.

I have noting against that. Shifts notes are too exaggerated and can be exciting but I think FM 4 has pretty realistic engine sound.

It makes you wonder if the FM team recorded a modded MX-5. The note details make it seem like it.

 

[patsu]

Not all people have good quality home theater sound. The responsibility is on the devs side.

Yes, and the devs can choose how best to represent their sound in a faithful way. Even cheap HT gears can boost their base. No amount of dev adjustment can make bad HT sound good though, plus like I said everyone has different ears and preferences.

 

[offshore]

You should be able to use your home theater to adjust the sound in your living room to your liking. Everyone has different setup and different ears. The game should try to mimic the "real" or raw sound.

But if the source sound in no good then there's nothing you can do. You can't boost frequencies that aren't there. To me, it always sounds like GT5 is missing a boatload of high frequencies, as they're the amongst the most fatiguing to listen to. Everything sounds dull, and I never get bored or fatigued of listening to GT5's sounds. And that's not a good thing.

Racing drivers wear ear-plugs for a reason.

 

[phosphor112]

But if the source sound in no good then there's nothing you can do. You can't boost frequencies that aren't there. To me, it always sounds like GT5 is missing a boatload of high frequencies, as they're the amongst the most fatiguing to listen to. Everything sounds dull, and I never get bored or fatigued of listening to GT5's sounds. And that's not a good thing.

Racing drivers wear ear-plugs for a reason.

Racing drivers also drive cars that have most of their seats removed along with most of the insides of a car leaving just basically the chassis and panels, so there is basically nothing to muffle the sounds. This just reminded me, one thing that GT5 does need to fix, is that cars with race mods, you know, like the ones described above, sound muffled just like standard production cars. So yeah, I stand corrected, GT does have some major things they can fix.

 

[eso76]

this 'forza cartoony lol' needs to stop.
It may not be up there with GT5 BEST moments, but it's very close at times.
Dusk and dawn do look too 'dramatic' and colourful. GT5 is just perfectly subtle with its orange and blue tints, but other than that (and the very colourful alps track) fm4 can look damn near photorealistic.
At noon, on tracks like indy, tsukuba, suzuka etc. they actually look very much alike.

 

[NullPointer]

Nothing out there sounds as amazing as Forza 4. Its blistering, powerful, and one of the best improvements over Forza 3. So yeah, in terms of audio, Forza 4 wins by a mile.

But visuals? I'd hope the gifs from the past few pages could leave this to rest. Forza 4 is much improved over Forza 3, but at least for me it still doesn't hold a candle to what GT5 is capable of rendering.

Been playing the shit out of Forza 4 and while everything is improved across the board, the AI is, well, its broken. Never thought I'd prefer to play with GT5 or even Shift 2 AI, but there you go.

Regardless, these two games need to have a baby.

 

[Majanew]

To me, GT5's lighting is more life-like and the skyboxes are better. FM4's audio and performance is better. FM4's environment detail overall looks better, but some tracks in FM4 don't look as good or realistic as others. The lighting on some of FM4's tracks look pretty good, though.

 

[patsu]

But if the source sound in no good then there's nothing you can do. You can't boost frequencies that aren't there. To me, it always sounds like GT5 is missing a boatload of high frequencies, as they're the amongst the most fatiguing to listen to. Everything sounds dull, and I never get bored or fatigued of listening to GT5's sounds. And that's not a good thing.

Racing drivers wear ear-plugs for a reason.

That means PD have more work to do in the sound department, but trying to skew the sound to suit a segment of users who don't have good home theater may be misguided. They should make it an option but not force everyone to use that setting.

 

[JWong]

Anyone know what Digital Foundry use to messure frames per seconds with?

It would be cool if there was someone who could count every frame. It's a lot harder than pixel counting. XD

 

[Angst]

It would be cool if there was someone who could count every frame. It's a lot harder than pixel counting. XD

Almost everyone here on GAF can do that without even having to concentrate on it.

 

[Snubbers]

Carrying on from the video posted yesterday..

The Digital Foundry 'Tech analysis' of FM4 is up, and it has some handy comparisons to GT5..

http://www.eurogamer.net/articles/digitalfoundry-tech-analysis-forza-motorsport-4

The end comparison on Page 3:

video:http://www.eurogamer.net/tv_video.php?playlist_id=109895&size=hd

The comparison is intriguing on a number of levels: it demonstrates that the aesthetic look of the two games is very, very different - especially in terms of the lighting, and that while Forza's performance level is wholly consistent at 60FPS, GT5 can drop frames and tear, even at 720p (with further frame-rate drops if running in 1080p or 3D modes). Also remarkable is how completely different the audio scheme is between the two games. On these courses at least, there's a pretty strong argument that Forza has considerably more track detail, but you can make up your own mind with this Forza 4/GT5 comparison gallery.

The overall impression we get from playing both games back-to-back is that each has its own strengths: in terms of the handling and audio Turn 10 could well have moved ahead of its competition while Polyphony's minute attention to detail (to the point of implementing working headlights/dynamic shadows in the cockpit view) plus the sheer range of different driving scenarios remains second to none. Both titles excel in their own ways - especially in terms of visuals - and we have the feeling that motorsport fans probably own both consoles and are equally invested in both games. We aim to take a look at how these games play with a number of different steering wheel peripherals in a future Digital Foundry hardware article.

A very balanced article, perhaps a bit too balanced at times, it's hard to equate games that have quite diverse strengths, but at least it's from a credible source.

 

[Yoshichan]

Regardless of the track detail or how vibrant the colours are, GT5 is the closest you'll get to photo realism on current consoles, which is what a racing sim should strive for.

What's going on here?

 

[SmokyDave]

I really like the 'boosted' sound in Forza and NFS: Shift. Sure, it's not realistic but then how realistic can you get without an actual exhaust pipe in your living room? It's exciting, and that's better than realistic, to me. I feel similarly about the colours, I prefer GT's muted style but I have no problem with the colourful 'pop' of Forza.

Surprisingly, inaccurate car models really piss me off.

 

[nib95]

Carrying on from the video posted yesterday..

The Digital Foundry 'Tech analysis' of FM4 is up, and it has some handy comparisons to GT5..

http://www.eurogamer.net/articles/digitalfoundry-tech-analysis-forza-motorsport-4

The end comparison on Page 3:

video:http://www.eurogamer.net/tv_video.php?playlist_id=109895&size=hd

A very balanced article, perhaps a bit too balanced at times, it's hard to equate games that have quite diverse strengths, but at least it's from a credible source.

A very interesting comparison. Also proves to me at least that Spec 2 definitely has a better or more consistent frame rate than on initial launch. pretty much always 60fps now except in close up jams or collisions, even then the drops are momentary.

Also, whilst GT5 looks more realistic overall, with a more muted colour palette and more accurate grading/lighting, I do think F4 takes it on track textures (most of the time). They seem to be a bit more detailed than GT5, though GT5 takes it on vehicle detailing and accuracy.

One thing I do disagree with DF on is the comment on handling. Imo handling and driving physics are still notably more realistic in GT5, namely because of the more advanced traction, suspension and road feedback.

 

[Angst]

]The overall impression we get from playing both games back-to-back is that each has its own strengths: in terms of the handling and audio Turn 10 could well have moved ahead of its competition while Polyphony's minute attention to detail (to the point of implementing working headlights/dynamic shadows in the cockpit view) plus the sheer range of different driving scenarios remains second to none. Both titles excel in their own ways - especially in terms of visuals - and we have the feeling that motorsport fans probably own both consoles and are equally invested in both games. We aim to take a look at how these games play with a number of different steering wheel peripherals in a future Digital Foundry hardware article.

I presume DF has never looked at this thread, youtube comments and comments on their own site? lol

 

[hirokazu]

I presume DF has never looked at this thread, youtube comments and comments on their own site? lol

Those aren't motorsport fans, those are console fanboys.

 

[Wazzim]

I presume DF has never looked at this thread, youtube comments and comments on their own site? lol

Many people in this thread and on GAF have both consoles, youtube comments etc might be another story of course.

 

[Angst]

Just read throught the DF article. Why do everyone that wants to compare F4 and GT5 use Laguna Seca instead of Le Sarthe or The Ring? :/

 

[SunhiLegend]

Just read throught the DF article. Why do everyone that wants to compare F4 and GT5 use Laguna Seca instead of Le Sarthe or The Ring? :/

Forza 4's ring doesn't hold a candle to the one in GT5, especially considering it has a day night cycle and weather

 

[Angst]

Forza 4's ring doesn't hold a candle to the one in GT5, especially considering it has a day night cycle and weather

Hence my question... Laguna Seca looks pretty bland and gets used for all comparisons.

 

[cutmeamango]

Hence my question... Laguna Seca looks pretty bland and gets used for all comparisons.

They are both slightly improved ports from the old games, so it makes a tiny bit of sense (even if one comes from ps2).

 

[Truespeed]

The DF article also pointed out some of the interesting optimizations that were made to Forza 4 to achieve that locked 60 FPS target.

Turn 10 also appear to be employing a load balancing mechanism which is a touch more subtle: while gameplay operates with a locked 60Hz update, rear view mirrors and reflections on the car are decoupled from the main renderer and refresh at arbitrary frame-rates: dropping to 30 frames per second if things are really being pushed.

From a rendering perspective, the game still operates at native 720p, but the locked 2x multi-sample anti-aliasing of the previous Forza titles has been altered to allow for an improved 4x MSAA implementation which we think is tied into the game mode selected: time-trial gives better edge-smoothing, while the more processing intensive race modes seem to be using the same 2x solution.

There are some other changes too, specifically in terms of environmental detail: while all of tracks from Forza 3 (bar two) have made their way across to the new game, the team have taken the opportunity to "remix" the circuits slightly. Detail appears to have been removed in some places, and added in others. You can't help but wonder whether this had something to do with the optimisation effort.

 

[RukusProvider]

Thank God atleast one racing game out there takes 60fps and no tearing seriously!

 

[Truespeed]

Thank God atleast one racing game out there takes 60fps and no tearing seriously!

If 60 FPS gives you these quality of shadows then I'm happy with 50-60 FPS and some tearing.

http://www.eurogamer.net/videos/forza-4-vs-gran-turismo-5?size=hd

This video also shows another "optimization" in regards to shadows. Look at how dynamic and undulating the GT5 car shadows are as they turn corners and react to the light being cast and compare it so the almost static shadows of Forza 4 and its fixed lighting source. I don't think I want to drive in this 60 FPS compromised world.

 

[ShapeGSX]

If 60 FPS gives you these quality of shadows then I'm happy with 50-60 FPS and some tearing.

http://www.eurogamer.net/videos/forza-4-vs-gran-turismo-5?size=hd

This video also shows another "optimization" in regards to shadows. Look at how dynamic and undulating the GT5 car shadows are as they turn corners and react to the light being cast and compare it so the almost static shadows of Forza 4 and its fixed lighting source. I don't think I want to drive in this 60 FPS compromised world.

The GT5 self-shadowing shadows on the cars are so low res that they end up flickering. Look at the license plate on the Ford GT flickering at the start of the video. Similar things end up happening on car doors where there is an air channel that casts a shadow. The tree shadows are very low resolution in GT5, as well, and they aren't smoothed at all.

With the next gen, I hope that both games can cast shadows, but at a resolution that isn't so very distracting.

 

[Apex]

This video also shows another "optimization" in regards to shadows. Look at how dynamic and undulating the GT5 car shadows are as they turn corners and react to the light being cast and compare it so the almost static shadows of Forza 4 and its fixed lighting source. I don't think I want to drive in this 60 FPS compromised world.

Yes, FM4 do a lot of shortcuts in order to achieve 60fps with more than 1 car on track.

-No real-time reflections in other cars except the player.

-No shelf shadows in other cars.

GT5 shadows even cast over other cars

-Less detailed higher LODs than GT5.
-LOD swaps at very short distances.
-Simplified interior lighting in all cars.
-Less AA.
-Etc.

 

[TUROK]

Yes, FM4 do a lot of shortcuts in order to achieve 60fps with more than 1 car on track.

-No real-time reflections in other cars except the player.

http://i.minus.com/iScLEMlGsiIMh.jpg

Uh, what the hell are you talking about?

 

[Metalmurphy]

http://i.minus.com/iScLEMlGsiIMh.jpg

Uh, what the hell are you talking about?

I think he meant that other cars don't reflect on each other, only the players car.

*edit* or maybe not.

 

[Ploid 3.0]

Wouldn't it be better to sacrifice the shading on the player's car instead of the other cars since most likely the player isn't looking at his car? Well unless it's a replay or 3rd person view. Not counting interior self shadowing.

 

[nib95]

For that extra realism in visuals, better lighting, car models, shadows, effects etc, a minor momentary drop in frame rate during collisions or extreme close ups is definitely something I'm fine with. You definitely notice a lot more detail on the cars up close in GT5 than in F4, and I think that adds a lot more to the game than the momentary 2-10fps drop (unnoticeable in gameplay) that you get as a result.

Doesn't seem to drop just by having lots of cars on screen either. It seems more to do with extreme close ups (not always mind) or collisions (assuming the physics/Ai reacts to such bumps and hits accordingly hence the frame rate takes the bump).

Point is, the frame rate seems to have come a long way since GT5's initial release. Average frame rate based on the comparison video's must be at near enough 58-59fps since the drops are few and far between and very short lived. Vast majority of the time it's 60fps.

 

[Apex]

http://i.minus.com/iScLEMlGsiIMh.jpg

Uh, what the hell are you talking about?

Check the other cars in external view and compare their reflections with yours at tunnels entrances, passing under bridges or near the rumble strips. They use some sort of predefined environment map and don't reflect the details like the player car.

Even the standard cars in GT5 reflect correctly the environment.

 

[2real4tv]

Check the other cars in external view and compare their reflections with yours at tunnels entrances, passing under bridges or near the rumble strips. They use some sort of predefined environment map and don't reflect the details like the player car.

Even the standard cars in GT5 reflect correctly the environment.

Not sure if this is true, I thought GT5 did not reflect other cars only the environment I thought Forza uses a technique acquire from the Project Gotham guys.

I do agree with the rest though. Also why did DF compare pics from TT in FM(one car and uses 4xaa) to regular gameplay in GT5(up to 16 cars and 4xaa)?

 

[Apex]

Not sure if this is true, I thought GT5 did not reflect other cars only the environment I thought Forza uses a technique acquire from the Project Gotham guys.

I do agree with the rest though. Also why did DF compare pics from TT in FM(one car and uses 4xaa) to regular gameplay in GT5(up to 16 cars and 4xaa)?

I'm not saying other cars, not GT5 and not FM4 does that. FM4 use a trick to map the game screen as a texture in the hood view like does PGR4 but that does not work for other views or replay.

What I mean is reflecting track elements, like fences, rumble strips, trees, bridges, etc... over the car.

For example in this picture you can't see the fences or trees reflected in the FM4 car side, just a generic plain horizon with little or no variation. Compare with GT5.

 

[Pepto]

For example in this picture you can't see the fences or trees reflected in the FM4 car side, just a generic plain horizon with little or no variation. Compare with GT5.

What about that other pic (http://i.minus.com/iScLEMlGsiIMh.jpg)?

 

[Graphics Horse]

Check the other cars in external view and compare their reflections with yours at tunnels entrances, passing under bridges or near the rumble strips. They use some sort of predefined environment map and don't reflect the details like the player car.

I think you are right.

What about that other pic (http://i.minus.com/iScLEMlGsiIMh.jpg)?

Same deal I think.. different track, different environment map.

 

[paskowitz]

After seeing Forza again, it looks great in certain instances... and just good in others. Much like GT5. I think if the lighting in FM4 was more subdued (less bloom) it would be more photo realistic.

As far as frame rate is concerned, there is no question it drops in GT5. If the alternative to that is the game looking slightly worse, its not worth it though.

One thing I noticed at Spa is that PD greatly improved the way you perceive trees. They are still 2D but they just seem more realistic.

 

[test_account]

It shouldnt be hard to check if the reflection in the AI cars are real time or not. Just drive i.e under a bridge or something and see if the bridge is being reflected. I tried to look for this in the Eurogamer video, but it is not a good enough angle for me to see it clearly.

 

[Snubbers]

Not that it matters, but bonnet view in FM should perfectly reflect the entire rendered scene as its a cheap trick carried over from FM3.

I think the last page pretty much sums up the two different sides... its almost a graphics over framerate dichotomy.

I think people should just accept that T10 put framerate first, and spread graphic detail beyond the car, PD put graphical flourishes ahead of frame rate and may simplify trackside detail and squeeze in night racing (based on T10s admission that night/weather would require track modelling simpllification)

The same goes for audio/physics and gameplay, the differences in priority and depth of each is quite different in both games, why are people getting a little silly by trying to argue their favourite game is perfect in every department? The evidence against both having opposing weaknesses and strengths is overwhelming.. IMO..

 

[RukusProvider]

I think the last page pretty much sums up the two different sides... its almost a graphics over framerate dichotomy.

I think people should just accept that T10 put framerate first, and spread graphic detail beyond the car, PD put graphical flourishes ahead of frame rate and may simplify trackside detail and squeeze in night racing (based on T10s admission that night/weather would require track modelling simpllification)

This is exactly right but this isn't the thread for common sense.

 

[patsu]

It should not be just about graphics though. The handling and physics are also very important.

 

[Apex]

It shouldnt be hard to check if the reflection in the AI cars are real time or not. Just drive i.e under a bridge or something and see if the bridge is being reflected. I tried to look for this in the Eurogamer video, but it is not a good enough angle for me to see it clearly.

Here it can be seen clearly.

 

[The Elite]

That Forza 4 sign is clearly being reflected.

 

[Apex]

That Forza 4 sign is clearly being reflected.

Check the other cars in external view and compare their reflections with yours at tunnels entrances, passing under bridges or near the rumble strips. They use some sort of predefined environment map and don't reflect the details like the player car.

.

 

[The Stealth Fox]

Bomba motorboats 4

 

[DJ88]

Yeah you can see the rumble strips and bridge reflected on the player's car, but none of that on that blue VW.

 

[TUROK]

It seems like the AI cars use a much simpler cube map in comparison to the player's car. It's still a real-time reflection, just a much simpler one.

 

[ShapeGSX]

Until we get real time ray tracing, every game is using tricks of some degree for reflections.

 

[Apex]

It seems like the AI cars use a much simpler cube map in comparison to the player's car. It's still a real-time reflection, just a much simpler one.

To me that seems a simpler environment texture not a simplified real-time cube map.

Check the other cars reflections inside a tunnel. If you drive close to them you can still see the backdrop texture used in the exteriors, mountains, sky, etc... clearly visible in Fujimi. Also the reflections seem fixed and don't seem to travel with the car when you place the camera near their car side and follow them. Mountains don't reflect on their roofs, etc... Clearly they are not reflecting anything in real time.

If T10 use that effects in the player car and not in the rest it has nothing to do with reflection tricks but degrading the graphic quality to keep a fixed famerate.

 

[NullPointer]

Love Forza but the AI does literally drive me up the wall (and off the track). It got so bad that I just hired a driver for my last race and watched all the stupid crashes that ensued. Its pretty damn bad.

Think I'll go back to GT5 once Spa it out in the US and play Forza for rivals stuff. Shift 2 still seems to be my sweet spot in terms of racing feel and AI fun, even if they are aggressive in that game as well. Funny enough they are not nearly as aggressive and blind as Forza 4. GT5 AI is just plain boring to race against.

So yeah, I guess in a Forza 4 vs. GT5 vs. Shift 2 my own playstyle comes down on GT5 for driving feel, Forza 4 for social/autolog stuff, and Shift 2 for the best racing feel and track selection.

 

[Shaneus]

If T10 use that effects in the player car and not in the rest it has nothing to do with reflection tricks but degrading the graphic quality to keep a fixed famerate.

Maybe PD should've done the same in GT5.