GT5 Sound Thread

  • Thread starter Marry_Me_GT
  • 5,638 comments
  • 597,631 views
I can clearly hear the difference in the video above, sorry. There's plenty of rumble unheard in GT5 even in "good" videos.

I said taking into account that we are comparing a real car vs a TV speaker... a bit like a live band vs a TV show of one...
so yeah obviously one will have more low frequency, but the character at the top end sounds much alike to me
 
I can clearly hear the difference in the video above, sorry. There's plenty of rumble unheard in GT5 even in "good" videos.

This reminds me; I never did manage to find that "tuba" video you mentioned ages ago now. Any chance you could re-post it?

I'm assuming that this "flutter" effect you mentioned back then is also missing in GT5's SLS? I'm keen to better interpret the sounds I'm hearing around me, and whilst I can obviously tell the difference between the off-screen SLS in the BestBuy demo and the sound of "the real thing", I cannot really put it into words - largely because I can't explicitly recognise the differences, just that they are different. :dopey:

I think there's more to it than just the recording environment. Although the GT5 demo's SLS does sound "like" the real thing, there are some interesting differences that I'd like to pin down - hopefully with a better recording. :)
 
This reminds me; I never did manage to find that "tuba" video you mentioned ages ago now. Any chance you could re-post it?



It may sound like I'm kidding, but this is the sort of rumble I'm talking about (or at least a major part of it) that can be felt in real life on sports cars with bassy exhaust tones.

I'm assuming that this "flutter" effect you mentioned back then is also missing in GT5's SLS? I'm keen to better interpret the sounds I'm hearing around me, and whilst I can obviously tell the difference between the off-screen SLS in the BestBuy demo and the sound of "the real thing", I cannot really put it into words - largely because I can't explicitly recognise the differences, just that they are different. :dopey:
I guess yes. It's really noticeable in the first 15 seconds of the real SLS video anyway, there's no real need to hear any more. I'm always under the impression that GT5 engine sounds are too "smooth", lacking of "grrrrrrowl" or "purrrrrrrr". The general tone however is there, and that why after all they sound similar to the real cars.

I think there's more to it than just the recording environment. Although the GT5 demo's SLS does sound "like" the real thing, there are some interesting differences that I'd like to pin down - hopefully with a better recording. :)
Yes, better recordings are needed, but I think we'll unlikely have them until the release date of the full game I fear.
 
I think I know what you mean. The problem with using samples is that they are recorded from outside the car, usually, such that you're only getting the airborne component.

When sat in the car, (I remember this being one of your bugbears for games) you get more than just the airborne bits. The engine is attached to a subframe, which is attached to the rest of the car. This attachment allows the vibrations / sound of the engine to be physically transduced through the car's structure to your butt, etc. In addition, the body and chassis "ring" at certain points (their anti-nodes) causing extra airborne sound - this is subtle, usually. A similar effect happens with the exhaust being excited by the gas pulses.

Modern cars have carefully designed mounts to avoid the transmission of engine "vibrations" to the chassis and hence the passenger cell, so they have a distinctive sound. Due to the fact that each car has a uniquely shaped chassis and body (more-or-less) each car has a different colouration to its sound, as heard when sat in it. Most people probably wouldn't be able to tell you what the difference was, but they would be aware of a difference in cabin sound, even with the same engine and overall sound insulation.

The louder the engine, though, the more the airborne component begins to take over - except that in race machines, the engine mounts are more or less solid, so the car feels like it's an extension of the engine. You get this impression with some performance cars, too - e.g. DC2 Type R, Lotus 7 derivatives etc.

The chassis / shell "ringing" can be faked with delayed filters, though it would be very tricky to get the correct frequency response (the nodes and anti-nodes are in different locations for different groups of frequencies - different delays / phases). If it were me making these sounds, I would probably hook up a pure-tone generator to a transducer on the engine block and record the resulting cabin sound, as well as using a pure-tone generator and loudspeaker (in different locations, e.g. engine, exhaust, tyres etc.) to record the resulting (muffled) airborne cabin sound, for a wide range of frequencies.

In addition, a second transducer could be used in the driver's seat for a dedicated "buttkicker" channel. This is all next-gen stuff, really, but some filthy hacks can be used (and be much less convincing) for this generation, especially since the PS3's Cell has those SPUs, which are ideal for this kind of DSP work - except that they're over-stressed as it is :D
 
The strange thing is in reply and with good surround gt5p sounds are great. I'll post you the proof later this week.
 
Last edited:
The chassis / shell "ringing" can be faked with delayed filters, though it would be very tricky to get the correct frequency response (the nodes and anti-nodes are in different locations for different groups of frequencies - different delays / phases). If it were me making these sounds, I would probably hook up a pure-tone generator to a transducer on the engine block and record the resulting cabin sound, as well as using a pure-tone generator and loudspeaker (in different locations, e.g. engine, exhaust, tyres etc.) to record the resulting (muffled) airborne cabin sound, for a wide range of frequencies.

Would Impulse Responses work in this situation?
 
Would Impulse Responses work in this situation?

Theoretically, yes. But it'd be the impulse response through the chassis and body, directly from the engine block, into the seat rails (or something) plus the airborne (e.g. ordinary mic, maybe stereo directional mics) recording in the cabin of the same (mechanical) stimulus. In addition, if you wanted to model the sound-proofing process in sounds coming from the outside into the cabin, you would use airborne impulses delivered by loudspeakers (or a spark, for "better" dynamic range) measured from the front, rear and the four corners.

I don't know if resonant (mechanical) structures can be properly characterised with an impulse response. I know that for RF / microwave resonant structures you typically use a frequency sweep to characterise their impedance; most artifical impulses are generated by using a very fast sweep (since an impulse is theoretically supposed to contain all frequencies) but there is always a bias in frequency content produced with this method, certainly for digital applications. I suppose if we are to assume that we are measuring the (complex) impedance from one position in the "network" to another (engine block -> mounts -> chassis / body -> secondary emission -> listener), then this method is valid as an approximation.
Maybe somebody with experience in this sort of vibration / dynamics stuff could clarify - in fact, now that I've finished writing this post, I've just stumbled across the term I was looking for: [WIKIPEDIA]Modal Analysis[/WIKIPEDIA], and [WIKIPEDIA]Modal testing[/WIKIPEDIA]. :P

However, given that a lot of the newer cars are modeled from CAD data (if only for reference purposes), the chassis and body structures can be thrown into some sort of acoustics simulator that could not only approximate the sound proofing, but could quite accurately describe the mechanical transduction and secondary emissions into the cabin. The same system could be used to calculate secondary emissions from the exhaust (which typically runs under the cabin). The main advantage being you're not relying on or having to compensate for imperfect transducers (including the microphones), and you don't have to worry as much about noise interference - but, the results may not be totally accurate.
 
From the front page


Sounds awful to be honest:(

Hah! Why does an L88 (or ZL1, I honestly can't tell) sound like an LT4? :odd:

And the 458 sounds like the F430 from Prologue. See? Like I've been saying....I held absolutely no reserve for audio improvement and I'm not the least bit disappointed.
 
That's a 350ci car. Not a big block. The sounds is still completely wrong though.

So an L46 then (I think, I can't remember off the top of my head....it's going on 3 AM give me a break :lol:), and Indeed it is completely wrong. As I said, kind of sounds like an LT4.
 
The grass is not always greener on the other side



Unnecessary post is unnecessary; I'm sorry, I do not care if Forza sounds aren't any better.

Why is it so hard for people to accept that GT does some things better, and some things worse than other games i.e. engine sounds?

IDC if the sound engine and processing is better it would make no difference if the sound samples were recorded by a microphone on
All Programs>Accessories>Entertainment>Sound Recorder

It's like using expired low-quality food in a brilliant food processor - the end result will still suffer.
 
Unnecessary post is unnecessary; I'm sorry, I do not care if Forza sounds aren't any better.

Why is it so hard for people to accept that GT does some things better, and some things worse than other games i.e. engine sounds?

IDC if the sound engine and processing is better it would make no difference if the sound samples were recorded by a microphone on
All Programs>Accessories>Entertainment>Sound Recorder

It's like using expired low-quality food in a brilliant food processor - the end result will still suffer.

Vid is not Forza. There's a lot more than 8 cars going on there :)

It's NFS Shift, PC version, playing it on headphones will tear up your eardrums!
Some here consider it to have "awesome" audio - I think not.
Very few games do.
 
So an L46 then (I think, I can't remember off the top of my head....it's going on 3 AM give me a break :lol:), and Indeed it is completely wrong. As I said, kind of sounds like an LT4.

Your almost right. It`s a 300hp L-48,its option code or name when installed in a 1969-70 vette is ZQ3. And was the standard V8 fitted to a 1969-70 vette.

A L-46 is a high perf L-48/ZQ3 with 11.0:1 comp and 350hp,fitted to 1969-70 vettes as a option.

Hopefully it does sound better when played at home and not in a noisy echo filled hall.
 
There is another vid of that car, 2nd half of the video, which everyone missed, but it still sounds like a hairdryer :)

 
Vid is not Forza. There's a lot more than 8 cars going on there :)

It's NFS Shift, PC version, playing it on headphones will tear up your eardrums!
Some here consider it to have "awesome" audio - I think not.
Very few games do.

That's not the point, point is when you people fail to justify GT's sound as good and realistic you guys begin to point at other games' audio flaws...
 
That's not the point, point is when you people fail to justify GT's sound as good and realistic you guys begin to point at other games' audio flaws...

The point is, all games have audio flaws. There are some good sounds, and bad ones in every game.
Much the same way all games have visual flaws and artifacts
 
Great... Judging something from an off-screen recording with all the crap thrown in...

Indeed. People criticising the sound of a demo played at low volume, through TV speakers, recorded on a cheap camera via the built-in mic. :banghead: :lol:
 
Vid is not Forza. There's a lot more than 8 cars going on there :)

It's NFS Shift, PC version, playing it on headphones will tear up your eardrums!
Some here consider it to have "awesome" audio - I think not.
Very few games do.



that's pretty much the best audio i've ever heard out of a racing game. the speed of the shifting animation is what gt5 should be too.
 
Last edited:
Indeed. People criticising the sound of a demo played at low volume, through TV speakers, recorded on a cheap camera via the built-in mic. :banghead: :lol:

You can still get a good idea of the quality of sound in this scenario.

As I said before (although with more Hyperbole) Shift sounds pretty good even on cheap tv speakers.

If the engine sample sounds incredibly simple and basic, no amount of hi-fi is going to get around this.

It's like saying you can't judge the graphics from low resolution youtube video, yes you can! Maybe not in fine detail, but you can get a good idea.
 
Indeed. People criticising the sound of a demo played at low volume, through TV speakers, recorded on a cheap camera via the built-in mic. :banghead: :lol:

It's the more sad that even through this poor audio chain you can obviously hear the engine sound still sucks big time. :crazy:
 
As I said before (although with more Hyperbole) Shift sounds pretty good even on cheap tv speakers.
.

I disagree by a lot. I've tried most of the cars in Shift and they sound terrible. They actually hurt my ears....
The Ferraris don't even sound anything like one.
The sound designer behind Shift is actually proud of all the distortion!
And they had to use the worst kind of distortion...

http://designingsound.org/2010/02/charles-deenen-special-need-for-speed-exclusive-interview/
 
Back