• Hey Guest. Check out your NeoGAF Wrapped 2025 results here!

GT5 News - Nurburgring update (pre-E3 edition)

Niks said:
11jqffa.jpg

20at6s2.jpg

21bq9ds.jpg

15zi5jk.jpg

kazwank.gif
 
blu said:
ps: one of the reasons i'm so enthusiastic about what GT5 brings to the table is that i was doing racing software for 4 years, in the span 2001-2004, covering engineering analysis workstations (used in both F1 and WRC), ECU worstations, dashboards (WRC), and wind tunnel control sw. from the POV of engineering, all (properly acquired) data is useful data.

blue, in your opinion, whats the state of current software for racing?
 
phil_t said:
I think you have to stretch your imagination quite far to be optimistic enough to say that this type of feature has no negative impact on the products development..

But if you want to throw practicalities out of the window, we'd never have anything to discuss, because unless absolute facts are known, everything is a 'maybe'...

We can't judge the inner workings of Polyphony before the game is even out because game development in general isn't exactly transparent. The fact of the matter is there are a bunch of features that money and time was spent on that will not make it into the final product... on any game, not just this one. There are also reasons to delay a game that have nothing to do with when something is ready or not, as is obviously apparent right now. I don't think it's optimistic to say that this had no real impact on when GT5 is coming out or the quality of the product outside of this feature for a general common sense reason. Do you see Sony putting the brakes on this 60 mil project that hundreds of people are working on so that they could make this little thing?
 
spwolf said:
blue, in your opinion, whats the state of current software for racing?
i've kept only a cursory tab on that industry since then (been doing entirely different things since then) - but i can tell you what it was 6 yeas ago.

it involved a lot of 'manual labor' on the part of the team's field engineers, to get the data in humanly-usable form, involving data 'refinment' and transformations - telemetry/log datasets are (a) practically always too rough to be used without some good dose of smart filtering, (b) often more useful in some 'derivative' form (e.g. 1st rate of change, 2nd rate of change, etc) - so that's what field engineering workstations would do - take 'raw' input and produce humanly-usable output. also, industrial racing simulators were a brand new thing back then, very few (maybe one or two) top teams could afford one, from what i recall. and none of them were mobile ; )

what particularly grabbed my attention from the GT5 announcements was the precision of the lap trajectory (apprently good enough to be used along sythesized replays) - i'm curious about
(a) what INS toyota have in there
(b) how much that feature works outside of tarmac-only circuits (where the game physics can potentially help with trajectory correction)
(c) what's the error rate there vs the true trajectory (there's always an error rate - no such thing as absolutely true-to-life logged trajectory) - if they did decimeter precision that would already be very nice.

basically, if i saw PD doing the same for WRC courses, i'd be floored (or INS tech advanced so much over the past few years that it's not tech problem anymore ; )
 
TTG said:
Do you see Sony putting the brakes on this 60 mil project that hundreds of people are working on so that they could make this little thing?
Do you think Sony has even the slightest control over Kaz any more? :lol
 
blu said:
i've kept only a cursory tab on that industry since then (been doing entirely different things since then) - but i can tell you what it was 6 yeas ago.

it involved a lot of 'manual labor' on the part of the team's field engineers, to get the data in humanly-usable form, involving data 'refinment' and transformations - telemetry/log datasets are (a) practically always too rough to be used without some good dose of smart filtering, (b) often more useful in some 'derivative' form (e.g. 1st rate of change, 2nd rate of change, etc) - so that's what field engineering workstations would do - take 'raw' input and produce humanly-usable output. also, industrial racing simulators were a brand new thing back then, very few (maybe one or two) top teams could afford one, from what i recall. and none of them were mobile ; )

what particularly grabbed my attention from the GT5 announcements was the precision of the lap trajectory (apprently good enough to be used along sythesized replays) - i'm curious about
(a) what INS toyota have in there
(b) how much that feature works outside of tarmac-only circuits (where the game physics can potentially help with trajectory correction)
(c) what's the error rate there vs the true trajectory (there's always an error rate - no such thing as absolutely true-to-life logged trajectory) - if they did decimeter precision that would already be very nice.

basically, if i saw PD doing the same for WRC courses, i'd be floored (or INS tech advanced so much over the past few years that it's not tech problem anymore ; )

thanks for the info... since PD is already working on similar data for their games, it is much easier for big software development like that to work on software and hardware automotive applications. I hope they do more work in automotive industry, it is sorely needed. Most of the software and hardware used in automative applications is stone age, for better or worse (there are some positives when it comes to reliability). Japanese automobile companies are definetly last and hopefully their initiative to develop standardized, new generation computer interface will bear fruits soon.
 
That Honda is sex. I do like how each region has their own styling, for example, EU made cars, JP, NA, etc. I do think EU and JP are way ahead of the curve here though.
 
spwolf said:
So you are saying that drivers do not need to learn the track and they dont use such tools? That Racing Simulators are used only for setup reasons not actually for learning the track? Interesting.

The race simulators you are referring to are merely games. They have there own set of physics which no matter how good they may be don't replicate real world physics. Sure, some drivers will use a game (or race simulator) to familiarise themselves with a new track.

But overlaying a 'ghost replay' generated from real life data into the game does not help a race driver in any way. Which is the only thing this technology is doing.

blu said:
it's not a 'pretty' replay, it's an engineering-log-precise (same resolution, same timeline) + point-of-view + clear picture + flexible-view-angles replay. still not seeing it?

Not doubting the precision of the data logged. But it's not possible to identify where a car is making or losing tenths or hundredths of seconds by staring at a GT5 replay with the naked eye.

This is possibly only useful to the average Joe who's just started out racing or track days and is not familiar with the concept of the race line.
 
MGR said:
The race simulators you are referring to are merely games. They have there own set of physics which no matter how good they may be don't replicate real world physics. Sure, some drivers will use a game (or race simulator) to familiarise themselves with a new track.

But overlaying a 'ghost replay' generated from real life data into the game does not help a race driver in any way. Which is the only thing this technology is doing.



Not doubting the precision of the data logged. But it's not possible to identify where a car is making or losing tenths or hundredths of seconds by staring at a GT5 replay with the naked eye.

This is possibly only useful to the average Joe who's just started out racing or track days and is not familiar with the concept of the race line.


Wrong!!

as an owner of an IS-F, and as someone who tracks their car vigorously at Infineon, and Las Vegas Motor Speedway, this technology has me on the edge of my seat with me and my team trying to contact Polyphony and Toyota to get this early, if only with Infineon as the race track.

My team and I predict this tech will knock off atleast 10 seconds from our laps, it could also help with fuel consumption and tire saving techniques.

AMAZING!!
 
KittonGotWet said:
Wrong!!

as an owner of an IS-F, and as someone who tracks their car vigorously at Infineon, and Las Vegas Motor Speedway, this technology has me on the edge of my seat with me and my team trying to contact Polyphony and Toyota to get this early, if only with Infineon as the race track.

My team and I predict this tech will knock off atleast 10 seconds from our laps, it could also help with fuel consumption and tire saving techniques.

AMAZING!!

C'MON dude, in the God of War Thread you were Zeus, In the Zelda thread you were Link and we know who you were in the super mario thread. I mean C'mon. Now if you serious I mean C'MON
 
Im just giving my perspective on the issue, dont know what else to say.

fellow says no one will use this, and I rebutted that statement with my own experience.

Seems like people would rather bash than praise, not my problem.
 
Well it looks like we finally got all the major JDM companies on board with there super cars.

Nissan - GTR
Honda - HSV-010
Toyota - LFA
Mitsubishi - Evo X
Mazda - Furai


Still waiting on what Kaz may confirm regarding the STi R205 / A-Line Type S
 
Raadius said:
Well it looks like we finally got all the major JDM companies on board with there super cars.

Nissan - GTR
Honda - HSV-010
Toyota - LFA
Mitsubishi - Evo X
Mazda - Furai


Still waiting on what Kaz may confirm regarding the STi R205 / A-Line Type S
Evo X is nowhere in their leagues and the HSV will not be a production car as Honda went through a loophole so that they could create that car and run it in the Super GT.

Unless Honda wants to be nice to us. That is one sexy machine.
 
enzo_gt said:
Evo X is nowhere in their leagues and the HSV will not be a production car as Honda went through a loophole so that they could create that car and run it in the Super GT.

Unless Honda wants to be nice to us. That is one sexy machine.
Yes, I'm aware of that. My post wasn't a comparison but more of a list showing the manufactures super car that have been confirmed.

Currently Mitsubishi doesn't offer anything than a Ralliart form of the Evo X (which still can't compete) and the UK's FQ series will not make it in the game.
 
Has any news come out of Famitsu this week? At present I am most interested in seeing what PD have done with the online considering GT5:P online was terrible however think I'm correct in saying they now have a separate team just for that.
 
MGR said:
Not doubting the precision of the data logged. But it's not possible to identify where a car is making or losing tenths or hundredths of seconds by staring at a GT5 replay with the naked eye.
of course you can't tell that with a naked eye - motoracing would be an entirely different sport if people could do that. but PD's tech could largely supersede the cockpit POV cam feeds, and, just like the latter, be used along with the rest of the timeline views run on an engineering station. the one problem is - you'd have to manually sync your 'virtual video' replay with the master time cursor of your engineering station, as the latter apparently would not know about your 'other station' AKA ps3. but you'd still be able to make the basic correlation: data point T in channel X corresponds to timestamp T in the 'virtual video feed', just by initially calculating the station's cursor to match the start of the video timeline, and then physically sync the 'time run' toggle on both platforms.

also, the more i think about it, the more i come to the conclusion that PD have what to offer in terms of lap trajectory precision to the industry as a whole, as the simulation can be used as a good measure of correction to the INS log. that could be used in two directions - allowing teams to use lower-class INS units and still obtain precise-enough trajectories in PD's simulation, or use high-end-class units, and achieve unprecedented precision. of course, that's purely conjecture on my part, as i have no idea what PD actually do there, and what tolerance they have toward the raw INS error. that's why i'm curious about the actual unit in the toyota, the ISA configuration, and exactly what channels logs & exports.
 
blu said:
of course you can't tell that with a naked eye - motoracing would be an entirely different sport if people could do that. but PD's tech could largely supersede the cockpit POV cam feeds, and, just like the latter, be used along with the rest of the timeline views run on an engineering station. the one problem is - you'd have to manually sync your 'virtual video' replay with the master time cursor of your engineering station, as the latter apparently would not know about your 'other station' AKA ps3. but you'd still be able to make the basic correlation: data point T in channel X corresponds to timestamp T in the 'virtual video feed', just by initially calculating the station's cursor to match the start of the video timeline, and then physically sync the 'time run' toggle on both platforms.

The technology as demonstrated by PD and Toyota does none of what you have written. Your entire post is speculation.

This from the official site:
http://www.gran-turismo.com/jp/news/d8925.html

Translated:

CAN export to USB memory location information and that information, "Gran Turismo" and loaded into a simulation engine, "Gran Turismo" was visualizes as a video replay, you can also play as a ghost, help you analyze your driving or you can.

CAN has the right to information about G Force and handle and brake, accelerator operations, and includes rotational speed of each wheel and engine speed, allowing a highly accurate reproduction of the combination of vehicle and location information .

The data logged is used to create a video reply / ghost file playable using Gran Turismo. Nothing more.

You can speculate all you like about what may or may not happen in the future but the tech as demonstrated has no useful purpose in motor sport.
 
KittonGotWet said:
Wrong!!

as an owner of an IS-F, and as someone who tracks their car vigorously at Infineon, and Las Vegas Motor Speedway, this technology has me on the edge of my seat with me and my team trying to contact Polyphony and Toyota to get this early, if only with Infineon as the race track.

My team and I predict this tech will knock off atleast 10 seconds from our laps, it could also help with fuel consumption and tire saving techniques.

AMAZING!!

You shouldn't be so transparent.
 
Raadius said:
Yes, I'm aware of that. My post wasn't a comparison but more of a list showing the manufactures super car that have been confirmed.

Currently Mitsubishi doesn't offer anything than a Ralliart form of the Evo X (which still can't compete) and the UK's FQ series will not make it in the game.

fast lap times = supercar? :D
 
MGR said:
The technology as demonstrated by PD and Toyota does none of what you have written. Your entire post is speculation.

This from the official site:
http://www.gran-turismo.com/jp/news/d8925.html

Translated:



The data logged is used to create a video reply / ghost file playable using Gran Turismo. Nothing more.

You can speculate all you like about what may or may not happen in the future but the tech as demonstrated has no useful purpose in motor sport.

no, what we are saying is underlaying technology is as sophisticated or more than what Motec uses for VCS system of theirs... and you will see same type of info you see in VCS with overlaid data logger - braking points, entry/exit speeds, rpms/gear, /throttle/brake/steering wheel input, etc... so whatever you get with VCS+data logger VCS overlay, you get here.
 
Raadius said:
Well it looks like we finally got all the major JDM companies on board with there super cars.

Nissan - GTR
Honda - HSV-010
Toyota - LFA
Mitsubishi - Evo X
Mazda - Furai


Still waiting on what Kaz may confirm regarding the STi R205 / A-Line Type S

Honestly, I am more interested in FT-86 G potential than some limited cars like LFA or non-production cars like HSV-010.

So...

Nissan - GTR, 370z
Mitsu - Evo X - whatever is the best special version
Subaru - Impreza - whatever is the best special version
Toyota - FT-86

All these will be serious sport cars that anyone could purchase, and have 300+ hp versions.
 
You know what would be awesome? If somehow you could record a real track with your GPS, upload it to GT and then the track editor generates the track for you. This way you can easily and quickly create real life tracks with great ease :D
 
SToRM said:
You know what would be awesome? If somehow you could record a real track with your GPS, upload it to GT and then the track editor generates the track for you. This way you can easily and quickly create real life tracks with great ease :D

That would be pretty nuts if you could create a start/ stop thing, creating a point to point race track from your work-home route would be awesome.
 
spwolf said:
no, what we are saying is underlaying technology is as sophisticated or more than what Motec uses for VCS system of theirs... and you will see same type of info you see in VCS with overlaid data logger - braking points, entry/exit speeds, rpms/gear, /throttle/brake/steering wheel input, etc... so whatever you get with VCS+data logger VCS overlay, you get here.

Even if the Denso/Toyota CAN system was the most accurate data logger on the planet, generating a GT replay / ghost with the logged data is of no benefit to motor sport applications.
 
MGR said:
Even if the Denso/Toyota CAN system was the most accurate data logger on the planet, generating a GT replay / ghost with the logged data is of no benefit to motor sport applications.

so Motec VCS is not useful at all for motor sport applications? You keep getting stuck on racing ghosts...
 
spwolf said:
so Motec VCS is not useful at all for motor sport applications? You keep getting stuck on racing ghosts...

Creating ghosts is the only thing the software (in this case GT) does.

The Motec VCS doesn't provide much useful data other than assisting engineers to identify anomalies found in the logged data (analysed with the i2 Pro Data Analysis Software). Video is able to capture instances where the race car may be baulked or encounter track debris / oil slicks etc.

Hypothetically even if the track (in this example Fuji) is absolutely perfectly replicated in GT the data logged then used to generate the GT replay is still not able to replicate any of the real world irregularities which occur during real life testing and racing.

I'd be much more impressed if you could log replay data using GT and then upload it into the real thing and have it drive the track for you while you sit in the passenger seat. :D

Not dissimilar to this:
http://www.youtube.com/watch?v=jBVgFJtvmaE
 
Famitsu? Soon? Maybe?

It will probably just be stuff about the GPS thing, and pictures of Tokyo R246 :(
 
spwolf said:
no, what we are saying is underlaying technology is as sophisticated or more than what Motec uses for VCS system of theirs... and you will see same type of info you see in VCS with overlaid data logger - braking points, entry/exit speeds, rpms/gear, /throttle/brake/steering wheel input, etc... so whatever you get with VCS+data logger VCS overlay, you get here.

It will be a nice little thing for us gamers to play around with. But for a racing driver that kind of information is basically bollocks isn't it.. The setup of a car, yes, then logging of various vehicle data is useful TOGETHER with the mentioned information to make context and meaning out of it the logged data (Engine data f.ex. is of course depending on where on the track the vehicle is, throttle input etc. etc.). But, to use it for learning to drive, if a race driver can't figure out racing lines and break points then perhaps he/she should rather be replaced with someone a bit more talented. No?
 
Not quite. If you can load different people's data into it, you can maybe see how you can push a braking point a little later, throttle a little earlier and overlay multiple laps onto each other.

Something I want for the game in general is not just one ghost, or best ghost, but overlay multiple laps lines onto the track at once, not even during gameplay, but maybe in a special race analysis view. So I can see on which of my past 10 laps I took that corner the best. Was that one attempt to brake a bit later a good idea? Can I try to brake even later?

Its not about learning apexes, but brake points are ever changing for many drivers. Even in F1 people compare themselves to Schumacher and others and see what their brake/throttle lines are and see that one of Michael's signature moves is brake later than anybody and for less time than anybody but harder. Theres a video floating around somewhere showing another F1 driver comparing his lines to Michael's.

I would love loading up Stig's and other celebrities' race lines and race them myself. Or learn from them.
 
Top Bottom