"Shokunin" Concept

  • Thread starter Gids
  • 70 comments
  • 7,104 views
Gids

Gids

106
United States
Fort Walton Beach, FL
GTP_Pavelow440
You know I really like this "Shokunin" concept of a master craftsman. I even like the "one person, one car" policy that PD has in place. It brings a loose connection to the high end cars of old and a few cars of today that were/are built by true artisans.

My only concern is the limited staff of PD and time it takes to craft these cars. Will we ever see all the standard cars as premium at the rate it takes to create a premium car by one person? Seems like Sony and PD should be hiring a small army to focus solely on transitioning all the standards to premium and then quickly move on to each and every car in the "Suggestions" thread in order of votes.

Anyone have the information on how long it takes to create a premium car from scratch, upgrade a standard to premium, and how many "Shokunins" dedicated to car modeling PD employs?

I did a search and did not see this addressed but if I missed it please lock/delete.
 
Last edited:
Interesting topic, indeed.
Truth is ... we don't need all the "Miatas" so ... I think GT7 will be actually with a smaller collection of cars then GT5/6.
As for the staff involved ... yeah ... 1300 people is crazy but if that's true ... then I don't see why PD is so slow on making all the cars be playable in the game.
My guess is making a car will take 1 month, just the 3d model and recording the sounds ... Then maybe testing the physics will be about 2 months because I'm sure that the car is tested by more then 1 person. So 4 months, imo. Otherwise it's just sloppy job.
 
I seem to remember an old interview with Kaz where he envisioned GT being almost an encyclopedia of automobile history. I agree the repetitive "Miatas" etc is unnecessary. However, the shear diversity of eras and types of cars people are interested in (as shown by the genius suggestions forum) would be an amazing addition to the GT encyclopedia.

I know people get bent out of shape with the talk of taking cars from one version to the next but that would be a true encyclopedia if you could.
 
Jose Mendonca
They didn't say each person only does one car. It's each car is made by only one person. Means one person can make 10 cars from start to finish for example.
Click to expand...

No that's obvious. The point was to do the math it takes one person to create a car. Then determine how many are "artisans" to give a rough guess on how long it would take to upgrade the standards and crank out new and suggested vehicles.
 
Gids
No that's obvious. The point was to do the math it takes one person to create a car. Then determine how many are "artisans" to give a rough guess on how long it would take to upgrade the standards and crank out new and suggested vehicles.
Click to expand...
This will of course increase the time to do one single car, but the overall manhours it takes stay the same. In the end it doesn't really matter if two guys do two cars in 6 months or one car in three months and then a second one in 3 more months...
 
Verbal
This will of course increase the time to do one single car, but the overall manhours it takes stay the same. In the end it doesn't really matter if two guys do two cars in 6 months or one car in three months and then a second one in 3 more months...
Click to expand...

Production line management would prove that the man hours to create a product are not the same. Typically, time is increased if one person is responsible for an entire project. Rather than the same small piece of the similar puzzle.
 
This topic sounds intriguing. It can also explain why PD is so fond of creating similar models as artists will want to find the easiest way of maximising their output [of individual cars]. It's probably got something to do with tradition, but at some point PD have got to adapt to higher demands and improve efficiency. Instead of using individuals to model one car, maybe get a team to do it? This sums up my thoughts on using Shokunin nowadays:

DingoDile
It was great for PS1 & PS2 GTs, but in PS3,PS4,... they should reconsider it. making a high detail car for PS3 and PS4 hardware isnt a one-man job IMO.
Click to expand...
 
While it is true that we don't need all of those cars which are very similar to one another or almost identical the fact is that those don't change the time required by much at all. It may take 6 months to do an MX5 but to do a second MX5 that is almost identical may take only a day or possibly even less depending on the differences. If they were truly identical as some people here at GTP say then it would take only a few minutes extra so if they really wanted the car count of premium models to be high then they would start with those cars and make them all premium.

As for the one person per car concept, I have no idea how they go about building the cars in house, perhaps they have some software that creates the wire frame for them and it is just a matter of created and applying all the textures and if so then one person per car makes sense. If they have to create the wire frames as well then it would make sense to have one person for the mesh and another for the textures since those are different skill sets.
 
I say just give us the cars we like, without having 30 different versions. You can have a base model, performance, higher performance and stuff like that. Such as V6 Camaro, SS, ZL1, and Z28. Then give us body kits and a livery editor. This will allow people to make those missing 20 versions nobody really wants.
An acception is that those companies who make performance versions, some of those are nice; but we still don't need all of them.
If PD will allow powertrain swaps and give us a livery editor, this can be used to replace cars they didn't build. Also allow drivetrain swaps people do in real life. Not just an R35 engine in an R33, but like an LS1 in a miata or RX7.
 
As this topic is my professional expertise in life; methodology and production management, I have one thing to say....

Are we really so desperate for content that we are killing time discussing the cultural methodology of PD? And what is it that we might discover as a result?

I assure you PD isn't that strange to the typical Japanese dev studio, a Sony studio, and longer term Japanese digital company... in fact I would argue its likely a carbon copy.

If you really want to get a stronger idea about these methods and processes as they pertain to Japanese culture I would recommend looking at any other studio that is a bit more open and then applying your new findings to PD. Capcom, square, etc.

But at the end of the day it still has really to nothing to do with the "why" of the PD enigma... And be carful to not let this confuse you as names and terms for process are 'buzz words" and most are just excuses despised as super advanced... Watchmacallits...

"I make space ships!!! 'Oh wow!' Ok guys, were is the light switch again?"
 
NixxxoN
People wanted very high quality and graphic detail... so the price to pay is that it takes a lot to create.
Enjoy your decisions, graphics loving gamers.
Click to expand...

I very much doubt other racing games who have the same if not better standards for car models follow the same approach.
 
I just think it would be a hell of a lot more efficient if PD had once person model each body panel on a car. A body panel can be made to very high quality and then pieced together with all the other body panels. We have a same (Or better quality product) for less time. (This can be created in the space of a work week)
 
The first time I read about this, the first thing that came to my mind was sounds. Kaz didn't say anything about this, but imagine if there is one person assigning sounds to cars since the first games, that would explain the inconsistencies in sounds ever since the early days of GT.
 
Personally i find pd's methods admirable and worthy. Let them be. I find most comentators arrogance in believing they know better or could do better laughable. Without fail everything pd does is slammed by a small core of bitter people. Pd do things the pd way, if we cant understand/comprehend that then there should be many many many other car games to play made by western "couldn't give a ****" developers. Also i strongly believe this kind of dedication is why pd have the best relationships with manufacturers
 
Crispy
I just think it would be a hell of a lot more efficient if PD had once person model each body panel on a car. A body panel can be made to very high quality and then pieced together with all the other body panels. We have a same (Or better quality product) for less time. (This can be created in the space of a work week)
Click to expand...
Hey Crisp, I know what you're getting at but the reality is quite different. Digital content is not the 'sum of parts' like real parts. Part of licensing does include cad art and game teams try to break it down into as few parts as possible. And despite digital source, there is still an interpretive component to it.

For instance they would build the whole body shell before cutting the seams for the panels. Digital cars are in fact much more accurate than real cars as things often don't line up on real cars... Look at the panel gaps of your own car, you'll notice they aren't all the same. Real cars are quite unique in that instance.

So it actually faster and more efficient if one guy does the bulk of the visual representation of the car.

You also have to consider there are tools in development built by others to aid those artists like animation of suspension etc.

And it's not black or white. If you break down a digital car it's more about properties than sections. And this is likely different people or tools. Like windows, flourishes like headlights and indicators, wheels and tires, interior/exterior, etc. separate from body.

Put a standard car and premium car next to each other and its clear to the trained eye exactly how they evolved their building process... But that's only applicable if they didn't change how do it over the years... And production process is the most organic and changing part of all creative content teams...

How to make a 'better game' often starts from defining a better process and methods, along with tools that support them. Note, "better" does not mean ridged, but more streamlined.

I hope that made some sense, as the topic of asset creation, static or organic, it infinitely more complicated than most think they know. Logic only works at the most simplistic levels.
 
Coming from a background in organisational design and management, it is quite funny to see so many people thinking that one modeler per car is bad management, inefficient, etc, etc.

The thing is, for any job requiring high levels of skill and/or craftsmanship, like building premium quality car models, dividing the job into smaller separate tasks is actually a BAD thing. Why, you ask? Because it creates a lot for need for coordination.

Look at it like this: Person A is modelling a car, together with person B. Person A will be modelling all sorts of parts that connect and even interact with parts modeled by person B. In order to make sure this connection and interaction is in perfect order, has exactly the right colour code, the exact right sizes & etc, they will frequently need to compare their work and even test a preliminary model to see if everything done so far is good. I hope it is evident that this costs a LOT of time.

If all of this is done by a single person, he does not need to do all that coordination and fine tuning, because he already knows all those details and how to fit them together!
In essence, organisations should minimize the need for coordination to be truly efficient.


And that is just the purely practical side of things. There is another reason to design work in this fashion: motivation and responsibility.

When a project is divided among several people to be done in parts, each separate person will not be completely attached to the project, because parts of it are out of their control.
When a project is done by a single person, that person essentially has ownership of the project. He/she will be personally and entirely responsible for its success, which leads to far greater motivation to do it right. There is after all, no one else to blame any mistakes on, and a good result directly affects their personal pride in their work.


In short, according to my professional knowledge, the division of labour within PD is a very effective one.
 
Last edited:
Lawndart
Hey Crisp, I know what you're getting at but the reality is quite different. Digital content is not the 'sum of parts' like real parts. Part of licensing does include cad art and game teams try to break it down into as few parts as possible. And despite digital source, there is still an interpretive component to it.

For instance they would build the whole body shell before cutting the seams for the panels. Digital cars are in fact much more accurate than real cars as things often don't line up on real cars... Look at the panel gaps of your own car, you'll notice they aren't all the same. Real cars are quite unique in that instance.

So it actually faster and more efficient if one guy does the bulk of the visual representation of the car.

You also have to consider there are tools in development built by others to aid those artists like animation of suspension etc.

And it's not black or white. If you break down a digital car it's more about properties than sections. And this is likely different people or tools. Like windows, flourishes like headlights and indicators, wheels and tires, interior/exterior, etc. separate from body.

Put a standard car and premium car next to each other and its clear to the trained eye exactly how they evolved their building process... But that's only applicable if they didn't change how do it over the years... And production process is the most organic and changing part of all creative content teams...

How to make a 'better game' often starts from defining a better process and methods, along with tools that support them. Note, "better" does not mean ridged, but more streamlined.

I hope that made some sense, as the topic of asset creation, static or organic, it infinitely more complicated than most think they know. Logic only works at the most simplistic levels.
Click to expand...


Thank you for that explanation/information. I am familiar with the manufacturing process of a couple different aircraft and have done test flights for a handful. However, I was completely unaware of how it is done in the digital world. Very interesting indeed.

Thanks again.
 
Well, this is tedious.
If a company can scan tracks with a laser to get accurate data don't you think it might be a good idea to use said laser and scan your cars?
It's not like the "designers, artisans or whatever you name the guys in charge to model a car" do have to draw a car from scratch. They scan the cars, take picture etc.
I don't what they do with their aquired data but there's a complete branch out there which bases on computer aided design (CAD). Of course the computer doesn't everything alone but it helps, anyone working in that scene knows that.

I'm for sure no genius but this might get an approach to go for.

On the other hand it is astonishing how people defend PD with such a verve, telling the guys criticizing GT they have no clue about what they are talking about. On the other hand those "defenders" do have the same insight as those who want a real game for their real money.

Why don't you stick to facts? Such as advertised DLC, additional tracks and cars, parts of the game still missing(course maker, online features, i'm sure i forgot at least one other point).

What they have proved with GT 6 is that they failed to provide the game to the announced date (come on, you know when your 15th anniversary will happen and how long before you have to start work to match your deadline) and they introduced a new, nearly unique feature: updates which increase the number of bugs already in the game rather than removing them.

I spend 69,- € (94 $, 57 £ or 108 AU$ ) for the Anniversary Edition in the hope, they did a step ahead of GT 5.
I might state they didn't.

I would be in rage mode if i had to spend additional money for a PS3.
And i think with the current experience and behaviour by PD i will think twice before i consider buying GT 7 and a PS 4; you can be sure that i'll wait till the first gamer experiences are available.
And my guess will be that this will not only be me.

Aside from my short and incomplete overview how i see PD i still hope the let live up the game to the expectations they already rised and finally provide the missing parts of the initial game.
 
Morphisor
Coming from a background in organisational design and management, it is quite funny to see so many people thinking that one modeler per car is bad management, inefficient, etc, etc.

The thing is, for any job requiring high levels of skill and/or craftsmanship, like building premium quality car models, dividing the job into smaller separate tasks is actually a BAD thing. Why, you ask? Because it creates a lot for need for coordination.

Look at it like this: Person A is modelling a car, together with person B. Person A will be modelling all sorts of parts that connect and even interact with parts modeled by person B. In order to make sure this connection and interaction is in perfect order, has exactly the right colour code, the exact right sizes & etc, they will frequently need to compare their work and even test a preliminary model to see if everything done so far is good. I hope it is evident that this costs a LOT of time.

If all of this is done by a single person, he does not need to do all that coordination and fine tuning, because he already knows all those details and how to fit them together!
In essence, organisations should minimize the need for coordination to be truly efficient.


And that is just the purely practical side of things. There is another reason to design work in this fashion: motivation and responsibility.

When a project is divided among several people to be done in parts, each separate person will not be completely attached to the project, because parts of it are out of their control.
When a project is done by a single person, that person essentially has ownership of the project. He/she will be personally and entirely responsible for its success, which leads to far greater motivation to do it right. There is after all, no one else to blame any mistakes on, and a good result directly affects their personal pride in their work.


In short, according to my professional knowledge, the division of labour within PD is a very effective one.
Click to expand...

However the refusal to outsource at all is not. There are qualified modelers out there that given the opportunity could do the job up to PD's standards and under PD's terms (one modeler per car).

What is also troubling is that PD does not give its users the ability to appreciate the work they put into each model (hacking interior view in photo mode, no free look mode, etc). Since they do such an amazing job one would think they would want to show it off.
 
paskowitz
However the refusal to outsource at all is not. There are qualified modelers out there that given the opportunity could do the job up to PD's standards.

What is also troubling is that PD does not give its users the ability to appreciate the work they put into each model (hacking interior view in photo mode, no free look mode, etc). Since they do such an amazing job one would think they would want to show it off.
Click to expand...
It would be nice if PD would outsource some model development.
And I feel like with new cars the manufactures probably have a model in their computers thats 90% game ready.
PD should focus on primarily features,tracks, and physics.
 

Latest Posts

Back