Budget Is NOT An Excuse

[Budgiecat]

Yes

What games you made? Maybe I played them. Maybe I didn't...

 

[LemonHaze]

I don't think I'll divulge that information, for obvious reasons.

A lot of factors are relevant though: Yu Suzuki and his friends were developers in a pretty different time in terms of rendering paradigms, concepts etc, and while that doesn't impact their abilities directly in any way, using a new engine, hiring a lot of new staff, securing physical distribution and marketing deals, effectively from scratch (which is what happened with YsNet and Neilo etc regarding Shenmue III), learning new methods from the past decade or so, potentially changing the core structure of the game by adjusting it to somehow welcome in new players to the franchise etc. These are just a few.

The difference being with someone like Hellblade's team is, they were not active in the industry at the same time as YS and co., and so these developers are already aware of these 'new concepts' whereas YS are not. This isn't something you can just shrug off either, the Japanese documentation for UE4 was barely complete, as was the English documentation, by the time they were even starting development of Shenmue III.

Nonetheless, I think it goes without saying that Hellblade and Shenmue III are two very different games and can probably only be slightly compared in the way of, "they're both 'open-world' games". Due to this, the development of both of these games differ tremendously.

 

[Budgiecat]

I don't think I'll divulge that information, for obvious reasons.

A lot of factors are relevant though: Yu Suzuki and his friends were developers in a pretty different time in terms of rendering paradigms, concepts etc, and while that doesn't impact their abilities directly in any way, using a new engine, hiring a lot of new staff, securing physical distribution and marketing deals, effectively from scratch (which is what happened with YsNet and Neilo etc regarding Shenmue III), learning new methods from the past decade or so, potentially changing the core structure of the game by adjusting it to somehow welcome in new players to the franchise etc. These are just a few.

The difference being with someone like Hellblade is, they were not active in the industry at the same time as YS and co., and so these developers are already aware of these 'new concepts' whereas YS are not. This isn't something you can just shrug off either, the Japanese documentation for UE4 was barely complete, as was the English documentation, by the time they were even starting development of Shenmue III.

Nonetheless, I think it goes without saying that Hellblade and Shenmue III are two very different games and can probably only be slightly compared in the way of, "they're both 'open-world' games". Due to this, the development of both of these games differ tremendously.

Actually the real reason was alluded by Yu Suzuki himself recently:
He said he really doesn't pay attention or play other people's games.


Thats why I solely emphasized by frustration with Epic for not making newer things known to Yu since they have that experience on their end. After all, we know they obviously helped him learn how to use the UE4.

Also, I'm not trying to 'compare' Hellblade and Shenmue 3.

I'm merely showing the methods of development strategies that were used in one that was able to get around certain financial restriction issues by thinking outside the box. That has nothing to do with the style of the game vs the style of Shenmue 3.

 

[LemonHaze]

Actually the real reason was alluded by Yu Suzuki himself recently:
He said he really doesn't pay attention or play other people's games.


Thats why I solely emphasized by frustration with Epic for not making newer things known to Yu since they have that experience on their end. After all, we know they obviously helped him learn how to use the UE4.

Not even just recently, in fact he's always been open to the fact that he doesn't play games at all. Playing games has nothing to do with ones ability to develop or know anything about games in general.

Also, if you're going to argue that fact, then consider that he didn't play games either when he made either of his previous games, yet you don't have a problem with those.

 

[Budgiecat]

Not even just recently, in fact he's always been open to the fact that he doesn't play games at all. Playing games has nothing to do with ones ability to develop or know anything about games in general.

Also, if you're going to argue that fact, then consider that he didn't play games either when he made either of his previous games, yet you don't have a problem with those.

The difference is though is as you said: we live in different times now.
Plus, the other difference is unlike the past, where he had a multimillion dollar company supporting him, much like how Kojima had Sony backing him for Death Stranding with practically a blank check, is now he has very limited funds and a project where he knew he had to put out something that would impress consumers as he did with the first 2 games 20 years ago. Very difficult high road to climb there.
That is why at this point, as opposed to when he was at SEGA, an open mind to learning newer strategies to help create the better game could have been utilized.
As in like I said, the difference between saying "aw shit we dont have this extra few million to hire this mocap studio" vs Epic telling them "well you know you dont have to; previously we worked with Ninja Theory and they developed a strategy to get around that financial hurdle"

 

[LemonHaze]

The difference is though is as you said: we live in different times now.
Plus, the other difference is unlike the past, where he had a multimillion dollar company supporting him, much like how Kojima had Sony backing him for Death Stranding with practically a blank check, is now he has very limited funds and a project where he knew he had to put out something that would impress consumers as he did with the first 2 games 20 years ago. Very difficult high road to climb there.
That is why at this point, as opposed to when he was at SEGA, an open mind to learning newer strategies to help create the better game could have been utilized.
As in like I said, the difference between saying "aw shit we dont have this extra few million to hire this mocap studio" vs Epic telling them "well you know you dont have to; previously we worked with Ninja Theory and they developed a strategy to get around that financial hurdle"

I think that's pretty much what happened, it's just that regardless, when you have a specific vision, unique systems, a particular art style and a very specific story to tell, it restricts quite a lot of what you can and can't do. It might not seem like it, but budgets are very much tied directly to exactly what makes your game unique. For example, NPCs in Shenmue games have always had that 'scheduling' feel to them, whereby each NPC actually seems to have a living schedule as opposed to just runs on a loop constantly. That alone can drastically affect your remaining budget simply dependent on how many NPCs you may have or how dense the area where they will navigate is, how detailed each model is, their respective materials and shaders, mesh properties, physics simulations (some will have complex hair simulation and cloth simulation, some won't etc).

The internal systems developed for this game compliments those found in the original two games, and also likely to the disagreement of many people, that includes the battle/combat system, however obviously some animations/move sets were ultimately restricted due to monetary budgets, it still manages to improve upon some things from the previous design of the underlying system.

I highly doubt that YsNet and co., are not aware of any of the commonly used practices, such as the aforementioned, but this still is only useful if you're making a game like Hellblade. Things are very different now, but the information that's out there is enormous.

In short, the development of any game is very much a different situation for every instance, yes there are similarities and there are some things that can be carried over between games etc, but this doesn't excuse anything. Both games have very different scopes.

 

[Budgiecat]

I think that's pretty much what happened, it's just that regardless, when you have a specific vision, unique systems, a particular art style and a very specific story to tell, it restricts quite a lot of what you can and can't do. It might not seem like it, but budgets are very much tied directly to exactly what makes your game unique. For example, NPCs in Shenmue games have always had that 'scheduling' feel to them, whereby each NPC actually seems to have a living schedule as opposed to just runs on a loop constantly. That alone can drastically affect your remaining budget simply dependent on how many NPCs you may have or how dense the area where they will navigate is, how detailed each model is, their respective materials and shaders, mesh properties, physics simulations (some will have complex hair simulation and cloth simulation, some won't etc).

Deady Premonition did the scheduling thing too. Far less NPCs but yeah. Also that game ran like ass..so yeah. But still, an example of it being done. Probably inspired by Shenmue no doubt. Other developers who do open world games though? They dare not touch that option.
If GTA V with its 250 million budget did the scheduling thing as well as fully voicing everyone in the game....well lol the budget. So yeah them and many others use the run on a loop method instead.
Moreover, in Niaowu in this game, you find a bit more of that run on a loop method for some of their NPCs so clearly Yu was working with less here in less time to do so.

But this has nothing to do with what I was specifically talking about which was mocap

The internal systems developed for this game compliments those found in the original two games, and also likely to the disagreement of many people, that includes the battle/combat system, however obviously some animations/move sets were ultimately restricted due to monetary budgets, it still manages to improve upon some things from the previous design of the underlying system.

There's no throws in this game. Probably more expensive and costly to add them in due to not having access to the VF engine. However, it looks like a lot of these moves were hand animated. Had they had access to inexpensiveness mocap procedure that NT came up with, they could have possibly circumvented this limitation and included them.

I highly doubt that YsNet and co., are not aware of any of the commonly used practices, such as the aforementioned, but this still is only useful if you're making a game like Hellblade. Things are very different now, but the information that's out there is enormous.

Disagree. Its motion capture. There's motion capture in Shenmue 3. Its just used sparingly. Because normally its expensive. And its expensive because they're going about it the common conventional way of approaching how to use it. Ninja Theory found an uncommon way to get around the problem of its expense by only spending a few thousand dollars. THAT is what I am saying. This has nothing to do with how 'different' Hellblade is from Shenmue 3. Both use tools. Mocap is simply a tool a means to an end to help portray your game. Its all about efficiency. This is why I brought up some of those other games because they spent millions yet their games still had problems. Because they werent being efficient. They were just throwing money around.

 

[LemonHaze]

Disagree. Its motion capture. There's motion capture in Shenmue 3. Its just used sparingly. Because normally its expensive. And its expensive because they're going about it the common conventional way of approaching how to use it. Ninja Theory found an uncommon way to get around the problem of its expense by only spending a few thousand dollars. THAT is what I am saying. This has nothing to do with how 'different' Hellblade is from Shenmue 3. Both use tools. Mocap is simply a tool a means to an end to help portray your game. Its all about efficiency. This is why I brought up some of those other games because they spent millions yet their games still had problems. Because they werent being efficient. They were just throwing money around.

You still seem to lack any understanding of games development. Funny how you never even answered my initial question and went straight to the "Are you?" response.

I wonder why you're suggesting they use Ninja Theory's method of motion capture, as opposed to say the likes of:

 

[Tomato Convenience Store]

@Budgiecat Is the argument you're making regarding Hellblade strictly about the mocap? A few things to consider:

1. All of the cutscenes, and essentially the entire game is based around 1 character so it's much easier to record and implement mocap in this case.
2. The actress was an on-site employee of the studio, so she's available 40 hours a week and there's no additional booking to bring her in, so that alone makes the mocap a lot cheaper.
3. Ninja Theory used Unreal since Enslaved, starting with UE3. They had experience using mocap on Andy Serkis and have the ability to streamline their process due to experience using the tech.

 

[Budgiecat]

@Budgiecat Is the argument you're making regarding Hellblade strictly about the mocap? A few things to consider:

1. All of the cutscenes, and essentially the entire game is based around 1 character so it's much easier to record and implement mocap in this case.
2. The actress was an on-site employee of the studio, so she's available 40 hours a week and there's no additional booking to bring her in, so that alone makes the mocap a lot cheaper.
3. Ninja Theory used Unreal since Enslaved, starting with UE3. They had experience using mocap on Andy Serkis and have the ability to streamline their process due to experience using the tech.

I'll give you 2 and 3 but not so sure I'll agree on #1. I mean, Shenhua was just as important in scenes as Ryo was due to their 1 on 1 convos and how frequent they were.
Also, I think they coulda still got around on #2. Just schedule and set up the whole thing differently of course.
Not sure if using UE3 means you know how to use UE4 perfectly but I dunno so I'll yield on that. In any case, even setting UE aside, the simple matter of figuring out how to inexpensively create your own mini mocap studio using GoPros, plantpots and other clever shortcuts they figured out so they could spend a few thousand instead of a few million was smart thinking.

 

[Tomato Convenience Store]

I'll give you 2 and 3 but not so sure I'll agree on #1. I mean, Shenhua was just as important in scenes as Ryo was due to their 1 on 1 convos and how frequent they were.
Also, I think they coulda still got around on #2. Just schedule and set up the whole thing differently of course.
Not sure if using UE3 means you know how to use UE4 perfectly but I dunno so I'll yield on that. In any case, even setting UE aside, the simple matter of figuring out how to inexpensively create your own mini mocap studio using GoPros, plantpots and other clever shortcuts they figured out so they could spend a few thousand instead of a few million was smart thinking.

The point with #1 is that if you're mocapping a cutscene you don't have to worry about booking multiple actors in a session, or figuring out the logistics of how multiple actors interact with each other in a scene. For #2, that means much higher costs since you need to book actors for a session at an additional cost each time you record mocap. For #3 it's incrementally building upon your experience. They were able to simplify their mocap process since they've been doing it for so long. If they were starting out from scratch and didn't have experience with mocap whatsoever it would have taken them longer to develop their process.

 

[Budgiecat]

The point with #1 is that if you're mocapping a cutscene you don't have to worry about booking multiple actors in a session, or figuring out the logistics of how multiple actors interact with each other in a scene. For #2, that means much higher costs since you need to book actors for a session at an additional cost each time you record mocap. For #3 it's incrementally building upon your experience. They were able to simplify their mocap process since they've been doing it for so long. If they were starting out from scratch and didn't have experience with mocap whatsoever it would have taken them longer to develop their process.

Conventionally, mocap was expensive because you're hiring:
one person to do the mocap
one person to voice them in the studio room (separate from mocap)
A huge expensive studio you rent out that specifies in mocap

IKEA WARDROBE POLES
$20 AMAZON LIGHTS
LOW END CHEAP MOCAP CAMERAS
GOPROS
COATS and regular GYM MATS for SOUND DAMPENING
cheap wireless headphones for dialogue instruction
4 people in set: (audio guy, mocap technician, director (filmer), and actor/actors)
etc

These are all clever and frugal ideas that coulda been adopted

 

[Ren]

Conventionally, mocap was expensive because you're hiring:
one person to do the mocap
one person to voice them in the studio room (separate from mocap)
A huge expensive studio you rent out that specifies in mocap

IKEA WARDROBE POLES
$20 AMAZON LIGHTS
LOW END CHEAP MOCAP CAMERAS
GOPROS
COATS and regular GYM MATS for SOUND DAMPENING
cheap wireless headphones for dialogue instruction
4 people in set: (audio guy, mocap technician, director (filmer), and actor/actors)
etc

These are all clever and frugal ideas that coulda been adopted

1) Kid Nocon is Korean not Chinese. You can actually talk to him.
2) Look into Yu Suziki latest games.. https://en.wikipedia.org/wiki/Yu_Suzuki
He been making mediocre mobile games these years since Sega got forced to Merge with Sammy.
3) Ninja Theory had a Team already and they had resources they owned from past games AAA games they made.
4) What did Yu had? Just resources from Shenmue from 2000...
5) Shenmue doesn't have proper motion capture part of the reason they didn't made throw moves into the game.

 

[shredingskin]

You can have mocap with a kinect and a modern cellphone (hell, you can photoscan anything with that even).
But there's not much point in doing mocap for a game like Shenmue apart from some combat scenes:

- Lipsync: probably close to millions of lines in 2 languages.Is just smarter to do a phoneme and dynamics conversion.
- Artstyle: To have good results you need many blendshapes on the face, and if you want to get any bang for your buck you'll need wrinkmaps, that doesn't fit the anime-like aesthetic.
- Time: even if you can have cheap mocap data, the data is not optimized for game development because it has thousands of garbage that needs to be cleaned up, and can be more time consuming than doing it by hand (and that's not taking into account the time getting the takes).

But I'll give BudgieCat one and that is that the technical ability of the team was lacking, they weren't even able to make consistent looking characters, they had to re-rig the characters mid development, the idle animations just look off in many cases, I don't think adding another layer of complexity would have done anything good to the game.

 

[Budgiecat]

1) Kid Nocon is Korean not Chinese. You can actually talk to him.
2) Look into Yu Suziki latest games.. https://en.wikipedia.org/wiki/Yu_Suzuki
He been making mediocre mobile games these years since Sega got forced to Merge with Sammy.
3) Ninja Theory had a Team already and they had resources they owned from past games AAA games they made.
4) What did Yu had? Just resources from Shenmue from 2000...
5) Shenmue doesn't have proper motion capture part of the reason they didn't made throw moves into the game.

My point was NT made public that you can make due and acquire said mocap equipment on the cheap instead of spending goads of money on flying out to expensive mocap studios in L.A., London or New Zealand. You and I could literally recreate what they did mocap wise for a couple thousand dollars. Not millions. They gave out the blueprints and Epic knew of it because they were involved. Epic coulda shared that knowledge to Yu.

 

[Budgiecat]

they weren't even able to make consistent looking characters

They had two art directors with vastly different opinions on where the art direction should go. One wanted to keep it classic the other wanted it more cartoony/anime (like Disney meets Baki the Grappler). Yu said they compromised in the middle. I woulda just....not hired the 2nd guy lol

I also felt it cheap to not show the masters actually doing legit martial arts moves (Sun, Bei, fat guy forget his name (Hsu?).
Sun just hopped around evading, Bei did basic arm slaps effortlessly, and fat Pixar panda guy just...moved his weight around...
Inconsistent since some characters actually were performed well (the first henchman for Lan Di. 2nd guy was mimicking Bruce Lee for the lols, and not sure wtf third guy did other than muscle grab Ryo). In fact other than that first guy, there wasnt anyone story plot wise who really showed actual martial arts prowess in battle. Red Snake boss was supposed to be an expert at Xin Yi Lui He Quan but his moves were just brawling strikes. He ended Ryo with a side kick. But there were mocapped moves of minor characters doing elaborate forms practice, so it was odd..

Even Lan Di's moves were pretty meh. Side kick was nice, as well as grab arm to double round kick to the face, but the basic as hell arm twist? Thats what ended Ryo? Bei at least pulled a Chin Na joint lock on him albeit it also kinda basic but still.

Of course I still liked the game overall but there was clearly some things that werent directed right here. Yu wore two hats for this game: Producer and Director. There's a reason those two positions are usually given to two different people...

 

[mjqjazzbar]

My point was NT made public that you can make due and acquire said mocap equipment on the cheap instead of spending goads of money on flying out to expensive mocap studios in L.A., London or New Zealand. You and I could literally recreate what they did mocap wise for a couple thousand dollars. Not millions. They gave out the blueprints and Epic knew of it because they were involved. Epic coulda shared that knowledge to Yu.

Epic probably has no interest in just sharing information with developers out of the goodness of their hearts. It’s up to Ys Net to do the R&D as well as hire experienced staff. UE4 is just a tool.

 

[mjqjazzbar]

The idle animations in this game barely make sense. Many NPCs look like they’re painfully writhing against their will or talking to themselves.

When you go through the Moby Games credits, you really will struggle to find many staff members who have a lot of experience.

Yu had to build a team from scratch and likely had to train many inexperienced people on a new engine.

They certainly had their work cut out for them...

 

[mjqjazzbar]

The other art lead worked on DQXI, which makes sense:

Shigeo Niwa - MobyGames

Game credits, biography, portraits and other game industry information for Shigeo Niwa - MobyGames

www.mobygames.com

 

[LemonHaze]

Everyone in this thread is acting straight outta the 2000s with the whole 'dude, why didn't they ask for that top-secret information from that big company LOL'.

Things don't quite work like that.

The engine is open source, even Hellblade's fabled mocap solution is publicly documented. Either way, saying "budget is NOT an excuse" is categorically incorrect and invalid, they had a restrictive budget, new processes to learn (which is entirely unavoidable, no matter how much information they may have been 'given'), new staff to train and ensure they deliver on your vision etc. These are all reasons as to why Shenmue III might not be 'completely up to scratch' in your eyes, but you cannot pretend like it's as bad as some of these points make it out to be.