COMMUNITY - FORUMS - GENERAL DISCUSSION
More on the SBS Development process

Snipehunter wrote: "A process that might take several months of total time to complete can be done in a third or a quarter of that time as a result of that parallel production and so, even though we’re a small team, we can maintain a pace similar to the larger-teamed MMOs I’ve worked on in the past, games that had teams that were literally 3 or 4 times as large as the team here at Soulbound."

"It might not be easy to see from the outside, but from where I sit on the inside? It’s the sort of smart planning that makes me think why didn’t we do this on that other game?! -- and I’m not going to lie, knowing we’re doing this when others never thought to do it?"

I decided to extract this from the recent Shiny thread to not clutter it with this discussion.

As a long time software project manager and developer (but notba game dev) I am always interested in ways to improve on a team's velocity and it sounds as if Snipehunter and team may have hit on something truly new.

I'd love to learn more about this, but I suspect I'd have to go work for SBS to really get more details so I guess I'll have to wait until Snipe writes his memoirs 15 years down the road.

But here's the thing, if these techniques truly are providing such efficiency, it's not unreasonable to see evidence of significant progress on multiple game systems, even some of the more complex ones, including seeing them in action in the actual game world.

The parkour demo was entertaining, but it was just a bare bones demo. The original Pax East video from back in 2016 looked more amazing even though that apparently wasn't the actual game world either.

We'll see how 2020 turns out, hopefully with a lot more visual evidence of actual in game footage of many of the systems in progress.


You must all be quick to listen, slow to speak and slow to get angry. JAMES 1:19 NLT

...
11/8/2019 10:05:43 PM #1

This may be derailing the topic a bit, but that is a really...ballsy... statement from Snipe given that virtually every release they have had, ever, has been late/delayed.

I love Snipe, I think he is amazing, but I just can't reconcile that statement with reality.

EDIT: I'll add a caveat to this. It may be that the process he is referring to simply doesn't apply to what has been delivered up to this point.

It may be that it has taken time to get their process down to the point it can be impactful, given they are a new studio.

However, when it comes to publicly demonstrable progress, I just haven't seen much evidence of this revolutionary approach they have referred to over the years.

Granted, it may just be that, without it, things would have been even more delayed than they were.


Imgur

11/8/2019 11:15:02 PM #2

Posted By Marovec at 4:05 PM - Fri Nov 08 2019

This may be derailing the topic a bit, but that is a really...ballsy... statement from Snipe given that virtually every release they have had, ever, has been late/delayed.

But both things can be true because missing dates has more to do with estimating effort (and communication) than the kind of productivity boost working in parallel can achieve (in the way Snipe described).


11/9/2019 12:49:28 AM #3

Posted By Hieronymus at 6:15 PM - Fri Nov 08 2019

Posted By Marovec at 4:05 PM - Fri Nov 08 2019

This may be derailing the topic a bit, but that is a really...ballsy... statement from Snipe given that virtually every release they have had, ever, has been late/delayed.

But both things can be true because missing dates has more to do with estimating effort (and communication) than the kind of productivity boost working in parallel can achieve (in the way Snipe described).

I do get what you are saying, and I alluded to that in my edit.

However, effective project management (or estimation of effort and communication, as you put it) is key to effectively executing any "process".

To me, they go hand-in-hand.

Maybe they shouldn't.


Imgur

11/9/2019 4:06:25 AM #4

I think the effectiveness of the approach that Snipehunter described depends very much on the context. In the case of CoE, the significant aspect of the context is a graphics interface that doesn't care what the actual graphics rendering geometries are in order to function. I don't think this is conceptually revolutionary, but it may be a new wrinkle in game development.

Conceptually, the scheme goes back to the 1970's, to the best of my recollection, and perhaps earlier. I was certainly trying to use it in the 1980's, though not always with the best success. It takes some skill and experience to successfully design truly transparent and portable interfaces, and there are many temptations to short-cut the development of them in order to achieve smoke-and-mirror demonstration capability earlier.


11/9/2019 5:34:04 AM #5

I barely looked at the "shiny" today beyond seeing the image. I don't know, I just keep hoping they'll drop something big on us soon and redeem a year that otherwise was very ... underwhelming.

11/9/2019 5:59:21 AM #6

I don't understand...this sounds like Project Management/ Production Coordination 101. Don't all programming studios work on the separate pieces in parallel?

Otherwise the teams would constantly be waiting for the most time-consuming and critical piece to catch up--while everyone else is sitting around waiting for them to finish.

That would be a crazy way to conduct business.


11/9/2019 9:30:03 AM #7

Posted By Runebrand at 06:59 AM - Sat Nov 09 2019

(...)Don't all programming studios work on the separate pieces in parallel?(...)

The way I interpret (and I have no clue if I didn't just misunderstand him) what he stated is that you'd usually work in sequential order. That said, in reality, it is still parallel, because with proper planning ( coughs ), the artists can start their work while the back-end and front-end engineers are working on non graphic related 'basics'. Hence, at the point when the first graphical assets are 'required', the artists would have finished the deliverable (I am going to not determine between the different steps of rigging and animation, since there is very little if any benefit in using a low poly version for that process) and already working on the next one. The only bottleneck is when there is any delay in that process, due to unpredictable difficulties or bad planning. Then again, the term 'programmer art' isn't exactly new and that is basically the core concept of PrElyria's low polygon assets.

The (only) benefit I can see is not in speeding up production, but in getting feedback earlier in the development cycle.


Sage willing to help with Purity (if you spot me on Discord and have some Plague on your account that could be nullified with a trade, drop me a message on Discord)

11/10/2019 1:34:34 AM #8

Posted By Runebrand at 9:59 PM - Fri Nov 08 2019

I don't understand...this sounds like Project Management/ Production Coordination 101. Don't all programming studios work on the separate pieces in parallel?

Otherwise the teams would constantly be waiting for the most time-consuming and critical piece to catch up--while everyone else is sitting around waiting for them to finish.

That would be a crazy way to conduct business.

It seems that what's claimed to be different about the SBS process is that it uses ad hoc modifications of low-resolution graphics to provide testing grounds for the development of all the various character actions and animations that are being worked on by the programmers. Evidently, waiting for high-resolution graphics updates by artists working in separate swim lanes is a common critical-path obstacle in game development. I've seen similar obstacles happen in the machine-control environments where I spent most of my career.


11/10/2019 2:59:58 AM #9

Does any of this pace also have to do with the fact that all the people actually building this game could all sit in my living room at the same time?

Even with this system to "speed things along", without more than one or two people doing each task it's going to take a loooong time to see any tangible results.

For example: It's taken how many weeks since a few thousand folks named their settlements and many still haven't gotten notifications about them being good or bad? Really? Is there one intern checking ten names off a list a day? An hour? At that pace the sun will have gone out before they finish.

To take that long to do such a simple process doesn't inspire a lot of confidence that they can run a whole world.

I do love the look of The Sailfin.

A whole lot.

But that now gives us a grand total of (approximately) 10-15 unique creatures after how many years of development now?

Unless there is a whole lot more they haven't shown us?

And if there is, it might just be time to start letting somebody see the work before the year ends and the community starts assuming that this is morphing into some game that nobody but the designers (and a few whales) will ever see in action.


We Are The Many... We Are The One... We Are THE WAERD !!!

11/10/2019 4:39:52 AM #10

Posted By kajoreh at 6:59 PM - Sat Nov 09 2019

But that now gives us a grand total of (approximately) 10-15 unique creatures after how many years of development now?

A lot more than that.... I know of at least the following:
battlecat
bees/cardinal bees
black raven
canis rabbits
carrier pigeons (or similar)
cats/black cats
chickens
conifer rats
cows
crocodile/aligator'ish
daemons
deer
direwolves
dogs/herding dogs
dovecotes
dryas elk
falcons
flower-cup porcupines
foxcelot
foxes/domino foxes
giants
goats/mountain goats/wooly mountain goats
horses/purebred coursers
krakens
fish/leatherfish
leffits
lemurs
liches (passed off as myth)
mann
monkies/versary monkies
mules
ogres
otterbears
owls/faceless owl
oxen
phoenixes
pigs
pteroguins
rabbits
rocs
sailfin
serpentaurs
sheep
shrubland wildcats
silkworms
song birds
tentakeels
trisons/war trisons
trollmouth salamanders
ursaphants
vampires (passed off as myth)
wildebeast
wolves


11/10/2019 5:42:20 AM #11

Most of those are drawings of the animal. I think Kajoreh means models of animals.


11/10/2019 7:18:46 AM #12

Posted By Vucar at 06:34 AM - Sat Nov 09 2019

I barely looked at the "shiny" today beyond seeing the image. I don't know, I just keep hoping they'll drop something big on us soon and redeem a year that otherwise was very ... underwhelming.

Something big gameplay wise yes I really want to see something big and be able to be involved in Prelyria. These Tiny bits of screenshots and small video bits don't do much for me . A decent video of the actual game is what i am waiting for.

*edit: lol someone disliked the fact that I would really like to see some actual gameplay. So there is someone who dislikes to see game play vids and dislikes any personal wishes anyone has.


11/10/2019 5:41:58 PM #13

What SBS does different from what i have gathered is that prelyria, the "throwaway" prototype, has much higher fidelity and can be used to test a much wider range of stuff than your ordinary "greybox" iteration.

So prelyria is different not because it is low fidelity, but because it is a middle ground that enables them to test stuff for a more accurate "feel" of systems. Meaning the cost of throwing away or completely changing stuff that feel wrong is much lower then if you had to wait for the full production version to realise your errors.

So, considering the small size of the team and the amount of new complex systems to be developed and tested, Snipe may very well be right that they are moving at good speed. It's hard to compare really without knowing details.

I also have a feeling that prelyria is a way to ensure high quality gameplay more then just pure speed. When the cost of iterating goes down, the developers will iterate more and eat up that time. In other words we get a better game for the same amount of dev time.

BTW recent delays have been more about world building, DSS and such, not prelyria. That was postponed by a decision to fill the world with more systems and content before alpha testing starts. However i really hope SBS prioritizes getting players into the game before iterating and polishing more as the patience is running thin around the communities.


11/10/2019 6:17:35 PM #14

@Protey

While I agree those are the fauna they have listed, or awarded, or sold, that does not mean they are actually "developed" (as in, have a completed in-game model, animation, AI, etc.).

As with most of what we "know" about CoE, they are largely conceptual.

Which seems to be the point people are trying to make.

I am still a HUGE fan of what CoE is trying to do, but as eager as I am, it still sits firmly in the "I'll believe it when I see it" category.


Imgur

11/11/2019 8:59:32 AM #15

While it may be faster to develop this way it comes with drawbacks. Less people means less potential talents/less people coming up with a different way of doing things.

How I understanded from the shiny it was more a way to get momentum in development and not get slowed down by hiccups in other parts of the chain.


...