Unity Abandoned Some Modeling Agency [v0.10.4e] [T Valle]

[ScottyLQ]

I... don't think Hogwarts Legacy was made in Unity. Not to mention you're talking about a multi-million dollar game from a big studio, is it really surprising that it is better optimized than this? Come on, bud.

 

[Farshid]

can you post the character of her please

well not exactly the same character but here goes nothing

 

[Farshid]

I... don't think Hogwarts Legacy was made in Unity. Not to mention you're talking about a multi-million dollar game from a big studio, is it really surprising that it is better optimized than this? Come on, bud.

yea it was made by unreal engine i guess which is more intensive

 

[Deleted member 1654379]

Although at least in this case your CPU isnt restraining you much


And the custom rooms from manniakk are (sadly) much worse in that regard. Definitely all CPU tho (and bad optimization). Lower resolution does nothing.
Would be sweet if the custom rooms could be improved in that regard otherwise its probably locked 30fps or stuttering and both of these are pretty meh

 

[Farshid]

Although at least in this case your CPU isnt restraining you much
View attachment 2455983

And the custom rooms from manniakk are (sadly) much worse in that regard. Definitely all CPU tho (and bad optimization). Lower resolution does nothing.
Would be sweet if the custom rooms could be improved in that regard otherwise its probably locked 30fps or stuttering and both of these are pretty meh

that performance for a game inside a cube and some lights and shadow is just pure garbage 80 fps lol xD

 

[b00marrows]

I cant remember now, but i think i had a discussion before about the games performance.
I think it was related to the physics? Both thread limited and complex (softbody tits?)

 

[Hephannon]

View attachment 2453699

NIIIIIIIIIIIIIIIICE!!
WHERE ??

 

[Dfgjjkl]

Let's consider what the game basically is: 2 characters interacting in a small room. I understand there's a lot of physics involved but the optimization is still terrible.

I don't really like the direction the game has been going to, at the start I thought you're gonna be able to have your own company, compete with others, that the women that you rate well would later be available to be hired by you etc.

Instead it's going in the direction of just another 3D sex simulator, and there's plenty of those made on Unreal Engine, a better engine.

I don't know, maybe I misjudged what the dev wanted to make, I felt that he wanted to make a game where the agency aspect is integral, but instead it's very bare bones in its current state

 

[Pchit]

SMA 0.9.7a was released this morning. Does anyone have access to the changelog to let us know please?

 

[COMMONFACTORY]

View attachment 2451590
This game needs some optimization can't even hit 60 fps my gpu and cpu kept around 40 percent usage. I want that smooth 60 fps animations but can't get it

What is the software for the benchmark?

 

[Gabriele D'Annunzio]

Has the dev ever said anything about adding a futa mode or variant to the game?

 

[Thora]

SMA 0.9.7a was released this morning. Does anyone have access to the changelog to let us know please?

Attached below.

So far Monkey is testing fine. I will try to add relief control in the next version as well, but the focus remains on clothing.

-T-

SMA 0.9.7a Changes

• A new emotion (relief) has been implemented to act as a pain buffer. As a result, dealing with the models' increasing pain is less difficult.
• Bribing a model increases her resistance to decreasing the relief emotion when these bribed interactions occur.
• The models will have alterations in their emotions according to their oxygen saturation (I can't give more explanations about this.., you know the r-word stuff.)
• The visual appearance of the worn-down holes varies depending on the size of the object that wears them.
• When the player undresses the models, they may cover their private body parts.
  

Fixes

• Holes were not decreasing pain gain as they wore down.
• Even after accepting removal of certain items of clothing, the models did not remove them.
• The shapes of the swimsuit, stockings. and sexy underwear have been improved.
• Some enhancements to the skirt skinning
  

Notes

• When the relief emotion (thin purple bar) is at its maximum, its bar is invisible, and when it is at zero, its bar is at its maximum. This was done to ensure that the player only sees important information on the UI.
  

Known issues

• Eyebrows won't display under her glasses.
• Skirts still need more skinning adjustments.
• The relief emotion description is still missing from the "How to Play" menu.

 

[b00marrows]

Attached below.

So far Monkey is testing fine. I will try to add relief control in the next version as well, but the focus remains on clothing.

-T-

Any thoughts on the games performance? i think we discussed before the main issue was related to physics? cant remember.

Any extra possible additions you could add to monkey for performance so some of these users might get a better experience?

 

[Farshid]

What is the software for the benchmark?

It's msi afterburner which comes with rivatuner

 

[Thora]

Any thoughts on the games performance? i think we discussed before the main issue was related to physics? cant remember.

Any extra possible additions you could add to monkey for performance so some of these users might get a better experience?

While Unity is not a very efficient engine compared to say Unreal, the main factor is 'how' you use the engine, and how much time you devote to optimizations.

I can't attach a profiler on the code, so it is impossible for me to know for certain how much work is being done each frame, and how much of that can be optimized.

Based on what I'm seeing in code though, it does appear to do an enormous amount of work each frame. So given time and focus this can likely be optimized.

In regards to Monkey, there isn't much I can do to improve core performance without essentially rewriting how the game works. You can already disable most GPU features and the game should auto adjust profile based on your hardware.

I suspect the core issue is the coupling of the simulation and render threads. My "guess" is the dev decided to force the simulation running at high speed (120fps+) because the physics solvers would not interpolate correctly (i.e. hand goes through the breast). Fixing that is not trivial, but would alleviate a lot of the performance issues people are seeing. Again, just a guess.

-T-

 

[b00marrows]

I suspect the core issue is the coupling of the simulation and render threads. My "guess" is the dev decided to force the simulation running at high speed (120fps+) because the physics solvers would not interpolate correctly (i.e. hand goes through the breast). Fixing that is not trivial, but would alleviate a lot of the issue people are having. Again, just a guess.

This is similar to my thinking too.
I must admit, whatever it is does work.
The most similar demanding game i can think of is VAM. Which is a glitchy and unstable mess at the best of times.
The physics simulation being extremely fast makes sense for stability.

Thanks for the input, i might do some digging into the software to see if i can crack into anything revealing.
I'm no programmer or real moder but who knows what i might find.
I should also do some performance checking so i know what I'm looking at.

Probably wait for the new monkey and release tho.

 

[Thora]

The most similar demanding game i can think of is VAM.

I briefly looked at VAM, and I couldn't find a way to interact freely with a model. You essentially select a set, than a 'pose' and have limited control over the animation speed that gets played.

From a technical standpoint, pinning two character in a known pose, then animating them using premade animation is orders of magnitudes simpler. The majority of the work is about creating the content. The CPU has little to do other than preparing a frame to the renderer.

All games I've reviewed recently use the same canned looped animation approach - this include HS, and ALL of the recent Unreal games that seem based on metahuman and free assets from the unreal store. If am wrong, or missed a title please pm me and I'll be happy to take a look.

Games do this because trying to solve n-body of different size and shape having interactive intercourse with proper collision and deformation during penetration like SMA does is stupidly hard.

-T-

 

[MavisFeatherlight]

SMA 0.9.7a was released this morning. Does anyone have access to the changelog to let us know please?

can someone upload the new update here please?

 

[b00marrows]

Games do this because trying to solve n-body of different size and shape having interactive intercourse with proper collision and deformation during penetration like SMA does is stupidly hard.

Yes, at the end of the day Softbody Physics is no small feat. Its not a well documented or established technology and has little use.
VAM is much like SMT in the technology department, having propper morphing and physics of breasts, ass, anus, mouth/lips and vagina.
But VAM is extremely tedious, takes a hell of a lot of time to set up and use. To add to the issue is unstable and the physics can cause the characters to completely spaz out into a spaghetti mess.
Similar to what happens in skyrim from time to time.

SMA is likely getting around that hurdle with the cranked up physics sim and slow movement. Tho its more theory than anything.

 

[Kraos]

Attached below.

So far Monkey is testing fine. I will try to add relief control in the next version as well, but the focus remains on clothing.

-T-

SMA 0.9.7a Changes

• A new emotion (relief) has been implemented to act as a pain buffer. As a result, dealing with the models' increasing pain is less difficult.
• Bribing a model increases her resistance to decreasing the relief emotion when these bribed interactions occur.
• The models will have alterations in their emotions according to their oxygen saturation (I can't give more explanations about this.., you know the r-word stuff.)
• The visual appearance of the worn-down holes varies depending on the size of the object that wears them.
• When the player undresses the models, they may cover their private body parts.

Fixes

• Holes were not decreasing pain gain as they wore down.
• Even after accepting removal of certain items of clothing, the models did not remove them.
• The shapes of the swimsuit, stockings. and sexy underwear have been improved.
• Some enhancements to the skirt skinning

Notes

• When the relief emotion (thin purple bar) is at its maximum, its bar is invisible, and when it is at zero, its bar is at its maximum. This was done to ensure that the player only sees important information on the UI.

Known issues

• Eyebrows won't display under her glasses.
• Skirts still need more skinning adjustments.
• The relief emotion description is still missing from the "How to Play" menu.

So, no fix for the auto-rating? I have given up on the complete profile maker getting fixed, but if auto-rating is messed up, it makes ever trying to play the game properly (minimal cheats) mute.