Would Second Life Benefit from an Engine Rebuild?

[Rachel1206]

1 hour ago, Qie Niangao said:

It's not only the cost of developing and deploying a change, but also the share of revenue lost when users are left behind because, for example, their PCs are so old they can't handle Vulkan, which I gather LL found to be a scary large share of their customers. (I don't know whether they've decided what to do about that.)

[ETA: This from Inara Pey's blog:

I'm realizing that my current PC might be over five years old. Not sure how to know if it would handle Vulkan, but may be a moot point anyway.]

Kronos lists supported hardware for the different Vulkan versions here: Conformant Products.

In general Nvidia's Kepler series, Radeon HD 7000 or newer can use Vulkan, provided drivers updated. VRAM of 4 GB or more a must, but again driver update dependent ( for example some 2 GB VRAM GPU will run Vulkan on Windows 10 but not on Windows 7)

Are we safe in the woods then, if our hardware is on the list? Well, now it gets complicated. NVidea have made professional GPU versions, which do not support Vulkan.

BTW, first time I seen "Undisclosed Product" appear on a list of conformant products! Guess Kronos is under a NDA on new hardware, but why list it public then.

 

Edited December 7, 2020 by Rachel1206

[Ardy Lay]

Excerpt from SecondLife.log

2020-12-06T18:46:34Z INFO # newview/llviewerstats.cpp(616) send_stats : Misc Stats: string_1: 'Vulkan driver is detected' string_2: 'Unused'

I suppose this might be telling.

[Lyssa Greymoon]

3 hours ago, Qie Niangao said:

I'm realizing that my current PC might be over five years old. Not sure how to know if it would handle Vulkan, but may be a moot point anyway.

I'd bet almost every PC that would have a problem with Vulkan is already having problems with SL. 

[Ardy Lay]

25 minutes ago, Lyssa Greymoon said:

I'd bet almost every PC that would have a problem with Vulkan is already having problems with SL. 

Seems likely.  I have a few old computers that fit that description.  Some that have Vulkan drivers suffer self-inflicted packet loss when trying to render some SL scenes so I call them too slow.  If there are a lot of “course updates”  coming in and the frame rate is low they let the packets hit the floor.  They also either take from several minutes to forever to render some mesh avatars and miss a lot of animations and sounds.  The computer I use today does not have any of these issues.  It is rather annoying to see people blaming Linden Lab and/or their ISP when they are using dimestore notebooks and computer history museum exhibits to access Second Life.

Edited December 7, 2020 by Ardy Lay

[Qie Niangao]

24 minutes ago, Lyssa Greymoon said:

I'd bet almost every PC that would have a problem with Vulkan is already having problems with SL. 

"Up to a third..."

Sounds about right! 😛

(Turns out my SecondLife.log file has the same note @Ardy Lay observed, and this GTX 660 is recent enough but only has 2 GB VRAM -- and my current SL experience probably would be enhanced by more.)

[Ardy Lay]

I think the last thing Second Life Viewer needs is yet another wrapper, as people often suggest when dealing with changes.

[Nikkesa]

I think it would entirely benefit. Heck, even sansar was looking to be something like that - integration with marvelous designer was genius, but the user interface and new-user-UX was so horrid it's not even funny.

I think it would HEAVILY benefit, but definitely not if it was done by linden labs. I think you'd need to hire an entire new team (obviously take some of the lindens, no idea who they are), ditch the board of governors, have a group of leaders who are passionate about possibilities and metaverse, and less passionate about shareholder-profits, have some rich billionaire who believes in the cause fund it with a few billion dollars as investment without expectation of profit-generation, and voila, you could probably have something pretty darn amazing after a good 8-12 years of development with constant review of upcoming technologies, cloud-capabilities, infrastructure design and several layers of abstraction and modularity to let them keep things updated.

The problem is that there are no billionaires willing to just casually throw money at something that wont guarantee profits in return, while also ignoring the fact that they could have spent that several billion dollars on building homes for the homeless, or investing in education, infrastructure etc in developing countries - stuff like that.

The other problem is that when someone like linden labs makes something like sansar, well, all you have to do is go to glassdoors and check out some the employee reviews of the company to get a fundamental understanding of how it really went wrong. Also the fact that we only see palpable updates to second life every 2-5 years... I'm hoping eventually they will buckle down and make something like looks and feels fundamentally awesome to play.

I feel like they'd get a lot more users if it was possible to switch between making it feel like a smooth first person shooter to a smooth third-person ARPG-style fast-paced combat-style thing, then back to chat simulator as easily as pressing buttons, with each of the different interface and input modes being a user-designed and customized experience.

More complicated, scripted animation controls, a proper 2D user interface API for people to make decent HUDS. I feel like the average person starts up second life and is immediately turned off by the fact that the avatar has the same default animations as they did in 2004, and moves around feeling like it did back then too. If they just made it fun to play the game - made it satisfying to do something as simple as run around without glitching out and crashing during a sim crossing, or even having a palpable minute amount of lag during a sim crossing.... If you started them off in a smooth, optimized, cool-looking city with a decent vehicle and showed them the possibilities, way more people would get into this game.

Edited December 7, 2020 by Nikkesa

[Nick0678]

1 hour ago, Nikkesa said:

The problem is that there are no billionaires willing to just casually throw money at something that wont guarantee profits in return..

We should do a fundraiser then!

[Elise Banx]

NO  

[Arielle Popstar]

At the end of the day it seems to me it would be nothing more then trading in the existing problems for an entirely new set of problems.

[Janet Voxel]

On 12/5/2020 at 6:17 AM, MrBalloone said:

Would Second Life benefit, If Linden was willing to do so, From a complete rebuild. Fixing the numerous problems with numerous aspects of SL. Like LSL. LSL, why doesn't LL allow the use of JavaScript, Python or what might be SL's codebase; C#, In addition to LSL (for compatibility). 

LL should do their best to dramatically improve SL, Since without such improvement, SL will keep declining in Active Users. A Rebuild of Second Life would reinvigorate interest in LL's platform, Giving Linden Lab more Resources and increased financial stability if something was to go wrong, like a recession. It would close any gaps in SL that competition could use to pull the metaphorical rug out from Linden's behind.  

So the question is, should Linden Lab rebuild SL? & would it be worth it for everyone involved?

It would literally have to be "The same thing but better" and the ability to take all your stuff with you for most people to be on board. I could see the need for a new engine and a revamping of backend stuff, but it's going to have a couple roadblocks. The first roadblock is going to be the average user, can their PC run it? If it can run on a Celeron, we're good. Can't change the UI too much, maybe just fix group chat. Fix that and all the the average user cares about is "Can I shop, does it look nice and do the **** animations look good?"

The next roadblock is going to be the creators, streamlining the process would be nice for clothing (even though its not bad now). Change it too much, like throwing Unity in there or something and you're going to have problems with them. If they decide they don't want to use Unity....well now the average user doesn't have the nice things either. Close up some of those complexity and LOD loopholes, that's about it. It would be cool if there was some kind of bridge from blender to SL for building. It's going to have to sell creators on it being easier and better.

So it's going to have to appear to be the same thing as SL in the beginning and gradually ween people off of a lot of things they're not used to. I don't think its going to work if it does a jump like windows 7 to Windows 10. The key thing is starting off like "It's SL but better."

[Penny Patton]

4 hours ago, Janet Voxel said:

and the ability to take all your stuff with you for most people to be on board

A large part of why SL runs so poorly is due to poorly made content. Besides, if LL is the one doing the developing and their goal is simply to get the existing SL userbase to migrate to a new platform, then they've already lost. They'd essentially be creating a new product with the goal of competing with their old product.

 If a next-gen SL is going to be successful it needs to draw in more than just longtime SL users. Sure, many SL users would eventually migrate to "SL but better", but hamstringing SL2.0 with old SL's problems before it even begins just to appeal to SL users wouldn't be how I'd go about it. I'd aim to make the next gen SL be the best it can be and count on that to draw in all new users, and draw back users SL has lost over the years. Some current SL users would migrate, some wouldn't. I expect in any scenario, classic SL would stick around indefinitely with its own hardcore userbase.

[MrBalloone]

On 12/9/2020 at 9:04 AM, Penny Patton said:

A large part of why SL runs so poorly is due to poorly made content. Besides, if LL is the one doing the developing and their goal is simply to get the existing SL userbase to migrate to a new platform, then they've already lost. They'd essentially be creating a new product with the goal of competing with their old product.

 If a next-gen SL is going to be successful it needs to draw in more than just longtime SL users. Sure, many SL users would eventually migrate to "SL but better", but hamstringing SL2.0 with old SL's problems before it even begins just to appeal to SL users wouldn't be how I'd go about it. I'd aim to make the next gen SL be the best it can be and count on that to draw in all new users, and draw back users SL has lost over the years. Some current SL users would migrate, some wouldn't. I expect in any scenario, classic SL would stick around indefinitely with its own hardcore userbase.

There would need to be a bridge. Since if you've spent a sum of money on SL1 and you aren't gonna get anything for it on SL2 then you aren't gonna move.

There are many ways LL could do this. One way is to have the amount of L$ in your account and the L$ your inventory is worth are given to you in SL2. So if I had lets say 40 items in my inventory all worth 4,000L$ and another 4,000L$ in my account then the 8,000L$ my account is worth, the account on SL2 would receive in an equal or equivalent amount. Any land in SL1 is also given to you in SL2, free of charge. So if my account collectively had 10,000m² of Land in SL1, then I'd get 10,000m² in SL2. All of this not affecting the inventories or land in SL1 directly. 

Essentially you can't have current SL1 users start from nothing in SL2. & to draw users in LL could celebrate the launch of this SL2 by having a promotion where for a set period of time (example 2 months after launch) any account created is entitled to a plot of land (with a limit on it's size of course. I feel something like 2000-4200m is appropriate) and a specified amount of currency. 

[Alwin Alcott]

23 minutes ago, MrBalloone said:

 the amount of L$ in your account and the L$ your inventory is worth are given to you in SL2. 

Any land in SL1 is also given to you in SL2, free of charge.

and how would this be a realistic idea?

What can you do with a token if there's hardly any replacable content?.. it will take YEARS for creators to make, and also optimized items to replace current inventories.
What about the value of items for creators in their inventories?
How about the value for items that aren't sold anymore?
How about creation by everyone as now?...
How about your friends that don't want to go over?
How about land renters? ..
How about the set up ?...rooms... experiences or land like now?
and i can add a complete page of "what abouts"

We seen the abilities in Sansar... want that debacle again? after several years a login number of 15 a day?
Guess i'll be one of the many who will wave goodbye if this happens.

Edited December 23, 2020 by Alwin Alcott

[MrBalloone]

7 minutes ago, Alwin Alcott said:

and how would this be a realistic idea?

What can you do with a token if there's hardly any replacable content?.. it will take YEARS for creators to make, and also optimized items to replace current inventories.
What about the value of items for creators in their inventories?
How about the value for items that aren't sold anymore?
How about creation by everyone as now?...
How about your friends that don't want to go over?
How about land renters? ..
How about the set up ?...rooms... experiences or land like now?
and i can add a complete page of "what abouts"

We seen the abilities in Sansar... want that debacle again? after several years a login number of 15 a day?
Guess i'll be one of the many who will wave goodbye if this happens.

Yes I can see that, my point was that you cannot have current users start from nothing. The best way would be to allow the transfer of an account and everything under it without large impacts to performance. As for the value, for purchased items it would be the L$ paid for them, any Uploads would get transferred. 

A possibility could be a sandbox to isolate SL1 items.

Creation

For creators, it would depend on the item. TO make it compatible with SL2, creators would have to update them, however if so desired then they could upload it to a LL software, the software would do all the necessary modifications, then showing the creator the result allowing them to test the item and then the creator could accept the result and the result would be uploaded to a separate marketplace for SL2. 

To encourage efficiency, Items would get a score, the score would be displayed on the item page. Items with a poor efficiency would still be purchasable however LL would take a cut of the purchase amount and various other reprimands' (efficiency would be based on the specifics of an item, so for example a shirt and a vehicle would not be graded on the same level). LL would have a software that could optimise various items.

LSL

LSL would remain, but only for compatibility. LSL would be redone, it wouldn't even be called LSL. I'll call it NS (Nova Script), from here on I'll refer to an LSL2 as NS or Nova Script.

SL2 would be not limited to NS, Creators would be able to use other languages such as C# and JS (JavaScript). Nova script would have some inherent design goals. Efficiency would be the main goal, Versatility 2nd and Portability 3rd.

Marketplace

The Marketplace would be separated into two, a legacy marketplace for SL1 (essentially the marketplace as it stands today) & a new marketplace for SL2. 

Marketplace owners would have the ability to pass on the responsibility of the store and related items to another resident or Open source all Items (see Inventories). If a marketplace has been closed or inactive for a long period of time would automatically have all items open sourced. Inworld stores would also be archived to allow residents to explore and receive any perks or purchase any items limited to the inworld store.

SL2 users would be able to purchase items from the SL1 marketplace if an equivalent is unavailable for purchase on the SL2 marketplace 

SL1

The OG SL would remain, sort of like an SL classic..

Development

As I said before LL would split into 2 internal teams. Team 1 would work on SL2 and comprise of 90% LL. Team 2 would work on SL1, comprising only 10% of LL, Team 2 would focus on improving SL1's performance and preparing it for the transfer of accounts, any new features that are not 70% complete would be scrapped and make their way into SL2 in some form if deemed appropriate.

SL2 would also be influenced by the community. The Development of SL2 would be open to suggestions and the community would get updates on its progress and be able to consult with LL about SL2.

Inventories and Account balances

Inventories would be transferred to the new SL2, any existing items that have been updated [For SL2] will be replaced with the updated versions, whenever an item is updated for SL2 this will happen regardless of time. Any items that have not been updated will remain in their current state (hence the need for an SL1 compatibility layer). Items may be classed as unavailable (items that are no longer sold), items of this nature would be updated by SL to have Full Perm unlocked for the object and any linked objects or contents such as script, making these Open-source in a way. 

Account Balances would be retained.

Land

Any land owned or entitled to in SL1 would be retained. Accounts would receive the same amount of land in SL2. Rented land would become a type of land, Rented land would be zoned as such and would be regulated. A new marketplace specifically for the sale or rent of land would be opened. Tennent's would be able to pay via a new Rent Manager. Tennent's would have a direct line of communication to their landlord and be able to defer payment among various other measures. Tennent's would only lose access to land if the due amount isn't paid within 6 months excluding deferred invoices. (if you don't resolve the issue within 6 months you lose the land, no more instant repossession) If a resident loses entitlement or ownership of land, the land is then archived to be restored at a later date and to allow viewing of the parcel (if it is commercial land)

Any land would be able to be Zoned by its owner. Land above a certain size would also be able to have multiple zoned areas.

Transition

After the Launch of SL2, a 36 month (3 year) transition period would commence.

Community

All community features would be optimised and expanded upon, any community features would require input from the current community.

Friends who do not participate in SL2 can be communicated with via SL Classic (SL1).

Groups would be transferred to SL2

Engine

The engine would get a rebuild from the ground up. As of current SL1 has many independent components (such as Havok & F mod). SL2 would run on 2 components, The Nebula Engine & The Lycra Engine. The Nebula Engine would be able deal with everything from physics, to rendering to calculation. The Nebula engine would developed to be efficient and versatile. Nebula would also be designed to be modular and upgradeable leveraging ever improving technologies.

The Nebula engine may also be developed into a standalone Development engine, to be used for other software and could be licensed similar to Unreal Engine and the Unity engine, Providing a sufficient and consistent source of revenue.

The Lycra Engine on the other hand would handle everything regarding SL1, this would be the Engine that handles Compatibility and Optimisation of SL1 content. This engine would not be licensed however it may be open sourced or Licensed at a later date. The Lycra Engine will only be secondary to the Nebula Engine and will only be used for SL1 content.

 

 

Want me to answer more, then lay all of your "what about" questions for me to answer.

[Alwin Alcott]

10 minutes ago, MrBalloone said:

Want me to answer more

no need, the work it needs will cost so much that even Bill Gates would go bankrupt. It will take years to complete, and be outdated before there's a 1.0 release.
The only way is a build from scratch with a clean start. And as said.. we seen the debacle with Sansar, and many other virtual worlds.

[MrBalloone]

1 minute ago, Alwin Alcott said:

The only way is a build from scratch with a clean start. And as said.. we seen the debacle with Sansar, and many other virtual worlds.

The Problem with Sansar is that it didn't take what makes second life, second life. Focused to much on things that didn't matter. Sansar began as SL2 before morphing into Sansar during development. Sansar became a spiritual successor of sorts but was more so an independent thing with little to no connections with SL and its original vision.

As detailed before, SL2 won't be developed behind closed doors. The SL community will have a part in the development of SL2 and make sure SL2 sticks to its vision and doesn't become another sansar.

SL2 will be built from scratch. And the products from the development of SL2 would be able to fund it, Such as the Nebula Engine. 

Since the Nebula Engine could be made into a standalone Product and Licensed, that will be precisely what LL would do. Since while the resources spent for SL2 won't recovered instantly, the resulting products from the development will. You'd be able to Use the Nebula Engine in whatever you are creating, The Nebula Engine could be Licensed in the same way Unreal and Unity are today. Your game would have the choice of the Unreal Engine, Unity Engine or the Nebula Engine. This would Provide LL with Revenue. Even if SL2 fails, the resulting products of it such as the Nebula Engine wouldn't have been made in vein. IF SL2 fails, LL could still license Nebula Engine.

1 minute ago, Alwin Alcott said:

the work it needs will cost so much that even Bill Gates would go bankrupt.

exaggeration

[Ardy Lay]

What a unique idea.  You could call it OpenSimulator.

[animats]

On 12/7/2020 at 7:00 AM, Ardy Lay said:

self-inflicted packet loss

Yes. Much "packet loss" in SL viewers is the viewer dropping packets because it can't keep up, not the network losing them.

[MrBalloone]

18 hours ago, Alwin Alcott said:

Sansar... want that debacle again?

Sansar was misguided. Instead of Second Life but better or Second Life 2.0,

Sansar went for VR Chat by Linden Lab

[iceing Braveheart]

On 12/5/2020 at 6:17 AM, MrBalloone said:

Would Second Life Benefit from an Engine Rebuild?

Yes

but second life is pretty much https://en.wikipedia.org/wiki/Abandonware & https://en.wikipedia.org/wiki/Pre-installed_software

They cannot optimize a rebuild would only result in what you see now they need to be carried https://aws.amazon.com/ https://www.unrealengine.com/en-US/

The servers are done, migrated to aws, just need that engine replacement and good to go.

unreal engine would also replace the existing server software that they can never seem to get working right in the first place https://youtu.be/dEkOT3IngMQ

[Taiga Cloud]

Hard to realize, as much as id love to create stuff with displacements and what not. Just alone the products who are rigged would need the same (primitive) bones + weighting sl currently has. Instead we should have major improvements on avatar bone hierarchy. (toes for example)... Creators of the brands they run would have to adjust those minor technical differences that will come up to have the product stilll be compatible. Also many products would need to be updated for a new shader engine to work or they would be dead weight on the new marketplace, just like old prims being sold for 2k still on marketplace because they creators abandoned sl in 2007 or so. 

In such a case we would also urgently need regulations for uploading mesh. That means that creators have to comply with industry standards and learn to do proper topology on their meshes, which 95% of "creators" are not capable of in SL.

But we can dream.

Edited December 31, 2020 by Taiga Cloud

[Furholio]

I'm all for LL trying to rebuilt the whole client from scratch since I have the impression they totally lost overview and thats the reason for the recurring bugs and the hesitation to add new features. There is that fear that changing a detail 'here' could break something 'there'. I can't think of another reason for missing basic features like skybox textures. Imagine putting a skyfield texture on the window of your spacestation, tick the 'make skybox' checkbox and the starfield will look infinitely far away. Its a feature thats been around since Doom from '93 and I bet its dirt cheap concerning rendering costs and would aid alot to immersion. But where is it? 

Edited January 2, 2021 by Furholio

[bigmoe Whitfield]

7 hours ago, Furholio said:

I'm all for LL trying to rebuilt the whole client from scratch since I have the impression they totally lost overview and thats the reason for the recurring bugs and the hesitation to add new features. There is that fear that changing a detail 'here' could break somethere 'there'. I can't think of another reason for missing basic features like skybox textures. Imagine putting a skyfield texture on the window of your spacestation, tick the 'make skybox' checkbox and the starfield will look infinitely far away. Its a feature thats been around since Doom from '93 and I bet its dirt cheap concerning rendering costs and would aid alot to immersion. But where is it? 

@Oz Linden

[Oz Linden]

On 1/2/2021 at 7:13 AM, Furholio said:

I'm all for LL trying to rebuilt the whole client from scratch

because that went over so well last time....