Sunday, February 05, 2012

Loki's Blog has moved

Due to Googles incessant need to merge all my services into one account, i have had to move my blog to a self published host.

From now on my blog will be at THIS LOCATION. Please update your bookmarks etc.

Friday, January 27, 2012

Display Cost of a Prim #SL

While working on my mesh cabin and rats i was wondering about 'Display Cost' . Display cost is the number calculated by the viewer to represent how much computing power is needed to view an object, The lower the number the more efficient your object is to be displayed.

I did a test on a normal default plywood Prim cube. I originally entertained the idea that by setting the sides to invisible, perhaps the display cost of the prim would shrink. Actually i found the opposite.

Default Prim has a display cost of 389
Setting the faces transparency to 100 added 15 making display cost 404

making just one face 100 transparency ads the same amount of cost
removing the textures deducted 120 from the display cost showing how much a texture can change
the cost your object has.
Changing the size of a prim seems not to change the display cost unlike a mesh object. 
Not sure what significance this is or wether it helps anyone, but i found it interesting none the less :-p

Wednesday, January 18, 2012

Making a Mesh: Recreating a Cabin #SL #SLmesh

Those following me on Twitter will have seen tweets as i set out to recreate my little Cabin on Escapades Island.

So far with Mesh i have grappled with creating clothing, zombie worms and an AirShip, but i had yet to try making a building out of Mesh. Compared to rigging clothes making a mesh building should be a lot easier :-p

Start with cube
Using the free Blender 2.6 i started creating the initial walls and built up the shape. I created several parts to the new cabin keeping things such as, the cooker, roof, windows and floor planks as separate 'Objects' within the blender file. Each part had it's own UV Texture map.

UV Mapping
UV mapping is a very important part between modelling and texturing. It's an art form in itself and can take up a lot of time to perfect. When you create UV maps for buildings, you want to try and keep UV shapes close to the shape of the walls. It's very easy to 'Un-wrap' a UV and end up with stretched or squished shapes. If this happens you can edit the UV map and reshape UV's to the shape of the surfaces to be textured.
In Blender you can export the UV map as a PNG file. This can be useful
for reference when texturing you model.
When it came to texturing the Walls i discovered i had made them badly making it rather impossible to UV-unwrap nicely so i had to pretty much delete the walls i first made and start again. 

This image shows how bad UV-Unwrapping was causing bad texturing.
This was due to bad modelling from the start.

I have recently been playing Star Wars: The Old Republic and had be interested in the texturing used in the game. They used quite low detailed soft illustrated textures and so i decided to try a similar style texturing for my cabin. I used reference textures of wood from google search and added bur while modifying with paint strokes to give a more illustrated feel to the textures. 

Another benefit of texturing mesh is you have more control over stuff like shadows and lighting. I was able to add a glow from the cooker onto the canon floor and shadows under the table and shelves.

Comparing Old to New
Part of this project was also to see the differences between my old cabin and a new mesh cabin, not just by its visual and prim counts, but also by how much it weighs on the server and your computer.

You have to realise that Mesh is not simply about creating awesome looking detailed objects. It is about creating stuff that is more efficient for you computer to download and display in the Viewer. 

In my previous blog post about uploading mesh i explained how i use normal old Prims to create
the Physics shapes. This is because it seems more efficient way to ad Physics to mesh, especially buildings.

It's the Count that Counts!

Linden Lab have added a new tool you may not be aware of. When you open the build/edit window, next to the Objects Land Impact is 'more info'. This opens up another window that calculates numbers to represent aspects of an object you never really think about.

Objects = how many grouped/linked items are selected.
Prims = how many Prims make up the selected item, that includes  mesh, sculpts and basic old prims.

Download = represents an estimation oh how much it costs to download i.e. how quick it is to load
Physics = represents how much physical energy it costs. i.e. how much energy used to roll a ball over.
Server = represents how much server attention costs. i.e. how much attention the sever has to spend watching your object.
Display = represents the cost to display your item in the Viewer. i.e. how much processing power your computer uses to display your item.

Land impacts = the new form of counting object costs, replacing Prim counts with Land Impacts.

So comparing the two cabins is quite eye opening. In every category the Mesh Cabin is less than half of the costs of the old cabin. If these numbers are anything to go by, once i convert most of my builds to Mesh, the SL viewer should display them with complete ease while the server runs as smooth as...  oh but to dream :-P.

Making a Mesh: Uploading to #SL

It's great to hear so many of my friends dabbling in Mesh creation now. 
They are all levelling up into a world thats going to offer more creative choice in the near future with 3D printing starting to hype it up.

Some of my friends have expressed frustration with the SL mesh uploader, so i have decided to blog how i do it and what i know in order to perhaps make my friends lives easier. There is nothing worst than slaving over a model in Blender only for SL to spit some error message at you on the final hurdle.

Beta Grid
To start with get on the Beta-Grid 'Aditi'. When starting in mesh, it's not kool if your first 100 practices are gonna cost you your Linden Credits. Until you have got a feel for how to optimise the mesh to cost as little as possible with low Land impact.

Mesh Uploader Window
There are a few things the SL viewer needs to generate when uploading you .dae Collada file to the grid.

Click to embiggen

1. Generate Levels Of Detail.
Each level should
be less detail than the
one above
As you move away from an object in SL, the further away you go the less detail you need to see because you are to far away. So 3D worlds and games use a system called 'Levels of Detail'. If you yourself are so far away that you can't see the object in detail then your computer need not waist valuable computing power to render all that detail so far away.

The first in the options tab is the Level of detail Generator. This is very important to how much Land Impact (prims) your mesh will cost/weigh. Proper Pro model creators will create and upload Levels of detail separately, but if you are like me an amateur self taught monkey then you will probably use the built in Levels of Detail Generator.

the LOD (levels of detail) options show four levels Lowest, Low, Medium & High. Each level has tabs of options, but the only options i change are the numbers on the end with the up and down tabs. The higher the number the more detail that level has. The lower the number the less detail and less computing power is needed to render it.

To better see changing a levels detail, click and change the 'Level of Detail Selection' button to the appropriate Level.

TIP FOR LOW LAND IMPACT: Select the Lowest Level of Detail in the LOD preview selection. now go to the Lowest level options and set the numbers to 0 (ZERO). The preview should show a small triangle. I always set the Lowest level of detail to 0. The reason for this i will explain in the next section XD.

Click to embiggen
Go through the other levels options and adjust the numbers as you feel best. What will happen is in SL as people walk away from your uploaded object the Viewer will drop down through the layers. As you walk towards the object the SL Viewer will raise up through the layers until you are stood in front of the object seeing the highest level of detail.... hope this makes sense :-)

2. Physics.
Adding a Physical shape to your object is like adding another invisible level of detail and as with the Levels of detail, the more detailed your physics shape is the more it will ad to the final land Impact (prim count). Using the Physics Options Tab you can select one of the Levels of Detail to use as the basis for your objects Physical shape. Pro Model creators will probably make their own specialised Physics shape to upload.

I have not had much success with Mesh Physics. I'm still finding them a bit crazy in their outcomes so what i tend to do is set the Physics shape to use the lowest level of detail. This basically makes the object an almost NON Physical shape (phantom).

Click to embiggen
TIP FOR LOW LAND IMPACT: Instead of making your mesh object an accurate Physics shape during uploading, use the lowest LOD as set in the previous TIP. "But I want my object to be physical!??", I use the old fashion PRIM. After uploading your low impact mesh you can link it to a normal SL PRIM with usually no additional Land Impact costs. Set the PRIM to 100% transparency and away yu go :)

IMPORTANT: Before moving on to the next section it is important that you remember after selecting your Physics shape to click 'ANALYSE'!

3. Upload Options.
In the Upload Options Tab there are some useful told for extra mesh fun such as Rigging and joints and... well stuff i don't totally understand yet. But the important thing for you is SCALE.

What ever the 3D modelling program you use, getting an idea of the scale of your object is tricky. When first trying mesh i would upload a t-shirt that was  big enough for King Kong or Godzilla.

Click to embiggen
The Scale section shows what the in world dimensions will be in metres once uploaded to the grid. Very handy for keeping costs down as you do not want to upload an object unnecessarily huge at a high Linden Credit Cost only to then shrink it down.

4. Calculating and Uploading.
Finally you are ready to click that Calculate Weights & Fees button. It will take a few seconds to calculate before giving you a price and land impact as well as some very handy costs Download/Physics/Server.

Click to embiggen
If you are happy with the calculation you can click upload and after a few seconds it should appear in your inventory.

Click to embiggen
TIP FOR UPLOADING AGAIN: Not everyone knows this but every time you upload a new Mesh Object your SL viewer will automatically save a file on your desktop. This file called a .slm file stores all the configurations for the mesh file associated with it. 

This is handy if you uploaded a mesh perfectly on the beta Grid and want to upload the same object with the same settings onto the Main Grid.

WARNING: Although handy for re-uploading previous Mesh Objects, the .slm file can totally screw things up if you modified your object and saved under the same name. In cases like these all you need to do is simply click 'Clear settings & reset form'. 

So thats the basics. 
It is the way i have learnt to use the Mesh Uploader and there is a lot of stuff i still don't understand yet. I hope this helps my creative friends and anyone else who wants to get into mesh as it really is a levelUP in the creative game. Most of all i hope this allows you to have more fun experimenting XD.

Some other tips about Mesh once uploaded that i have learnt...

1. Changing the Size of the object changes the cost in Land impact (prim cost) dramatically.

2. You can change a mesh objects physics shape to be PRIM, Convex Hull or to NONE.  This changes how the physical shape interacts with the world and also dramatically changes the objects land Impact.

3. When making vehicles it's land impact will change dramatically as the object changes from a static object to a physical object. A way around this is to set your mesh objects Physics to NONE in the features tab of the Objects edit options.

Friday, December 30, 2011

Loki's Predictions for #SL 2012

I spent more time in SL this year
than last year
Predictions for SL have started popping up and for the most part they paint a rather pessimistic view.  
I think last year most of my friends and i were tired of the stagnation, but blimey what a year 2011 has been!

So i've decided to list my predictions for SL which will express my optimism that has grown through out 2011.

What ever happens next is built on progress gained from 2011….

1. Content creators will create the biggest and strongest change.
Is it me or are people forgetting that SL is made by US!. When Rodvik first arrived he expressed that Second Life is OUR world and the most interesting part of it is OUR creative energy. If SL is going to grow it'll be in some way down to us creators and what we build. It's up to us to embrace the tools LL have been working on through 2011 and to use them as best we can to create fantastic places, games, experiences, role-play havens or social environments.

2. Market place will become even easier and more efficient.
The new magic-boxless Marketplace is almost here allowing anyone to sell their creations on marketplace directly from their inventory. This will hopefully remove dramas such as crashing before deliveries etc. In world stores will find it difficult, and i saw some very old stores pack it in this year. It's a changing world, the game is changing just like in the real world.

3. More land will be used to host games and experiences to subsidise tier costs.
It was easy to rent out land for people to set up shops. It's not quite that easy to rent out land to set up micro-payment games and experiences. I still do not know exactly how the new gaming tools will be implemented. Will they be new scripting rules, or will they be options available in build edit or land tools? In any case from what Rodvik has blogged the Lab seem to be working on making it easier for you and i to create interactivity in what we make. It'll be up to us weather we use the tools to  charge people for entertainment.

4. SL feed integrated into other SL features.
Currently you can follow people, tag your current location and upload snapshots. It would seem silly to have the SLfeed and not have it tied into the destination guide, or on marketplace. I assumed that LL would simply ad a 'love button' by now. It would also be nice to have a simpler way to 'LOVE' your current location with one click.

5. LL will create a mobile project that allows us to share creative aspects with the mobile world.
Cant really predict what this will be. It's interesting what ever it is. The last new product from the Lab was a gated business orientated conference flop and did not involve the current SL user base. What i hope is that what ever Rodvik and his team unveil in 2012 it will thrive on the freedom of creativity the current SL users spill out.

6. Linden lab will fix notifications.
Ok so this is a HOPE more than a prediction. I hate the viewers notification system, i think its dreadful and has been for the past 2 years. I basically use my email client to store and organise offline notifications because SL Viewer lumps it all together in a tiny corner box and sometimes not at all. It just feels like a broken mess, so I'm hoping LL are working on new ways to present, store and organise notifications from groups, receiving items, offline messages, SLfeeds etc.

7. Destination guide will have to evolve into something more universal and dynamic.
Currently Linden Lab see the destination guide as a way to present places of interest to new users. But anyone can see that it is one of the best ways to find a place to visit no matter how many years you have been in SL. 

If Linden Lab want people to build and submit more games and experiences, then they will need to create ways to better manage visitors. I can't have over 2000 people arriving at my private region without it causing major lag and a terrible experience. The destination guide will have to change somehow or new users from the login screen will be continuously forwarded to places so full of people they will have the worst first experience.

8. Old users will loose out to those who take hold of new features and brave the lag of progress.
Not much i can really ad to that.

Im very excited about 2012. I do not see SL disappearing, i have to many projects on the go, to many friends enjoying their time in SL. If you think SL is not going to survive 2012, maybe you are just not interested anymore and are just predicting the demise of that interest. And from what i have witnessed through my attempts at making games in SL, there are plenty of people who are still exploring what SL has to offer.

Thursday, December 29, 2011

Another Year of Change in My Second Life #SL

This is not predictions for next year but rather a comparative look at what i wrote this time last year.

1. Land. I spoke about how the old system of renting shopping space to help pay for over priced tier was gone thanks to market place and linden homes. 

So what happened during 2011? nothing really, except i have worked really hard to make gaming experiences that bring money back into the land to pay for the tier. Escapades saw a major drop in donations but thanks to the Secret underground base, the Eggmen hunt,  Zombie worm massacre and the Dark Aether Falls game i managed to hit Tier targets.

2. Shops. I said last year that Marketplace killed the in world shopping system, and i still believe that. Specifically affiliate sellers. 

During 2011 i re built my clothing brand and delved head first into Mesh clothing. I've seen record sales and gained a lot of positive feed back regardless of the deformer drama. But i am aware that i am now working with stuff far beyond simple linked prims. The game of creating a shop, selling quality items is a lot harder now.

3. Viewer2. As you can probably guess i basically said it was crap. I also touched on viewer fragmentation as the long time users refuse to use V2.

So has it been fixed? Well viewer 2 is now Viewer 3, or simply called 'Viewer' now. LL took a good look at making things simpler with a basic viewer mode, then integrating that basic idea and allowing a nice customisable UI which i liked.  But fragmentation of the viewer is still as annoying as ever. 

If you make stuff in mesh you have have a disclaimer that it's not viewable on all third party viewers, if i use Shared Media i have to have a disclaimer that its not viewable on all viewers. It leads to such frustration as people blame me for not being able to see the virtual world correctly when it is their decision to stick to out of date viewers.

The probable reason third party viewers are faster might have something to do with them not rendering the whole world.

4. Newbies VS Oldbies. I said "Viewer2 is the wedge that deepened the divide between the New and long time residents of SL. "

During 2011 i became almost anti-Newb, i started to see them as a pest. While creating games it was near impossible to cater for them and while using destination guide to promote my experiences it was impossible to get away from them. As a result experiences i made slowed to a crawl. 

Oldbies continued to stick to what they know and complain about the changes, never has this been more clear than in 2011. Phoenix viewer finally right at the end of 2011 added mesh support, only for it to be noticeably slower and crash making it a completely pointless viewer. If anything it has convinced many that mesh is evil.

5. Communication. I talked about the group chat FAILs and the horrible notification system.

What changed in 2011? Well they fixed the group chat system which is great, well done for that. They also added the SLfeed which most third party viewer users are oblivious of. Possibly the greatest addition after Mesh the SLfeed allows to say to friends whats been going on, link places, and post pics direct from SL. I did not think i would check out my SL feed as much as i do seeing as i have Facebook and twitter as well, but it really has become a daily need to catch up on what my friends are doing. If you are not on SLfeed then you are BORING.

During the UI facelift the terrible notification system still remains unchanged, all be it is now at the top instead of the bottom. It now shoves SLfeed notifications into it as well to be lost for ever. There is also annoying issues of getting IM's via email and them not appearing once you login.

All in all 2011 saw communication for official SL Viewer improve lots, but the way we receive and display notifications still needs serious makeover.

6. My Island and my friends. i said that 'My biggest aims will be finding a balance between making content & experiences in SL and Doing work in real world.'

Well i could not foresee the problems that would crop up during 2011 but I'm really REALLY grateful for the support i got from my SL friends. A lot of stuff never happened in SL that had been planned but a lot of unplanned stuff did happen and Escapades is still here.

During 2012 i would like to bring the Escapades gang together more as a family and help each other to build Escapades concentrating on what we want the island to be.

I tried roleplaying games on the island, and i tried mini games on the island, made a shop, was placed on destination guide five times. I buried myself in work on SL and failed to make it balance with real world work, so maybe it's time to accept that SL is my work in 2012.

7. Linden Lab. I said i look forward to what those left at LL after the great cull of 2010 come up with in 2011. For me at least it has been a fun year watching the Lab churn out new stuff, trying new ideas. The lab is a lab now, and everything they seem to change keeps changing and improving. The cogs are moving and SL is steaming forward, to where we don't care just as long as it moves! :)

Wednesday, December 28, 2011

llSay(0,"Roleplaying Games and Trolls and Lag OH MY") #SL

It's that week between christmas when you sit back and relax with left overs and rubbish tv…. 
well thats what it's supposed to be. In stead i have been popping into SL every couple of hours to check on SIM performance and dealing with user questions regarding the Dark Aether Falls game.

The Dark Aether Ends
Lets step back a few weeks first. Early december the Dark Aether Falls role-play was in full swing after two months build up. The Mini clue hunt game was finished and served as a side plot for a story explaining who Jason Moriarty was and present a few ideas for why he was so hell bent on the destruction of the town.

The residents of New Babbage loved this small game as a prelude to the final chapter and end of the story which this year entailed a ritual, race against time hunt amongst the clouds to stop the Dark Aether from falling. 

Everything was automated this year making it easier for me to do talking parts between NPC's while the machinery clanked away behind me revealing the final monster. This was all thanks to my continued efforts in grasping SL scripting by communicating with brilliant minds on the SL scripting forums.

The story that unfolded was everything i hoped for with the Cloud angels, Moriarty, the Van Creed, the canisters all converging for a climatic end with a building collapsing around the players in a panic. Everything went to plan except as hard as i tried to kill off the main character the players refused to let him die :). So as i close this chapter of my creative story it has been taken out of my hands by those who are so involved with the cannon created that they can't bare to let it die.

The feed back from the residents of New babbage has been phenomenal and extremely positive and although i have said i will not be returning next year with an rp on that scale, I'm sure i will will have something because what i learnt this year is that creating an RP story for a small group is very different and sometimes more rewarding than making a game experience for a mass of people.

The Dark Aether Continues
So what happens when 80 people an hour arrive to play your gaming experience? Let me take another step back to explain how its possible to end up having 80 people an hour arriving at your region.

When i first announced the Dark Aether Falls mini game at the start of december i had only just submitted my game to Destination Guide. Linden Realms was being promoted on the main login screen and all over the front of the Destination guide. I thought at first as you do that they were just using their godly powers to be ahead of everyone else. I now believe it was necessary for them to understand how the destination guide herds n00bs to places en mass and to gage how regions cope with the strain.

I myself enjoyed being placed in the Destination guide at a position that saw low but manageable levels of interest from people who were experienced in SL.

What happens when 80 people an hour arrive at your region?
Then came boxing day and suddenly New Babbage was seeing thousands of people a day arriving to play the Dark Aether Falls mini game. It was soon discovered that the Dark Aether had been placed on the number one slot of the login screen and New babbage was suffering. The SIM crawled to a halt as people continued to arrive at the HUD shop making it impossible to do anything. Not only does movement become impossible but script also start being delayed and so the game HUDS stopped working.

And this is where i am today, overwhelmed by the interest caused by the Login Screen Destination guide, trying to help those i can who have managed to get through the lag to play the game they purchased in SL. For many this is their first experience of SL after going through the orientation and i so wish it wasn't. I feel helpless to do anything. 

I requested the destination guide editor to remove the Dark Aether from the number one slot to try and lower the number of visitors which they did and New Babbage is once again playable. Thanks to the LL editor being quick :).

'Be careful what you wish for'
yes i wished for lots of people to be able to experience what i make, and why shouldn't i? I haven't been put off by whats happened in New Babbage, instead i view it as highlighting a very important area in which the Lab NEEDS to be aware of and strive to fix. 

The lab has avoided this issue with their Linden Realms game as they created twelve regions. Im not sure how many people can afford £3500 a month tier costs (not including VAT) to run a game in a similar fashion, but the drop off from destination guide will kill you out right. I would like to know how many are still playing Linden Realms now it has dropped off the number one slot. 

Linden Lab want us to use new gaming tools to build games in order to bring value back into owning land and encourage new and current users to stick around longer. Combine that with the current destination guide, all you will get is bad experiences like i'm currently going through :).

Egnog shows a fragmented future.
Dark Aether aside, i did one final gaming experience before the end of the year. Egnog the christmas Troll is a character i created for my SL friends. He is not of any faith and simply gives you a present if you want it. Originally created 4 years ago he was made crudely with 24 sculpt prims and this year i decided to recreate him and his winter grotto in MESH. He is now only 10 prims and more detailed.

The grotto is perfect example of all that i have learnt this year in SL. From Mesh modelling and texturing to automated scripted machines and gaming HUDS. Unfortunately the free Egnog game HUD does not work so well, but again i have learnt a lot from making it. 

It has highlighted an issue that will plague me and those working in Mesh to make detailed, efficient builds throughout 2012. That is the issue of fragmentation between viewers.

Part of the decision to make Egnog in mesh was because the Phoenix viewer now has mesh support, but the feed back i was getting was that many could not see Egnog because they had changed back to the old phoenix simply because it's a faster viewer. If you remove the ability to render objects, textures sounds and lighting, then yeh your viewer is going to be faster of course, but then you can't complain when you don't see things.

We should be embracing mesh viewers and demanding that LL make it work better.

Anyway Egnogs winter grotto is open to who ever visits Escapades Island or shares between friends and will be around until New Year.

Tuesday, December 06, 2011

The Dark Aether falls Mini Game of Clue Hunting and Bug Zapping #SL

This weekend will hopefully see my latest attempt at a game added to SL destination guide (fingers crossed). 

Part of a much BIGGER story.
Part of the fifth New Babbage Chronicles Roleplay game in Second Life is is designed as a simple clue hunt game to uncover a small part of the overall story The Dark Aether falls. The Roleplay has already been unfolding for almost two months and is nearing its end dead line and while the residents of New Babbage have been writing lots and lots about their reactions to the story and contributing their own spin off encounters i have been working on this mini game.

The aim of the game is to uncover the truth about why a man is found hanging in the Red Dragon Opium Den. Following Clues and solving riddles while watching machinima cut scenes that explain the story. It is a way for people to get into the spirit of New Babbage who's motto is 'what could possibly go wrong'.

The end is somewhat open ended allowing those who are interested to stick around, talk to other residents of New Babbage and perhaps become addicted as i have to the steam filled smoggy city of Steampunk.

Building on experience.
The game itself is a combination of what i have learnt from past games such as the evil zombie worms and last years The Black Heart. It uses a HUD to allow the user to play the game. It holds all the parts of the story and you basically search the town for the clues which when you click activate the necessary part of the HUD.  The HUD also comes with the Dapper Zapper with which you can dispatch annoying evil Aether Bugs that run around in the sewers.

The Dapper Zapper and HUD together cost just L$100 which is about  USD$0.50 and i hope people will feel the experience is worth it.

Shared Media woes continue a year later.
It is a shame the new LL gaming tools have not been released yet as they might have helped make the initial start of the game easier for those with little SL Skill. Also i found it really hard to get the Shared Media to work across all versions of SL viewer and on windows.

For some reason on my Mac using Viewer V3 the machinima cut scenes worked perfectly in world. But on windows on Viewer V3 the Videos appeared just grey. Wether this is a Flash plug in issue, or a windows Webkit issue i do not know as I'm not a software engineer. 

The shared media video does not work at all on the popular out dated Phoenix viewer so an external video version of the HUD was included allowing people to click a button to pop the video outside into a web browser.

image from Nat's record of battling Aether Crabs

The Writer who's writing a story backwards