SL project updates week 3 (1) Server and viewer + misc items

Simulator User Group meeting (stock)
Simulator User Group meeting (stock)

Server Deployments: week 3, 2014

There are no server deployments this week on either Tuesday 14th (Main channel) or Wednesday 15th January (RC channels). The latter was anticipated to be the case, after Maestro Linden indicated there were no maintenance or other projects in the RC pipeline during the Server Beta meeting of Thursday January 9th.

While it had been thought that issues relating to STORM-1831 were not sufficient to prevent the RC package deployed in week 2 from being promoted to the Main channel, the decision has been taken to allow the package to run a further week on the RCs.

Week 4 Deployments

It now seems as if the plan is to deploy the RC maintenance project to the Main grid in week 4 (week commencing Monday January 20th), and to issue a new server maintenance project on the Wednesday of that week, although details on this have yet to be finalised.

SL Viewer

The Project Interesting viewer with the viewer-side interest list changes (notably the changes to how region information is cached and re-used by the viewer) gained a further update on Tuesday January 14th, with the release of release candidate 3.6.14.285213, which includes a number of additional bug fixes – see the release notes and download page.

STORM-1831: LSL Syntax Highlighting

This contribution by Ima Mechanic will allow LSL syntax highlighting in the viewer’s LSL editor to be updated and defined using a file obtained directly from the simulator the viewer is connected to, with the intention to eliminate issues of the viewer using outdated versions of the file. The server-side updates for this work are currently deployed to all three RC regions. However, as of the end of week 2, additional work was required on the syntax file itself.

In addition, there are a number of issues which are currently being addressed with the viewer-side code. These are listed on the JIRA for the project – STORM-1831, and require resolution prior to the code being issued  in a release candidate viewer. Most of these are currently being worked on, and are not seen as major show-stoppers, but will delay the code’s appearance in the short-term.

Other Items

Values for brush constants in calls to llModifyLand

One issue which has (again) been raised through STORM-1831 is that the values for brush constants in calls to llModifyLand, used for scripted terraforming operations, are out by 1 each (they are set to 1, 2, 3 and should be 0, 1, 2). This means that, for example, a call using LAND_SMALL_BRUSH will edit an area of 4×4 metres, not 2×2. The wiki documentation for the function notes the error as well. As a result of discussing this issue in the Open-source Developer’s meeting on Monday January 13th, a STORM JIRA was raised to (again) highlight this issue (STORM-2006).

However, as the issue itself is actually related to either STORM-1831 per se and because it will actually require server-side changes, the issue is liable to be moved to a BUG status and triaged from there.

Snapshot uploads to the SL Feeds

“It has to do with authentication under the hood when getting the credentials and then using them to fetch the snapshot config info,” Simon Linden said at the Simulator User Group meeting on Tuesday January 14th. The problem is still being worked upon for those still affected by it (I am), but there is currently no ETA as to a complete fix for the matter.

With thanks to Mona Eberhardt for the meeting notes.

SL projects update: week 2 (2): Fitted mesh, breedable issues and [L4L]-Gestures & Walkers (Freebies) <3

Week 2 Deployments – Recap

  • On Tuesday January 7th, the Main channel  received the server maintenance project that had been on the RC channels for the past few weeks. It contains a single bug fix, related to vehicles becoming stuck in the ‘sat upon’ state (which prevents parcel auto return)
  • On Wednesday January 8th, all three RC channels received a new server maintenance project, which contains some crash fixes and the new LSL functions for uniformly scaling linksets, all of which are immediately accessible using current viewers (see part one of this week’s report for details).  This project also contains updates related to STORM-68 and STORM-1831, both of which require viewer-side updates which have yet to be released by the Lab.

LSL Syntax Highlighting Updates

Related to STORM-1831, these updates, which were deployed to the RC channels as noted above, will eventually see the viewer able to fetch LSL syntax highlighting rules directly from the simulator. However, there are currently some errors in the syntax file as deployed to the RC channels this week (repeating text, bad whitespace, or inaccurate definitions of functions), which require further corrective work. As there is currently no publicly available viewer which can use this new capability, it is unlikely this issue will prevent the server maintenance project from being promoted to the Main channel in week 3.

The file is designed to be cached by the viewer (once the viewer-side updates are released), and is around some 600KB in size. This means that the file should only ever be downloaded and updated if the LSL editor is open, and the viewer detects a version number difference between the file it has cached and the file held by the simulator to which it is connected.

It is unclear when the viewer-side updates for this work will appear. While planned to arrive in a release candidate viewer containing a number of Snowstorm updates, it appears there are still a few bugs in the STORM-1831 code, such as with function arguments being presented in an incorrect order in the tool tips, which may delay its inclusion.

Week 3 Deployments

It is likely the server project currently on the three RCs will be promoted to the Main channel in week 3 (commencing Monday January 13th). However, it is also likely there will be no RC updates for the week, as there are no server maintenance updates ready to go, and no other projects (such as group ban lists) are in a position to be deployed.

The first Server Beta UG meeting of 2014
The first Server Beta UG meeting of 2014

Viewer News

Release Viewer

Thursday January 9th saw the PackageFix viewer (dated January 2nd, 2014) promoted to the de facto release viewer. As per pervious notes in this blog, this viewer has no SL-related functional updates. Rather, it corrects an issue introduced with the 3.6.12 code base whereby the Windows executable name was changed from “SecondLife” to “SecondLifeViewer”, without removing any executable using the “SecondLife” name from the installation folder. As a result, any shortcuts pointing to the “old” executable would allow it to run if used, thus potentially triggering further auto-updates.

With this fix, any executables using the old name (“SecondLife”) are removed from the installation folder, so any shortcuts created to them will cause Windows to display an error message, and the user can then remove them or modify them to point to the correct executable.

Fitted Mesh Progress

Oz Linden is anticipating a release candidate of the Fitted Mesh viewer Real Soon NowTM. There are currently no open issues at present, and work is underway to move it to a release candidate status. Exactly how soon “Real Soon Now” might be, however, is a little up in the air, as Oz went on to note at the Open-source Dev meeting on Wednesday January 8th, saying, “there are a bunch of steps [still to be taken] and I don’t know how long they’ll end up taking.”

Continue reading “SL projects update: week 2 (2): Fitted mesh, breedable issues and [L4L]-Gestures & Walkers (Freebies) <3”

SL projects updates week 2 (1): server releases, AIS v3

The first Simulator User Group meeting of 2014
The first Simulator User Group meeting of 2014

Server Deployments: week 2, 2014

As always, please refer to the week’s forum deployment thread for the latest news and updates.

Main Channel: Tuesday January 7th

The main channel received the server maintenance project that had been on the RC channels for the past few weeks. It contains a single bug fix, related to vehicles becoming stuck in the ‘sat upon’ state (which prevents parcel auto return).

This issue is related to vehicles getting into a “bad” state if they lose the passenger right at region crossing. The vehicle is left with what is effectively a “ghost rider” sitting in it, which defeats parcel auto return, leaving the vehicle in-world.

Release Candidate Channels, Wednesday January 8th

All three RC channels should receive a new server maintenance project, which contains some crash fixes and the new LSL functions for uniformly scaling linksets.

The uniform scaling project was one of the last projects Andrew Linden worked on prior to departing the Lab at the end of 2013. The new functions comprise:

  • integer llScaleByFactor(float scaling_factor)
    • Uniformly resizes the linkset by the given multiplicative scale factor (e.g. 2.0 to double the scale in all dimensions)
    • Returns TRUE if rescaling was successful or FALSE otherwise
    • This function only succeeds in non-physical objects
    • Rescaling is subject to Linkability_Rules and prim scale limits
  • float llGetMinScaleFactor()
    • Returns the minimum multiplicative scale factor which can be used by llScaleByFactor(); i.e. such that every prim in the linkset is >=1mm in scale
  • float llGetMaxScaleFactor()
    • Returns the maximum multiplicative scale factor which can be used by llScaleByFactor(); i.e. such that every prim in the linkset is <=64m and the resulting linkset meets linkability requirements

In addition, the RC project includes server-side support for custom default permissions on newly created objects. This is again some of the last pieces of server-side work Andrew Linden was involved in prior to his departure, and forms the server-side work related to STORM-68.  The viewer element of this work is a third-party code contribution from Jonathan Yap, and should be appearing in a Snowstorm RC viewer in the near future.

Finally, the RC project also contains the server-side work to allow to fetch LSL syntax rules directly from the simulator, rather than downloading them from a manually maintained file. This should enable up-to-date syntax highlighting within the script editor. This work forms a part of STORM-1831, which is a third-party viewer project developed by Ima Mechanic. Again, the viewer-side updates should be appearing in a Snowstorm RC viewer in the near future.

Sunshine AIS v3

During the Content Creation meeting on Monday January 6th, Nyx Linden reported at the next round of server-side changes for the Sunshine / AIS v3 updates are currently being “queued-up” ready for deployment – although not day is as yet available for when they’ll actually see the light of day.

In the meantime, the viewer-side changes are progressing, and are “nearing readiness for release as a test viewer. Those who have been following my AIS updates will know that the Lab has been encouraging TPVs to adopt the viewer-side code updates for use in experimental versions of their viewers, as LL would like to run a pile-on test to test the new AIS updates and inventory handling using regions on the Beta (Aditi) grid. There had been hopes within the Lab that such testing might take place prior to the end-of-year holidays, so expect to be hearing more on this in the near future.

Other Items

Snapshot uploads to the SL Feeds

As many are aware, there have been numerous issues with the SL profile feeds (my.secondlife.com), including issues around uploading snapshot to the feeds.

The problem, which first surfaced prior to the Christmas break, has not affected everyone using the feeds, but those who have been impacted have found it to be most irritating. Simon Linden poked at it, and as a result some people who were impacted were once again able to upload their images. However, for a good number, the problem still remains. As of January 7th, the situation remains that the Profiles team is still attempting to resolve the issue.

SL projects update week 50 (3): miscellaneous items

Server Deployments week 50 – Recap

As always, please refer to the week’s forum deployment thread for the latest news and updates.

  • Tuesday December 10th, 2013 saw the main channel updated with the server maintenance project that was on the RC channels in week 49.  This project includes a few miscellaneous bug fixes
  • Wednesday December 11th, 2013 saw the three RC channels updated with a new server maintenance project containing a single bug fix.

The RC channel bug fix is related to vehicles getting into a “bad” state where they appear sat upon / selected even if they have lost their passengers as a result of a region crossing. When in this state, they defeat the parcel / region auto return.

As the fix for this issue is currently only on the three RC channels and will not be promoted to the main channel until 2014 due to the code freeze / no change window, and because this issue may lead to vehicles accumulating in regions where there is a lot of traffic, Maestro Linden has requested that support move any main channel regions which are affected to an RC channel.

2014

These deployments were the last scheduled deployments for 2013. They also mark the last scheduled rolling restarts for the year as well, as there are no plans to restart any channels in either week 51 or week 52. The only exception to this will be if any major issue / fault occurs within the grid which necessitates a restart.

The next scheduled server-side deployments will take place in week 2 of 2014, the week commencing Monday January 6th, 2014.

Some information on updates which will be forthcoming in 2014 were given during the final Server Beta meeting of 2013. These include:

  • Andrew Linden’s work on “Uniform Scaling” LSL Functions for linksets (see part 1 of this week’s report). There are a couple of basic rules with the new functions: no prim in the linsket can be smaller than 1 cm or larger than 64m, and all prim centres must be within 54m of one another in order to be linkable. There will be two new functions as a part of this work:
    • integer llScaleByFactor(float factor):  uniformly scale a linkset by the specified factor (e.g. 2.0 to double the scale).  Returns TRUE if successful, FALSE otherwise
    • float llGetMaxScaleFactor(),   float llGetMinScaleFactor(): return the maximum / minimum scale factors that will work in llScaleByFactor due to limits in place by prim scale and linkability distance restrictions
  • An update to llLoadURL, so it will have a 0.1s delay instead of 10s (although there is still a throttle in place to prevent someone from spamming somebody else in order to crash their viewer)
  • llGetObjectDetails() & OBJECT_STREAMING_COST will be amended to return data when targeting an agent. Currently, OBJECT_STREAMING_COST doesn’t give you much of anything when targeting an agent; once this change is in place, it will return the sum of the streaming costs of all worn attachments (excluding HUDs)

SL Viewer

The “Project Interesting” (viewer-interesting) RC viewer updated on Thursday December 12th to version 3.6.13.284757, which includes a number of fixes from the initial release:

  • BUG-4675 Viewer crashes while reading chat history
  • SH-4606 Interesting: Small objects do not load until they are very close.
  • SH-4627 [INTERESTING RC] “Object out of range” is not detected on teleport.
  • SH-4631 [INTERESTING RC] Parts of linked objects are not shown in new release Second Life 3.6.11
  • SH-4641 Interesting: Incorrect amount of system memory detected on Mac

This may be the last RC update for 2013, unless something like the Google Breakpad RC is slipped out for another round of testing on Friday December 13th.

STORM-68

As noted in part 2 of this week’s report, STORM-68 is a third-party contribution which will allow a builder / creator to specify the default permissions applied to a new prim object (cube, cylinder, torus, etc.) on creation. Further testing has been taking place since my last update, and a number of issues and bugs have been resolved as a result. The test plan for the changes is complete and the viewer-side changes now look ready to go to LL’s internal QA. Again, this change requires server-side updates, so it is unlikely to appear in an RC viewer until the server updates are reasonably available on the grid, although this will hopefully happen in the early part of 2014.

SL projects update week 50 (1): server and viewer updates

A typical Simulator UG meeting (stock)
A typical Simulator UG meeting (stock)

Server Deployments week 50

As always, please refer to the week’s forum deployment thread for the latest news and updates.

Second Life Server (main channel) – Tuesday December 10th, 2013

The main channel was updates with the server maintenance project that was on the RC channels in week 49.  This project includes a few miscellaneous bug fixes. These include a fix for BUG-4431, which Maestro Linden, speaking at the Server Beta meeting on Thursday December 5th, described as:

A fix for avatars with crouch / crouchwalk animation overrides. Previously, the llGetAgentInfo() LSL function would only return AGENT_CROUCHING if the avatar was playing the default crouch or crouchwalk animations, so if your avatar had an AO which replaced those animations, (either with llSetAnimationOverride() or possibly with classic AOs too), scripts couldn’t tell when you’re crouching. But with the fix, the function is looking at whether you’re actually crouching, regardless of which animations are playing.

This should be the only “visible” fix within the package.

Second Life Release Candidate Channels – Wednesday December 11th, 2013

All three RC channels should receive a new server maintenance project containing a single bug fix related to vehicles becoming stuck in the ‘sat upon’ state (which prevents parcel auto return).

This issue is related to vehicles getting into a “bad” state if they lose the passenger right at region crossing. The vehicle is left with what is effectively a “ghost rider” sitting in it, which defeats parcel auto return, leaving the vehicle in-world.

SL Viewer

The SL release viewer was updated on Tuesday December 10th to version 3.6.12.284506, formerly the NameUpdater release candidate. This release does not contain any updates to the viewer’s functionality, but does change installer naming and fixes an updater issue.

Code Freeze / No Change Window

Week 50 marks the last week for deployments to release, main and RC channels by the Lab for both the viewer and the servers prior to the Christmas / New Year code freeze / no change window commencing on Monday December 16th. The no change window extends through until the start of January, and will see no significant releases (other than possible emergency updates, if they are required), although there may still be updates to project viewers.

The main reason for the no change window is to allow Linden staff, notably support personnel, have a decent break over the holiday period without the risk of having to deal with significant issues as a result of a change or update immediately prior to the actual holiday period. To ensure this is the case, the code freeze starts ahead of the actual holiday period so that the Lab can ensure those final releases for the year which are made are robust and stable and unlikely to give a major cause for concern while still having staff available to deal with matters should things get a little higgledy-piggledy*.

Other Items

“Uniform Scaling” LSL Functions

Andrew Linden has been working on a set of “uniform scaling” LSL options which would allow an object / linsket to be rescaled via a single LSL call – such as uniformly increasing or decreasing its size by a factor of 2. The work is still experimental and won’t be deployed until 2014. However, commenting on the work at the Simulator User Group meeting on Tuesday December 10th, he said:

One problem with scaling by multiplicative factor is that the scale operation might fail for a number of reasons; there are four categories of failures: hit the “prim is too small” limit. (2) hit the “prim is too big” limit (3) violation of linkability rules for linked set (when making bigger) (4) misc failures because of navmesh or other things.

The current API includes three calls: (A) llGetMinScaleFactor(), (B) llGetMaxScaleFactor() ; (C) llScaleByFactor(float). I currently handle failures (1) through (3), and the llScaleByFactor() will return FALSE if it fails, but it won’t tell you why it failed. You can use llGetMaxScaleFactor() and friend to ask what the max/min multiplicative factors are possible for reasons (1) through (3). Needs some work, but it is in progress.

A further concern / failure point was raised at the meeting: rescaling an object containing mesh and hitting the land capacity for a region as a result of the LI value increasing. One suggestion for avoiding this would be to have a function which could determine how far an object can be scaled prior to hitting the capacity limit for a parcel (e.g. returns the potential LI for a given scale value before an object is rescaled; however how this might be accurately achieved is unclear, particularly as LI scaling with mesh objects can be subject to a range of factors.

It will be interesting to see how this progresses.

*For those unfamiliar with the term “higgledy-piggledy”, I offer the following explanation:

Higgledy-piggledy explained courtesy of Mr. Berke Breathed
Higgledy-piggledy explained courtesy of Mr. Berke Breathed

SL projects update: week 49 (2): Oculus Rift Support “soon”

Maestro Linden (foreground) leads the Server Beta meeting. The colony of bats to his left is Voidpointer Linden, who is working on the Oculus Rift project
Maestro Linden (foreground) leads the Server Beta meeting. The colony of bats to his left is VoidPointer Linden, who is working on the Oculus Rift project

Server Deployments week 48 – recap

As always, please refer to the week’s forum deployment thread for the latest news and updates.

  • Main channel, Tuesday December 3rd: received the maintenance package deployed to BlueSteel and LeTigre in week 47

Issues with Main channel deployment

Two issues were discovered post-deployment of the Main (SLS) channel  updates:

  • BUG-4637 “”Can’t rez object at { x, y, z } because the owner of this land does not allow it”when rezzing any object from Library”
  • BUG-4635 “”Selected / sat upon:” incorrectly shows objects that are not actually selected or sat upon. “

Maestro has verified a fix for the latter issue, which he described as occurring with vehicles which get into a “funky” state, ” The vehicle gets ‘bad’ if it loses the passenger right at region crossing,” he said by way of explanation, leaving them appearing to have somebody sitting on them per parcel accounting rules, but who is effectively a “ghost rider”.

It is hoped that these fixes will form a RC release together with some additional small updates prior to the no change window / code freeze kicking-in on Monday December 16th.

Animation fixes

Commenting on the llGetAgentInfo() update deployed to the RC channels at the Server Beta meeting on Thursday December 5th, Maestro Linden said:

The only change which should be visible normally is a fix for avatars with crouch / crouchwalk animation overrides. Previously, the llGetAgentInfo() LSL function would only return AGENT_CROUCHING if the avatar was playing the default crouch or crouchwalk animations, so if your avatar had an AO which replaced those animations, (either with llSetAnimationOverride() or possibly with classic AOs too), scripts couldn’t tell when you’re crouching. But with the fix, the function is looking at whether you’re actually crouching, regardless of which animations are playing.

He went on to note that there is a similar issue with ground sit, wherein if you sit on the ground, the viewer only presents the ‘stand’ button if your avatar is playing the default ground sit animation. Originally, llsetanimationoverride() allowed the ground sit animation to be replaced with something else, but this led to situations where a seated avatar could not stand up.

To fix this latter issue, Kelly Linden implemented a workaround for this problem by making “ground sit” play two animations, the default ground sit and any custom ground sit specified by the user, with the priority of the default ground sit hopefully being low enough not to clash with any custom animation also used. The change was viewed as a compromise to make the AO system compatible with viewer 2x/3x, and is why the SL wiki alludes to in ‘ State “Sit on Ground” will play the default animation in addition to any override set. This is required for correct viewer behaviour. ‘

SL Viewer

The Fitted Mesh project viewer was updated to version 3.6.12.284458 on December 5th.  The update addresses:

  • LL internal JIRA MAINT-3311 (Skinning to some collision volumes is broken)
  • STORM-1985 (Mesh garments don’t adapt to changes in avatar shape)

In addition, it includes the updated avatar_lad.xml and avatar_skeleton.xml  file developed by Jeremiah Linden in accordance with his notes on FITMESH-2 (notes dated December 2nd, 2013).

Oculus Rift Support

Oculus Rift: release of a "feature complete" viewer with Rift support "soon"
Oculus Rift: release of a “feature complete” viewer with Rift support “soon”

Also attending the Server Beta meeting, Voidpointer Linden reported that support for Oculus Rift is feature-complete and should be released “soon”.

There will be a formal announcement when a viewer with Rift support is released (no date as to when this will be as yet), however, a few clues were given out during the meeting:

  • The same viewer can be used in both a “normal mode” and a “Rift mode”
  • There will be no apparent changes to the viewer / UI when in “normal mode”
  • Frame rates when in “Rift look” will be very much down to the user’s own hardware  (unsurprisingly). Voidpointer apparently attended the meeting using a Rift headest and reported that he was getting frame rates ” pretty comparable to normal,” but also noted he has a good machine on which to run SL.

Details on the presentation of the UI, etc., were not provided, as these are apparently still under wraps. In the past, it had been indicated that the UI had been set to be floating “overhead”, just outside of your normal point-of-view, so you had to look up to see them. Whether this is still the case, remains to be seen.

There have been reports of people using the Rift (in general, not just with SL) suffering from nausea and motion sickness. Commenting on this, VoidPointer said, “I’ve been using it for a while now and I don’t really have problems with nausea at this point. [But] the Rift is very sensitive to frame rate, vsync, and other things.   Before the rendering was fully hooked up or optimized,it wasn’t as fun, I’ll say that.” He also revealed the Rift headset can be somewhat adjusted so it can be worn over glasses, if necessary.

As the Rift is not currently commercially available, those with the headset and SDK will be able to make use of the new viewer once released, and will require a DVI cable connected to the headset and their system for the video output and a USB connection for the head tracking capability (so the screen view follows the wearer’s head move to present them with the expected view). Commercial versions of the system will use HDMI for the video.

Rod Humble tries out Oculus Rift in a photo released on July 18th
Rod Humble tries out Oculus Rift in a photo released on July 18th

There was a lot of additional talk about possible future options for presenting in-world views with the Oculus Rift, however, as Voidpointer advised, “Heh, let’s get Rift support first, then talk about more :).”

Whether support for the Rift will be announced before or after the end-of-year break remains to be seen.