[Opensim-dev] Ruth vs Gas Clouds: Bug or feature?
Paul Fishwick
fishwick at cise.ufl.edu
Wed May 20 14:04:58 UTC 2009
Re: RC Version 1.23 Viewer
I've delayed responding to this until we could check it out on various
platforms.
For bots connecting to the LL servers, this suggestion below
(Client.Appearance..)
appears to work fine in conjunction with re-texturing the eyes as long
as we recompile
with the latest libomv.We are not sure why one eye texture worked and
another did not, or
whether the act of re-specifying the eye texture did the trick.
For bots connected to OpenSim servers, there are issues: the latest
libomv may be
incompatible with the latest opensim. We have filed a mantis on this.
Also, it would
appear that normal avatars do not rez unless one changes appearance
(each time on
login). The latest libomv might fix this issue.
-p
Dahlia Trimble wrote:
> should be just Client.Appearance.SetPreviousAppearance(true);
>
> On Thu, May 14, 2009 at 4:21 PM, Paul Fishwick <fishwick at cise.ufl.edu
> <mailto:fishwick at cise.ufl.edu>> wrote:
>
> Do you know of any sample code that does this? I wrote this up a
> while
> ago, but it doesn't contain those baking functions:
>
> http://opensimulator.org/wiki/Building_a_bot
>
> -p
>
> On May 14, 2009, at 6:58 PM, Dahlia Trimble wrote:
>
>> There is a baking method in libomv, but it is not as
>> feature-complete as the one in the LL viewer and may fail if
>> incompatible layers exist, or not all layers are available such
>> as skin/tatoo layers or texture based eyes. Hence, many bot
>> programs will not use the baking code but rather rely on viewers
>> for compositing. These are the bots that appear as clouds in the
>> 1.23 viewer; other bots which can successfully use the libomv
>> baking method appear as normal avatars.
>>
>>
>> On Thu, May 14, 2009 at 2:57 PM, Paul Fishwick
>> <fishwick at cise.ufl.edu <mailto:fishwick at cise.ufl.edu>> wrote:
>>
>> perhaps the bot code (libomv) can be changed to send baked
>> layers to the server (LL or OpenSim) ? I think the libomv folks
>> may be looking into this. Short of that, I'll be sticking
>> forever with
>> 1.22 (as far as I know, the regular LL viewers do not time
>> out like
>> the RCs).
>> -p
>>
>> On May 14, 2009, at 5:27 PM, Melanie wrote:
>>
>>> I believe I can explain this.
>>>
>>> JCool, the creator of the Emerald Viewer, has been able to
>>> shed some
>>> light on this, since his viewer has been using a similar
>>> feature
>>> from the start.
>>>
>>> Bots largely depend on viewer compositing, this means they
>>> send the
>>> individual clothing layers, but no baked ones, and the viewing
>>> user's viewer composites them into a viewable avatar.
>>>
>>> Starting from 1.23, LL viewer no longer send, or expect,
>>> compositing
>>> layers. Avatars no longer carry texture information about their
>>> individual clothing items. This is to combat theft, avatars now
>>> carry only the 3 baked textures. Therefore, Copybot can no
>>> longer
>>> rip clothing from an avatar it sees.
>>>
>>> Because of this, the "Cloud" stage is perceived to be longer
>>> on 1.23
>>> users, as they don't render until the bakes have been pushed
>>> to the
>>> server and then downloaded by the other clients. An avatar
>>> counts as
>>> unloaded if it's visual params are the default ones OR it
>>> has no
>>> baked textures. Previously, it would count as unloaded only
>>> if it
>>> had no textures at all, or all default visual params.
>>>
>>> Since bots don't bake, they will never render for users of
>>> the 1.23,
>>> since it neither creates not expects compositing layers.
>>>
>>> Conversely, users of the Emerald Viewer, which features
>>> clothing
>>> protection, appear as light grey avatars until the baked
>>> textures
>>> are pushed to the server.
>>>
>>> Melanie
>>> Arthur Valadares wrote:
>>>> Paul,
>>>>
>>>> True, many issues are coming with the 1.23, but what really
>>>> scares me is
>>>> that this issue is happening on 1.22 and Hippo 0.5.1. So it
>>>> would be
>>>> nice we could fix 1.22 issues before cracking heads on the
>>>> 1.23 ones
>>>> (and they are coming from what I've heard, that's for sure)
>>>>
>>>> On Thu, 2009-05-14 at 16:47 -0400, Paul Fishwick wrote:
>>>>> This has been reported on JIRA at Linden Labs, and seems
>>>>> to occur, at
>>>>> least for me, only with the new Release Candidate (RC)
>>>>> 1.23. Version
>>>>> 1.22
>>>>> does not exhibit the problem.
>>>>>
>>>>>
>>>>> With RC Version 1.23 and an LL server, all bots are rendered
>>>>> as clouds. I noticed this first with Kristen's S17 viewer
>>>>> which is
>>>>> built
>>>>> on 1.23, but the regular RC has the same problem.
>>>>>
>>>>>
>>>>> With RC Version 1.23 and an OpenSim server, all avatars
>>>>> (including
>>>>> bots) are rendered as clouds.
>>>>>
>>>>>
>>>>> So, this is indeed a major issue with the upcoming LL
>>>>> client. For
>>>>> reference
>>>>> see:
>>>>>
>>>>>
>>>>> https://jira.secondlife.com/browse/VWR-13386
>>>>>
>>>>>
>>>>> There are many related issues, including the clothing one
>>>>> you mention.
>>>>>
>>>>>
>>>>> -p
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> On May 14, 2009, at 4:28 PM, Arthur Valadares wrote:
>>>>>
>>>>>> Hi everyone..
>>>>>>
>>>>>> I've been looking into a "bug" where if you get a brand
>>>>>> new avatar
>>>>>> and
>>>>>> create pants, shirt, shape and skin and wear all of them,
>>>>>> you turn
>>>>>> into
>>>>>> a gas cloud. After lot's of research, it seems this is
>>>>>> "kind of"
>>>>>> expected behavior. Let me try to explain what I think is
>>>>>> happening.
>>>>>>
>>>>>> After SL Viewer 1.20, Ruth no longer existed. If you took
>>>>>> off all
>>>>>> wearables, you would now become a cloud of gas.
>>>>>> Apparently, the
>>>>>> viewer
>>>>>> thinks that if you use all those 4 wearables without changing
>>>>>> anything,
>>>>>> you have become Ruth, and so you should now be a cloud of
>>>>>> gas. If
>>>>>> you
>>>>>> make the 4 wearables, but change ANY parameter, you no
>>>>>> longer become
>>>>>> a
>>>>>> cloud of gas . The thing is on OpenSim you start as Ruth,
>>>>>> and not as
>>>>>> a
>>>>>> cloud of gas! Which is kind of confusing that the viewer
>>>>>> thinks you
>>>>>> become Ruth when you're actually trying not to.
>>>>>>
>>>>>> So I wanted to gather some ideas on how to circumvent
>>>>>> this, if my
>>>>>> assumptions are correct, or get corrected otherwise. I
>>>>>> can only
>>>>>> think of
>>>>>> creating default clothing, like Second Life does.
>>>>>>
>>>>>> Thanks all!
>>>>>>
>>>>>>
>>>>>> --
>>>>>> Arthur Valadares <arthursv at linux.vnet.ibm.com
>>>>>> <mailto:arthursv at linux.vnet.ibm.com>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> Opensim-dev mailing list
>>>>>> Opensim-dev at lists.berlios.de
>>>>>> <mailto:Opensim-dev at lists.berlios.de>
>>>>>> https://lists.berlios.de/mailman/listinfo/opensim-dev
>>>>>>
>>>>>>
>>>>>
>>>>> Paul Fishwick, PhD
>>>>> Professor
>>>>> Director, Digital Arts and Sciences Programs
>>>>> Computer and Information Sciences and Engineering Dept.
>>>>> PO Box 116120
>>>>> Bldg. CSE, Room 301
>>>>> Gainesville, Florida 32611
>>>>> Web: http://www.cise.ufl.edu/~fishwick
>>>>> <http://www.cise.ufl.edu/%7Efishwick>
>>>>> Ph: (352) 392-1414
>>>>> Fax: (352) 392-1220
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Opensim-dev mailing list
>>>>> Opensim-dev at lists.berlios.de
>>>>> <mailto:Opensim-dev at lists.berlios.de>
>>>>> https://lists.berlios.de/mailman/listinfo/opensim-dev
>>>>
>>>> _______________________________________________
>>>> Opensim-dev mailing list
>>>> Opensim-dev at lists.berlios.de
>>>> <mailto:Opensim-dev at lists.berlios.de>
>>>> https://lists.berlios.de/mailman/listinfo/opensim-dev
>>>>
>>>>
>>> _______________________________________________
>>> Opensim-dev mailing list
>>> Opensim-dev at lists.berlios.de
>>> <mailto:Opensim-dev at lists.berlios.de>
>>> https://lists.berlios.de/mailman/listinfo/opensim-dev
>>>
>>
>> Paul Fishwick, PhD
>> Professor
>> Director, Digital Arts and Sciences Programs
>> Computer and Information Sciences and Engineering Dept.
>> PO Box 116120
>> Bldg. CSE, Room 301
>> Gainesville, Florida 32611
>> Web: http://www.cise.ufl.edu/~fishwick
>> <http://www.cise.ufl.edu/%7Efishwick>
>> Ph: (352) 392-1414
>> Fax: (352) 392-1220
>>
>>
>> _______________________________________________
>> Opensim-dev mailing list
>> Opensim-dev at lists.berlios.de
>> <mailto:Opensim-dev at lists.berlios.de>
>> https://lists.berlios.de/mailman/listinfo/opensim-dev
>>
>>
>> _______________________________________________
>> Opensim-dev mailing list
>> Opensim-dev at lists.berlios.de <mailto:Opensim-dev at lists.berlios.de>
>> https://lists.berlios.de/mailman/listinfo/opensim-dev
>
> Paul Fishwick, PhD
> Professor
> Director, Digital Arts and Sciences Programs
> Computer and Information Sciences and Engineering Dept.
> PO Box 116120
> Bldg. CSE, Room 301
> Gainesville, Florida 32611
> Web: http://www.cise.ufl.edu/~fishwick
> <http://www.cise.ufl.edu/%7Efishwick>
> Ph: (352) 392-1414
> Fax: (352) 392-1220
>
>
> _______________________________________________
> Opensim-dev mailing list
> Opensim-dev at lists.berlios.de <mailto:Opensim-dev at lists.berlios.de>
> https://lists.berlios.de/mailman/listinfo/opensim-dev
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Opensim-dev mailing list
> Opensim-dev at lists.berlios.de
> https://lists.berlios.de/mailman/listinfo/opensim-dev
>
--
Paul Fishwick, PhD
Professor and Director, Digital Arts and Sciences Programs
University of Florida
Computer & Information Science and Eng. Dept.
Bldg. CSE, Room 301
P.O. Box 116120
Gainesville, FL 32611
Email: fishwick at cise.ufl.edu
Phone: (352) 392-1414
Fax: (352) 392-1220
Web: http://www.cise.ufl.edu/~fishwick
More information about the Opensim-dev
mailing list