Exploring Exodus

This week the Exodus team joins us to discuss why they created the Exodus viewer, how 1.23 based viewers negatively impact Second Life, graphics and more!

This week’s topics include:

Guests:

The Exodus Team are developers of the Exodus viewer, a specialized third party Second Life viewer geared towards gamers and visual artists based the Viewer 3 series and packed with useful features to enhance the experience of Second Life.

  • Clix Diesel is the voice of Exodus Viewer. He leads the project and brings it all together. He also leads the Second Life military, Ark, and strives to bring the best possible combat viewer experience to everyone that engages in player versus player activities in Second Life.
  • Ash Qin is Exodus’ Linux developer, he also provides various viewer related documentation and server related stuff! Ash also owns and runs the popular Sci-Fi sim Deshima. He is a great motivator and constantly pushes for perfection with his superb administration.
  • Geenz Spad is Exodus’ graphics guy. He is also responsible for the slick shiny Rubber Infusions range of avatar modifications. Geenz works hard to bring brilliant rendering features and improvements to Exodus, adding an additional layer of visual “wow” for users, making Exodus ideal for photographers looking for an extra degree of control in their viewer.
  • Ayamo Nozaki (not present on the podcast) is the lead developer for Exodus. Responsible for most of the combat functions that make Exodus stand out. He is a Second Life combatant with a understanding of what the community wants and expects from a viewer specially made for competitive play in Second Life.

William Reed Seal-Foss (Reed Steamroller) is a 3D artist and virtual world content creator for Sand Castle Studios, LLC.

Bernhard Drax (Draxtor Despres) is a musician, new media producer, and machinima journalist.

Karl Stiefvater (Qarl Fizz, formerly Qarl Linden) is visual effects artist, software developer, and interactive designer extraordinaire.

 

Thank you to our sponsor, Pretty Feet!

About Kim/Gianna

Kimberly Winnington (SL: Gianna Borgnine) is the Emmy nominated owner and CEO of Sand Castle Studios, LLC, a company dedicated to helping organizations maximize the full potential of virtual worlds and social media by creating interactive, social, and 3D experiences. For the last 5 years, Kimberly has helped SCS clients stay ahead of the curve and is a respected resource for information on current and developing trends, social media, and immersive experiences.
This entry was posted in Podcast and tagged , , , , , , , . Bookmark the permalink.

14 Responses to Exploring Exodus

  1. Crap Mariner says:

    Excellent show… great that y’all got the Exodus folks in as guests, and they brought up a few things that were eye-openers. V1.x viewers harming sim performance might end up becoming the next witch hunt though… but then those legacy protocols are holding performance back, no? *shrug*

    (I’ll probably give Exodus a try this weekend… been using LL Dev viewers for a while to check out the bleeding edge.)

    Looking forward to more guests like that… maybe go for the Hard Alley folks or other recent Sim Deathwatch types… or perhaps someone who’s still doing well and is willing to share what keeps them going/their strategy?

    -ls/cm

    • Kim/Gianna says:

      Working on a bunch of guests :) Those are all good suggestions as well!

      • Crap Mariner says:

        Follow-up… been using Exodus.
        Fast. Stable.
        And I love the Quick Preferences button (similar to Phoenix and FireStorm)
        Glad to see they recognize and know of a good thing where others don’t… look back in your comments threads here to have a wry smirk…

        -ls/cm

  2. Vanessa says:

    - wow, all those British accents, how refreshing! The passion of the devs is wonderful to hear. And so terribly polite too – Drax could learn from this
    – Lindenlab should hire Clix as their PR spokesperson. He is spot on, Secondlife is not a game, it is a platform where you can be anything you want.
    – I think Ash has understood why the vampire photos on the SL front page are so bad; it reflects the minimum quality that so many users have. Most games will show unrealistic graphics eg sims online, raising people’s expectations that fail on logging on. At least LL has not given a false sense of expectation for new users. Photoshopping those photos would not reflect an honest representation, although I agree, it could have been made sparkly with the use of windlight
    – I agree with Reed, the Exodus team are a breath of fresh air. I hope Reed’s reinvigorated enthusiasm convinces him to go back to his brilliant building
    – its quite fun to hear Drax trying to interrupt, and being over talked by the experts
    – since when does ‘penis’ need to be censored?

    Overall, a great show Kim. Thank you for arranging to have the Exodus developers on today, it was very much a learning experience for me, I hope it was for your greater listening audience too.

  3. One constant theme with Metareality podcast lately has been looking down at people who prefer v1 UI based viewers. This episodes goes even further and in addition to the usual snark it adds some faux technical reasons to the mix.

    One of the Exodus developers even comes with a claim that v1 viewers hurt sim performance, supposedly because they use older method of fetching inventory. Nonsense. He’s wrong on 3 separate levels in that one statement. He assumes that all viewers with v1 UI use older protocols and don’t update which is not true, and even if it was true fetching inventory doesn’t impact sim performance in any significant way, and even if it did, in order to see any sim impact it would require all v1 users to teleport into the sim where he supposedly did the testing, then promptly log out, then clean their cache, then log back in, because no viewer fetches inventory constantly. The only time that’s done is on the first login after the cache is cleared.

    Another thing is repetitive FUD tactics about Linden Lab’s supposed plans to block v1 from the grid. This has no basis in reality, Linden Lab has made no statements to that effect. I mean I understand that Firestorm and Exodus developers want to promote their viewers, but there is really no need to spread false information to do so. Emphasize strong points of your viewers, point out advantages over other viewers, stick to the things you know and leave out the myths and misinformation.

    As for why a lot people still prefer v1 interface over v2 and even the new FUI, I can point out the reasons why I find it more usabe:

    * Communications: I talk to a lot of people while logged in. I’m often in situations where I participate in local chat and at the same time communicate with several people in private and group IMs. I find it very useful to be able to quickly switch between the conversations. In v1 I can combine local chat and IM conversations in one place and simply switch between them with alt-left and right arrow. When I see new IM notification I simply press ctrl-t which brings it up, another ctrl-t to close it down. To achieve the same things in v2/3 i need to keep mousing around and finding where exactly to click.

    * Notifications: When someone sends me a teleport request, or there is a group notice in v2/3 those get displayed shortly and then get buried into IM windows. Then people ask me why didn’t you respond to my teleport request, and I didn’t even noticed it unless I was watching SL window very closely. Another thing that irritates me lately in v3 is that they have added giant “BLOCK” button on every dialog a scripted object presents. And while operating an object that is menu driven you go through a series of dialogs with menu options. I had to deal with many customers that come to me and say “hey that item you sold me stopped working” just to discover that they managed to mute it while operation the menus by accident. That’s just bad UI design.

    * Building: most things take more clicks to do in v3. My pet peeve is removal of the ruler mode from the edit floater. While building I often switch from local to global mode depending on which sort of prim I’m currently editing. Now in v3 that requires opening an additional floater because they had to jam in their precious PE/Land Impact thingie in there.

    And I could continue like this for 5 more screenfuls, but you get the point: it’s not that I’m resistant to change; I simply find that the tasks that I perform most often are easier and quicker to perform in a modern v1 UI viewer than in v2/3, even with the fancy FUI which is admittedly a great improvement over stock v2.

    • Kim/Gianna says:

      Hi Latif! I appreciate your comments and I’m sorry you think we’re being unfair, but I did want to explain a little.

      First, I’m not saying that Linden Lab will “block” 1.23 based viewers (I can’t see them blocking their own product), but they have repeatedly stated for over a year now that as they will not support 1.23 and will be “breaking” it as they introduce new features. Here’s a post from the Phoenix team (who use 1.23) where even they state “within the next year or so, LL will be essentially breaking all 1.23 and snowglobe based viewers…” as they were told before viewer 2 was released. Recently, Linden Lab have even removed the 1.23 viewer from their download wiki page. In my opinion, these are all signs that the Lab would like to eventually phase out 1.23 viewers, but they haven’t yet because of the large amount of users who still prefer to use them. Also, I can’t see the creators of Firestorm spreading false information about 1.23 based viewers because Phoenix and Firestorm are developed by the same people so they would essentially be putting down their own viewer.

      As far as why you prefer 1.23… I understand and I want things to be easy to use as well. It’s just my opinion that we need to be working on improving the future, not the past. It’s not a judgement of anyone or their preferences, I just want everyone to be able to experience and enjoy the new features that are coming.

      • I am very well aware of the fact that the Phoenix/Firestorm team has been spreading these rumors about LL’s intentions. They use it as universal answer when their team gets asked why they’re “abandoning” Phoenix in favor of Firestorm. I have also called on them to stop spreading false information about alleged “blocking” of v1. See their latest blog post where they tone it down significantly.

        And it’s sounding lately like LL is starting to take a less aggressive stance towards V1 than what we were initially told nearly a year ago. By the sounds of it, LL has no immediate plans now to break v1 by turning off server side functionality.

        Again, they are opensource developers, being one too I fully respect that they have total freedom to work on whatever they please. There is just no need to justify your choices by spreading FUD. I’m glad to see that they are talking that into account in their latest blog. I’d wish Exodus developers would just do that, and say “hey, we like v3 FUI, that’s what we’re focusing on”, and not say that v1 based viewers are going to lag your sim to hell which has no basis in reality.

        And I’m not talking about stock 1.23 either. I personally use Singularity, the viewer that has a rendering engine from the latest v3 but with what is to me a more usable UI. It uses all the latest communications APIs too, like the new inventory system. Henri is also keeping Cool VL Viewer up to date with all the latest innovations Linden Lab produces.

        So all this talk about obsolescence and resistance to change make no sense to me. Choice is a good thing.

        • Ash Qin says:

          > He assumes that all viewers with v1 UI use older protocols and don’t update which is not true

          I said legacy protocols (even the early viewer 2 versions were using some of these now legacy protocols), not v1 viewers. I also noted that newer viewers (like Firestorm) on the podcast can use these legacy protocols. Please stop misrepresenting what I said.

          > in order to see any sim impact it would require all v1 users to teleport into the sim where he supposedly did the testing, then promptly log out, then clean their cache, then log back in

          Are you implying this isn’t a common occurrence on SL? You may find this surprising, while anecdotal evidence, I noticed when users experience any sort of issue, they try clearing their cache and relogging as an attempt to fix their issue. I would be surprised if you didn’t encounter users doing this on a regular basis.

          I have noticed more issues though through legacy texture fetching, particularly on a sim that is already under performing (either due to grid issues or other issues).

          I have done extensive testing previously on the matter using statistics tracking graphs like http://www.subnova.com/secondlife/deshima/html/day.htm

          Rather than give you graphs people will claim I made up, I invite you to test on my sim (Deshima) and read those statistics (updated periodically on the link I provided) for yourself with legacy protocols. On a bad day on the grid, legacy protocols do really appear to affect sims badly.

          • Siana Gearz says:

            Singularity has shipped with HTTP texture fetch enabled by default for an eternity, since almost the very first versions. This puts us at disadvantage regarding crash rate stats compared to Phoenix and Imprudence, which ship with HTTP texture fetch disabled, but i decided to keep it that way, and instead work towards limiting the bad factors of HTTP fetch, such as opening and closing a lot of connections in a short timespan, and tried to compensate against this by being lowering crash potential by other means. Considering we’re so close to others apropos crash rate that the order on TPVD of top 3 viewers is nearly random, i believe i have done well.

            Next version will also ship with HTTP inventory fetch enabled, over newest protocol.

            It is our goal to provide the newest technology, highest compatibility to first party client, and fairest use of resources underneath the classic UI, and you can’t just go around dropping hints which are INTENDED to make the opposite impression and disrespect our effort, i won’t leave this standing like that. I’d expect an apology at this point.

          • Ash Qin says:

            > you can’t just go around dropping hints which are INTENDED to make the opposite impression and disrespect our effort

            Speaking plainly. I am only speaking about ‘legacy protocols’, which is the exact phrase I used on the podcast. I spoke as plainly as possible in the podcast. I also pointed out on the same podcast that modern viewers can use legacy protocols (AKA: Not only 1.x viewers). Stating this is a statement targeted at a specific viewer that we didn’t even mention or hint at is really far fetched in my opinion.

            > I’d expect an apology at this point.

            You want an apology for misrepresenting what I said after it’s already been clarified to not be the case in a previous post already?

            From my point of view, I don’t see any legitimate reason to give one.

  4. Pingback: Exodus Viewer (12.01.02.1) - Page 6 - SLUniverse Forums

  5. Good writeup and good interview! It’s good to see TPVs get some attention! There’s a couple of minor points that may need clarification, however.

    Around 37 minutes in the implication is made that some other next-generation third party viewers cause sim lag due to optionally supporting older V1-era protocols for inventory fetching, texture fetching, etc. This implication is highly misleading! Firstly those newer protocols are on out of the box by default on modern third party viewers such as Firestorm. Secondly, while newer protocols do work well for the majority of users, there are a significant number for whom they do not work at all, on any viewer, including the LL official V3 viewers. This is typically due to a user’s inventory count or distribution, certain types of internet routers, or character of their network connectivity. For these users, their only practical way to experience SL is to fall back to older protocols, while waiting for LL to develop a better fix . LL has admitted many times that their new-protocol HTTP server and viewer code has problems with certain scenarios and they do intend to fix it at some point in the future. Until then, we should be careful not to blame users for bugs that are out of their control, or blame modern viewers who provide fallback methods for those users who cannot have a useful SL experience any other way.

  6. Vitolo says:

    I know many residents that still use older versions of SL with “patches’.
    1.15, and 1.18.3.
    Although the download is off the LL website, viewer 1.23 is available widely on many sites, and I personally refuse to give it up as long as it functions at all!

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>