Archive for the 'Testing' Category Page 3 of 4

Progress Update : Taketori Katana

Just wanted to talk a little about the progress on the new sword, which has been "under development" for far too long

This sword has taken me far longer than any previous weapon I’ve made, for a number of reasons.  Probably foremost is the fact that I’ve just not had the time to devote to getting it done faster, but other reasons include the fact that I had decided that this new weapon absolutely required new animations. 

Since I made the Wave Katana I’ve learned quite a bit about creating Second Life® animations, and I felt that I could do a far better job this time around.  Previously I had tried to give a more "real-life" feel to my animations, which were inspired by my own real-life study of kenjutsu, but the animations I ended up with didn’t really convey the dynamism that such simplistic and direct movements have when performed by a real-life master such as my former instructors (or this guy). 

This time around, I thought that it would be good to exaggerate the movements and make them more "Hollywood-style", which I hoped would convey more of a feeling of power and speed.  While I still have a great deal to learn about animating and am far from an expert (hey, animations are hard to do well), I’m fairly pleased with the anims I’ve got this time around, and it’s very gratifying to me to be receiving so many compliments on them.

Another thing I’ve done this time around is spend a great deal of time playing with the control code, perhaps even too much time.  I’ve tried a few other melee combat systems in Second Life, and while C:SI is already the most dynamic and responsive in my opinion, I just really wanted to try out some ideas to see if I could get a more consistent responsiveness profile across a wider range of sim conditions.  It’s still kind of unproven whether I’ve achieved anything worthwhile in this area, but those people who are beta testing have good things to say for the most part.

Making most of the weapon out of sculpted prims didn’t help it get done any faster either, of course.  This sword was my opportunity to really see where I can go with Second Life’s new sculpted prim type, and although I’m far from an expert 3D modeller, I think I’ve managed to do a few things on this sword that I’m pretty proud of. 

So what’s left before I can put it on the market?  There are several things that still need to be done, here’s what comes to mind right away:

CRITICAL ISSUES/TASKS

  • Fix the "kick bug".  Linden Lab has thus far not been helpful in resolving the issue with Havok 4 breaking the push on kicks, so I guess we’ll need to explore this ourselves to make up for the lack of assistance or relevant information.
  • Figure out just what I’m going to use for the special attack on this blade.  I’ve had several ideas banging around in my brain, it’s time to settle on one of them and get it done.
  • The customization code needs to be heavily tweaked for this new weapon, and it’s possible I’ll have to redo some textures to make them more "colorizable".
  • Create an update server.  This is pretty easy, but is necessary before putting anything out.

"NICE TO HAVE" FEATURES

  • I would like to be able to resolve the problems with the block-rolls and moving slashes, and have made some progress in this area.
  • Better drawing/sheathing animations.  The ones for the Wave are just not good enough, but it’s not absolutely necessary to do this, could change them in a future update.
  • I would love to do a new product box.  Mine are the plainest in C:SI, and I’d love to do something more imaginative than what I have, but again this is not a "must have".
  • I’m sure there are at least another 50 things I can put here, but it’s taken me too long to write this post and my RL boss just called :)

This is probably a good time for me to point out that the prototype that people have been using is NOT an update to the Wave, but rather is the anims and scripts for the new sword in a Wave body just to make it quicker and easier to get a prototype out to testers.  The final product will be an entirely NEW sword and new product, not an update to the Wave.  I’ve had some people recently buy a Wave thinking that they are seeing my testers using an update about to be released, and it’s an entirely different product.

I have moved the existing scripts and animations over to the new sword and sheath builds this morning, and everything appears to be working correctly.  Those of you who are beta-testing the prototype and are providing me with regular feedback can expect to receive a new sword sometime in the very near future.

It’s coming, I promise :)

Popularity: 46% [?]

Flabbergasted: Why don’t users report issues?

bugs-icon First of all, I do know that, despite the inflammatory headline, many users do in fact report bugs, and those users are a valuable asset and are appreciated.  But without any way to know specific numbers, I’d guess that maybe 75% of the bugs that are experienced by C:SI users go unreported.

This is especially frustrating because I have on many occasions heard people - either first-hand or through stories related by very close friends - saying derogatory things about C:SI developers (and me specifically) because of the bugs.  This is pretty unfair in my opinion, and such people are a part of the problem they are complaining about.

Not everyone says nasty things, of course, probably most people simply don’t say anything at all.  I guess I can’t fault people too much for that, it is after all not their job to spend their own time helping us with a product they had to spend money on, though in some cases it is only the end users who are in the position to see a problem.

For instance, there was apparently an incident last night (on Meiji?) that was reported to me by my buddy Zai, where several people had their C:SI weapons break in a short period of time.  This seems to be the same problem that has been discussed several times recently both here on this blog and the C:SI forums, and which is so sporadic and uncommon that no dev has been able to reproduce it. 

Every time I’ve written about it, I’ve asked people to report when it happens, with as much information as they could give, and only a few people have responded to the call (a special THANK YOU to those that have!!!!), certainly fewer people than have experienced the issue.

For whatever reason, though, of the supposedly several people who experienced the problem last night, so far as I know at this point only Zai has reported it.  It may be that the others have reported the problems to other devs, I don’t know yet, but I do know that only one person reported it to me personally.

Because I know Zai in real life, I was able to discuss the problem with him and ask him questions in real-time over the phone, and I think that the information he gave me was very enlightening.  For instance, he is the very first person to my knowledge to describe the actual symptoms of what it means when the "weapon broke". 

According to him, after the script errors occurred, people could still attack with the swords, but while the weapon would still play the attack animations, there was no damage, no blood sprays, and no sounds. In addition, it didn’t seem to be happening to the Wave (or at least not to him personally).  I have yet to find another person to validate that information, but if it is true and is what most people who experience the problem are seeing, then I believe that it tells me at least which one of the scripts is most likely to be the culprit.

If the info he reported is correct, then I am pretty excited to know that I can localize the problem to a (hopefully) single script and start really looking into it.  This is the first big break we’ve had on this bug in terms of locating the offending code, though thanks to Darien we might know *what* is happening, just not where.

So I’m asking again, please, we really do want to fix this bug.  If you experienced this issue or experience it in the future, please send us a notecard or IM with as much information as you can remember about what happened.  Did it happen during a fight?  Did it happen drawing the sword?  Could you still attack, was there blood, sounds, etc?  You never know what important details might be critical to fixing the issue, so please please please send us any information you can.

Popularity: 37% [?]

Mono Update 6 Now Available For Testing

Linden Lab® recently posted that the Beta grid has a new Mono update deployed and available for testing :

The Mono regions on the beta grid are now running on shiny new simulators. This update includes fixes for the following:

  • SVC-1414 llBase64ToString does not work properly
  • SVC-1397 Unicode character difference between LSL2 and Mono
  • SVC-1867 Unable to upload changes made to scripts in inventory
  • SVC-1323 llPlaySound / llTriggerSound fail
  • SVC-1827 if (key) always returns false
  • SVC-1857 llDumpList2String corrupts key values
  • SVC-1451 Chopper Hoverboard doesn’t work in Mono
  • SVC-1707 Keys cause invalid IL exceptions
  • SVC-1825 Control event triggers runtime error

Just looking at that list of fixes gives me hope.  At least 4 of those 9 items were blocking issues for C:SI/Mono testing, and without those being fixed there was no possibility of going further.  Yay for fixes!

As usual, I’m way too busy at work to be able to take the time out for immediate testing, but I’ll definitely try to get around to it as time allows and will report my findings here.

Popularity: 38% [?]

Testing Tools : Fight Results Report

This morning I locked my previous post on the device that I was previously calling the "Duel Data Monitor", due to a disturbing recurring theme that people seem to be worried that it will somehow adversely affect the C:SI system, which I have absolutely no desire to do.  Rather than attempt to keep up with the comments on that post, which I am feeling FAR too sick for (I suffered a relapse this morning dang it), it is easier for me to just state my position here.

My desire is to create a device that records fight data as an aid to the C:SI developers in their goal of creating the best melee combat system Second Life® has to offer.  I have no desire to do anything that will reduce hard work and training to a "set of numbers".

There is a great deal of data that would be useful to C:SI developers if we could gather it, and I strongly believe that it is in everyone’s best interests to do so.  There is, apparently, a lot of concern over publishing that information, so it will not be available publicly.

The only people that will have the ability to record fight data will be beta testers and the C:SI developers themselves, and the devices that allow such recording will be strictly limited to those two groups of people.

I had thought that it might be interesting and instructive for a much broader group of people to be able to review the results of their own duels (note that the device could only ever be used to record the fights of the device’s owner anyway), but I cannot at this point come up with a way to reconcile that possible benefit with the notion that it might serve to lessen the accomplishments of others, so I have given up on that idea for the time being.

I hope that will address any concerns people have.

Popularity: 43% [?]



Bad Behavior has blocked 143 access attempts in the last 7 days.

Close
E-mail It