00:00
00:00
Zelazon
Retired music designer and retired tournament gamer.
Commentator for current state of gaming.

Age 37

Joined on 1/14/24

Level:
2
Exp Points:
32 / 50
Exp Rank:
> 100,000
Vote Power:
2.33 votes
Rank:
Civilian
Global Rank:
> 100,000
Blams:
0
Saves:
0
B/P Bonus:
0%
Whistle:
Normal

Zelazon's News

Posted by Zelazon - 3 weeks ago


Post development, the characters that had not returned was the following:


Zimilon (Missing since 2004)

David Lega (Died in the original story, did not return subseqently)

The Twin-Swords Sibling (Missing since 2005)

Grant Lega (Pre-timeskiped character that did not return from the old timeline)

Omega Final (Missing since the timeskip and not in the subsequent versions)

Old-Crenval (Missing since 2005)

The Dark Elementals (Missing since timeskip since 2005)

Talvix (Missing since 2018)

Zelazon (Died 2019)


Posted by Zelazon - 3 weeks ago


The current architecture for the old shadow gods bosses in the story (not as rendered as the prime three but still a good starting point):


Ferculin:


iu_1398728_20717386.jpg


Zodium:


iu_1398724_20717386.jpg


Fiste:


iu_1398725_20717386.jpg


Fate:


iu_1398726_20717386.jpg


Unknown:


iu_1398727_20717386.jpg


All are very hard to defeat in the story.


Posted by Zelazon - 3 weeks ago


My story's enemies keep getting stronger and stronger.


Onthra:


iu_1398711_20717386.webp


Onthra


Crenval:


iu_1398712_20717386.webp


Crenval


Talvix:


iu_1398710_20717386.webp

Talvix


These three don't show up in the story until really in the 2nd half of the story with most of their power intact. The character that fight against them are at 4th or 5th form by the time they are at this omega level, but to think that I would have to continue to look at them meant that I had to take their design seriously. I have long since removed myself from editing the story, but it's been interesting to see that implementation has gone perfectly.


The three evils are essentially designed for maxed out characters, but I just see now that with movement AI, I was kind of right to keep it in the back of my head until I could further evaluate them. Impressive how far AI technology has come, but troublesome to see that these character need a bit of an design adjustment as they would be too challenging from their original design.


Overall, I think the pacing with the technology is great that it allowed me to evaluation my original characters, but I do think that today, I see that I don't really like it as much as I used to. I think Crenval and Talvix specifically have inherent problems with their abilities and damage values that would take too long to adjust (backend things with the design from the story concept).


Then there's Zelazon themselves that I'm like... oh boy. That character has to be redesigned overall... Sigh.


Posted by Zelazon - June 29th, 2024


Pretty good.


http://sndup.net/szw2r


http://sndup.net/wk6gr


They refused to let me share one of my vocal recordings for testing, so here is it.


Additional:


After listening to it, some of the ideas that I was intending with these songs were kind of changed to fit a different feel than I was intending. I like them, but as someone who is extremely proficient at creating song, it's only as good as the AI can provide, but I think it's just easier for an experience to create something better.


I just see too many limitation that are worse then Soundgoodizer or OTT (don't know if they're still used these days), but I just don't know how you can beat an NGADM artist with AI songs.


Posted by Zelazon - June 16th, 2024


How the game even got published on steam is anyone's guess.


These days, it's pretty easy to draft up custom work and generative music for a custom game, but as I continued with the game, I noticed myself that the formulas that the game used is not working at all.


The main problem is the default damage formula, which is the following:


a.atk * 4 - b.def * 2


The "a" points to the entity or person using the attack, and "b" points to the target of the attack.


The formula at first seemed pretty basic, and doesn't really draw any sort of ire from any direction. The formula looks to be solid enough to get through a play through without much problems. What I ran into first is that my stats on initialization were so poorly forthcoming that the damage required massive changes to the character themselves to be able to damage properly.


You may ask me why this is even a problem to worry about, which I will get to.


So, if you have problem with a +10 sword, you should not get a 300 attack damage buff by having a +10 sword vs. a +1 sword. The formula was so out of wack that I had to divide the formula, add extra calculation, and do a whole lot of changes that would be a nightmare for a mathematician because I had though I got the formula correct for late game, but when I changed my defenses to be reduced to no defenses, my character was receiving no damage.


For some reason, their formula seems to be increasing the damage that the character received upon receiving more defense.


I don't know why this happened, but RPG Maker XP (many years ago) did not have this problem. It used to be solid as designing your own RPG, but now it's just that the formula is damaged to the point of being unplayable.


Why on an advance scale this matter is because with AI technology, I was able to develop a slot machine in the game that could calculate the value of each roll and result in a certain winning with the following formula:


Control Variables: #0001 Slot1 = Random No. (1..5)
Control Variables: #0002 Slot2 = Random No. (1..5)
Control Variables: #0003 Slot3 = Random No. (1..5)

Text: Slot Results: \V[1] \V[2] \V[3]

Conditional Branch: Variable [Slot1] == Variable [Slot2]
 Conditional Branch: Variable [Slot2] == Variable [Slot3]
   Control Variables: #0004 Prize = 0
   Conditional Branch: Variable [Slot1] == 1
     Control Variables: #0004 Prize = 100
   Else
     Conditional Branch: Variable [Slot1] == 2
       Control Variables: #0004 Prize = 200
     Else
       Conditional Branch: Variable [Slot1] == 3
         Control Variables: #0004 Prize = 300
       Else
         // Add more conditional branches as needed
       Branch End
     Branch End
   Branch End

   Text: You win \V[4] gold!
   Change Gold: + \V[4]
   // Add other prize logic here
 Else
   Text: You lose!
   Control Variables: #0004 Prize = 0
 Branch End
Else
 Text: You lose!
 Control Variables: #0004 Prize = 0
Branch End


It's not complicated in it's own right, but it is something that need a little assistance in order to figure out how a slot machine would work (mostly for discovery). Where it went wrong is in the following formula to weight the amount of a single number would populate:


Common Event
◆Script:var weights = [1, 1, 1, 1, 1, 2, 2, 3, 3, 4, 4, 5]; // Adjust the array to increase the chances of certain numbers
◆Script:var randomIndex = Math.floor(Math.random() * weights.length);
◆Script:$gameVariables.setValue(1, weights[randomIndex]);

Slot Machine Event
◆Common Event: Weighted Random
◆Control Variables: #0001 Slot1 = Script: $gameVariables.value(1)
◆Common Event: Weighted Random
◆Control Variables: #0002 Slot2 = Script: $gameVariables.value(1)
◆Common Event: Weighted Random
◆Control Variables: #0003 Slot3 = Script: $gameVariables.value(1)

◆Text: Slot Results
◆Text:\V[1] \V[2] \V[3]

◆If:Variable [Slot1] == Variable [Slot2]
 ◆If:Variable [Slot2] == Variable [Slot3]
   ◆Text:You win!
   ◆Control Variables: #0004 Prize = 100 // Example prize
   ◆Change Gold:+ \V[4]
 :Else
   ◆Text:You lose!
 :End
:Else
 ◆Text:You lose!
:End


For some reason, the formula not only bugged out, but it will not register the common event which call the event for the random number to generate a value from it so that the event can register the number correctly. This is a huge problem and a huge barrier because you cannot generate weighed randomized number without having the events populate the correct number from a set of numbers.


What this additionally means is that you essentially get weighted down with using static randomization because the number will only take in bare-scripted numbers of 1, 2, 3, 4, 5, 6, and so on, versus having the flexibility of weighed calculation and formulas that could have made for a more wholesome experience by having a meaningful formula that could have supported the game.


If you mess up the formula, the game is unplayable. Essentially, the scaling will be destroyed and there is no chance of repair because the game needs a fundamental rework.


Why I'm not considering upgrading is because if this was the problem, then RPG Maker MZ would have the same (or worse) flaws because XP didn't have the kinds of issues I am seeing today.


It's unfortunate that stuff like this kills the game.


Posted by Zelazon - June 15th, 2024


While I write from a back-up monitor I hoped not to used, I need to go over some things with the newest version of Stable Diffusion 3:


iu_1222299_20717386.webp

The old model that I had been using for attempting my so somewhat attempt of a game in RPG Maker MV, I found with Moonbeam, that it's been relatively useful, though quite a bit bashed at coming up with new monster design that AI can think of (it seems to be good a randomly generating monsters names that seems to populate an image).


However, with the newly release Stable Diffusion 3, these monsters though highly detailed are almost in the eyesore category versus the seamless designs that old configuration that many gambles would allows to permeate the AI image platform. Though I am little trusting of the AI monsters (please see aforementioned), I am quite on the fence about where the designed point of view is coming from due to having these now odd designs become the standard footprint:


The Fallen King (Boss) old design:


iu_1222300_20717386.webp


New Stable Diffusion:


iu_1222301_20717386.webp


A part of me wants to like the newly updated imagery, but considering that both architectures are asking for the same blueprint, I just don't know how I could fit the new design in. I have a little trepidation on how the change are not really taking a serious evaluation as to what they are expecting on the developer side, but just trying to incorporate as many details as possible without a meaningful approach to what the design is supposed to be. It just felt like too many people are trying to make something cool versus have designs that have proper implementation.


I'm just writing my thoughts as I was reviewing what it is today, and how it's changes dramatically over a single year. I don't really know what to make of the new designs, but I just think that maybe they should slow down and really think over the incorporation of the designs versus just trying to upgrade for no reason.


That's just me.


Posted by Zelazon - May 12th, 2024


Last week was rough...


I took a week off from my job to take a long-needed break from work, but something gave me the feeling that I should upload as many 8k videos as possible for some reason. So, I managed to upload like 10-15 8k videos and only two did not processed in the adequate format, which I utilized an external storage website to store the videos if viewers needed to see the videos.


I am so glad because I had a lingering worry that the tv that I was using, the LG Nanocell, wasn't quite up to task with keeping up with the resolution. One of the problems was that mediocre games like Phantasy Star Online would still keep the heat from the TV at a super-high temperature, and I had to stop using it from time to time so the TV could cool down. Last week, I was encoding the videos into a .webm format to make the rendering more efficient for playback on YouTube, when the lines started to form on the TV horizontally.


I am currently getting rid of the TV.


I had only had the TV for less than two years and this was not quite the first time it happened but is the first time where it has been entirely and reasonably overdramatic because the TVs at 8k resolution are too expensive to replace. When I got my Nanocell, I spend over $3,500 on it in prep for the resolution, and that TV is not even available at 65 inches. I had to save up a few months doing professional work, but it's been entirely jaw-dropping that I am directly experiencing that TV at this price range have this kind of problem.


I've tried to reason with myself on whether to spend the kind of money to replace the TV, but after my current experiences at this level, I just cannot think about it at this point without a monumental shift in reasonable prices and some sort of magic trick that will make it slightly feasible to go through this effort again. Seriously, I'm just taken aback at this point how the whole experience has been.


My first attempt to fix the problem was to buy a "8k" project which I am currently using. It's not good, I can barely see the words, and the resolution does not even show in 8k, which means I didn't waste too much money, but my projector is severely inefficient at the task expected at "4k" resolution.


I may still use the projector, but I'm not really going to hold my breath. If these companies cannot get their act together, I'm stuck in a bind because I just don't want a 65-inch monitor, and I need a better alternative for an 8k tv or monitor. I just don't have any good options, and it's sad to be in this state, but I took a risk and at least got the videos that I had made out. Better than nothing, I suppose.


I do not know when the fix could be, but it could take a few years.


Posted by Zelazon - April 29th, 2024


YouTube is good at many things, but horrible at uploading videos.


I've been using it for nearly over 15 years, and the problems have always been the same problems. Things take too long to upload, by they time they do, the video is not in the right format, and etc.


Today, It's hard trying to get around the fact by 2024, the problem has gotten significantly worse by providing no options for standard consumers who are looking to just post videos, and only want to offer pretend user services that cannot actually comply with their own advertisement.


It takes about 4+ days of uploading an 8k video (which at 10 minutes does average above 5 gigs of data for a video). Though the video is extremely large, it's just pretty sad that the video service has not come up with a good way of resolving this time disparity. Maybe I'm one of the few individuals who can just actively upload a video if necessary, but this problem has not changed since I started in late of 2022, and you wonder with these multibillion-dollar industries why they cannot get this problem corrected.


YouTube and similar companies always brand towards influences, but for influences, how can it go on this long that the company has not been notified of the lag it takes for uploading? It's one thing to be a college student uploading cool videos to friends, but at this kind of scale, this can ruin a lot of companies if they cannot get their resolution at the current standard of today. It is depressing that I am one of the few individuals who are actively pointing to this problem.


At some point, it has to be noticeable that something has failed. What do you do when you're talking about the primary company that's supposed to be state of the art, and they are still on square one with this issue? Maybe I'm being nitpicky, but the optimization should not be taking a week or more for a ten-minute video.


Maybe it will get better in 2028.


Posted by Zelazon - April 29th, 2024


Didn't know this existed (secretly):


Gamers' Bill Of Rights (gamersbillofrights.com)


iu_1196164_20717386.webp


Posted by Zelazon - April 23rd, 2024


iu_1193903_20717386.webp


The things you can make with AI has exceed expectation of energy and wit.


I cannot believe it. It is done.


Edit:


A SBS photo of space


iu_1194330_20717386.webp


1