• Sign up for TFS Icon Competition! The prize includes Otland Premium time and a medal, as well as obviously taking part in TFS history.
  • There is NO official Otland's Discord server and NO official Otland's server list. The Otland's Staff does not manage any Discord server or server list. Moderators or administrator of any Discord server or server lists have NO connection to the Otland's Staff. Do not get scammed!

Do not use/buy OTC V8

Nekiro

Legendary OT User
TFS Developer
Joined
Sep 7, 2015
Messages
2,367
Solutions
108
Reaction score
1,464

Olddies

Veteran OT User
Joined
Nov 21, 2009
Messages
1,125
Solutions
12
Reaction score
269
Location
Rep.Dom
I see its very closed though (issues are closed). No idea whats the reason for that, maybe he dont want yet to deal with people at this early stage of development. I like that repository. I hope it will not die out and hopefully @fabian766 gonna reach for some help regarding that.
It would help to develop the client more faster but like u said maybe he dont want to deal with people with the issues
 

zippoxer

Active Member
Joined
Mar 18, 2008
Messages
106
Reaction score
42
Location
Israel
its not otclient and for me it is better than otc you can try it download it here Upload files for free - Pre-Alpha.rar - Uploadfiles.io (https://ufile.io/8fsyu258) and create an account here that is just for test this client dont use for play the server

Cool!

I've taken a glance at the code, and it seems like performance is a priority, which is great. Seems like it would break OTClient in performance. This guy knows what he's doing. On the other side, it doesn't seem like it's gonna be very customizable, since scripting isn't supported and the architecture doesn't seem to prioritize for easy customization (like OTClient's easily swappable modules).

Still super excited though 😃
 

Olddies

Veteran OT User
Joined
Nov 21, 2009
Messages
1,125
Solutions
12
Reaction score
269
Location
Rep.Dom
Cool!

I've taken a glance at the code, and it seems like performance is a priority, which is great. Seems like it would break OTClient in performance. This guy knows what he's doing. On the other side, it doesn't seem like it's gonna be very customizable, since scripting isn't supported and the architecture doesn't seem to prioritize for easy customization (like OTClient's easily swappable modules).

Still super excited though 😃
this client is not for bot servers you know what i mean
 

zbizu

Legendary OT User
Joined
Nov 22, 2010
Messages
2,841
Solutions
10
Reaction score
1,909
Location
Poland
It would help to develop the client more faster but like u said maybe he dont want to deal with people with the issues

it's in early development and he probably doesn't want to deal with people flooding him with "this and that doesn't work"
I'd suggest either forking and requesting pulls or not interfering at all until our help is welcome
 

Peonso

Godly Member
Joined
Jan 14, 2008
Messages
1,636
Solutions
26
Reaction score
1,281
i never use OTC. all i hear is bugs here bug there. Get a tibia client and a editor like object builder and got yourself a good running client.
That is the worst display of thread necromancy I ever saw. Totally clueless comment about a issue you don't comprehend and has no relation at all to the thread discussion.
 

CesarZ

Active Member
Joined
Sep 20, 2012
Messages
192
Reaction score
32
At the end of the day the man is right. Somebody tell one of these otc people to get it together.

Give me like a couple of years. i be helping out.
Tell @Nekiro to be ready he is gonna have to keep up with me on tfs 57.1 develop.
 

Farlavor

New Member
Joined
Jun 20, 2009
Messages
50
Reaction score
4
ok. at first, I thought OTClient was an improved OTserver distribution project like TFS, but after seeing the stuff and analyzing, I found out it was an unoficial tibia client that uses vanilla client protocols, can login on ANY OTserver using IP/port and it supports mobile devices.
to be honest, I am agree that it must be opensource. sadly entire OT-server comunities, or at least the majority, have interests in asking for offers.
I had some problems getting a lousy map of ToL (Temple of Light) and almost ended paying cash for it, this is how evil the capitalism really is.

I would love it if this would become opensource, and not having to do the coding from scratch. if all of this ruckus were becuz certain user don't follow the code of "OPEN TIBIA" comunity. converting into a "CLOSED TIBIA" that only the richer persons can afford and enjoy.
What the heck can we do?... we might not consume that material. but. the creator will not make it opensource just for the likes of us.

let's hope if other contributors don't fall into the capitalism and make free stuff for the comunity and for the Tibia fans.
 

Xikini

I whore myself out for likes
Senator
Joined
Nov 17, 2010
Messages
5,145
Solutions
310
Reaction score
3,258
ok. at first, I thought OTClient was an improved OTserver distribution project like TFS, but after seeing the stuff and analyzing, I found out it was an unoficial tibia client that uses vanilla client protocols, can login on ANY OTserver using IP/port and it supports mobile devices.
to be honest, I am agree that it must be opensource. sadly entire OT-server comunities, or at least the majority, have interests in asking for offers.
I had some problems getting a lousy map of ToL (Temple of Light) and almost ended paying cash for it, this is how evil the capitalism really is.

I would love it if this would become opensource, and not having to do the coding from scratch. if all of this ruckus were becuz certain user don't follow the code of "OPEN TIBIA" comunity. converting into a "CLOSED TIBIA" that only the richer persons can afford and enjoy.
What the heck can we do?... we might not consume that material. but. the creator will not make it opensource just for the likes of us.

let's hope if other contributors don't fall into the capitalism and make free stuff for the comunity and for the Tibia fans.
OtLand's fork of the Original OTC (Open Tibia Client), that is still being maintained
 

oen432

Legendary OT User
Joined
Oct 3, 2014
Messages
1,150
Solutions
39
Reaction score
1,169
Location
Poland
GitHub
Oen44
Why? Why do you want OTCv8 to be open source? So you can look at the code and... do nothing? Honestly 98% of people that are bitching about v8 are inexperienced, unskilled or without basic knowledge of C++. You wouldn't be even able to edit them so what's the point? Do you people don't realize that modules are so powerful, you can literally do 95% of edits there?

My first days working with OTC, I've been doing 99% of the stuff by editing sources. Even @Don Daniello pointed out that this doesn't make sense with OTC. Back then I had no idea about modules and what they really can do. Now? All of these client modifications could be done in modules and would be even BETTER to maintain.
Here is the thread and his replay to my post.

"Do not buy"
Sure, you don't have (need) to.
"Do not use"
Look at me, I'm such a rebel!
mocking big brother GIF by globaltv
 
Last edited:

Source

Banned User
Joined
May 31, 2020
Messages
74
Reaction score
61
Location
OpenTibia .dev
GitHub
source61
Nothing wrong with making some money, but at some point, like after probably having 20, 50, maybe even 100+ solo indie dev customers each paying literally $1000 for nothing but really a sometimes overhyped improvement of a one-man game client project (edubart's original otclient), you should reconsider your subscription model (and price) unless you're only in for the money and couldn't give less of a f about the community.

And plenty of pros has been openly negative about otclientv8, the most active one being Nightwolf who made us the first ever AI sprite generator in a very short time (first real AI anything in this community) and works as he says in tech for a multi-billion dollar company.
Not sure if you're satisfied with that though, might take Bill Gates himself to come in before you're happy...?
 

oen432

Legendary OT User
Joined
Oct 3, 2014
Messages
1,150
Solutions
39
Reaction score
1,169
Location
Poland
GitHub
Oen44
Nothing wrong with making some money, but at some point, like after probably having 20, 50, maybe even 100+ solo indie dev customers each paying literally $1000 for nothing but really a sometimes overhyped improvement of a one-man game client project (edubart's original otclient), you should reconsider your subscription model (and price) unless you're only in for the money and couldn't give less of a f about the community.

And plenty of pros has been openly negative about otclientv8, the most active one being Nightwolf who made us the first ever AI sprite generator in a very short time (first real AI anything in this community) and works as he says in tech for a multi-billion dollar company.
Not sure if you're satisfied with that though, might take Bill Gates himself to come in before you're happy...?
So much wrong with your post.
First of all, if you think kondrah sold more than 5-8 copies, you are mistaken. Price is so huge that not every OTS owner can afford it.
Overhyped improvement? So you are literally just following the pack and yelling "OTCv8 is bad" without even checking this client and what it can do.
He can even work in multi-trillion dollar company, they have janitors too, should we ask them for the opinion? Also it's not Artificial Intelligence per se (AI is more advanced) but Machine Learning and being so common in 2020-2021 that you shouldn't be too excited because someone did it. You can google it, study for few weeks and in few days get very similar results, that is if you have proper skills already.
 
OP
Night Wolf

Night Wolf

I don't bite.
Joined
Feb 10, 2008
Messages
390
Solutions
5
Reaction score
565
I was not going to comment here ever again but it's so funny watching every comment from @oen432 that I couldn't hold myself.
Also it's not Artificial Intelligence per se (AI is more advanced) but Machine Learning and being so common in 2020-2021 that you shouldn't be too excited because someone did it. You can google it, study for few weeks and in few days get very similar results, that is if you have proper skills already.

AI is an area that goes from (AI basic algorithms > Machine Learning Algorithms > Neural Networks > Deep Learning), at least at the moment.
I could write here for hours about why AI seems easy but in a certain extent is so obvious that I'll let you figure it out on your own.

So with this in mind, while I understand the root of your comment, it's very misleading to think USING AI and UNDERSTANDING AI are the same thing. See, we have been using Neural Networks for quite some time now but to understand how they learn and how to improve their learn is something yet unknown and an active area of research. The sprite generator didn't achieved the level I wanted it to be, in terms of development I feel comfortable in confirming that was just a proof of concept, still, neither you nor anyone will find about that in google (except for my two topics here in otland, probably). It'll take at least 5 years, probably way more, until you see anything close to that being created/researched ever again. Probably way more time until you see that adjusted to Tibia world.

A few data to back my claims so you don't need to trust my word on this one:
1611780428637.png

If you don't know gartner, they basically make those 'hype cycles' anually to see what are the trends in technology.
If you still don't understand, I literally combined two papers that were released and launched a near end-product with the timeframe of a month while the most optimistic prediction was to have that applied in 5 years. Not only I'm mature enough to see oportunities where no one else was seeing, I was also able to put my idea into practice before anyone else could even think about this being a reality.

Now back to the sprite generator, I used the stylegan2 as base. It's a project released by Nvidia team, they were trying to do it to generate more frames per human-faces using this technique and achieving a more human-like graphics in rendering. It's a small upgraded version than the regular stylegan released 2 years ago. (stylegan2, stylegan)
You probably will recognize as this made nearly all deepfakes possible, and started other companies to work on deepfake dection algorithms.

OK, but how do you take one project that was built SPECIFICALLY to work with 1024x1024 images and was trained for MONTHS using all available latest state of art GPUs and technology and modify it to work with 32x32 images that don't resemble at all a human face?
First of all you need to understand the code, which isn't easy. Just browse to regular stylegan2 repo and you'll notice IT'S HUGE. And it was built by probably the most advanced company in Deep Learning area: Nvidia themselves.

If you were a deep learning expert, you would still struggle a LOT to understand its architecture, let alone improve it further. It took me nearly 24/7 of over a month to understand the basic concepts and see what I could improve empirically. Ok, next thing is: You know how to run it, you have an idea of what can be improved.. now you need a RTX2080 to prove you're right, and the answer take months because it has 4238527 parameters with infinite possibilities of combinations. I've read all papers related to GANS from 2010 onwards, even built a few myself to get more knowledge.
Ok, if you are a beast like me, you probably should have it finished around the same time (~3 weeks). Next step is repeating the same thing, now with another technology called GANSpace (ganspace).

Besides this, you need to enhance your dataset, understand latest releases that are coming out and watching nearly all papers released related to it to make sure you're always updated in latest findings. As I've mentioned before, this is an active area of research and the answers aren't something you will find at google. It will be hard even to find them in sci-hub. I had to spriting a few corrected versions myself so I could retrain the dataset with more data available, that before a paper explained how to edit the architecture so I could have better results with less data.

Will ever be someone willing to do the same thing? Probably not until more libs are out and people can skip some of those processes. I myself wouldn't waste my time/money again on this, let alone for OT community. As technology evolves (cheaper and better GPUs) deep learning will probably become more popular and there will more people researching, discovering and improving current known algorithms. As well as more people doing it for fun just to test the technology.

Next time try to search the technology on computerphile, I think I have even posted this link in one of my previous topics about this. Knowing the basics would avoid you to make such bold claims that this can be achived as easily as any other already built-in machine learning lib for python.
Post automatically merged:

First of all, if you think kondrah sold more than 5-8 copies, you are mistaken. Price is so huge that not every OTS owner can afford it.
Overhyped improvement? So you are literally just following the pack and yelling "OTCv8 is bad" without even checking this client and what it can do.
being a good developer doesn't make him a good leader. He clearly lacks vision to expand his 'product'. The high cost literally is limiting to only few tibia-like servers that can match it. I've tried to give different directions but when someone can't see one palm in front of its nose, there's no point in showing them a sign.

The client is amazing, but it was already amazing with edubart version, let's not forget that.

He can even work in multi-trillion dollar company, they have janitors too, should we ask them for the opinion?
This is pure prejudice being destiled, it's not because someone is in a lower position that they don't have a valid and sometimes better vision/opinion. When I was an intern I knew I could be leading a team, it took 2 years for I start to be recognized and enter leadership positions in my company. My old boss used to think like you and now he's unemployed.
Post automatically merged:


Lastly, having it open source could expand the otcv8 past Tibia community and probably bringing more developers/testers to use this. He could have potential buyers of all different functionalities, similarly how unity asset store operate.
Otc it's great and you can achieve a lot of features by justing changing modules (Lua) as you mentioned, but there will be a point if you're going custom that you'll require the source code and for an indie dev having Xk$ isn't a viable option.
Don't even need to wait much time, want to build a League of Legends OT? You can only achieve certain things starting with src edit, so it's either having to make several improvements that are already made in otcv8 in otc yourself (wasting dev time) or paying.

It's not atractive at all to migrate to TFS source if you have way better sources out there for who want to go custom, such as unity or unreal.
 
Last edited:

skulls

Member
Joined
Aug 28, 2010
Messages
26
Reaction score
24
+1 in everything wolf said. Also, claiming that kondrah is a good developer is wrong by itself. ;)
 
Top