About tmpxyz (twitter: @tmpxyz)


Ludum Dare 37
Ludum Dare 36
Ludum Dare 35
Ludum Dare 33
Ludum Dare 32
Ludum Dare 31
Ludum Dare 30

tmpxyz's Trophies

tmpxyz's Archive

A Minimal Board RPG

Posted by (twitter: @tmpxyz)
Monday, September 5th, 2016 3:02 am

This is my 7th LDjam, I decided to make a challenge for my prototyping efficiency.

And here I made a RPG game for this jam, with minimal narrative / control / event / battle.

Stayed up whole night on last day, cut many extra features, pushed to the last minute, and finally finished this little game.

Hope you would have some fun time with that :)







Almost failed to finish

Posted by (twitter: @tmpxyz)
Tuesday, August 30th, 2016 5:05 am

[PLAY MistIslands]



Phew, that was too close, almost failed to finish the game.

Recently I’ve been reading the designs of classic boardgames, like Twilight:struggle and Babarian prince. So I decided to try making a boardgame style RPG this time.

It feels a little dry and lacks real challenges now, I didn’t have enough time to populate enough interesting content at last.






Though kind of late, there’s one thing I have to say

Posted by (twitter: @tmpxyz)
Friday, May 6th, 2016 3:18 am

My fellow Unity game developers,

Please don’t package those 100M+ .pdb files into the compressed files again.

It adds the package size for no meanings.

Have fun making games,

Have a play with BattleMath

Posted by (twitter: @tmpxyz)
Saturday, April 30th, 2016 11:38 am



BattleMath is a bloody spice for old rigid math. Have fun solving math with brute force 😀

Greenlight Concept Page

Have a play with “BattleMath”

Posted by (twitter: @tmpxyz)
Monday, April 25th, 2016 8:38 pm

View post on imgur.com

Play Here

You can solve math with brute force now. :)

Have a play with “BattleMath”

Posted by (twitter: @tmpxyz)
Friday, April 22nd, 2016 9:44 pm

View post on imgur.com



This is some blood & spice added to rigid math. Have fun people. 😀

PostJam summary for BattleMath

Posted by (twitter: @tmpxyz)
Friday, April 22nd, 2016 11:30 am

View post on imgur.com

PLAY BattleMath

At start, I intended to try making something traditional for this jam,

Because I just finished “Witcher3” last month, so I decided to make some melee combat theme game. And I just have a set of digit models at hands, so I added hands & feet for them and skinned the mesh with blender.

There’re only a small set of animation clips to make, (4way walk + run, player attack, player evade), I made these clips in Unity with my own Skele plugin.  (A shameless self-promo here 😀 )

Although I used Navmesh for AI pathfinding, but didn’t drive them with transform but with Rigidbody, as I expected to make some collisions and I worry that frequent state changes would be too error-prone. Also a nasty gotcha of NavMesh took me some hours to crack.

I’d been playing with the PandaBT from assetstore for a little while, its examples are quite interesting. From the perspective of programmer, I like it more than RAIN, but RAIN has a better UI and debug interfaces.

But during this jam I felt that my AI scripts should keep more high-level, I’m not that familiar with the BT syntax and when things got messy, it took extra time to debug. I think I need do more experiments with BT later.

The player’s control & state transistions took longer than I expected, directly resulted in I cut off extra attack modes for player. It took quite some effort to tweak the physx.


Do some math with impact

Posted by (twitter: @tmpxyz)
Tuesday, April 19th, 2016 9:22 am

Play “BattleMath” Now


It’s time to do math with brute force 😀


It’s time to do math with brute force

Posted by (twitter: @tmpxyz)
Monday, April 18th, 2016 11:37 pm


Play “BattleMath” Now

You’re going to make simple math here, the only obstacle is that the numbers are resisting you.


13-hours milestone

Posted by (twitter: @tmpxyz)
Saturday, April 16th, 2016 9:31 am

Just setup some placeholders and basic controls & NPC;

Took more time than I expected, must speed up tomorrow.

View post on imgur.com

Some experience learned so far

Posted by (twitter: @tmpxyz)
Sunday, August 30th, 2015 12:42 am

One of the best thing of GameJam is that, unlike the full games, most games in Jam are created aiming at only one or two main points, just like dishes made with fewer kinds of ingredients. It made easier to pick out each ingredient, taste it, think about how the cook made it, and why it tastes so.

Of course, the skill of “cooking & combining ingredients” plays a large part in the final result. But as a cook, it is very beneficial to know the taste and characteristic of every ingredient before cooking.

Here are something I found in these entries:

1. Youkai

Violence is an instinct-level desire within people. Blood, brutality, screaming, broken limbs, etc. could stimulate such illusion of violence, raising the dopamine level.

Youkai uses the human enemies, it increases the feeling of violence too.

Youkai has equipped a dash-based killing system, player needs to rush through the enemies to take them down, and in the same time player needs to keep notice of other enemies around who are ready to attack. The pace of the game is relatively fast, and become faster with more and more enemies.

The game grants an invincible buff when players got 10-kills combo, the combo reward is good in itself, but IMHO, the buff is granted too frequently and breaks the FLOW state of player.


2. Beats of Nightmare

This is a well-made reflex game, hard, fast-paced, with strong beats, simple graphic is perfectly programmed to match with the music beat. Also have high requirement of precise control and unforgiving death.

All of this helps to bring player to focus on the game and enter FLOW state. Of course it could scare off many players too.

It reminds me of the “Super Hexagon”, but this one has more features and the difficulty is lower.


3. Tentacle Rescue : Summer of Love

Well, we all know that instinct-level desire Freud paid a lot of attention to.

This game is quite straight to the point: keep the starfish away, the basic action is click and drag, when the starfish is red, the dragging will rip a piece of clothes.

Yeah, it fits the theme well. :)


4. Coin Block Clicker

It looks like a plain idle game, its name suggests an idle game. But it’s not.

You will get no spoiler from me.

This game is not aiming at instinct-level or manipulative-level pleasure. Its main feature is humor and surprise. Such features are often one-time use and make them very low cost-effective.

But good humor and reflections-level stimulation might be more effective to make people to remember longer time.


At last, a shameless self-promotion of my own game, hope you would have a play and give some opinions. :)

0. Hammer Destroyer

Timeline of development

Posted by (twitter: @tmpxyz)
Wednesday, August 26th, 2015 10:52 am


PLAY “Hammer Destroyer” 


This time I’ve been using Toggl to track what I’m doing during the whole development period.

This timeline could help me to review the time distribution and efficiency.

This LD I’ve spent less time at the starting brainstorming process. But it costs me several more hours working on things that not show up in the final build.

Again the last day is the most efficient day, I even added an AI and a scene in the last 2 hours.

I should make better plan and allocate the time in a better way next time.


Date Duration Action
08-22 2:09:39 Brainstorming
08-22 1:57:50 Prepare assets
08-22 1:29:22 Prepare
08-22 1:20:43 make scene
08-22 0:39:55 work on base mechanism
08-22 0:16:01 make a bunch of building models
08-22 2:23:28 make a bunch of building models
08-22 0:17:11 Prepare ground material
08-22 1:06:54 debris
08-22 0:38:45 runtime gen building
08-22 1:16:49 runtime gen building

08-23 0:38:25 runtime gen building
08-23 0:49:46 runtime gen building
08-23 0:41:22 loading screen
08-23 0:15:36 loading screen
08-23 0:37:25 HUD UI layout
08-23 0:45:54 player Data model
08-23 0:07:51 choosing BGM
08-23 0:24:49 Building hp
08-23 1:40:06 player control: Locomotion
08-23 0:08:20 player control: camera follow
08-23 0:30:45 fix: gen building
08-23 0:42:58 player control: turn around with mouse
08-23 0:50:19 fix: tails jittering
08-23 0:56:51 player control: punch
08-23 0:21:51 gen building collider
08-23 1:47:29 IK swing
08-23 0:43:20 IK swing : collider model

08-24 0:45:27 fix transparent rendering
08-24 0:24:43 fix idle action
08-24 1:48:08 IK swing: weight
08-24 1:09:21 collision stops player moving : collider on mesh
08-24 1:12:05 collision stops player moving
08-24 0:07:20 make arm bigger
08-24 0:29:04 make arm bigger when punch
08-24 0:00:06 adjust punch dir with cam
08-24 0:59:38 adjust punch dir with cam
08-24 0:40:39 attack building
08-24 0:29:01 attack building
08-24 1:40:53 prepare sfx & vfx
08-24 0:18:24 redo ik-collider
08-24 0:43:13 sabotage buildings
08-24 1:27:43 SettleUI
08-24 0:36:53 test basic play
08-24 0:16:17 fix footstep sound

08-25 0:00:25 fix hand rotation
08-25 0:34:09 add combo
08-25 1:07:04 fix hand rotation
08-25 0:15:30 fix hand rotation
08-25 0:26:14 road
08-25 0:41:07 doodad
08-25 0:43:53 title
08-25 0:57:25 fix hammer render
08-25 0:53:33 fix GS_Watcher
08-25 0:26:20 change IK mode rotation
08-25 2:12:09 helicopter

My idea of adding incentive for rating games

Posted by (twitter: @tmpxyz)
Monday, May 11th, 2015 1:18 am

I think these goals are desirable and good for the community of LD:

1. We need people to rate more games;
2. We need people to give more help/suggestion/comment to less appealing games too.
3. We would like people to give more insightful opinions/suggestion than “Great, nice_work, great_graphics, etc”
4. [Optional] Discourage behaviors that like giving false rating or just give random rating without playing.

I think the best method to achieve that is to provide more **INCENTIVE** to people;

Current rating system already provide much incentive for people to keep rating other games, but the drawback here is the incentive will fade quickly after receiving more than 40+ rates, you will have much less chance to be in the first page, and that’s when self-promotion blogs become more effective. (Of course, Post-mortem is valuable, we should keep the incentive for people to keep writing that)

It’s hard to design good incentive to drive people to do things, but luckily, we’re all game developers, and that’s what we do daily.

I believe many of you could come up with good ideas for that (with low development cost).

Let me just give my own idea first:


(1) The “Top-Page Featuring” hidden-ticket;
Add 6~8 special feature slot at the top page;
and every 15min, for each slot, the featuring ticket will be regenerated and hidden in a random game entry, the guy who first comments that game after that will dig that ticket out, and gain the 15min “special-featuring” chance.

[Optional] add some fun for digging, give hints like “VeryFar/Far/Near/VeryNear”;

This plan might give incentive for people to play random games and give love to those games less appealing.

(2) Add some gamble elements

We all have a sorting-score for sorting in the pages. Use that to boost the incentives.

Even if you didn’t dig out the ticket in (1), you would get some random credit for rating a game;

People could place some credit into the slot machine and might earn a boost of the score, or even a ticket.

(3) Improve the comment quality

(1)&(2) aim to add incentive to drive people to keep playing & rating; but we also need to give incentive for people to improve the comment quality.

An upvote system for comment could do, and the upvote will boost sorting-score.

[Optional] every one could only give X+D+5 upvotes, X is the upvote he received, D is day in judging. This is to prevent vote rigging.


Please help rate my Post-LD game on Kongregate

Posted by (twitter: @tmpxyz)
Wednesday, December 24th, 2014 8:32 am

Merry Christmas, LudumDare Community!

In LD30 compo, As my first LD entry, I made a physics game called “Universal Gravitation”.  I had a lot of fun in making it!

After that, I decided to take a try to make it into a full-length indie PC game, so I have put several months of work in this project since then.

Every Friday, I upload new maps and fixes to Kongregate to collect feedback from players. The rating has been slowly rising, that’s a good sign.

However, the game has only received 30+ ratings and 1200+ plays by now, that means it’s not yet eligible to be picked up by Kongregate’s recommendation system.

So, I’m here to request help from the LudumDare community, please give a hand to play & rate an indie game, your support is VERY IMPORTANT for my development.











(Current game might take 2~4 hours to beat through, your progress is auto-saved in browser, you can also jump to specified levels if you lose the saves file)

Tweak the Difficulty…

Posted by (twitter: @tmpxyz)
Thursday, December 18th, 2014 10:54 pm

I have made some gameplay tweaks to make the game easier.
So now you can beat the game and add comments to previous players like:

“What? I found nothing hard to beat this game.”
“I think this game is easy, maybe just me.”

Well, Okay, here’s the truth, I’ve only got 43 ratings by now, That’s just too few. And I only want it to at least reach 50.
So, I would be very thankful if you could spare some time to play and rate it. Thank you.






Want to take some pure logic challenge?

Posted by (twitter: @tmpxyz)
Sunday, December 14th, 2014 5:55 am

I have played a lot of interesting action and reflex games in LD, I believe you have played a lot too.
Well, before taking next batch of control challenges, why not have some changes to play a pure logic game?


Play Detective vs Killers

[cache: storing page]