Skip to main content

Indie game storeFree gamesFun gamesHorror games
Game developmentAssetsComics
SalesBundles
Jobs
TagsGame Engines

ROBOT-RANDOM

378
Posts
8
Topics
3
Followers
6
Following
A member registered 35 days ago

Recent community posts

(2 edits)

Update: To get the FPGA Sega CD acknowledge I had to create a folder called "Prehis", then put a ROM in it called called "Prehis.bin" with a cue sheet called "Prehis.cue" with the following text:

FILE "PrehistoricIsle4MB.bin" Binary
TRACK 01 AUDIO
 INDEX 01 00:00:00

Then, instead of playing the folder like a Sega CD game, I had to open the folder and launch the bin directly. Only then would it detect the FPGA Sega CD core.

First of all, the new version is very stable. Thank all of you for your time, talent, and hard work.  I have always wanted to see this on the Genesis.

- Game version (shown on the title screen): v5.3

- Platform: emulator + version / flashcart / real hardware + model: Sega Genesis Model 1 VA3 NTSC (non-TMSS), MEDPRO, Everdrive MD v2.6.

- What happened: When foot soldiers are knocked down, instead of sliding slightly, like in the arcade game, they slide far away, sometimes off the screen, as though on a hidden conveyor belt. It does not look natural and seems bugged. It should be an easy fix to simply have them slide half as far.

- What you expected to happen: I expected the foot soldiers to slide the distance they do in the arcade game and not half way across the screen of off the screen.

- Steps to reproduce: Just play the game.

- How often (always / sometimes / once): Guaranteed to occur. 

- Screenshot or video (optional): Too difficult to record it happening while holding a phone. The Genesis is not connected to a capture device. 

This seems to be fairly stable at this point. No clue how you fit all the audio in that tiny file.

The base game has become a lot more stable in this recent build. The side quests, not as much. But the base game is improved considerably.

I'm not the one who negged the developer that was probably M (who is actually an alt of another user).

(2 edits)

I used to play this on the Commodore 64 and it played nothing like this (nor did it look like this). Is it not compatible with 6 button controllers? All buttons cause the plane to explode (except the machine gun). Looks nice though. 

0.5.1 needs to be padded to support bank switching for basic Everdrives.

New version works on Sega Genesis Model 1 VA3.

40 years ago, they had this arcade game at the Eilson Air Force Base Exchange. Thanks for bringing back old memories. 

(1 edit)

V0.8 - (Rester knows the hardware)
First, graphics are greatly improved. Control as well.  A missing area was restored under the waterfall.

Bugs:
(1) Same bugs that were photographed before are present, though the railing partially goes behind the sprite in level 2. So that is improved.

(2) Level 2-1 infinitely spawns enemies until the level crashes. In the arcade game, when the DL-0411-10003 security chip was missing there would be infinite enemy respawns and the gold drops were nearly flattened. Is that related to this problem? Who knows. Worth mentioning.

(3) When the game starts and the sky pans downward, there are some corrupted graphics.  When the mage first appears at the brige on the waterfall she has some corrupted graphics. The first boss still starts with a disembodied arm that should not be visible as it is a reserve animation part.

1. Se trata de una animación poco común que en la versión arcade solo aparecía de vez en cuando, pero ahora aparece con casi todos los soldados de a pie. Deberían ajustar la configuración para que solo ocurra en situaciones muy puntuales.

Please test for PAL. https://drive.google.com/file/d/1v9d8kMRSrsd7yGT73rxzs53FA54LZdGq/view?usp=drive...

Developer already fixed it. Look in Slash's responses. You will see a link to a file. That is the bug fix. It still has problems, but the graphical corruption was resolved (I think). As you are on PAL, it would be helpful if you would test this bug fix version as well.

So far, I've played through all of level's 1-3, then skipped to level 5 and played both sections. The worst errors seem to have been fixed.

* There was some kind of garbage at the end of level 5-2, but otherwise the level was fine (see image).

* The spears that were invisible on level 5-1 are still invisible (see image).

* However, when you knock the foot soldiers down and the lay flat, they sometimes slide half way across the screen and disappear. They should only slide about half that far or not at all. It doesn't look like they are being knocked away; it looks like they are on an invisible conveyor belt.




(1 edit)

Edit: Pointing out technical issues with a homebrew game is not an attack on the user who provides it. They do not have a corporate staff of debuggers & testers and rely on other users to tell them.

Bug: Sometimes at the title screen the game plays a strange note that sounds like machine noise. This does not appear anywhere else. The music in the game is fine. 

There is s difference between PAL & NTSC. Because PAL has a significantly slower refresh rate, and thus a larger VBlank window. it has a larger safe zone, which is likely why the team didn't notice these errors themselves. However, user ZeroVolt IS seeing errors. We don't fight through a level "in a certain way" to avoid the level's graphics becoming corrupt.  If this is happening, it is an issue and the only way the team can know about it is if users report it to them. We aren't going to sabotage the project by telling them. It is the only way they can know there is an issue.

The difference in timing between NTSC (60 Hz) and PAL (50 Hz) is the reason these VRAM errors show up differently on NTSC and PAL, but WE AGREE that the errors are present.

(2 edits)

UPDATE: THIS IS NOT AN ATTACK ON THE DEVELOPERS OR AN ATTEMPT TO UNDERMINE THEIR HARD WORK. THEY DO NOT HAVE A CORPORATE TEAM OF BUG TESTERS ON THEIR PAYROLL, SO THEY NEED TESTERS TO LET THEM KNOW IF THERE ARE PROBLEMS. THUS FAR, THEY HAVE RESOLVED ALL ISSUES PREVIOUS TO THIS BUILD. WHAT THEY ARE DOING IS EXTREMELY DIFFICULT.

Game version (shown on the title screen): V 5.0 CRC 59E15DFC

- Platform: emulator + version / flashcart / real hardware + model: Sega Genesis Model 1, 2, 3 (NTSC), Model 2 (PAL), Evergenesis v3, Medpro, Everdrive MD.

- What happened: Graphical and logical errors randomly appear throughout game. Errors are different between PAL & NTSC, but happen on both PAL & NTSC. The difference in timing between NTSC (60 Hz) and PAL (50 Hz) is the preason the VRAM errors show up differently between consoles but there is an agreement between two testers of real hardware that the VRAM errors are present.

A single frame lasts 20 milliseconds on PAL and only 16 milliseconds on NTSC, so the errors are worse on NTSC, but still appear on PAL as well.

- What you expected to happen: Not have errors.

- Steps to reproduce: Play through game, use level select to go to level 2 & level 5-2 specifically for the most noticeable problem

- How often (always / sometimes / once): Every time. May take longer to show up on PAL, but does show up.

- Screenshot or video (optional): First photo is on PAL console. Second photo is on NTSC console.

(2 edits)

Again, your post CONFIRMS graphical errors, but suggests  the errors are different on PAL than on NTSC. People who do not read all the way through will think you are disagreeing with me. You are getting different glitches because you are on PAL. The entire game is bugged and needs to be recompiled. THAT DOES NOT MEAN THE ENTIRE GAME HAS TO BE REPROGRAMMED FROM SCRATCH. THAT JUST MEANS SOMETHING WENT WRONG IN THE COMPILE OF THIS PARTICULAR UPDATE. The team will simply go back and figure out what went wrong. This is NOT an attack against the dev team. These are just honest observations to help the team.

(1 edit)

EDIT: UPDATE: THIS IS NOT AN ATTACK ON THE DEVELOPERS OR AN ATTEMPT TO UNDERMINE THEIR HARD WORK. THEY DO NOT HAVE A CORPORATE TEAM OF BUG TESTERS ON THEIR PAYROLL, SO THEY NEED TESTERS TO LET THEM KNOW IF THERE ARE PROBLEMS. THUS FAR, THEY HAVE RESOLVED ALL ISSUES PREVIOUS TO THIS BUILD. WHAT THEY ARE DOING IS EXTREMELY DIFFICULT.

Confirm. Weird errors like this on the same level. It looks like it just needs to be recompiled. Also, starting on level one, when enemies are knocked down, sometimes they slide away as though on a conveyor belt.

So, you are reporting you also get the graphical glitch, but a little later in the level? People that don't read all the way through are going to miss that. It happened on both the Everdrive MD & the MEDPRO on a Sega Genesis Model 1 VA3 & a Sega Genesis Model 3 VA0.
Also, you are reporting a glitch as well, just happens slightly differently. Are you on NTSC or PAL? What model?

If it  had been released in the US in 1989....

Latest version ran on real Sega Genesis Model 1 hardware.

Thank you for porting this fantastic game to the Mega Drive.

Bug:
Because none of the extra buttons trigger the standard PC Engine soft-reset when pressed with Start, the "Select" input line is unmapped in the Sega Genesis code, making it physically impossible to enter a cheat code should one exist. This may be a problem in future ports. Button C or A could be mapped to select.

(2 edits)

The non-existent level select for Bonk is misreported all over the internet. Sites like Gamefaqs and Mobygames have been attributing the level select of the Amiga, NES, etc. to the Hucard version of Bonk 1 for over 20 years. It does not actually have a level select. There are game genie-like codes (that work on the Everdrive), but those would not work for your port as it has different code (it is no longer an 8-bit game).

Additionally, the Sega Genesis does not have a "select" button, which the NES and PC Engine both have. So, if you unlocked the debug menu (which was never made accessible in either the US or Japanese release) , it would be impossible to use it.

Tested working on Everdrive MD on Sega Genesis Model 1 VA3 (non-TMSS).

(5 edits)

Suggestion: Debug level select for bug testing all levels in next build. Something like the Konami code at the title screen.  Note: Gamefaqs and Mobygames incorrectly attribute the NES / Amiga Bonk level select to Bonk 1. It does not actually have a level select trick and the debug menu buried in its code is NOT functional. Additionally, the Genesis does NOT have a select button.

Current version seems stable! Been playing for about 20 min. No problems.

Still enthusiastic. Once the bugs are smashed, it will be the ultimate TMNT port.

Unfortunately, it got worse! The Evil Krang corrupted the new level you created! He's terrible, that Krang.

(1 edit)

EDIT: UPDATE: THIS IS NOT AN ATTACK ON THE DEVELOPERS OR AN ATTEMPT TO UNDERMINE THEIR HARD WORK. THEY DO NOT HAVE A CORPORATE TEAM OF BUG TESTERS ON THEIR PAYROLL, SO THEY NEED TESTERS TO LET THEM KNOW IF THERE ARE PROBLEMS. THUS FAR, THEY HAVE RESOLVED ALL ISSUES PREVIOUS TO THIS BUILD. WHAT THEY ARE DOING IS EXTREMELY DIFFICULT.

It looks like the Full v 5.0 was not quite ready for prime time. New Level 5-2 had graphical corruption & there were disappearing weapons in level 5-1. Probably other problems as well.

(2 edits)

EDIT: UPDATE: THIS IS NOT AN ATTACK ON THE DEVELOPERS OR AN ATTEMPT TO UNDERMINE THEIR HARD WORK. THEY DO NOT HAVE A CORPORATE TEAM OF BUG TESTERS ON THEIR PAYROLL, SO THEY NEED TESTERS TO LET THEM KNOW IF THERE ARE PROBLEMS. THUS FAR, THEY HAVE RESOLVED ALL ISSUES PREVIOUS TO THIS BUILD. WHAT THEY ARE DOING IS EXTREMELY DIFFICULT. WE THANK THEM FOR THEIR HARD WORK IN BRINGING US THIS FANTASTIC PORT.

BUG REPORT TEMPLATE (copy and fill in): Bug: Level 5 graphical corruption.

- Game version (shown on the title screen): v5.0 (Build with CRC 59E15DFC)

- Platform: emulator + version / flashcart / real hardware + model: Sega Genesis Model 1,2,3 & MEDPRO, Everdrive MD 2.6. Note, this was tested on multiple consoles and on BOTH the MEDPRO & the Everdrive MD 2.6. 

- What happened: The new level (5-2) graphics are corrupted.

- What you expected to happen: The graphics to not be corrupted.

- Steps to reproduce: Play the game.

- How often (always / sometimes / once): Evertime. It is fine for about a second, then the level corrupts as you travel to the right side.

- Screenshot or video (optional):

It doesn't look that bad in the photo, but after that point ALL of the graphics start to fall apart.


(1 edit)

BUG REPORT TEMPLATE (copy and fill in):

- Game version (shown on the title screen): tmnt_megadrive_port_0.5.0_full (Build with CRC 59E15DFC)

- Platform: emulator + version / flashcart / real hardware + model: Sega Genesis Model 1, 2, 3 & MEDPRO, Everdrive MD v2.6

- What happened: Foot soldiers on level 5-1. Their spears became invisible between version 4.2 and verion 5.0.

- What you expected to happen: Expected spears to be visible.

- Steps to reproduce: Play game.

- How often (always / sometimes / once): Always.

- Screenshot or video (optional):
Here is the guy from v 5.0

Here is the same guy from v 4.2

Should be simple fix, right

So, it could happen in emulators as well? Glad we caught it then.

Good luck bug hunting.

(2 edits)

BUG REPORT: MEMORY FAULT ERROR (POSSIBLY FIXED)
Edit: Again, this is not to be negative; it is to help the user fix their game. To prevent a crash, this would have to be recompiled. Rester159 has some experience porting PC Engine to the Genesis and might be able to help (if they have the time or inclination), but I'm afraid that is all the assistance I can be. 

I'm NOT A PROGRAMMER, but according to Google:

"The crash screen shows a Line 1111 Emulator (Exception $0B) fault. This error triggers when the Motorola 68000 CPU attempts to execute an instruction opcode that starts with the hexadecimal digit $F (bits 15-12 are set to 1111). Because the 68000 has no native hardware instructions starting with $F, it automatically traps the execution and forces a crash. Looking at the exact register dump and stack values from the crash screen, we can trace precisely what broke in the code:

• The Culprit (The Program Counter): The crash happened at an Offset of F00008. On a standard Genesis, memory addresses starting with $F belong to Work RAM / System RAM ($FF0000 to $FFFFFF).

• The Trap (What went wrong): The CPU was forced to jump directly into Work RAM to execute code. Look at the stack dump data at +3C:—it contains the hex string 393C FFF0 0000 FF00.... The byte string FFF0 starts with $F. The CPU tried to read FFF0 as a machine-code instruction, panicked because it is an invalid $F opcode, and tripped the Line 1111 exception.

This might be one of three programming oversights:

• Function Pointer Corruption (Most Likely): The game was trying to read an object's behavior routine (like a specific enemy AI, an animated background tile, or a collision bubble in that lava zone). The pointer variable holding that routine's address got corrupted or filled with RAM mirror junk (F00008) instead of a valid ROM address ($00XXXX).

• Trashing the Stack via RAM Copy: The engine frequently uses fast block copies or DMA to clear RAM or update graphics. If a loop size calculation was off by even a few bytes, it overshot its boundary and completely overwrote the active function return stack. When the game executed an rts instruction to go back to the main loop, it pulled F00008 off the broken stack and jumped to its death.

• Executing Uninitialized RAM: If an array or an active entity slot was cleared out or freed up but the game logic didn't properly remove it from the active update loop, the CPU will attempt to execute empty RAM bytes, which often default to pattern values that trip Line 1111 or Line 1010 traps."

Again, that is Google.

iAsk stated:

"The error shown is a Motorola 68000 "Line 1111 Emulator" exception, which occurs when the Sega Genesis CPU accidentally tries to execute invalid data or an uninitialized memory address as code. This crash commonly stems from a null function pointer call, a stack overflow corrupting return addresses, or a mismanaged DMA transfer overwriting program memory. To resolve it, you must cross-reference the crash offset and stack trace with your compiler's symbol map to locate the offending function or data corruption in your source code."

This is the error code. This time it happened at level 2-1, not level 1-3.
This is the exact spot it happened when it happened the first time.

(1 edit)





Unfortunately, all of those ignore many types of bus faults, even Blastem. It appeared to be an illegal instruction exception error and there was a black screen with illegal instructions shown.

It was the stage where you go into the mouth of the green dinosaur and are swimming around inside of the pools of stomach fluid. One of the last pools, a head bonk caused a page fault.