The Guru is proudly supported by....

25th December 2020 A new undumped Sega C2 game called Head On Channel found its way into my inbox today, delivered personally by Santa. Get it from the usual place.... My Google Drive Shared ROMS Link I could be wrong and this is just a wild guess but since there is no mention of this game on the net I assume this is like previous C2 dumps and is another unreleased prototype Sega C2 game being hoarded. Guys, stop the shenannigans and just release the stuff so people can enjoy it. Here's something to think about in the future... release it, take the credit and be the hero instead of a zero. As with previous unreleased C2 dumps, to make it easy to run in MAME I've already renamed the ROMs inside the zip and named the zip so you can see it is Head On Channel running as tantr. Just download the zip and rename the file to tantr.zip and then run 'mame tantr'. This is of course just a revamp of Sega's Head On. Nothing really special, so play it for a few minutes until you get bored and then forget about it like all the other crap prototype arcade games that failed arcade testing LOL! If you are interested to play with some other unreleased Sega C2 prototypes, check the news below on 26th November 2018 and 2nd December 2018. Enjoy, and Merry Christmas :-) 3rd November 2020 If you are a fan of Hyper Neogeo 64 hardware you will know that there are a few I/O boards and they only work with specific games. Specifically the 1st I/O board which is used for the driving games (also known on the net as the rev 1 fighting I/O board) will only run the two Samurai Showdown games but not the other later Hyper Neogeo 64 fighting games like Fatal Fury Wild Ambition and Buriki One. Running the later fighting games results in a 'MACHINE CODE ERROR' on boot up, so there's something going on..... I recently spent some time repairing all of my Hyper Neogeo 64 main boards and carts. I once owned all 7 of them, 2 were sold years ago. Repairing is maybe not the right word though. Basically I just re-assembled them and put all the ROMs back onto the boards which have been sitting in bags for many years. Then once I had some working games I updated the Hyper Neogeo 64 hardware documentation that is in the mame source. While doing this I checked the I/O board situation more closely and I discovered something that I missed before..... with this new info it is now possible to convert a rev 1 I/O board to run ALL of the Hyper Neogeo 64 fighting games. The updated Guru-Readme(tm)(c)(r) will be added to the MAME source code soon. In the meantime, here's a video showing Buriki One (game 007; the last game made which normally requires the rev 2 JAMMA I/O board) and it is running on the first version Hyper Neogeo 64 LVS-IOJ I/O board. This I/O board normally runs the driving games and only the Hyper Neogeo 64 Samurai Showdown games; Samurai Shodown 64 / Samurai Spirits 64 and Samurai Shodown: Warrior's Rage / Samurai Spirits 2: Asura Zanmaden. In reality this board works just fine with all of the fighting games when the I/O board is modified :-D 12th September 2020 I just received a very rare Namco System 256 game called 'Moto GP'. According to system16.com it was developed in China and had a very limited release also only in China and was then cancelled by management due to poor income performance so no other country got to see it. We all know from past history that higher management in large companies are not known to be the smartest people on the planet and that was clearly the wrong decision as this type of racing game is very popular in westernised countries and especially in those countries that host the official MotoGP races. system16.com says it runs on Namco N2 hardware, but this is obviously a guess and now proven to be wrong. This was purchased and sold in China. Big thanks to 'ekorz' for helping out with funds and shipping. Even though the game was cancelled, someone at Raw Thrills thought it was a good idea because there is a newer release Moto GP arcade game made by Raw Thrills running on PC-based hardware but this is definitely not the same game. This System 256 game is obviously based on the PS2 version as that came out in 2000 and this game is dated 2007 on the title screen. For those who don't know, System 256 hardware is basically just a cost-reduced single board PS2. It appears this game is only a twin cabinet release and it displays a permanent message 'Link Error' since this is only half of the complete system. The game runs off a HDD. The original drive is a Western Digital WD800BB 80GB hard drive. The actual item purchased was the hard drive and the security dongle. I dumped the drive and there's only about 2.5GB actually used with the remainder of the drive zero-filled. I don't have any arcade driving controls for this game and it is connected using the standard System 246 JAMMA I/O board. Controls don't work, not even coin or test, so I can't start the game. I can enter test mode using the DIP switches, but again the controls don't work so I'm not sure if the 'Link Error' message can be turned off or if the game can be set to single cab mode. The I/O board for this game is unknown. A manual would help to fill in some gaps but there is virtually no chance of obtaining one since the game is rare and there probably aren't that many around now, 13 years after the release date. So basically what you see is all you get ;-) Here's a pic of the game using my standard System 256 box and a video showing the full attract mode for about 20 minutes. For this test I'm just sitting the HDD on top of the DVDROM drive. The real thing would have mounted the HDD in the provided frame where the DVDROM drive is now. In the video, the hiss in the background is caused by the crappy fans used on all Namco System 246/256 boxes so that is 'normal' ;-) 31st August 2020 A few months ago a friend gave me a non-working bootleg Galaga (Gallag) PCB. This is the common blue PCB type with DG-01 written on the CPU board and DG-02 written on the video board. On power-up it just showed static garbage. Additionally it looked like one of the colors was missing. I first looked at the color signals on the video board. Probing with my logic probe showed activity on red and green but nothing on blue. I turned the board over and discovered why... Damn cowboys again! Instead of a connector there are wires soldered into the holes. I say the word 'soldered' but they were basically just sitting there on a blob of solder LOL! I resoldered the wires and the colors looked good. Later I'll remove the wires and fit a connector but for now I'll move on. Probing the main Z80 at location 6B on the clock pin showed it was present measuring 3.072MHz. The reset pin was moving fast low/high continually. I powered on while holding my logic probe to the reset pin and there was no delay after the pin went high, it immediately toggled to low then high very quickly, and repeated. This means there's no way for the code to run as the CPU is constantly being reset. I traced the reset to a logic chip LS259 at location 3N. Comparing to the datasheet and truth tables the chip appeared to be working correctly with active inputs and outputs. I noticed just above this chip at location 3M there was a 74LS37 and it was branded Fujitsu!! This is a fairly simple chip with a simple truth table. It is a Quad 2-INPUT NAND Buffer. Here's the datasheet... Basically there are 2 inputs (A,B) and 1 output (Y) times 4. If any of the inputs are low the output is high. If both inputs are high the output is low and that's all there is to it. The inputs were either low or low with a clock present (yellow LED on logic probe is flashing). Of course the outputs where just static high so the chip was bad... no surprise there hahah! I removed it and replaced it with a good chip and powered on.... Now the game is running but there's a permanent coin-up sound and it jumps into the game as if start was pressed. This is caused by the 4th Z80 not running correctly. The 4th Z80 'emulates' the MB8841 4-bit microcontroller on the original PCB and controls coinage among other things. This Z80 has 1Kb of SRAM connected to it. I removed and checked the 2x 2114 SRAMs (1k x4-bit.... x2 = 1k x8-bit). There's no easy way to check RAM other than put it into something that works that uses the same RAM and power on that device and if it works the RAM is ok. The good old Commodore 64 just happens to use a 2114 SRAM for the color RAM, so I used my C64 with the Dead Test Cart to test it. The screen on the left is what it should look like and the screen on the right is what I got... After swapping the RAMs for good chips the game now works 100%. Now I wanted to do a couple of other mods to make it better. One is to fix the colors so they are correct. The white text comes out slightly pink. This is fixed by changing 4 resistors. Basically R3, R4, R5 and R7 on the video board near the video connector need to be changed. The nice thing is 2 of the resistors are correct but in the wrong place so to do this mod only requires a 220 Ohm and 470 Ohm resistor and the other 2 can be re-used. R3 should be changed to 100 Ohm (move it from R7 to R3) R4 should be changed to 220 Ohm (move it from R3 to R4) R5 should be changed to 470 Ohm R7 should be changed to 220 Ohm This mod is documented here. I also removed the RGBS wires and fitted a connector onto the video board. The final mod is to change 2 EPROMs with EPROMs from the original Namco Galaga so that the copyright year and Namco name/logo shows as well as it saying GALAGA instead of GALLAG. This is as simple as just burning a couple of EPROMs with the correct code. On the Gallag CPU board at position 4C, replace that ROM with ROM GG1-2.3M (CRC32 43BB0D5C) which fixes the GALAGA text and copyright text. On the Gallag video board at position 5R, replace that ROM with ROM GG1-9.4L (CRC32 58B2F47C) which shows the red NAMCO logo. Here's some pics showing the changed ROMs and the board fully working.... 16th June 2020 A few years ago I bought a Dlink DAP-1155 802.11n wireless bridge which has just been lying around unused for years..... This little box allows older PCs with a network card to receive the internet through an ethernet cable from a WIFI signal coming into the box. It can also act as a wireless access point when connected to a router either by ethernet or WIFI. These are really handy for all kinds of old computers and consoles like getting a faster WIFI connection on an old laptop that only has 802.11b/g (my original use for it), getting an internet or network link on Windows 95/98 PCs, uploading game images to a NAOMI configured in network mode (read my tutorial here), FTP'ing games onto the internal HDD on a cracked XBox, connecting the first model XBox360 to the internet and even getting the internet onto Amiga computers like the A1200/A2000/A3000/A4000 that have an add-in network card. Basically it's just a great way to get wireless internet on a device that could never have wireless internet. To get it to work all you have to do is set a static IP address on the network card in the same range as whatever is set in the bridge and it connects to the network. You also have to set up the wireless network SSID and security mode/password then you will have the internet on the device that the bridge is plugged into. It's a pretty painless experience really and it usually just works. A few years ago I updated my router when I got the NBN (broadband internet) and it uses 10.1.1.1 for the Gateway but DLink typically uses 192.168.0.50 for their wireless devices so it didn't work because they are on incompatible networks. Either device can have the IP reconfigured but it is better to leave the router as-is and change the bridge IP. I set the network card on my old Windows XP machine to 192.168.0.55 initially so I could get into the bridge firmware. I went into the firmware on the bridge (192.168.0.50 or http://dlinkap) and configured a static IP to 10.1.1.50 and gateway to 10.1.1.1 and set my network card to 10.1.1.55. All the settings were correct but the bridge couldn't see the internet even though I could ping 10.1.1.1! It could see my home network and shared drives and I could even copy files from any PC on the network but the bridge didn't see the internet. I figured maybe I had an incorrect setting in the firmware so I did a factory reset and then re-entered all the settings again but that didn't help. I figured maybe the latest firmware update would help so I updated the firmware to the latest version found on the DLink web site using the built in maintenance firmware update mode through the GUI via a web browser. Unfortunately this is where things went south. The firmware update completed without any issues but now I couldn't even get into the bridge firmware! I did a factory reset by holding a button on the back of the bridge and reset my network card to 192.168.0.55 but I still couldn't get into the bridge firmware. There was one last option.... DLink provides a recovery mode which can be accessed by holding the factory reset button while powering on. It then shows the following screen when going to the default DLink IP.... This allows uploading a firmware image to restore the device to working condition. I uploaded the latest firmware, waiting for it to complete and reset but it still didn't work. I also found some older firmwares on the net and tried several of them including V1.00 which was the factory version when shipped. At this point I was screwed. The recovery mode didn't work and there was no way to restore it to working condition. Maybe it could be repaired. Inside there isn't much, just two Realtek network chips, a RAM and the firmware EEPROM on the other side.... Unfortunately without a real firmware backup read from the physical EEPROM a fix wasn't possible. At least a firmware fix wasn't possible assuming the firmware was actually the problem. I really wanted this thing to work so I could use it on my home network on an old Windows 98 PC I recently put together so I went onto ebay and found one sealed and brand new in box for $25.... That's a pretty sweet deal when you consider this thing is still useful today in 2020 with current internet speeds, at least here in Australia. My connection is 50Mbits/sec (anything faster is very expensive here) which means I can theoretically download at 6.25MB/sec. The fastest real speed I downloaded something was about 4MB/sec and that was using an ethernet cable wired directly to the router so I will never max this thing out no matter what. Only an internet connection above 150Mbits could do that and the reality is it wouldn't download at the full speed regardless. I always laugh when people toss out useful things like this just because they 'think' it is obsolete.... same with 'upgrading' to the latest mobile phone which does exactly the same thing as the old one in most cases. Manufacturers really love those people that blindly buy the latest thing just because it exists even though they probably only used 50% or less of their older device anyway. I know a few people locally that do that every year and when I ask them what the new one does that the old one didn't do they just give me a dumb look and try to change the subject LOL! The inside of the new one is exactly the same as the old one.... And here's the chip I'm concerned about.... This is a Macronix MX25L3206E 32Mbit serial EEPROM. Fortunately my EPROM programmer supports the chip so I read it and saved a backup of the ROM... I then removed the EEPROM from the non-working bridge and erased and re-programmed it with the image I just took from the new bridge and re-mounted the chip back onto the PCB. Looks just like new and you can't even tell I messed with it hehe! So does it work? I powered on, loaded a browser and went to 192.168.0.50 and this is what came up.... So yeah, looks like it's fixed hehe! BTW, this is a new XP-compatible browser based on Firefox that works with modern web sites. Unfortunately it has a stupid name 'MyPal' (for obvious reasons it makes you think of the other 'pal' rip-off company that terrorizes online buyers and sellers). I re-configured the bridge to work with my network (10.1.1.50 / 10.1.1.1 and network card 10.1.1.55 etc), rebooted the bridge and it still works. I went to my web site and that works too so this is fixed 100%! There's one caveat... because the EEPROM is a 1:1 copy, the MAC address of the old bridge now matches the MAC address of the new bridge. I could probably look through the firmware image and find it and change it with a hex editor, but I will never be using both bridges at the same time so it doesn't matter. Looking back, I didn't do anything wrong when updating the firmware so I suspect that newer firmware screwed up the bridge to the point where it couldn't recover no matter what.... maybe even on purpose by DLink to force an upgrade. The last firmware was 4 years after the device was made so anyone upgrading would be doing it on a device that is well out of warranty, meaning when it screws up you have no recourse except to buy a new one. This is also a classic example of the perils of upgrading firmwares and operating systems.... afterwards there's usually no noticeable difference or the device becomes slow or gets screwed up so unless you are actually having an issue and the update will fix that specific issue, don't mess with the firmware or do OS updates. One proven example of not doing updates is I'm still running the original release of Windows 7 on my laptop and this thing just works without any issues. When everyone updated to Windows 10 or was force-updated I was only laughing :-D Anyway, good try Dlink attempting to brick my bridge so I would have to buy one of the very latest and expensive AC1200 bridge devices (AC is useless in my case anyway) but it wasn't successful. I guess you didn't do enough research into planned obsolescence and the repercussions if caught LOL! Final result: Guru 1, DLink 0 11th June 2020 Here we have a fairly standard Cabal PCB that has been sitting on my PCB rack for a few years in non-working condition.... Or maybe not.... the astute Cabal fan will notice that the top board is on the bottom and the bottom board is on the top LOL! This is of course the wrong way around, the larger board should be on the bottom. So a quick swap around and.... WOW! It's working! The only problem appears to be no sound. I wonder why? Here's what it is supposed to look like.... and here's what I actually have..... Well there's your problem right there hahah! OMG!! Some cowboy has made a real mess of this trying to hack in some other type of amp. To fix the sound I will have to undo all of that crap and put the board back the way it should be. On the bottom there were 2 tracks cut so I patched those first.... Notice those 2 cut tracks actually come from the HB-41 and go directly to the JAMMA speaker connection (which is the final amplified audio signal) so the full power amp audio goes into the HB-41. This is quite unusual because normally the power amp is the last place where the audio ends and then it goes directly to the JAMMA connector. I also replaced the amp chip with the correct type (LA4460) and replaced the 2 small volume pots. The pot nearest to the JAMMA connector is 500ohm and the other one is 2K. I didn't have a 500 ohm pot so I used a 1K pot which should work just fine. I powered on but no sound. I turned up the volume pots just to be sure but still no sound and touching a (wet) finger on the amp pins didn't make any noise either. I checked the sound section for other amplification-related parts and saw a strange chip labelled MB3614. Comparing to my other working Cabal it turns out that this chip is a LM324 op amp. Since there was no way to test it in-circuit I removed it and put it into one of my other working PCBs that I keep for testing parts (many chips are socketed) and the chip failed to make any sound so it was bad. I removed it and replaced it with a common LM324. Now when I powered on I got some sound effects (progress!) but no music. I also noticed some crackling and hissing noises and the volume was very, very low which means the power amp IC is not doing anything. Continuing on with the music issue, I probed the YM2151 and it was active and changing occasionally on the digital output (pin 21) so appeared to be working. The YM2151 is usually connected to a YM3012 DAC but in this case I didn't see one. Comparing to my other working Cabal board that has a YM3012, the chip at the same location is labeled 7105, so a 7105 is a YM3012. I checked the output pin on the DAC using my logic probe but there was only a static high signal. The problem with DACs is the analog output can't be checked with a logic probe, it requires an oscilloscope or just swap out the chip for another. I opted for the latter and replaced it with a known good chip. I powered on but still no music (I later tested the original 7105/YM3012 in a different PCB and it was good). I looked more closely at the YM2151 and noticed one leg was soldered. Since I couldn't check the chip in-circuit anyway, I just decided to pull and replace it. After desoldering and removing the chip this is what I saw.... Hahah!! One leg was broken off and that same damn cowboy tried to resolder the leg to the tiny stub that was remaining (which of course doesn't work!). One way to fix a chip like that if you don't have a spare is to dremel away the plastic to reveal part of the copper leg and then put the chip into a machine-pin socket and solder the broken pin using a small piece of wire to the exposed copper leg stub. This will permanently fix the leg and allow the chip to work again. Here's a pic showing the completed chip repair.... It worked fine in my test board but I decided to swap it with a known good chip just to be sure. I powered on but still no music. I decided to just change all of the caps since they all looked pretty beat up. About half of them were out of specification but after changing them and powering on there still wasn't any music! At this stage I had changed almost all of the sound related parts. The next piece of the 'audio puzzle' is the custom HB-41. This does some pre-amplification of the music and mixes the sound effects and music into one audio signal using a JRC4560 dual op amp and JRC2060 quad op amp and there's some caps and resistors for filtering. In typical 'engineer' fashion, this approach is way over the top for this application... the music is mono and the sound effects have already passed through the LM324 quad op amp so another op amp like a LM324 could have been used to mix the music and sound effects together. Since I have a working Cabal I just removed the HB-41 custom from the board with no music and soldered it onto the working board to test it. I was surprised to hear both sound effects and music but also that same hissing noise I heard before.... obviously the HB-41 is partially bad but still works enough to get music. So then I soldered the good HB-41 onto the Cabal PCB that has no music. When I powered on I could now hear the music (yay!) and no hissing (yay!) but the volume was still very, very low so there's still an amplification issue somewhere. I was almost out of ideas so I looked up the pinout of the LA4460 amp IC and saw that pin 7 was the output. I was thinking I might remove the amp and replace it with a small AMP PCB kit that I have which uses a LM380. Just for kicks I used my logic probe and checked if there was an active signal on pin 7 of the amp IC. To my amazement there was and I could hear the audio on the logic probe piezo speaker! But the sound wasn't getting to the speaker for some reason. Using my multimeter I traced pin 7 to a small via above the amp at pin 2. Because of the caps I couldn't see where it went further but figured it connected to the HB-41. After some time probing around looking for continuity, I couldn't find anything joined to that via. I removed a couple of caps (again) and then I saw what the problem was.... Son of a bit.... That damn cowboy again! The output track coming from the amp IC was cut under the cap! Holy COW! And yes it connects to the HB-41 custom (on pin 3) but only for some more filtering. Pins 1, 2 and 3 of the HB41 are connected to some caps inside so it just does extra filtering. Pin 3 of the HB-41 also goes directly to the JAMMA connector audio output as well so in reality the AMP output could just join directly to the JAMMA speaker output pin.... again way over-engineered for a simple low quality mono audio signal. Anyway, I patched up the track.... Then replaced the caps and amp IC then powered on and now sound effects and music are working 100%. Problem solved. Both Cabal PCBs are now sold so the new owner can replace the HB-41 to get rid of the crackling noises if he chooses and the other Cabal PCB will go from working 99.5% to working 100%. The moral of this tale is 'Beware of cowboys, they can be deadly!' 15th May 2020 I finally got some real arcade controls hooked up to my Dirt Dash PCB (except brake which is just a 5k pot wired up so I have no brake in the game... it's always off). Here's the full attract loop and game play run through all of the tracks. The general idea is this is a reference video to help improve MAME emulation. Youtube isn't the best venue for quality videos and the game runs at 15kHz interlaced which makes it not so easy to video either but I suppose it's better than nothing. I got the frame rate right though... notice how the video is solid, not having the usual dark horizontal band moving vertically down the screen... that's because when I set my camera to 25p for some reason it syncs perfectly. It might also have something to do with the power frequency here which is 50Hz. My camera is a Nikon Coolpix B700 which can do up to 20MP pics and 25p, 50p, 60p and 4K video so I may experiment with other rates later. One thing that is not so obvious in the video is the shadow under the cars is actually multiple horizontal lines separated by a small grey space. It is not just a flickering solid box like in MAME, although it does flicker but not all the time. It's not the easiest thing to photograph either ;-) In the 2nd video on the right I go through all of the screens in the test mode. One thing that doesn't show up is a calibration screen. This can be accessed by holding 'service credit' and pressing the test button.... If you are trying to calibrate PC controls for use with MAME and your controls don't work by default or are erratic, you probably need to go into the TAB menu then 'Analog Controls' and check to see if your controls are actually working. If not you probably need to enable them by changing the controls in mame.ini in the 'Core Input Options' section and add 'joystick' for 'paddle_device', 'pedal_device', 'dial_device' & 'positional_device'. By default MAME has the wheel set to 'mouse' and the analog gas and pedal controls are n/a. When 'joystick' is added in mame.ini the analog controls will change to the joystick axes but they may not be the correct axes and it's likely they won't be properly calibrated for the game. You must go into the TAB menu then 'Input (This Machine)' and re-set the control axes for steering wheel and gas and brake pedals to use the correct wheel/gas/pedal axes by selecting the item with the cursor then pressing enter and then moving the control you want to re-set. Once that is done, go into the game and while pressing SERVICE CREDIT (i.e. 9 on keyboard) press TEST (F2 on keyboard). If you did it right you will see the calibration screen, if not, while holding those keys also press reset (F3 on keyboard). With all controls at default positions and untouched and with the steering wheel centered press SERVICE CREDIT and an offset bias will be applied to the controls so they are fully calibrated. They are correct when the number on the far right (VALUE) is close to zero. Note on my real PCB the value sometimes moves slightly up and down about 0001 but that is normal. Press F2 to set test mode off and now the controls should work. Note my board has a small gfx glitch at the top of the high score screen that wasn't there a couple of days ago. Sadly it seems common that these boards develop random graphic faults due to 25 year old mask ROMs going bad and the only way to fix it is to replace them with an equivalent flash ROM. Fortunately I dumped all the Namco Super System 22 games and the ROMs are available so that's not a big problem :-) 10th May 2020 A few weeks ago I noticed some ROMs in Namco's Dirt Dash had been flagged as bad. After re-checking my PCB it turned out my PCB had the same fault so there wasn't much I could do since one of the ROMs was internally bad. A couple of days ago I got hold of a different version of the same game and dumped the ROMs. After comparing, only one gfx ROM was different. As well as getting a good dump of the Dirt Dash graphics ROMs, as a bonus we get the Japan version of Dirt Dash :-D I was also playing around with TX1 in MAME. It has the potential to be a great game but the controls are horribly broken. While the game is normally very hard, in MAME the game is impossible to control and the car is always spinning off the track or has a hard wheel lock to the left or right. This is because of a tiny bug in the way the controls were configured. While both the gas and brake values go from 0x00 to 0x0F, the steering should go from 0x00 to 0xFF. Even in the manual it says the wheel movement goes from 00-FF. In MAME it was set to 0x0F so it could only move in large steps. Simply changing the wheel to have the full 00-FF movement gives the car more turning steps (256 instead of 16) and thus smoother animation which makes the game a lot easier. Although even with this fix, as I said it's still a very hard game but maybe now with a real 360 degrees steering wheel it might be possible to get further than the 1st road split haha! I'm watching out for a youtube video where someone gives it a try :-) Anyway, both of these changes should hit MAME 0.221 at the end of this month, or you can download the src from Github and compile your own. With some 3rd-party artwork TX1 looks fantastic and now with the controls fix it might even be playable too! ;-) In other news recently, on the 3rd May 2020, after a long hiatus the lamer page got a small update LOL! 27th April 2020 I've been working on something completely different for a few days. Here's a quick work-in-progress demo of it in action. Look closely at the bottom right corner of the emulation screen and you can see an arm.... :-) This game has never been really fully playable in MAME, even now. The controls are completely broken and implimented as a dial without any limits so it wraps, which is competely wrong. Basically the arm/motor emulation that was 'fixed' at MAME 0.104u3 is a big hack. As far as I know it was never possible to win any match... until now ;-) To improve the emulation some info is needed from the real arcade cabinet. If you have a working Arm Champs II arcade cab or know someone who does, go into the test mode and then to I/O Port Check and look at Arm Volume. I need to know the range of the Arm Volume when at the left and right limit switches and the middle limit switch. Also, I need a description of the arm movement at the motor test check when the game is powered on. Thanks! 6th April 2020 Here's a repair log for some stuff I've been working on recently. First one is a bootleg Terra Cresta PCB that was not working. As usual it sat on the pile of non-working crap for a few years. This is the nice YM3526 version with better sound and is virtually a 1:1 directly copy of the original Nichibutsu version. Initially it was dead. Looking at the top board I noticed the 68000 pin 32 was hacked. I pulled out the 68000 and verified that pin 32 was broken. Normally I would replace the entire socket but I did not have a proper DIP64 socket in stock (obviously the bootleggers didn't have one either hahaha!!) and despite how hacky it looked, the existing socket was mostly ok so I just pulled out the old hacked socket pin and replaced it with a new pin taken from another socket, then replaced the 68000 with one that has all 64 pins ^_^ That did not get the board working so I pulled the program ROMs to check them and found that all of the ROMs and sockets were rusted. I cleaned the ROMs with some fine sandpaper but the sockets were too far gone so I replaced all of the ROM sockets on the top board. The game now booted up. The background graphics and sounds were good but all of the sprites were partially missing, showing only random pieces of sprites on screen. The sprites are generated on the bottom board. The sprite ROMs tested ok. There are 8x 2148 RAMs for the sprites so of course I started looking at the RAM but I didn't find any issues, although they were running very hot. Piggybacking another RAM on top didn't help so I pulled and tested all the RAM but they were not faulty, so I added sockets and replaced them back on the board. The rest of the board looked ok and there were no easy 'Dead Fujitsu' chips on the board so I decided to take a break and leave it for a day. The next day when I powered it on the game worked perfectly!! But after a few minutes the sprites started to fade slowly and after 5 minutes there were only random pieces of sprite showing on screen. After spending several days slowly checking every logic chip I found no problems.... very frustrating! The board seemed to have a heat-related issue so I used some freeze-spray on various chips on the board while the game was running but it didn't make any difference. I decided to take extreme measures and put the whole boardset into the freezer for 5 minutes. Then I plugged it into the power and to my amazement it again worked perfectly! Note don't run the board like this for too long because the coldness will turn into water droplets on the board as condensation takes place... so just a quick test is all that is necessary. I figured the fault was on the lower board since that is where the sprites are generated so this time I put only the lower board in the freezer. After 5 minutes I pulled it out, plugged in the lower board and powered on and the sprites were still bad! That means the fault must be on the top board. I tested the theory by putting the top board in the freezer for 5 minutes then repeating the quick power up and as predicted the sprites came back to normal. I let the board dry then ran it again for longer and after 5 minutes the sprites were flakey again. Now was the time to check every logic chip on the top board with the freeze spray, but amazingly it made no difference. The strange thing was none of the logic chips were getting hot so the faulty chip is going to be tricky to find because it works when cold and when it goes bad it only partially fails. This is the worst kind of fault and is best checked using an oscilloscope. But I was impatient and didn't want to waste time scoping out potentially dozens of logic chips. I put the board in the freezer again to 'fix' the sprites. Then plugged it in and powered up the game board, and while cold, used my heatgun to warm up the board. When I hit the top right corner of the board the sprites started to glitch out but that was only after I had heated the board a lot. I used my multimeter to trace the output circuit from the lower board through the 50 pin flat cable and found that it came onto the top board through 4x 74LS174 Motorola branded logic chips which just happened to be located in the top right corner of the top board :-D On the scope the outputs didn't look bad, but to be sure I pulled all 4 chips and tested them out of circuit and they passed the test! Urgghh! Now you would think... 'ok, just put them back and keep going'. No no. With this kind of strange fault, the chips work when they are cold and there's actually no way to test that situation except maybe using a heatgun to warm up the chips and re-test when hot. But there's an easier way. I learned a long time ago when faced with this problem to replace the chips with different chips ^_^ So I found some 74LS174 chips on a scrap board and replaced those chips. I powered on and got this..... Obviously at least one of the 74LS174 chips was suspect but it's time to move on. I played the game for about 20 minutes and everything was still ok so I'm calling it fixed :-D Second one is a quickie :-) This is a Sega Repulse bootleg that was dead. On the top board, the Z80 reset signal was missing. I replaced a broken electrolytic capacitor in the reset circuit and now the reset was happening but the reset was not getting to the Z80. The first chip in the line is our unfriendly enemy, the nasty 74LS04 Fujitsu logic chip! I checked the inputs and outputs with my logic probe and of course the outputs were all dead. I replaced that chip and then the game booted up. It had other faults in-game and the lower board was littered with nasty Fujitsu logic chips. The fix is probably as simple as removing the Fujitsu logic chips and replacing them with good chips. At this point someone wanted the board so I gave it away and this is now his problem ;-) The third and final board in this log was relatively easy just tedious and time consuming. This is a bootleg Twin Cobra that works but has big sprite problems. Looks like it was stored in a damp place for a long time and some traces are corroded and broken under the RAM. Removing the RAMs reveals the carnage... There were 7 broken traces. I patched the traces neatly and directly on the PCB using micro wire and then put the RAMs in sockets. The board now works fine and from just looking at it you can't tell it was repaired :-D 12th March 2020 I've had this for a while but since a few people sent it to me, here it is.... My Google Drive Shared ROMS Link This link is also added to my main menu under 'Guru Misc'. As well as the previously mentioned Sonic Bros, you can also get OOParts. It is a prototype Arkanoid rip-off made by Success that also runs on Sega C2 hardware. This is not supported in MAME and may never be due to arcade scene hoarders not wanting their secret prototypes in MAME and having a pissing contest with the rest of the scene and desperately trying to hold onto the crown of 'King Pisser' (and failing), but you can run it as 'tantr' and it works fine. To make it easy I've already renamed the ROMs inside the zip and named the zip so you can see it is ooparts running as tantr. Just download the zip and rename the file to tantr.zip and then run 'mame tantr'.... problem solved.... 23rd February 2020 I picked up a fairly rare Sega System 32 ROM board called 'Soreike Kokology'. In MAME 'Soreike Kokology Vol.2' is dumped but not the original. I have dumped it and it will be added to MAME soon. 10th February 2020 Something priceless just arrived for dumping.... no words needed ^_^ Expect this to hit MAME very soon!