Fire Shark Sprite Problem

[caius]

Hi guys, I'd be highly interested in any results or successful repair attemts to this mentioned Fire Shark PCB. I, myself unfortunately failed in repairing, so someone (hopefully) might be better than me.

By the way, I DID resolder the customs pins...

Greets to everyone,

Mike McBike

Hi, I followed your attempt to repair this board, very interesting, I think you have done your best.So, to recap, can you tell exactly your inteventions done?
Last chance could be the replacement of custom FCU-2 which is the sprite controller, you said signals of Sprite ROMs are good until they reach custom BCU-2 but, according to MAME source, this is the Tile controller and, we know, PCB has a sprite and not tile issue.Have you checked all the RAMs addressed by these two customs?

[mikemcbike]

Hi guys, I'd be highly interested in any results or successful repair attemts to this mentioned Fire Shark PCB. I, myself unfortunately failed in repairing, so someone (hopefully) might be better than me.

By the way, I DID resolder the customs pins...

Greets to everyone,

Mike McBike

Hi, I followed your attempt to repair this board, very interesting, I think you have done your best.So, to recap, can you tell exactly your inteventions done?
Last chance could be the replacement of custom FCU-2 which is the sprite controller, you said signals of Sprite ROMs are good until they reach custom BCU-2 but, according to MAME source, this is the Tile controller and, we know, PCB has a sprite and not tile issue.Have you checked all the RAMs addressed by these two customs?

Well, I think so. Signals in data and addresses seem to be valid, but you'll never know, what is right and wrong. I also triggered on every second scanline just to detect if there is any difference in data bus behaviour. But unfotunately there was none... I already noticed the difference between tile and sprite custom controller, but I could not detect any problem in the outputs of the sprite controller and it seems, that sprite data gets combined with tile data in the tile controller. Can this be confirmed by anyone else?

MMB

[caius]

Well, I think so. Signals in data and addresses seem to be valid, but you'll never know, what is right and wrong. I also triggered on every second scanline just to detect if there is any difference in data bus behaviour. But unfotunately there was none... I already noticed the difference between tile and sprite custom controller, but I could not detect any problem in the outputs of the sprite controller and it seems, that sprite data gets combined with tile data in the tile controller. Can this be confirmed by anyone else?

MMB

Sadly there are no schematics of this pcb, the only documentation about is the MAME source, take a look here:


http://mamedev.org/source/src/mame/vide ... an1.c.html

[mikemcbike]

I already had a look on the source code, but, unfortunately, emulation does not always help you in hardware failures... In the beginning, some data busses seemed corrupt for me, because of strange data levels, but this revealed to be o.k. after i checked the chip select lines... it's just tri state behaviour.
As you can see on my website, I'm equipped with some fine measurement devices, like a 500 MHz digital oszilloscope and a fast logic analyzer - what a waste of technology on that beast...

MMB

[caius]

I already had a look on the source code, but, unfortunately, emulation does not always help you in hardware failures... In the beginning, some data busses seemed corrupt for me, because of strange data levels, but this revealed to be o.k. after i checked the chip select lines... it's just tri state behaviour.
As you can see on my website, I'm equipped with some fine measurement devices, like a 500 MHz digital oszilloscope and a fast logic analyzer - what a waste of technology on that beast...

MMB

Yes, I've seen your amazing equipment.I think you've eperienced all the best it can be done.As I said, the only possilbity woud be the custom (one or both) replacement.I have a scrap Outzone PCB with same ASICs, If you want I could give you them but I don't know if they are good.

[mikemcbike]

This might be a decision of gbaplayer, because he has to pay for it... By the way, the tile controller is a fine pitch package, so it's a pain to desolder / resolder... But thanks for your offer anyway.

MMB

[caius]

This might be a decision of gbaplayer, because he has to pay for it... By the way, the tile controller is a fine pitch package, so it's a pain to desolder / resolder... But thanks for your offer anyway.

MMB

OK.I use a simple hot air station to desolder and resolder these QFP ICs, if you got patience and steady hand the good result is assured.
Obviously if you got a infrared station , job will be better and more simply.

[mikemcbike]

One may do this with good results with an electronically regulated hot air device like this one mentioned on my website:

http://www.wolfgangrobel.de/arcade/loet.htm

It's cheap (approx. 80 Euros) and may also be used for removing old paint...

MMB

[caius]

One may do this with good results with an electronically regulated hot air device like this one mentioned on my website:

http://www.wolfgangrobel.de/arcade/loet.htm

It's cheap (approx. 80 Euros) and may also be used for removing old paint...

MMB

Good device.I use this 2in1 instead:

http://www.ebay.it/itm/STAZIONE-AD-ARIA ... 1643wt_906

[mikemcbike]

wow, that's a cool price... I'm not so much into italian, is there a air pump included?

Thanks for the hint...

MMB

[caius]

wow, that's a cool price... I'm not so much into italian, is there a air pump included?

Thanks for the hint...

MMB

Ye, it 2in1 station.It has a soldering iron ( from 200° up to 480°) and air pump (seven velocities and up to 500°).A real bargain IMHO.