Sunday, 22 February 2015

Some Phillips FTR 9955 action!

Is the grass greener on the LCD/Plasma/LED side?

Back in the late 90's, it was in the opinion of AV specialist that progressive scan was the way of the future. The most one can expect was 480i at the time, so manufacturers came in droves to implement field double technology (aka 100/120Hz scan) from the traditional 50/60hz refresh rate. The result was mixed: 480i material was less shaky when it come down to fine details and a softer image was a result., where as 240p (retro console resolution) looks digitised probably due to some image processing and that is the crutch of the whole change. The most common problem with the set is red crawls on white background. All the sets required the motion compensation circuit to be activated to reduce the problem. This adds to input lag on a weird image for gamers unfortunately and it broke all light gun games too.

Earlier Plasma or LCD panels had their input tuner/image processors as a separate console, while feeds the signal (usually as a DSUB/VGA output) into the panel. I am interested in the earlier revisions of these devices because the extent of the scaling is limited to 640x480 50/60Hz which in all respect is just line-doubling and may not deinterlace the image.And that is what I got from the Phillips FTR 9955.

To my knowledge, the 9955 does not incorporate the Pixel Plus technology. That was introduced in the 9965 model and I do not know whether that feature can be turned off. The reason is the technology will introduce lag and various image distortions as a result of inserting extra frames of interpolated images, plus it makes real life video scenes like it was shot in a porno studio.

The FTR 9955 (pictures courtesy of eBay seller ik8zsz)

Right off the bat a lot of retro gamers will be salivating over the sheer amount of inputs this line-doubler has: 1 x RF, 4 x RGB enabled SCART inputs, 1 x YUV input and a DSUB bypass for PC or a freaken VGA enabled Dreamcast!

Personally, I will not bother with scaling images on a flat panel as I would much prefer a computer CRT. Even after the image is perfectly scaled, different flat panels will interpret the signal "differently" where as the CRT will bang out the signal as it was intended. But my CRT is in storage and I have a crappy DELL one in the gameroom.

But, before all that can happen, the 9964 is set to only work with certain Phillips branded flat panels. So I proceeded to do some research and opened it up to access the service menu.

And it worked the first time too, so lucky me. The menu can be navigated by the consoles' front buttons. But I suggest you are in a good frame of mind before you begin as its not very intuitive.

I changed all the EXT to Game, which I presume means RGB which I confirmed with a c-sync video (via a sync strike) and also zeroed the sharpness and removed all the digital motion circuitry.

The sheer amount of overscanning/underscanning options is quite impressive as well. With other scalers, the user has to undergo tedious fiddling to fill the image to the screen. 

So how does a 240p image look like?

And the cherry on the top, the SLG3000

Needless to say, it was quite nice :)

Youtube videos

When the SLG is incorporated into the signal cable, the DELL seems to be coping with the image quite well when the image is still but blurs a bit when the image move. I am 100% sure this problem wont be exhibited on a computer CRT.

480i images are line-doubled with the interlacing effect intact, which totally negates the SLG function. Maybe the FTR is true to its source afterall. 

Friday, 20 February 2015

A quick video of the new toys and Harness modification on the I/O

My new toys:

So after running through various solutions to the Kick Harness issue on the I/O, I decided the fault is the I/O itself so it should be the subject to the modification.

I used my trusty dremel and went all dentistry on the header. Thankfully the quality of the plastic was friendly with friction and allowed some bits of it shaved off without melting or cracking. Now the Kick Harness is held (by friction) into the header.

Everything is fine now, the I/O is working like it should.

 And all my toys are working like they should. All ended well this Friday, hope the same for the upcoming week.

Thursday, 19 February 2015

Naomi madness in the house! Whoop whoop...

What started out as a form of curiosity has turned out to be hours of research and emails back and forth, but finally, I have what I need to get two Naomi 2 machines up and running.

The loot:

Firstly, I bought two of these JVS I/O from here Aliexpress I/O as it is a copy of the famous CAPCOM I/O people rave about. The reason why I chose this over a used SEGA or CAPCOM one is because it is brand new and the price is reasonable. The I/O draws current from the JAMMA harness so it will be used as often as the cabinet is powered on, the mileage of a second hand one can vary and could die at anytime.The package came with the wrong RCA cables (RCA to 3.5mm headphone jacks were dispatched instead...which is awesome because I needed a couple of those, score~) and a set of JVS power plugs which will make sense if there is a header next to the DSUB plug, which is not the case here. Any how, it is best practice to power the Naomi 2 motherboards with another power supply instead of relying on the I/O. The other function of the I/O include amplifying the sound from the motherboard to JAMMA level, JVS controls to digital JAMMA standards (a bit on this later), passing through video signal (RGBHV/RGBS) to the JAMMA edge and allowance for a CPS2/3 Kick Harness to accomodate the extra inputs on certain games (again, a bit on this later). What is unique about this I/O versus the SEGA one is that it can allow up to 4 Players on certain games like PowerStone, but the trade off is you can not use any analog controls on the board. The build quality is pretty good, no hot glue for jumps or bad soldering jobs. The only gripe I have with it is the misdirection on production and the lack of an information leaflet.

This is how it looks like once interfaced with the Naomi 2 (or 1) with a Naomi 1 cartridge inserted. The JVS powerline is present on the Blast City's -001 loom so the two connectors plug directly into the motherboard. Then its a matter of plugging the serial USB plug, RCA audio and DSUB into the I/O. At this point you might want to toggle the jumper on the filter board to output 15KHz as that will be what the cabinet is expecting at  the JAMMA edge. As the I/O is more or less a passive adapter, it will not matter whether you power on the Naomi 2 or the I/O first. In this case, both are on at the same time as power is coming from the same power supply. The Naomi 2 fan is comparable to the CPS2 one so its rather loud, but unlike the CPS2 one, you really need them running with the fan on as the chips get very hot over time.

With luck, you should see this going on the screen. I do not think the 15KHz image directly from the Naomi 2 is ideal for me as the image is interlaced, so outlines are blurred and the image is not 100% stable. The better way to display this is setting the Naomi 2 to 31KHz on the jumper and plugging the DSUB directly to the monitors DSUB input. That way the image is in progressive mode, plus you can reintroduce scanlines if you wish with any SLG (ScanLineGenerator) on the market or build one yourself.

And the same here for the second setup:

I was very comfortable with the setup, until this happened:

As you can see, both player 1 and player 2 has Shot 6 activated eventhough I had nothing plugged into the Kick harness port. How odd and frustrating. I cross checked my setup on a second cabinet as well to make sure the problem is not an isolated case and it was happening on whatever setup I had. So my next step is to obtain some help from the sellers of the various components. 

The best hint I got was from Chloe from Highway Entertainment where I got the Naomi 2 motherboards from. Her advice was that all was not what it seem on the JAMMA pinout of the I/O, which made sense as I know of the infamous Blue Elf reincarnations which capitalised on the unused JAMMA pinouts for the extra Shot buttons.

So compare this:

To this:

The last 2 pairs of pins on a JAMMA standard are both grounded together and that corresponds to Button 6 for both players. So logic has determined that the Chinese manufacturers of the I/O will follow the Chinese made Blue Elf boards. So its time to test the theory.Que Macgyvers theme...

And what do you get?

Have not diffused a bomb but equally satisfying.

I tested the Kick harness with a makeshift wire and paper clip combo as I was in the Macgyver mood and all the Kick buttons are working as they should. Could this be a redundancy precaution taken by the manufacturers of the I/O should the kick harness kick the bucket? Nobody will ever know.

And finally, the Kick harness and header issue. Honestly, I do not care what was the designers motivation here? Does he/she dislike propriety plugs? Dislike standard JAMMA conventions? Though people will not use the connector because the ChAMMA convention was the standard?? There is no way to get this plug to work unless I void the I/O's warranty or is there?

Stay tuned.