alex_79

How many digit is the score?

Actually, I think it's very hard, I had to be very concetrated to reach that score. But it's addicting!

Great improvement! My best score so far: (PAL60, Starting at Level 6, left difficulty A)

Yeah, levels 1 and 2 are very easy with the precise trackball control, while 5 and 6 are insanely hard. I can last a while only if I started on a easier level, so to have a reserve of points when level 5 starts. In this case I can let the aliens destroy some of the cities so I can concentrate defending a smaller area. But if I start directly on harder levels, (where there are only 800 points and every destroyed city costs 200) I die before being able to "warm-up". Maybe you could make the aliens less "smart" so that they sometimes bomb also already destroyed cities? And/or start with more points in harder levels. A more gradual speed increment would be ideal.

Finally tried it on real hardware. I wasn't familiar with this game, but I think it's very interesting and fits perfectly with the trackball. Probably the first levels are a bit too easy, but since you have the option to start with the later ones, that's not a problem. So far I played the PAL60 versions (atari mouse and trackball variations) and I have no issue or bug to report. Are the PAL50 ones slower? I will surely play more of it. Thanks again for the great work with these hacks!

Actually I'm bothered about that too. When modding a console I try to keep everything reversible. I don't cut and solder directly to the pins of the chips (unless they're common parts still produced today), and instead trace the board and desolder and lift (so I don't lose them) discrete components to obtain the same results. If it's not possible, I cut traces on the pcb in areas where can they be easily repaired. Instead of drilling the case I prefer routing cables through existing holes (for example the one for the rf cord) and use in-line sockets. And I don't buy mint condition, boxed consoles to mod them. I purposely seek cheap ones in untested or non-working condition, which usually are also in poor cosmetic shape. Restoring them is part of the fun for me. and if they cannot be fixed they'll be source for parts.

NTSC is 525 lines interlaced at 60Hz. (PAL/SECAM is 625 lines at 50Hz). You have 60 fields per second each with 262.5 scanlines. That half scanline is what allows interlacing because cause every other field to be offset vertically so that it's drawn in between the scanlines of the previous one. Each field uses about 240 lines for the active picture, so the total resolution in about 480 lines (483, if I remmeber coreclty). It takes two 60Hz vertical refreshes to scan all those lines, so the framerate is 30Hz. To display it on a progressive scan TV such as a LCD it must be deinterlaced, and to do so both fields need to be received first. On a CRT the electron beam just moves in sync with the signal. Old consoles and computers used a non-standard signal by slightly altering vertical synchronization timing, so that the fields don't have that half scanline. (for example they can all be 262 or 263 lines). Without that half scanline, what in standard NTSC are the fields are now drawn always at the same position and become actually progressive frames. Framerate is 60Hz and vertical resolution is about 240 lines. This is a non standard signal, but it happens to work on CRTs. On the other hand, many digital TVs or generic upscaler don't recognize it. (you need a dedicated upscaler like the framemeister in that case). Being progressive, each frame can be upscaled individually (at 60 fps). The terms 480i and 576i actually refers to the digital versions of the analog standard definition 525 and 625 lines; anyway they are often used also to designate analog signals. By analogy the terms 240p/288p are also commonly used, altough they don't really exist as a standard in digital nor analog video.

I see your point. In my post I was referring to older pre-crash consoles, particularly the 2600, with non-interlaced output (often referred to as 240p/288p) . I can totally see that with less ancient hardware emulation can be worse, and the CRT filters might not make such a big difference to the visuals as in early games with low resolution and sparse graphics. On those, as I said, a simple "scanline" effect isn't good enough for my tastes. And while in the future some of the phosphors glow and persistance might be emulated (albeit with some approximation), I doubt you can implement those on an external scaler without introducing lag. Interlaced video on CRTs does flicker a bit, which can cause eye-strain and is more evident on computer generated images than on photographic ones (like TV shows). Digital displays are progressive so they eliminate the flicker. But the video needs to be deinterlaced first and this can generate artifacts in object that moves quickly horizontally. CRTs displays interlaced material natively with no artifacts as the 2 fields are not shown at the same time, but sequentially. Also the digital TV (or the upscaler) needs to wait both fields in order to process the image, which takes the equivalent of 2 frames of the progressive output used in old consoles. Yes, but the TV might do extra processing even when you feed a signal at its native resolution via hdmi, which might add lag. Some have a "game mode", some others require you to manually turn off every filter. It takes some efforts to setup everything for best performance. CRTs are "plug and play".

I didn't vote because none of the options seem to reflect my opinion. To the question "Original Video Quality vs. Best Video Quality?" my answer is "best video quality" hands down. But I don't consider digital signals and displays to be "best quality" for retro consoles. Not even close, altough they surely are for modern devices. I don't care too much for originality: I keep old consoles because I like playing them, not as collector items to be put on a shelf, so I A/V mod them. The fact is that "best video quality" is a matter of personal preference. When dealing with old consoles (I mostly refer to pre-crash hardware and the Atari 2600 in particular) I want to match the video technology that is, analog signal and CRT display, and at the same time I want to improve the quality the best I can, sending the signal to the display with the minimum of video conversion needed, and of course using a good quality display in the first place. I don't like rf interference, dot crawl and/or color bleed, or geometry and/or focus problems. On the other hand, the glow and persistance of the phosphors, the scanlines and the shadow mask/ aperture grille are not (to me) "defects" or "limitations due to the technology", but a fundamental part of these old video games. (It's like the grain on movies shot on film, or the texture of the canvas in paintings: it's part of the medium). Removing them (as it happens when using digital connections and displays) decreases the video quality compared to an optimal analog setup. This is a quite common opinion and is the reason why emulators developers spend time and efforts to include CRT filters, or why retrogaming dedicated upscalers include at least simple scanline effects. Many people wants those because they believe they enhance the picture. There are also emulator filters that can simulate rf interference, color bleed and dot crawl and some people like them because they want to recreate the experiece they had back in the day as close as possible. Again, that's a matter of personal preference and it's nice to have that option too. I aim to the best crystal clear picture, which to me means quality CRT display and RGB connection. An old game displayed on a HD digital TV, with perfectly square pixels just looks flat and dull. And that's assuming you spent a lot of money for highest quality video mods and upscalers; else the results are even worse. Honestly, I find the video from a CRT with some rf interference to be better quality than an emulator-like (with no crt filters) output, just like I think that an old movie with dirt and scratches looks better than one digitally restored with a too aggressive filter that removes not only dirt/scratches but also the film grain. No one "needs" to use 40 years old video game consoles in 2017. And in that case, no one "needs" to use an analog TV. But there are many good reasons to do so, not only video quality. This is my list of methods to play old games ranked from best to worst video quality: real console with RGB mod + CRT real console with S-Video mod + CRT emulator with CRT filters (to simulate the above two setups) + HD digital monitor/TV real console with composite mod + CRT real console with rf + CRT emulator with simple scanline filter + CRT computer monitor emulator with simple scanline filter + digital TV/monitor emulator without filters + CRT monitor emulator without filters + digital TV/monitor Real hardware with digital display doesn't even qualify, in my opinion. It takes a lot of efforts to just minimize the issues and the results are always inferior even to emulation.

Yup. This video covers it: https://www.youtube.com/watch?v=JUafi3EWBY8 (the star destroyer part is at 12:46)

Small regression in the new core. Keystone Kapers shows some problems (ball positioning, I guess), resulting in black lines on the left edge.

I wasn't aware of the latest release. Upgraded to 1.3 now.

You only need a dpdt switch and a new cable. The switch connects the pot and button to either paddle 1 or paddle 2 inputs of the port. You can use a genesis extension cord which has all pins connected, or build your own using a cable with at least 6 wires and a 9-pin female connector. A standard connector with shell won't fit on all consoles (especially heavy sixer and 7800, where the ports are quite recessed), so you need to modify them. I posted some pitures of the method I use in the videopac.nl forum: http://videopac.nl/forum/index.php?PHPSESSID=544fa181217704b34e199770b3418383&topic=2330.msg22880#msg22880

Pots are 1 Mohm, linear (not logarithmic!)

For a NTSC console you can use the "Color Bar Generator" From the manual:

Yes. Anyway, there could be some game that have paddle 1 and 2 reversed (as they're not marked in any way). So if you have some soldering skills, a better solution would be to wire a new cable and add a switch on the controller to select which player you want to use. In that way it can also be used for alternating 2-player games. This is mine (you can see the switch on the side of the controller):

I disassembled and cleaned all my paddle sets about 10 years ago, and they're still working great. They're electrically so simple that a replacement can be built by anyone with very basic electronic skills using cheap off-the-shelf parts, so I don't think it would be worth the effort to design a digital alternative.

I like to have the best picture quality possible, so I'm all for modding the consoles for RGB or s-video. I try to avoid composite, unless the console/computer already has an output for it and the quality is good. I've seen pictures of consoles modded like that and connected to digital TVs (with and without a dedicated upscaler like the framemeister) obtaining a picture almost indistinguishable from an emulator output without filters enabled. I find that totally awful, and I'd rather not play the games anymore if that would be the only possible way. The CRT look is (an important) part of the gaming experience, in my opinion. I like the glow of the phosphors and the persistance of them that make bright objects leave a faint trail behind them while moving on a dark background. I like to see the thin scanlines and the shadow mask or aperture grille that give texture to the low resolution image. And I like that pixels are not a bunch of square flat tiles, but instead bright spot of light without a perfectly defined edge (which doesn't mean the image is blurry: with a quality video signal like s-video or rgb every single pixel is perfectly visible. The crisp edges are an artifact of the digital display which was never there in the first place). Lag and lightgun support are additional issues that have been already mentioned. I only play old consoles on CRTs (and living in Europe, where basically every TV sold since the '80s has RGB input, surely helps) and I also have a PC connected to the large 50" LCD in the living room, where I can play the games through emulation using adapters for original controllers. With the PC set to the same native resolution as the TV and crt filters enabled I can enjoy the games on a big screen without most of the drawbacks of using a real console with it. A modern PC can generate the display and apply filters in a fraction of the time needed by real hardware to output a frame, while an upscaler has to wait that the entire frame is outputted by the console, then convert it to digital, then apply filters and it surely doesn't have the CPU power of a PC. So, it's not that I use CRTs because I'm a "elitist" (whatever that means), but because digital displays really detract from my enjoyment in playing old games. Emulation is a compromise I find somewhat acceptable if I really want to play on a big screen, but I consider real hardware connected to digital TV (with or without external upscaler) an unacceptable compromise, so upscalers or HDMI mod are totally worthless to me.

The b/w switch is just an input for the console so it's behaviour depends on the specific game. Early games used it to actually select an optimized b/w palette, later they used it for other purposes (for example as a pause switch) or just ignored it. Those two games don't use the switch at all, so what you see is normal.

I agree. I think it would be better to entirely ditch any reference to the actual models CX22 and CX80 and just call them Atari Trak-ball and Atari ST mouse protocols.

True. The fact is tha "CX22" and "CX80" in Stella refer actually to different encoding methods and not to the actual trakball models. "CX22" stands for the original Atari Trakball encoding (used by both the cx22 and cx80 models!) "CX80" stands for the Atari ST Mouse protocol (The misleading name comes from the fact that the first cx80 trakball which was documented in the Stella Mailing List was actually a modified unit that outputted Atari ST mouse compatible encoding and this led to believe that all cx80 were like that) While testing the 2600 Trackballs hacks, some of us used third party trackballs and mice using adapters and/or modifying them to change the output protocol. The 50% ratio applies to trackballs with the same mechanic characteristics (for example a standard CX80 and one modified for ST mouse output). http://atariage.com/forums/topic/243453-atari-2600-trak-ball-games/page-9?do=findComment&comment=3374393 It makes sense to use that same ratio in Stella when emulating one or the other protocol with the PC mouse.

Thanks for the hacked roms. I managed to take better pictures (I purposely misadjusted the color pot on the board in order to get a crisp B/W image). Hardware: PAL 7800 (board C300633-001 rev. C), TIA chip produced in 1989. Missiles only: Players only: Hope it helps.

I tested "Snoopy and the Red Baron", which in the stable Stella version has problems displaying the cloud and mountains, both made of a combination of players, missiles and playfield graphics updated and repositioned every line. Screenshots of the game in various Stella versions with and without "debug colors": real hardware: 5.0.0pre3 is closest to real hardware, the only issue seems to be a missing pixel in the left mountain. Real hardware screenshot is from my 7800 which has composite mod. It shows some color bleed, so it's difficult to see every pixel. Unfortunately my S-video modded 6-switch (which has a much clearer video out) needs some servicing. Maybe an hack of the game that only displays Players OR Missiles would help in debugging the problem. I used PAL version of the game, but apart for the different colors, the NTSC has the same behaviour.

There are two versions of Yars Return: the first one was released with the Flashback II console, and it had many problems including heavy flicker and unstable picture on many TVs. A fixed version was made by Dennis Debro (I don't know if he was able to fix the original code or just rewrote the game from scratch) and was included in the Flashback II+ and subsequent releases. See this post: http://atariage.com/forums/topic/156823-flashback-2-for-preorder-on-ataricom-get-a-free-t-shirt/?do=findComment&comment=1924667 Here's a video of the first one: https://www.youtube.com/watch?v=1gdX4hMTvNI&t=3m37s Here the fixed one: https://www.youtube.com/watch?v=PEnuhIbPDxc

Great review. Saw the movie yesterday and I enjoyed every bit of it. I liked it more than "The force Awakens", toward which I have mixed feelings. I need to rewatch it when it will be out in home-video, but right now I consider it my 3rd preferred movie of the saga (after StarWars and The Empire Strikes Back)