+stephena Posted March 23, 2013 Share Posted March 23, 2013 OK, then it is a light 6'er after all (which is what I suspected, but wasn't sure). Quote Link to comment Share on other sites More sharing options...
Omegamatrix Posted March 23, 2013 Author Share Posted March 23, 2013 I'm actually not sure if I have a light or heavy 6'er. I'm sure it's a 6-switch, of course. Is there a way of differentiating? I do know that the starfield effect in Cosmic Ark is different on my console than in Stella. In fact, I released a version of Stella that duplicated what's on my system some years back, but reverted it because it wasn't what most other people were seeing. BigO said it! Here is a picture I grabbed years ago (I think from the Wiz) which sums it up best for me. Just look at the shapes of the trim on these 2 consoles. The light sixer is thinner, and more angled. The heavy sixer is rounded, and thicker. For anyone just joining the thread, we are conducting some compatibility tests to see if Circus AtariAge will be able to run on different consoles. The current testing needs your help, and details can be found on the previous page, here! http://www.atariage....25#entry2720683 Quote Link to comment Share on other sites More sharing options...
Andromeda Stardust Posted March 24, 2013 Share Posted March 24, 2013 (edited) Wood Grain 4-switcher (S.N. 81817660, manufacture date = 1980): two columns for the text test. Edit: Something truly bizzare happened. I turned the TV off for a time while I went to find the adapter to access the memory card inside my phone. I left my Atari running in the meanwhile. I came back (~40 minutes later) and turned on the TV to find a third column of text has magically appeared where one was not before hand. The truly weird thing is that both photos are from THE SAME ATARI!!! Wood Grain 4-switcher (S.N. 81817660, manufacture date = 1980): three columns for the text test. Same console, same Harmony cart, same ROM, ~40 minutes time difference, two different screens. I turned it off and turned it back on and it still had three columns of text. Perhaps a cool TIA is more likely to display two lines of text while a warm TIA is more likely to display three? I would attempt to chill my Atari to see if it goes back to two columns, but I'm too afraid of possible damage from condensation. EDIT2: I just finished giving my Atari a 15 minute chill in the fridge. I plugged it in and once again, I got two columns of text. After a couple of minutes powered on, it changed back to three columns of text. My theory is confirmed: a cool TIA chip will display 2 columns; a warm TIA will display 3 columns. This tells me it's not a matter of Batch A versus Batch B, but subtle variations in part manufacture will dictate whether or not it displays 2 or three rows. at a given temperature. Could this be an issue of bus conflicts, perhaps? Some early NES mapper hardware could potentially suffer from bus conflicts if the game wasn't programmed correctly. Basically, if two outputs are tied to one input, then whichever output is stronger will overpower the weaker one. The DK "Complete" Edition ROM which got leaked in 2010 exhibited bus conflicts which caused graphical corruption on random NES consoles. Impedance and voltage drop change subtly with temperature, so that could be one explanation why my Atari displays both two and three column results depending on whether it's running cool or warm. Those of you with two column consoles, leave it plugged in for a while and report back if it changes from two to three. Or if you've got a hair dryer, a can of CPU coolant, and an exposed TIA chip handy, you can try freezing and cooking it... Edited March 24, 2013 by stardust4ever 2 Quote Link to comment Share on other sites More sharing options...
Omegamatrix Posted March 24, 2013 Author Share Posted March 24, 2013 Wood Grain 4-switcher (S.N. 81817660, manufacture date = 1980): two columns for the text test. Edit: Something truly bizzare happened. I left the TV on for a time while I went to find the adapter to access the memory card inside my phone. I left my Atari running in the meantime. I came back (~40 minutes later) to find a third column of text has magically appeared where one was not before hand. The truly wierd thing is that both photos are from THE SAME ATARI!!! Wood Grain 4-switcher (S.N. 81817660, manufacture date = 1980): three columns for the text test. Same console, same Harmony cart, same ROM, ~40 minutes time difference, two different screens... This might be due to thermal effects, as Joe noted a few posts back. I think you will have some understanding in this since I know you are taking Electronics in school. I read about your dedication to get a perfect square wave. I respect that. It is not easy to filter it just right. Did you try the balloon test rom? I would also be curious what would happen if you tried leaving the balloon test running for 40 minutes as well, on that same console (if you are willing). Just make sure it's fully cool, fire it up, and try the balloon rom. Leave it on, then come back in 40 minutes and see if it is responds the same. Quote Link to comment Share on other sites More sharing options...
Andromeda Stardust Posted March 24, 2013 Share Posted March 24, 2013 (edited) It only took a couple of minutes last time before it went back to displaying three lines. I literally glanced away from the TV for a few seconds then noticed that the third line had appeared. The balloon test appeared normal when I loaded it but the Atari was already warm. Every row had one missing balloon aligned diagonally, except for the bottom and top. I just put my Atari back in the fridge. This time I'm leaving it in there for 30 minutes instead of fifteen. Probably a dumb thing to do IMO, but it's all in the name of research. A couple of facts about electronics and temperature: Voltage drop of semiconductor junctions decreases as temperature increases. This has been known to cause a thermal runaway effect in certain classes of audio amplifiers. A small decrease in Vdrop with temperature increase causes a large increase in current, which causes the transistor to get even hotter, pulling more load in the process, until it burns up. Electrical resistance of non-semiconductor materials generally increases with temperature. This includes metallic conductors as well as other conductive materials such as carbon. Propagation delay of semiconductor logic generally decreases with decreasing temperature. This is why most attempts at breaking world CPU speed records are generally performed using liquid nitrogen (or in some extreme cases, helium, ouch!). AMD FX/Bulldozer rocks; 8Ghz, baby! http://hothardware.c...g-FX-Processor/ My guess here is that the propagation delay of some of the logic between the CPU/TIA chip may be on the cusp of just-in-time execution. The exact tolerances of one part compared to the next can often vary by up to 5, 10, or even 20%, which means that some parts may be faster/slower draw slightly more/slightly less current, etc, so some Ataris may exhibit this behavior and others might not, especially if you're using undocumented op-codes. Another possibility besides just the ambient temperature which may have a bigger impact is the 5V regulator. It's possible that one Atari may be running at 5.2V while another Atari may be running 4.8V. Much like a car, semiconductor logic also responds more quickly when you give it more juice (under- or over-volting a CPU can effect it's ability to operate reliably over a given range of speeds), so a logic chip running on 5.2V may have a slightly shorter propagation delay than the same chip running on 4.8V. It seems the issue at hand is the result of one clock cycle of difference regarding communication between the CPU and TIA, so the difference in either ambient temperature or voltage may have an effect. However, voltage and temperature won't have any noticeable impact on the timing crystal, because timing components are generally designed to have .001% tolerance across a very wide range of operating conditions. My timer just stopped; time to take the Atari out of the fridge and see what happens... EDIT: Nope, the Balloon demo looks just fine on a chilled TIA... EDIT2: That was coooool... I watched as the TIA warmed up. It took about 5-10 seconds total for the 1st column to transition from off to on. lots of random sprite flicker for a few short seconds. If can find my camera, I'll attempt to record it... EDIT3: Booyah! http://www.youtube.com/watch?v=2LzadQv77gc Skip to 1:20 to see the transition effect from two to three columns. Goodnight people; I'm hittin' the ol' haysack... Edited March 24, 2013 by stardust4ever 5 Quote Link to comment Share on other sites More sharing options...
BigO Posted March 24, 2013 Share Posted March 24, 2013 EDIT2: That was coooool... I watched as the TIA warmed up. It took about 5-10 seconds total for the 1st column to transition from off to on. lots of random sprite flicker for a few short seconds. If can find my camera, I'll attempt to record it... It was like that column was beamed in a la Star Trek. Quote Link to comment Share on other sites More sharing options...
Andromeda Stardust Posted March 24, 2013 Share Posted March 24, 2013 It was like that column was beamed in a la Star Trek. Beam me up, Scotty! Quote Link to comment Share on other sites More sharing options...
+SpiceWare Posted March 24, 2013 Share Posted March 24, 2013 I've seen something like that with the stars in Stay Frosty on my heavy sixer. 1 Quote Link to comment Share on other sites More sharing options...
iesposta Posted March 24, 2013 Share Posted March 24, 2013 The good news is that testing indicates your game runs on all hardware whether or not it shows two or three columns! The flashback 2+ was tested just because I could and really doesn't count - it was not intended to run carts and their numbers are few. 6th test: Tested ok on 4 switch Taiwan Atari. 2 columns and no balloon problems. 1 Quote Link to comment Share on other sites More sharing options...
Omegamatrix Posted March 25, 2013 Author Share Posted March 25, 2013 Thank you for your continued support, Iesposta. I'm really glad that you did test the flashback 2+. It would have been nice if it worked, but, c'est la vie. At this point I'm feeling cautiously optimistic about the kernels, but I would like some more heavy sixer's and 7800's tests for sure. I have a couple of Jr's myself (one is a PAL unit, thanks SvOlli!), and the other is a late model which is screwed up for Thunderground's score. Both passed the balloons trial by fire. I'm also really interested in the more exotic clones and adapters. The Gemini is right at the top of that list. Does anyone have a 5200 to 2600 adapter? I read there was also an Intellivision adapter. I don't know if anyone ever uses these but the Gemini and Colecovision adapter I'm sure some people use. I'm glad your Colecovision test passed. Sometime this summer I hope to reclaim my Atari collection. It's several hundred kilometers from me right now. I have a couple of 7800's, Coleco (with an adaptor), a Funvision clone, and a couple of 4 switchers? (can't remember). It'll be good to have my dumpers back at least. I'd love to fill a few of those holes in the PAL roms, if I can get carts cheap enough. Quote Link to comment Share on other sites More sharing options...
BigO Posted March 25, 2013 Share Posted March 25, 2013 (edited) At this point I'm feeling cautiously optimistic about the kernels, but I would like some more heavy sixer's and 7800's tests for sure. Tested on Sunnyvale H6, S/N 79394H: 2or3 showed 2, cleanly and consistently. BalloonTrial seems to have worked flawlessly. I note that in the posting where you tell how to test, the picture shows sticklowns in the balloon gaps. I see nothing in the gaps. Edited March 25, 2013 by BigO 1 Quote Link to comment Share on other sites More sharing options...
Omegamatrix Posted March 25, 2013 Author Share Posted March 25, 2013 You can see the players by pressing the select switch. Pressing the fire button scrolls the balloons along. I actually debated whether I should show any players or not in this demo... If the players covered up bits of the balloons appearing where they shouldn't then it would kind of defeat the purpose. On the other hand they visually give a great perspective of why the balloons are missing. I was going to do some collision detection for the players in case the balloons showed up where they shouldn't, but in the end I decided to just make a default of having no players to begin with. I figured most people would scroll through it at least once like that anyway, and the players are still available if they want to test with them too. Quote Link to comment Share on other sites More sharing options...
BigO Posted March 25, 2013 Share Posted March 25, 2013 (edited) Ah, sorry. I guess I missed that part. Anyway, the test was a success. Good luck with your project. I'm looking forward to seeing it completed. I assume you're writing this from scratch as opposed to hacking the original. Either way, would you mind giving me an opinion on the feasibility of making a version of Circus that plays slower than the original? Kid's version...low gravity...circus on the moon sorta thing. I haven't really dug into the original source code to see if I can figure out how that might work. One of these days... Edited March 25, 2013 by BigO Quote Link to comment Share on other sites More sharing options...
accousticguitar Posted March 25, 2013 Share Posted March 25, 2013 I tested out my spare 7800 with similar results. The two or three kernel showed two columns of text most of the time, but occasionally would show three columns. The balloon kernel seemed to work fine. I will dig out some heavy sixers to test next. 1 Quote Link to comment Share on other sites More sharing options...
accousticguitar Posted March 25, 2013 Share Posted March 25, 2013 First heavy sixer worked good. The two or three kernel showed two columns of text every time (I tried it ten times). The balloon kernel tested fine with no problems. 1 Quote Link to comment Share on other sites More sharing options...
accousticguitar Posted March 25, 2013 Share Posted March 25, 2013 Next heavy sixer displayed two columns of text the first time I turned it on and three columns every other time I tried it. The balloon kernel worked fine. 1 Quote Link to comment Share on other sites More sharing options...
accousticguitar Posted March 25, 2013 Share Posted March 25, 2013 The last heavy sixer showed two columns of text every time. The balloon kernel worked fine. 1 Quote Link to comment Share on other sites More sharing options...
Omegamatrix Posted March 25, 2013 Author Share Posted March 25, 2013 I'm loving those heavy 6'ers tests. Thanks Accousticguitar!! I assume you're writing this from scratch as opposed to hacking the original. Either way, would you mind giving me an opinion on the feasibility of making a version of Circus that plays slower than the original? Kid's version...low gravity...circus on the moon sorta thing. I haven't really dug into the original source code to see if I can figure out how that might work. One of these days... Yep, it's an all new game. I like the idea of moon gravity though, maybe I'll consider it for an option! As to the original, there was lots of space left in the rom, so hacking it should be easy. I believe it just used tables for the arching positions, so you just have to find and update those. I would help you with this, but I got way too many things on the go. Nukey has hacked Circus Atari lots though, and he might now the locations right away. Quote Link to comment Share on other sites More sharing options...
Andromeda Stardust Posted March 25, 2013 Share Posted March 25, 2013 Did anyone else test the "2 or 3" demo and get both results on the same Atari, or have a better explanation as to why it happens? See the transition video I made in post #55. Really bizarre... Quote Link to comment Share on other sites More sharing options...
iesposta Posted March 25, 2013 Share Posted March 25, 2013 Did anyone else test the "2 or 3" demo and get both results on the same Atari, or have a better explanation as to why it happens? See the transition video I made in post #55. Really bizarre... Yes. Both acousticguitar and myself have 7800's that get both results. Your video of the third column appearing is epic! Quote Link to comment Share on other sites More sharing options...
Omegamatrix Posted July 27, 2013 Author Share Posted July 27, 2013 Hi everyone! We are still a few months away from a playable build... probably be next year , but I thought I'd post a rom of the current build running the new kernels. There is not an interactive demo, but you can press ALT-comma to see all the kernel masking and switching going on. CAA_072713.bin So far each row has a different speed, and personality. The red balloons are really slow while the green balloons zip by really quick. The blue balloons go to a beat all their own, and like to bounce around. The new kernels allow fairly smooth scrolling while never masking the jumper, and never flickering the jumper. I tried to keep the 30Hz flicker to a minimum through low luminosity in the color selections. The score and lives are now color striped. Hope you enjoy!! Omega 4 Quote Link to comment Share on other sites More sharing options...
iesposta Posted July 27, 2013 Share Posted July 27, 2013 A Carney's life is dangerous, as taught to me so very long ago by Circus Atari. Why, I remember as if it were yesterday when my player missed the end of the teeter totter and died, his head smashed flat, legs still a-twitching. Such a shocking introduction to a videogame portraying violent death! Quote Link to comment Share on other sites More sharing options...
Omegamatrix Posted July 28, 2013 Author Share Posted July 28, 2013 A Carney's life is dangerous, as taught to me so very long ago by Circus Atari. Ha ha, yes it is. Why, I remember as if it were yesterday when my player missed the end of the teeter totter and died, his head smashed flat, legs still a-twitching. Such a shocking introduction to a videogame portraying violent death! That was one of my favorite things, and the humor is what made it great for me. I'm hoping to have a little blood spilt when the jumper goes splat , but right now it is all about getting the core routines done. Progress has been painfully slow up to now, but after collisions are in place things should pick up steam. I made a second build today, build B. I lengthened the time that the bottom row swings, and I made the middle row all dive at the same time instead of toggling. I also changed the animation in the middle row, to make it more simpler. It looks better to me, but I still might revisit it several times until it feels "right". Not too sure. In this build pressing select will toggle between having all the balloons on, and set pattern of balloons on. CAA_072713b.bin 1 Quote Link to comment Share on other sites More sharing options...
Wickeycolumbus Posted July 28, 2013 Share Posted July 28, 2013 This looks amazing! Can't wait to see how it progresses. Quote Link to comment Share on other sites More sharing options...
Andromeda Stardust Posted July 28, 2013 Share Posted July 28, 2013 Did you ever work out the "2 or 3" bug where some systems will occasionally display different numbers of sprites? I posted a video a while back showing how the transition can occur even on the same system due to changing latency with thermal fluctuations, so whatever code caused the initial sprite glitch ought to be avoided. Hopefully you worked this out. Quote Link to comment Share on other sites More sharing options...
Like us on Facebook
Follow us on Twitter
AtariAge on Instagram
Watch us on YouTube
Chat on Discord
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.