+x=usr(1536) Posted January 10, 2021 Share Posted January 10, 2021 (edited) OK, seeing something I hadn't noticed before. For reference, here's the directory tree on the 64GB card I'm using with the Concerto: / ├── 2600 │ ├── 0-9 │ ├── A │ ├── B │ ├── C │ ├── D │ ├── E │ ├── F │ ├── G │ ├── H │ ├── I │ ├── J │ ├── K │ ├── L │ ├── M │ ├── N │ ├── O │ ├── P │ ├── Q │ ├── R │ ├── S │ ├── T │ ├── U │ ├── V │ ├── W │ ├── X │ ├── Y │ └── Z ├── 7800 │ ├── HHDP_20200603 │ │ ├── 0-9 │ │ ├── A │ │ ├── B │ │ ├── C │ │ ├── D │ │ ├── E │ │ ├── F │ │ ├── G │ │ ├── H │ │ ├── I │ │ ├── J │ │ ├── K │ │ ├── L │ │ ├── M │ │ ├── N │ │ ├── O │ │ ├── P │ │ ├── Q │ │ ├── R │ │ ├── S │ │ ├── T │ │ ├── U │ │ ├── V │ │ ├── W │ │ ├── X │ │ ├── Y │ │ └── Z │ └── Retail_v2_0 │ ├── NTSC │ │ ├── A │ │ ├── B │ │ ├── C │ │ ├── D │ │ ├── E │ │ ├── F │ │ ├── G │ │ ├── H │ │ ├── I │ │ ├── J │ │ ├── K │ │ ├── L │ │ ├── M │ │ ├── N │ │ ├── O │ │ ├── P │ │ ├── Q │ │ ├── R │ │ ├── S │ │ ├── T │ │ ├── U │ │ ├── V │ │ ├── W │ │ ├── X │ │ ├── Y │ │ └── Z │ └── PAL │ ├── A │ ├── B │ ├── C │ ├── D │ ├── E │ ├── F │ ├── G │ ├── H │ ├── I │ ├── J │ ├── K │ ├── L │ ├── M │ ├── N │ ├── O │ ├── P │ ├── Q │ ├── R │ ├── S │ ├── T │ ├── U │ ├── V │ ├── W │ ├── X │ ├── Y │ └── Z └── WIP ├── 2600 └── 7800 ├── A ├── M └── U Under /2600/C/ , page numbers appear as 'Page 0000 of 000?'. '000?' translates out to 10 pages in total. Other directories may also be affected; that's just the one that I noticed it under. 2600 directory structure is based off of the ROMHunter pack; 7800 structure is Trebor's 7800 pack. The WIP directory follows a similar structure to both of them, but is one that I made in order to keep WIP ROMs separate from the 'official' releases. Edited January 10, 2021 by x=usr(1536) Quote Link to comment Share on other sites More sharing options...
Trebor Posted January 10, 2021 Share Posted January 10, 2021 Another A3 console I have is the most promising one yet. I need to perform many more and thorough tests; however, menu is flawless and many games so far loaded and played beautifully. There are some exceptions so far with results similar to @John Stamos Mullet. The graphical corruption on top of Choplifter for example is well more extensive than it should be normally. Tower Toppler loading crashes with letters and symbols replacing the normal two digit numbers. But again, many things (I will provide specifics in the near future) are running wonderfully. I'm also going to be testing an A1 unit either tomorrow or Monday. A3 Serial 'LH' specifics - To distinguish it from the UAV one: No chips socketed - resistor present - C025233-002 Motherboard MARIA: (Four lines pic)821 V N2522 2049C0002 C025349-30 KOREA A SALLY: C014806-12 (Atari Fuji) 11151-12 MEXICO E1 8831 2398-1 RAM: LH5116-15 SHARP JAPAN 8827 3EB (x2) BIOS: "Universal Mod" O.S. w/ASTERDS 7800 2/21/89 NTSC/PAL ATARI 187A TIA: iMP (SYMBOL diff than previous system) 8832LEA C010444-31 (c)ATARI 1980 I10350-01 RIOT: C010750-12 R6532-15 MEXICO K3 8740 G876 1 Quote Link to comment Share on other sites More sharing options...
+mksmith Posted January 10, 2021 Share Posted January 10, 2021 2 hours ago, DrVenkman said: Found another issue with corrupted loading - the most recent (week old) ARKANOID WIP version loads, and the intro screens and menu display properly, but the game playfield is corrupted, the game does not actually start or play, and there's no sound at all. I just verified via A7800 that there should be an AtariAge jingle plus sound in the intro screens. 2 hours ago, x=usr(1536) said: Confirming that I'm seeing the same behaviour. Thanks guys! I suspect that the issue relates to the pokey initialisation much the same as what happened with Popeye (as both are based on 7800basic). I'll need to investigate how Darryl fixed Popeye to see what I can do for a specific Concerto build. My Concerto currently doesn't get past loading much at all on either of my PAL 7800s currently but will see if I can test it here first. I should crack both of mine open and list some things as Robert is doing. 1 Quote Link to comment Share on other sites More sharing options...
Swami Posted January 10, 2021 Share Posted January 10, 2021 3 hours ago, RevEng said: Yep, that's what I meant. Game loads but is messed up. I noticed a lot of mentions of socketed Sally's with these reports, so I was wondering if there was a pattern there. Thanks for confirming it's not. Like I said, Alien Brigade works fine with my old Concerto board with three year old firmware Quote Link to comment Share on other sites More sharing options...
Swami Posted January 10, 2021 Share Posted January 10, 2021 1 hour ago, Trebor said: Another A3 console I have is the most promising one yet. I need to perform many more and thorough tests; however, menu is flawless and many games so far loaded and played beautifully. There are some exceptions so far with results similar to @John Stamos Mullet. The graphical corruption on top of Choplifter for example is well more extensive than it should be normally. Tower Toppler loading crashes with letters and symbols replacing the normal two digit numbers. But again, many things (I will provide specifics in the near future) are running wonderfully. I'm also going to be testing an A1 unit either tomorrow or Monday. A3 Serial 'LH' specifics - To distinguish it from the UAV one: No chips socketed - resistor present - C025233-002 Motherboard MARIA: (Four lines pic)821 V N2522 2049C0002 C025349-30 KOREA A SALLY: C014806-12 (Atari Fuji) 11151-12 MEXICO E1 8831 2398-1 RAM: LH5116-15 SHARP JAPAN 8827 3EB (x2) BIOS: "Universal Mod" O.S. w/ASTERDS 7800 2/21/89 NTSC/PAL ATARI 187A TIA: iMP (SYMBOL diff than previous system) 8832LEA C010444-31 (c)ATARI 1980 I10350-01 RIOT: C010750-12 R6532-15 MEXICO K3 8740 G876 I buy a couple more to test again, but they always have the horrible hiss I can't get rid of, even with the audio pot, unless I do a AV mod. I sometimes think that's how you get the cheap ones that are "working". Quote Link to comment Share on other sites More sharing options...
+batari Posted January 10, 2021 Author Share Posted January 10, 2021 2 hours ago, x=usr(1536) said: OK, seeing something I hadn't noticed before. For reference, here's the directory tree on the 64GB card I'm using with the Concerto: / ├── 2600 │ ├── 0-9 │ ├── A │ ├── B │ ├── C │ ├── D │ ├── E │ ├── F │ ├── G │ ├── H │ ├── I │ ├── J │ ├── K │ ├── L │ ├── M │ ├── N │ ├── O │ ├── P │ ├── Q │ ├── R │ ├── S │ ├── T │ ├── U │ ├── V │ ├── W │ ├── X │ ├── Y │ └── Z ├── 7800 │ ├── HHDP_20200603 │ │ ├── 0-9 │ │ ├── A │ │ ├── B │ │ ├── C │ │ ├── D │ │ ├── E │ │ ├── F │ │ ├── G │ │ ├── H │ │ ├── I │ │ ├── J │ │ ├── K │ │ ├── L │ │ ├── M │ │ ├── N │ │ ├── O │ │ ├── P │ │ ├── Q │ │ ├── R │ │ ├── S │ │ ├── T │ │ ├── U │ │ ├── V │ │ ├── W │ │ ├── X │ │ ├── Y │ │ └── Z │ └── Retail_v2_0 │ ├── NTSC │ │ ├── A │ │ ├── B │ │ ├── C │ │ ├── D │ │ ├── E │ │ ├── F │ │ ├── G │ │ ├── H │ │ ├── I │ │ ├── J │ │ ├── K │ │ ├── L │ │ ├── M │ │ ├── N │ │ ├── O │ │ ├── P │ │ ├── Q │ │ ├── R │ │ ├── S │ │ ├── T │ │ ├── U │ │ ├── V │ │ ├── W │ │ ├── X │ │ ├── Y │ │ └── Z │ └── PAL │ ├── A │ ├── B │ ├── C │ ├── D │ ├── E │ ├── F │ ├── G │ ├── H │ ├── I │ ├── J │ ├── K │ ├── L │ ├── M │ ├── N │ ├── O │ ├── P │ ├── Q │ ├── R │ ├── S │ ├── T │ ├── U │ ├── V │ ├── W │ ├── X │ ├── Y │ └── Z └── WIP ├── 2600 └── 7800 ├── A ├── M └── U Under /2600/C/ , page numbers appear as 'Page 0000 of 000?'. '000?' translates out to 10 pages in total. Other directories may also be affected; that's just the one that I noticed it under. 2600 directory structure is based off of the ROMHunter pack; 7800 structure is Trebor's 7800 pack. The WIP directory follows a similar structure to both of them, but is one that I made in order to keep WIP ROMs separate from the 'official' releases. Thanks, looks to be a bug in the code that converts decimal numbers to ASCII. The "?" should be "9". 1 Quote Link to comment Share on other sites More sharing options...
+darryl1970 Posted January 10, 2021 Share Posted January 10, 2021 8 hours ago, mksmith said: Thanks guys! I suspect that the issue relates to the pokey initialisation much the same as what happened with Popeye (as both are based on 7800basic). I'll need to investigate how Darryl fixed Popeye to see what I can do for a specific Concerto build. I will let you know. Just note that it does not fix the sound completely. The music sometimes modulates off key, especially the tunes. It's better than nothing though. 1 Quote Link to comment Share on other sites More sharing options...
+x=usr(1536) Posted January 10, 2021 Share Posted January 10, 2021 8 hours ago, mksmith said: Thanks guys! I suspect that the issue relates to the pokey initialisation much the same as what happened with Popeye (as both are based on 7800basic). I'll need to investigate how Darryl fixed Popeye to see what I can do for a specific Concerto build. My Concerto currently doesn't get past loading much at all on either of my PAL 7800s currently but will see if I can test it here first. I should crack both of mine open and list some things as Robert is doing. Just to clarify: I don't have a POKEY installed (yet). It'll be a couple of weeks before I can get to that point. 1 Quote Link to comment Share on other sites More sharing options...
Trebor Posted January 10, 2021 Share Posted January 10, 2021 So, I tested one of my A1 serials, we will call 'A1 BE' (Best Electronics purchase with Simple Composite Mod installed). It is performing exactly like my A3 Serial 'LH' console. Between the two consoles, every single chip is different. There is also a difference in what is socketed and what is not. The A3 Serial 'LH' has the full expansion port, the A1 Serial 'BE' does not have one. The items they share in common is the same Motherboard revision C025233-002 and a resistor is present. That has provided the best compatibility for me so far. What is also worth noting is the other A3, Serial A3 'UAV', which isn't performing as well with the Concerto (and is performing like my Serial AT84 - no resistor either), had the resistor present originally, but was removed when the UAV mod install was performed. Removing it, fixed graphical glitching between one version of a Pole Position cart and a Ballblazer retail cartridge. There is a video about it by @-^CrossBow^- (who did a excellent job with the UAV mod). A1 Serial 'BE' specifics: MARIA and SALLY socketed only - resistor present C025233-002 Motherboard MARIA: VTi (SYMBOL) 427 V 2402 VC 2049B CO34674-30 KOREA-A SALLY: NCR F815108 G 8504 C0148060-29 (c)ATARI 1983 RAM: SONY 6G30 CXK5816PN-10L JAPAN (x2) BIOS: MCR2333-30 F813292F8429 CO24922B-29USA (c)ATARI 1984 TIA: AMI 8645MGT C010444D-01 C04075 (c) PHILIPPINES RIOT: (Globe Pic) 8642C UM6532 In addition to the four consoles I am currently leveraging, I do have two more consoles, a second "A1" unit, and a third "A3" unit. I can utilize them as well if deemed needed. Also, wanted to say *thank you*, @batari!!! If it was not clear to me before how daunting of a task it is to have Concerto working well across all systems, this has made it crystal clear. And thank you to the community for coming together once again and providing necessary feedback and data! 3 Quote Link to comment Share on other sites More sharing options...
Trebor Posted January 10, 2021 Share Posted January 10, 2021 So, after some moderate testing with the A1, I was feeling pretty confident. I thought to myself, I will install the POKEY chip and place the shell back on the Concerto PCB, as already having one of the new(er) Concertos with Firmware 0.95 installed, it's unlikely that I will need to remove the POKEY chip for updates. I first installed the POKEY as shown in choice "A" below, thinking to match the small notch on the socket to the large indentation notch on the POKEY chip. Tried it out, and the cart won't load. Received a 'fry' like result on screen. Decided the chip may be upside down, I then installed it as seen in choice "B" below. Same results. I have now completely removed the POKEY chip, but it has not restored functionality of the Concerto. I'm still obtaining console 'fry' like results across all systems. Before I proceed with doing anything else, @batari, what would be best to do at this juncture? And to assist others, which is the correct orientation of the POKEY chip for those installing one...Is it choice "A" or "B"? Quote Link to comment Share on other sites More sharing options...
Jinks Posted January 10, 2021 Share Posted January 10, 2021 There should be a picture of how the pokey should be installed and if there are variations where pin 1 etc is backwards or not. Sounds like you may have bricked it for real.. ? I hope not.. I have never installed my pokey yet to scared it might get all blowed up. ? Quote Link to comment Share on other sites More sharing options...
+DrVenkman Posted January 10, 2021 Share Posted January 10, 2021 @Trebor, pic A is correct. 3 Quote Link to comment Share on other sites More sharing options...
Trebor Posted January 10, 2021 Share Posted January 10, 2021 8 minutes ago, Jinks said: Sounds like you may have bricked it for real.. I'm pretty certain Fred stated you can't brick a Concerto and it can be restored regardless via USB in the other/original big thread. Nonetheless, waiting for his response for next best step(s). Quote Link to comment Share on other sites More sharing options...
Trebor Posted January 10, 2021 Share Posted January 10, 2021 9 minutes ago, DrVenkman said: @Trebor, pic A is correct. Perfect, thank you! That made the most logical sense to me and what I did first, but obtained the aforementioned results. Unsure though why removing the POKEY completely has not restored it back to how it was prior. When Fred has the opportunity, I'm sure he will provide some additional insight. Quote Link to comment Share on other sites More sharing options...
+-^CrossBow^- Posted January 10, 2021 Share Posted January 10, 2021 1 hour ago, Trebor said: What is also worth noting is the other A3, Serial A3 'UAV', which isn't performing as well with the Concerto (and is performing like my Serial AT84 - no resistor either), had the resistor present originally, but was removed when the UAV mod install was performed. Removing it, fixed graphical glitching between one version of a Pole Position cart and a revision of a Ballblazer retail cartridge. There is a video about it by @-^CrossBow^- (who did a excellent job with the UAV mod). I added a 100nf cap inline of the A15 off the CPU to Maria and eventually to the bios to correct the graphical glitching. The only other thing that might have been done on yours ( let me check the pics of the board real quick...)... and yes confirmed. Capacitor 64 was clipped out of circuit on yours to improve compatibility with Activision games and some homebrew demos. But as I received your 7800 mainboard separately from rest of the case, it actually arrived to me already clipped. I actually de-solder one leg of the cap out instead of clipping it, but that would be the only difference. If you wanted to or felt comfortable with it, you could certainly take off the shielding, remove the small piece of black heat shrink tubing that covers the capacitor, and then apply a dab of solder down onto where the leg was clipped to but it back into circuit to see if it makes a difference? Although now that I look at the pic more closely, it appears that whomever had this mainboard and tried to repair it before me, shrunk the entire cap instead of just on the end allow it to be pulled off easier. Hmm.. You could even just open it up and with everything plugged in, press down on that cap so it makes contact and then power it up to see if it makes a difference? The spot in question is circled in this photo below of your mainboard in your 7800 I did. 2 Quote Link to comment Share on other sites More sharing options...
Trebor Posted January 10, 2021 Share Posted January 10, 2021 Thank you for the background and detailed info! Once my Concerto is functioning again, I will try the press down on the cap. The solder work I'm not comfortable with doing, as my "X" serial served as tribute several years back. Not my best moment(s). 1 Quote Link to comment Share on other sites More sharing options...
+-^CrossBow^- Posted January 10, 2021 Share Posted January 10, 2021 Well that heatshrink would have to be trimmed back so expose what is the left of the capacitor leg before you can press it down to make contact with the mainboard again. Assuming there is enough component lead off that cap to do so. But I don't know as it arrived to me that way in a non working state and I had to replace the 6502 and the RIOT at some point in the past to get it working fully again. Along with the other small stuff I did late like that extra cap I added on the bottom side of the PCB for the graphic corruption on PP2 and BallBlazer. Quote Link to comment Share on other sites More sharing options...
+batari Posted January 10, 2021 Author Share Posted January 10, 2021 3 hours ago, Trebor said: So, after some moderate testing with the A1, I was feeling pretty confident at this point. I will now install the POKEY and place the shell back on the Concerto PCB, as already having one of the new(er) Concertos with Firmware 0.95 already installed, it's unlikely that I will need to remove it for updates. I first installed the POKEY as shown in choice "A" below, thinking to match the small notch on the socket to the large indentation notch on the POKEY chip. Tried it out, and the cart won't load. Received a 'fry' like result on screen. Decided the chip may be upside down, I then installed it as seen in choice "B" below. Same results. I have now completely removed the POKEY chip, but it has not restored functionality of the Concerto. I'm still obtaining console 'fry' like results across all systems. Before I proceed with doing anything else, @batari, what would be best to do at this juncture? And to assist others, which is the correct orientation of the POKEY chip for those installing one...Is it choice "A" or "B"? There is a notch on the chip that you match up with the notch mark on the PCB (below the socket.) Typically the socket also has a notch. Match the notches up for correct orientation. 1 Quote Link to comment Share on other sites More sharing options...
zezba9000 Posted January 10, 2021 Share Posted January 10, 2021 Starting to test some stuff with mine. I updated from 0.92 to 0.93 then 0.94. Ms Pac Man working for 7800 & 2600 (first game tried). 0.94 is the latest firmware correct and I didn't miss one? Like others I saw in the form had issues with SD compatibly. I did as well on the first two SD cards I tried. Sharing the one that worked for me. Quote Link to comment Share on other sites More sharing options...
+Karl G Posted January 10, 2021 Share Posted January 10, 2021 20 minutes ago, zezba9000 said: 0.94 is the latest firmware correct and I didn't miss one? 0.95 is the latest. Maybe the first post of this topic could be edited with a link to the current latest firmware? 1 1 Quote Link to comment Share on other sites More sharing options...
+DrVenkman Posted January 10, 2021 Share Posted January 10, 2021 21 minutes ago, zezba9000 said: 0.94 is the latest firmware correct and I didn't miss one? 0.95 is the latest version, posted somewhere in this thread. My Concerto arrived last week with that firmware version. Quote Link to comment Share on other sites More sharing options...
zezba9000 Posted January 10, 2021 Share Posted January 10, 2021 (edited) Any suggestion on where / what POKEY chip to get / use? Is this a good price or better shop for these: https://www.ebay.com/itm/PokeyONE-Atari-POKEY-Chip-Replaces-Atari-P-N-C012294-137430-001-NEW/124343070444 Edited January 10, 2021 by zezba9000 Quote Link to comment Share on other sites More sharing options...
Trebor Posted January 10, 2021 Share Posted January 10, 2021 2 hours ago, batari said: There is a notch on the chip that you match up with the notch mark on the PCB (below the socket.) Typically the socket also has a notch. Match the notches up for correct orientation. Thanks @batari, I did that first. However, it resulted in a 'frying' console looking screen. Thinking it was perhaps mounted incorrectly, I reversed POKEY chip orientation and received the same result. I then removed the POKEY chip completed; however the Concerto is still producing a 'fried' console looking screen when powered on across all my systems. Any recommendation for what I should do as next/best steps to restore the Concerto? *EDIT: I went ahead and from the firmware update post, followed the directions provided. Despite keeping and running an up to date version of Windows... Edition: Windows 10 Home (x64) Version: 20H2 OS Build: 19042.685 ...I still needed the Windows driver from the driver post for the Harmony Software to recognize my cart. Without it, Windows recognized the cart as a "USB Serial" device when attached, but the Harmony software did not see it. After the driver install, the recognition of the cart, as well as a successfully 100% downloaded eeloader_ntsc.bin file, went smoothly. I placed the 'hbios_concerto_0.95_SDcardfix.bin' file to the root of the SD Card. Power on the console, but still see a 'fried' screen (sometimes a blank screen - similar randomness when "frying"). Tried without the SD card inserted, and a different SD card. Same results. Any other suggestions or guidance is appreciated. Quote Link to comment Share on other sites More sharing options...
zezba9000 Posted January 10, 2021 Share Posted January 10, 2021 1 hour ago, Trebor said: ...I still needed the Windows driver from the driver post for the Harmony Software to recognize my card. I feel like something changed in the driver sub-system at some point in Win10 but can't remember when. Around the time they made it so if graphics drivers don't have special flags, Windows will replace the driver you just installed with their old ass one in their DB. Quote Link to comment Share on other sites More sharing options...
tep392 Posted January 10, 2021 Share Posted January 10, 2021 I've tested my A1 and A3 s/n machines with the 0.95 firmware and they generally work very well. I don't have any problems with the menu's. Most 7800 games and homebrews work well except for the games that most are having trouble with like Tower Toppler. I have the problem with fire buttons not working on some games as described by others. My A3 machine is stock. My A1 was stock until I removed the infamous timer circuit. I did this when testing the XM earlier this year. Removing the timing circuit cured some glitch's we had with the XM RAM testing. I'm posting hires photo's of the mother boards to document the chips rather than listing them out. EDIT: I forgot that I also modified the clock circuit. It used to have the original two transistor buffer design. I modified it to be the same as the A3 model with a single transistor. This cleaned up the clock signal a bit. See removed Q4 with added jumpers. There were a few other componenets removed with this mod. This is the A1 with timing circuit removed (red rectangle around spot on board) and modified clock circuit (Q4). This is the A3 unit. 1 Quote Link to comment Share on other sites More sharing options...
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.