Concerto firmware thread

[RevEng]

4 hours ago, evg2000 said:

I ran the test program on my updated concerto, and I saw a graphics glitch on one screen and at least one error I thing on the cart integrity test.  First two rows are unmodified A78s, the third row is my composite modded A78.
Not sure if the error is from the cart, or if it could be my system?

Yeah, looks like a rom load error. If you shutdown the console and launch the same rom from Concerto, do you get the same result in the exact same place, or does the bad rom location move? It might be good to give the console slot an alcohol cleaning, just in case.

 

It would be good if everybody reporting an issue could try out the "internal ram test" menu item in the same 7800 utilitycart rom, and advise if it was clean or not with your report.

[+ZeroPage Homebrew]

I had some time to test the Concerto cart on my 7800 today. I was testing with two SD cards (see pic below):

 

Transcend 2GB - Didn't work on Firmware 0.92, 0.93 or 0.94, no file/directory listings

Patriot 16GB Class 6 - Worked on Firmware 0.92, 0.93 and 0.94

 

I do experience quite consistent freezing while navigating through menus on both firmware 0.93 and 0.94, forcing me to reboot each time, but no problems once a game was loaded. Looking forward to trying the new firmware fix for the issue as it will be noticeable for me once it's fixed.

 

As others experienced, pressing the button to 'Reboot' after upgrading the firmware from 0.92->0.93->0.94->0.93 never rebooted the 7800, it just sat there with a rapidly flashing/changing square in the bottom left corner (see pic below).

 

- James

 

[+x=usr(1536)]

5 hours ago, RevEng said:

It would be good if everybody reporting an issue could try out the "internal ram test" menu item in the same 7800 utilitycart rom, and advise if it was clean or not with your report.

 

Here's mine for reference (and I'll run a fresh one with each report sans image):

 

 

FWIW, external RAM also passed with Cartram Found.  Would it be worth tracking Console Info as well?

 

Edited December 21, 2020 by x=usr(1536)

[RevEng]

20 minutes ago, x=usr(1536) said:

Would it be worth tracking Console Info as well?

If someone has "unknown bios found" without actually having installed an alternate bios, I'd totally be interested in hearing that. But for Concerto troubleshooting purposes the Console Info screen can be skipped, I think. A mention of the console region is always useful, of course.

[mimo]

8 hours ago, ZeroPage Homebrew said:

 

 

As others experienced, pressing the button to 'Reboot' after upgrading the firmware from 0.92->0.93->0.94->0.93 never rebooted the 7800, it just sat there with a rapidly flashing/changing square in the bottom left corner (see pic below).

 

- James

 

 

 

On 12/15/2020 at 11:52 PM, batari said:

 

 

From version 0.92, you can update from the menu. Download the .cbi file, copy to your SD card, boot Concerto and press fire. Wait a moment until you get the confirmation screen. (auto rebooting isn't working yet, so manually power cycle.)

 

 

 

 

[+evg2000]

21 hours ago, RevEng said:

Yeah, looks like a rom load error. If you shutdown the console and launch the same rom from Concerto, do you get the same result in the exact same place, or does the bad rom location move? It might be good to give the console slot an alcohol cleaning, just in case.

 

It would be good if everybody reporting an issue could try out the "internal ram test" menu item in the same 7800 utilitycart rom, and advise if it was clean or not with your report.

So the location did move.  I did a good cleaning of the connector and now it reports as good!

So after getting it working on all 3 of my machines I then installed a pokey.  The test do not show the pokey, but ballblazer has sound as does bentley bear.  So I believe that means the pokey is working? Just the test didn't recognize the chip?

Also, is it common to have a capacitor conned to a couple of the pokey pins?  I removed it to get it to fit, but can reattach if there is a good reason for it to be there.  Pulled from a 5200 MB I had lying around.

thanks!

[RevEng]

3 hours ago, evg2000 said:

So the location did move.  I did a good cleaning of the connector and now it reports as good!

Awesome news! Flash carts in general seem to be more sensitive to cart pin resistance than regular rom carts, so cleaning is always a good place to start.

 

The utility cart software doesn't know how to look for a write-only pokey@450  (technically it can't be detected, it just needs to be written to blindly by the software) so yeah, your lack of pokey detection in the utility cart is normal.

 

I don't know how common it is to have a cap there on pokey, but it appears to just be a power decoupling or conditioning cap. (it's connecting v and gnd) I wouldn't sweat adding it back on, but we'll see what batari says about it.

[+batari]

35 minutes ago, evg2000 said:

So the location did move.  I did a good cleaning of the connector and now it reports as good!

So after getting it working on all 3 of my machines I then installed a pokey.  The test do not show the pokey, but ballblazer has sound as does bentley bear.  So I believe that means the pokey is working? Just the test didn't recognize the chip?

Also, is it common to have a capacitor conned to a couple of the pokey pins?  I removed it to get it to fit, but can reattach if there is a good reason for it to be there.  Pulled from a 5200 MB I had lying around.

thanks!

Keep the capacitor. In the more recent Concertos I do have a sizable capacitor between these lines right underneath the POKEY.

 

It is a good idea to add one as the POKEY would interfere with the data lines on older Concertos. Interesting that Atari found the same sort of thing back in the day.

[+ZeroPage Homebrew]

On 12/20/2020 at 1:48 PM, RevEng said:

It would be good if everybody reporting an issue could try out the "internal ram test" menu item in the same 7800 utilitycart rom, and advise if it was clean or not with your report.

Using Firmware v0.93

 

- James

 

 

 

 

[Swami]

I received my Concerto on Saturday and tested it a little. With the Concerto from the first run, I grabbed up five 7800s of different major revisions and kept the three that worked best, to varying degrees. The one I started with here worked perfectly with the first run Concerto if it made it to the intro screen, so, no surprise, it works with all the 7800 games I tried like BBCQ (with POKEY!)  and Popeye demo. It also worked with the simpler 2600 games and homebrews, but only less complex than those like Bosconian, which failed to load. I can try some more tomorrow and a different 7800. I did experience a menu freeze and a high pitch noise after power cycle once each.

 

Used 0.93 and I used a SanDisk Class 4 32GB SDHC card with no issues.

[Shawn]

Original fully socketed ntsc model 7800 here and these are my results so far.

 

Had to remove the eeprom bios and go back to the stock prom bios.

 

Kingston 32gb SDHC Class 10 (2 of them) won't read.

 

Sandisk MicroSD SDHC Class 10 card with adapter (hipstreet, not that it means anything) does work kinda, 

2600 games run just fine but 7800 games do not.

 

I got Dig Dug 7800 to boot once but not again since. Every other 7800

binary will not run. Double Dragon crashes at 90 percent on the loading screen.

Commando fully loads but crashes before starting. Ikari, MIA, and others see the

proper binary type for the game size and extras but all crash before starting the

game.

 

I updated to version .93 via the sd card to get 2600 games to run. Updated without issue.

I have not updated to .94 as of yet as I didn't read anything about that being of any help

but would update to that version if suggested.

 

All retail and homebrew games play normally on this machine as does the 2600 UNO SD Cartridge.

 

 

[Shawn]

I changed out all the brains and it now works like a charm. I will go back one chip at a time to see if there is a direct culprit to this. I'm thinking it's could be a ram issue. The stock model didn't have the resistor from pin 21 of the bottom ram chip to ground as seen on later model 7800's so when transplanting the ram I I also add the resistor.  Every game listed above that I tried and failed with now all work perfectly. Will report back and I narrow down the previous issue to be able to share with everyone in case they can make use of the info.

 

EDIT: The ram was a brand change as well, from the Hitashi shown in the post above to Sony with a resistor, FWIW. I'll snap a pic if anyone wants but I'm pretty sure you already know what I'm speaking to.

 

[mimo]

3 hours ago, Shawn said:

Had to remove the eeprom bios and go back to the stock prom bios.

Is it working with eeprom bios now? I have the PAL asteroids eprom in my NTSC machine and it works fine

[Shawn]

3 minutes ago, mimo said:

Is it working with eeprom bios now? I have the PAL asteroids eprom in my NTSC machine and it works fine

 

That will be one of the things I check as I go back over it. I'd like to go back to my dual BIOS with the NTSC PP2 & PAL Asteroids as it's just fun to have. Also nice to skip the splash screen with Asteroids BIOS.

 

[+batari]

1 hour ago, Shawn said:

I changed out all the brains and it now works like a charm. I will go back one chip at a time to see if there is a direct culprit to this. I'm thinking it's could be a ram issue. The stock model didn't have the resistor from pin 4 of the bottom ram chip to ground as seen on later model 7800's so when transplanting the ram I I also add the resistor.  Every game listed above that I tried and failed with now all work perfectly. Will report back and I narrow down the previous issue to be able to share with everyone in case they can make use of the info.

 

EDIT: The ram was a brand change as well, from the Hitashi shown in the post above to Sony with a resistor, FWIW. I'll snap a pic if anyone wants but I'm pretty sure you already know what I'm speaking to.

 

What pin does that resistor connect to on the chip?

 

Glad to hear the changes worked, as this is the sort of thing that could help a lot of us with compatibility. 

[Shawn]

8 minutes ago, batari said:

What pin does that resistor connect to on the chip?

 

Glad to hear the changes worked, as this is the sort of thing that could help a lot of us with compatibility. 

 

That was my hope in sharing these results. It's connected to pin 21 and it's just a 1k resistor.

When I have a chance I'll go back over it and see if this resistor "fix" works with the hitachi

ram as well.

 

[+batari]

The issues with some SD cards freezing at boot and the random menu freezes have been identified. They have the same cause: some SD cards don’t behave as expected when you ask them to release the bus. I verified this both with test code and with a scope. 

 

Forthcoming is a new version that works with these cards so that freezing at boot and the random menu freezes should go away. Trying to get this done before I ship more Concertos so that the cards people already have will work.

[Shawn]

Pin 21 is WE on these XX16 sram chips. Interesting that it's grounded with resistance but only on the bottom chip and not the top one. This is where I'll tap out and be happy you can do something with this info as with the 7800 specifically this is where I turn back cause the water is too deep and the talk is about to get outside of my pay grade.

[Swami]

I have one Atari branded POKEY chip from Best Electronics with original part number C012294 and two I got from somewhere with "Philippines" on them with part number C012294B. Any idea if they behave differently? I seem to recall having better luck with the Atari branded one, but it was a long time ago, so, I may have accidently used different 7800s with each. It could be something I check unless it is known they are functionally identical and a waste of time.

[Trebor]

55 minutes ago, Swami said:

I have one Atari branded POKEY chip from Best Electronics with original part number C012294 and two I got from somewhere with "Philippines" on them with part number C012294B. Any idea if they behave differently? I seem to recall having better luck with the Atari branded one, but it was a long time ago, so, I may have accidently used different 7800s with each. It could be something I check unless it is known they are functionally identical and a waste of time.

 

POKEY Versions
By part number:

-C012294 — Used in all Atari 8-bit family computers, including the Atari XEGS, as well as the Atari 5200 console. The suffix on the chip refers to the manufacturer:
    -C012294B-01 — AMI
    -C012294-03 — Signetics
    -C012294-19 — National Semiconductor
    -C012294-22 — OKI
    -C012294-31 — IMP

 

-137430-001 — Part number sometimes used in Atari arcade machines for POKEY.

 

-137324-1221 — Quad-Core POKEY used in Atari arcade machines Major Havoc, I, Robot, Firefox, and Return of the Jedi.

[mimo]

I've never noticed any difference in any of the POKEY behaviour in my 8bit computers. I don't think it's ever been discussed in the forums unlike SID chips which can give quite different sounds IIRC

[+DrVenkman]

4 hours ago, batari said:

The issues with some SD cards freezing at boot and the random menu freezes have been identified. They have the same cause: some SD cards don’t behave as expected when you ask them to release the bus.

I bet @flashjazzcat could have had some good insight here, as he’s been been wrestling with code to read SD cards for the A8 SIDE3 cartridge project for many months now. Unfortunately this is probably another case where one corner of the retro-universe isn’t really paying attention nor aware of what’s going on in another corner. Not surprising, of course, just unfortunate. 

[Shawn]

18 hours ago, batari said:

Keep the capacitor. In the more recent Concertos I do have a sizable capacitor between these lines right underneath the POKEY.

 

It is a good idea to add one as the POKEY would interfere with the data lines on older Concertos. Interesting that Atari found the same sort of thing back in the day.

 

So the newer concerto's don't need to add this cap cause you have one in the design now correct? If it is advised to add this cap, what flavor is it please?

[tep392]

5 hours ago, Shawn said:

Pin 21 is WE on these XX16 sram chips. Interesting that it's grounded with resistance but only on the bottom chip and not the top one. This is where I'll tap out and be happy you can do something with this info as with the 7800 specifically this is where I turn back cause the water is too deep and the talk is about to get outside of my pay grade.

It will be interesting to see if this is a factor.  I have two 7800's that work perfectly with my Concerto.  My expansion port model has the Sony RAM with 1k resistor.  My later model has OKI RAM and the 1k resistor is now on the motherboard.

[Shawn]

The Hitachi ram that I first posted is what was in the machine from the factory and it did not have the resistor "hot fix" nor was one installed on pcb. The Sony ram with the resistor fix was from a C025233-002 REV A board that I stripped that had many other issues that made it more work than it was worth to continue fixing again.

 

The full motherboard pic I posted is from a COA25234-001 from 1984. That is the one that had the Hitachi ram with no resistor. 

 

 

I have noticed that Missing In Action will not work right. It loads fine, title screen for the game is fine, picking a level is fine but as soon as you enter the level, no matter which one, it crashes hard. Tried it without the level skip to just start level 1 normally and it still does the same thing. I'm not super worried if MIA is the only title it's not gonna play cause as fun as MIA can be the game is pretty broken even at it's best do to it only being 85 percent complete.