Anonymous
User login
Please support the defense of Ukraine.
Direct or via Unclutter App
Active forum topics
Recent content
Navigation
No Ads.
No Trackers.
No Social Media.
All Content Locally Hosted.
Built on Free Software.
We have complied with zero government requests for information.
You don't need Linux to make it work. Just use small, simple and cheap SDcards of up to 32GB. The SDcard standard specifies the type of factory formatting. All MicroSD cards up to 32GB (SDHC) come with FAT32 formatting from the factory. No formatting required, they will work out of the box:
https://www.sdcard.org/developers/sd-standard-overview/capacity-sd-sdhc-sdxc-sduc/
I recommend the slightly better "class 10" or "UHS 1" speedgrade. Class 2, 4 or 6 also work, but result in longer loading times.
The problems only start when you try to use newer, larger, more expensive cards - SDXC or better, above 32GB. These come factory formatted with the newer "ExFAT" file system, which the DAN][ Controller doesn't support. You can try to reformat these with FAT32, but FAT32 only supports up to 32GB, so Windows is trying to be smart and split these into several partitions. I have never tested this, but can see that this might cause issues.
Cheap & simple is better in this case. The maximum usable size is per card is 4GB, so anything much larger isn't really helping anyway.
FAT32 supports at least 2 terabytes. The problem is that the Windows UI doesn't let you format anything larger than 32 GB with FAT32. However once the SD card is formatted, it is more than happy to read or write to it no matter the size. The reason so many people mistakenly believe that FAT32 is limited to 32 GB is this formatting idiocracy.
However there are other utils that let you format. You can use DISKPART from the command prompt or you can use third party free tools with UI, like this one: https://fat32format-gui.en.lo4d.com/windows
I have tried the Dan ][ Controller with these two 256 GB FAT32 formatted SD cards and it worked with both of them:
MicroSD cards2.png
If my iiplus only had 48k, I would add one of the modern Saturn compatible cards. Lower power. Low cost.
https://www.tindie.com/products/garrettswrkshp/ram2gs-ii-gw4201d-8mb-ram-for-apple-iigs/
Concerning power-consumption, the DAN][ Controller can easily compete with most similar devices. The ATMEGA is very efficient. A DAN][ Card without the Ethernet module and one SD card consumes about 30mA. It might rise slightly when you write to the SD card, since SD cards consume more during writing, but generally power-consumption is negligible. It would actually be the same or less than the device linked above. An DAN ][ doesn't need 100mA.
In another topic I measured the power consumption of various cards. The Dan ][ Controller had the lowest consumption of all of them.
Here is the +5V rail current consumption for most of the cards in my collection:
Apple IIe PAL version with all slots empty: 1.08 A
---------------------------------------------------------
CFFA3000 Card + MicroSD Adapter with an SD card + USB flash drive: 500 mA
Apple Mouse Interface Card + Mouse: 280 mA
Apple Disk ][ Interface Card + Floppy: 260 mA (doesn't matter if the floppy is reading/writing or not)
Z80 Card (clone): 250 mA
Taxan RGB II Card: 250 mA
ESP32 SoftCard: 190 mA
Printer Interface Card: 150 mA
Apple 64K AUX memory/80 col card: 120 mA
ProDOS ROM-Drive Card: 50 mA
MultiROM card: 30 mA
Dan ][ Controller: 10 mA
Here is the +12V rail consumtion of the Apple Disk ][ Interface Card:
Apple Disk ][ Interface Card with the floppy at rest: 0 mA
Apple Disk ][ Interface Card while the floppy is loading: 680 mA
I was suggesting a Saturn type memory card, multiple banks of 16k, over a language card with a single bank of 16k . This is a lower current draw than the old language cards.
I plan to put a Dan][ into my iiplus. It is great device.
I continue my learning expedition,
the card is working with a 512 Byte eprom, and It is fine with a AT28C256, but it is not working with a AT28C64 (double bip at startup)
is it due to the address range ? can I do someting with the eprom firmware ?
Thanks
Vincent
The Atmel 28c64 works fine. I used several of these with the DAN][. Make sure to program the 512byte firmware into the lowest address range.
Which type of 82c55 did you use, which resistor values did you use for the address line pull-downs and which jumpers did you set? Post a photo of your card showing the top, so we can see the components. That's the quickest method to get help...
I use 10k Pull resistor,
IMG_5126 Large.jpeg
Ok, remove the "PROGRAM" jumper (JP6) for normal operation. I haven't used any of the Renesas 82c55s, but a while ago we already noticed with another brand, that the 10K pull-ups are too weak (also according to the datasheet). Some 82c55 types require stronger pull-down resistors. 4.7 Kohm works reliably and matches the numbers in the 82c55's datasheet. It's enough if you replaced R2, which is the resistor for jumper JP2/BNK (which must remain set). You can also test this by temporarily soldering a second 10K resistor in parallel (e.g. on the bottom side), to see if it makes any difference.
Otherwise I do not see any reason why this AT28c64 shouldn't work (if programmed correctly). I use the same on mine.
Happy new year 2024,
I have done some further testing,
Here is my results :
- 1/ 82C55 with AT28C256 : I see DANII press return, I press return and I access to the list of PO
- 2/ 82C55 with AT28C64 : I see DANII press return, I press return, it blink and I see DANII press return with some time ??? and wired stuff
- 3/ Adding a 10k on R2 : no changes same as 2/
- 4/ Changing to another ref of 8255 with AT28C64-> same as 2/
- 5/ Changing to another ref of 82C55 with AT28C64 and adding 10k same as 2/
nothing as to be changed on ATMEGA with 64 instead of 256 ? Do i have to change something with the programming of the eprom ?
this is really strange ;)
Vincent
No, the ATMEGA doesn't care about the EPROM size. Neither does the Apple II: only 512bytes are used. You could use any ROM/EPROM/EEPROM with at least 512byte of memory - as long as it is properly programmed (and working).
Make a dump of the slot ROM area, as it is seen by the 6502. Assuming you have the DAN][ Controller installed in slot 7:
CALL -151C700.C7B7C0F3: FAC0F3: 01C700.C7B7What do you see? You can compare the dumps with the hex file for the EPROM (or post two photos here, showing the dump of the two pages...)
Happy 2024! :)
Ok this is the original eprom.bin file
IMG_5133 Large.jpeg
Now this is the first page
IMG_5132 Large.jpeg
I have checked for C700 to C7FF and it seems to be the same,
When I enter :
C0F3: Fa [enter]
C0F3: 01 [enter]
and then C700.C7B7 I have exactly the same result as page 1 ...
How do I switch to page 2 ? I must miss something
After Rechecking:
I found one error from the dump of the rom :
IMG_5134.JPG
It should be a B1 instead of B0
I have double check what is read from the TL866 twice and I have a B1 in the ROM chip, and in the file it is as well a B1
Why is the Apple II having a B0 instead of B1 ?
it is maybe not the only error...
The other card with the AT28C256 is switching to page 2 when issuing :
C0F3:Fa
C0F3:01
You initialize the port controller (once is enough):
C0F3: Fa [enter]
Then you can switch between the two pages:
The content of $C700..$C7ff will differ between the two pages. The first few bytes are similar though.
If that's not working then compare with the other ROM. And you'll need to investigate a hardware issue with your card (maybe an issue with a solder joint or a briged connection).
Ok i will investigate hardware issue, the card with the AT28C64 is never swithcing to page 2
Back,
I have checked and maybe solved soldering issue.
Now I have the same behaviour with the 2 card:
CARD A & CARD B are working with AT28C256 and not working with ATC28C64
When using AT28C64, I have a second bip with the line on screen :
IMG_5135 Large.jpeg
Should I redo a check on the ROM content from the Apple perspective ?
You need to verify that the Apple is able to see both pages (correct content, page switching works). The only difference between the AT28C256 and the AT28C64 is that the C256 has two additional address line, which are unused ("NC") on the other device. So there is no good reason why one should work, but no the other.
One thing you could do is use a multimeter to check the voltage of the page selection bit. You can measure this at the resistor we discussed earlier (10k vs 4k7). When you switch between the two pages, you should see this line toggling between 0V (page 1) and some value above 3V (page 1).
Of course there is always the possibility that your AT28C64 chip is defective or marginal. If I'm seeing this correctly, the date code of the AT28C64 suggests it's an older chip, not a new one. So maybe something had happened to it.
It is this kind of thing that can make you crazy !
I have 10 x AT28C64 and all does not work ! In that case you could assume the problem comes from elsewhere !
But, I decided to order a bunch of IC on Mouser and I order new AT28C64 ! and IT WORKS !
All the 10x AT28C64 are not working like expected ! (forgot to mention that it is coming from Aliexpress).
My key takeaway -> NEVER ORDER CHIP ON ALIEXPRESS !!!
Thanks for the very good support ! I feel a bit stupid btw ;)
Vincent
Well, not quite. You simply need to take under consideration that whatever you order could be fake or taken from scrap and damaged in the process. However there are many repro chips that are not available anywhere else, except from scrap. The Apple IIe keyboard encoder is one example. This situation is a problem for mission critical electronics, but for us it's a gold mine, since very often we can get some really valuable retro chips for a buck or two.
Also never order quantities above 2 or 3. I have seen so many complaints on YouTube where someone would order one retro chip and it would be good, but when they return for a quantity order from the same seller they would get a bunch of fakes.
And finally always ask for a full refund when the chip doesn't work. The sellers are fully aware of what they are selling and they usually approve it right away.
I went back to this project after a few months of doing other work and now I have the external card working. What did I find out? The connector headers on the 2 boards (DAN ][ and the external card) do not match. Can't just use a 10 pin ribbon cable to connect. It is flipped on the external SD card. But now that I have that working, I have another little issue. The network card does not want to work when connected to the external SD card board. When I have it connected, it wont even display the ProDos volumes on the SD card in slot 1 or 2 on the external board. Will keep truing some things and will also do a firmware upgrade to see if this solves anything. Anyone have ideas? I now have 5 cards for my own use in all my Apple II's (2 IIgs, IIe and II+) and one ready for Apple III when I get the rom board.
I tried to do the flash of new firmware 3.4.1 using a IIgs and I kept getting a No Sync error message to try again. I went to a IIe and it worked fine. Is there some compatibility issues with the IIgs to flash firmware? I used the same DAN card and SD card on both machines. Thanks in advance for the support.
I know a thing or two about Apple IIs, but admittedly not much about IIgs. I have barely every seen one.
The firmware update on the Apple II relies on the fact that a Ctrl-Reset triggers a RESET of the 6502 _and_ at the very same time, also of the I/O cards. It expects the Apple II to execute the "warmstart hook", before the ATMEGA on the DAN][ card has left its bootloader. The bootloader is waiting for a fraction of a second, to see if there was a flash request, before starting the normal application.
I have no idea how a IIgs starts. Does it start in 6502 mode, so timing is more or less identical to a normal 6502? Or does it start execution in some IIgs-specific 16bit environment, and it will only switch back to 8bit/6502 mode after a longer delay due to some IIgs-specific init sequence?
I suspect the latter is the case - since, when it was starting up in 6502/8bit mode, it probably couldn't even switch back to 16bit. So it probably has to start in some IIgs specific way, which means the timing is going to be very different on the IIgs. That could explain why the update utility wasn't working on the IIgs.
Hi all, ive been having stability issues since upgrading from 3.4 to 3.4.1, it seemed to work well initially and solved the mouse text problem that I discussed with macfly previously however over last few weeks of heavy use (filling in the pages with images) ive noticed its hanging , the slot led both flashing and just shows DAN II or sometimes it will only load a few from the menu and just stops with the sd card led light on? Ive had to reflash the atmega back to 3.4 and it’s working perfectly (except for the mouse text issue) .
Ive built 2 cards and both have same issue with 3.4.1? has anyone else experienced issues with 3.4.1?
My setup is Apple iie enhanced, 64k/80col card and danII in slot 7 no other cards.
Thanks again :-)
ok just a quick update as this was confusing me, looks like it could be related to the recent a2vga card acquisition? after going back to ver3.4 and testing everything working i reinstalled the a2 vga card (slot 4) and after a couple games/reboots the issue is back? its corrupted the atmega on the dan2 card so had to reflash it and works fine (via composite) will keep using it without vga card for now ..
will have to do some more testing as the vga card is a new toy ;-)
Exactly which type of A2VGA card are you using?
Yeah, it would have been strange if the difference between the 3.4 and 3.4.1 firmware images had caused the issue. The only difference between these is a single instruction to reset the ALTCHAR (aka mousetext) soft-switch.
However, I recently asked in another thread on this forum, if it was known that there were IOUs which did not reset all the soft-switches during a hard reset. That thread suggested this was not the case. Since, however, it was clearly the case with your machine, when you reported the issue, this could mean:
a) something might be wrong with your machine. For example, a broken trace - like RESET no longer connected to the IOU chip.
b) a broken IOU chip (so it does see the RESET, but somehow no longer clears the soft-switch).
c) nothing is wrong with your machine - but the forum was wrong - and there were IOU variants after all, which did not clear all soft-switches.
I just received a batch of Dan ][ PCBs... Can someone point me to a link to the latest firmware version?
I did a git pull on the repository at:
https://github.com/profdc9/Apple2Card
Is the file firmware/firmware.bin the most up to date to write to the 28C256?
Depends which version you want. This is Dan's original version.
I had made a few additions which can be found here https://github.com/ThorstenBr/Apple2Card . There is also a ZIP file under "Releases" with all the binaries. In my repo I have renamed the file for the EPROM to "eprom.bin" (to avoid confusion between the Arduino firmware vs EPROM binary).
Has anyone else run into issues with some brands of 82C55 not working in a Dan ][ card? I have run into Samsung and Harris chips which cause the computer to hang with garbage on the screen. The same card with an Oki chip in it works.
Yes, I believe there were issues with some CMOS chips, mentioned on page 5 and page 6 of this thread: https://www.applefritter.com/comment/102461#comment-102461
Thanks for that! Which 10K resistors should be swapped out for 4.7K? R21 & R23 or R18, R7, R3, R2, R22? Looking at the placement diagram I've been using to assemble cards it looks like those were previously 4.7K?
I am also trying to understand what the hell were we talking about! Maybe they were 10K at that time, but then they were changed to 4.7K after?
This image posted by McFly on page 5 of this thread:
https://www.applefritter.com/files/styles/95-percent/public/2023/02/09/Apple2Card_Placement2.png?itok=r2iKvNMu
I just checked my cards - they are 10K. But then I looked at the KiCad schematic I have on my machine and I see them as 4.7K:
Rev1.1.1.png
So I guess try replacing them with 4.7K and see if this helps.
Yes, there was a bit of confusion with the resistors. Dan had originally used 4K7 pull-down resistors, but replaced them with 10k resistors in his schematics, shortly before his initial release. The KiCad PCB design, however, was never updated, and kept using 4k7 pull-downs.
Then, much later, we noticed that the 10K resistors won't work with some 82c55. Indeed, the 82c55 datasheet shows that stronger pull-down resistors are required. So, the initial choice of 4k7 resistors is actually better/correct. According to the datasheet, this should work with any 82c55.
If you already have soldered boards then it's enough the swap R2 to 4k7. R2 is the pull-down for JP2 (the "bank " jumper), which is the only jumper which is closed (so connected to the 82c55 and needs to be strong enough). If you make new boards, then it's better to use 4K7 for all pull-downs (R2, R3, R7, R18, R21, R22, R23).
I have updated the schematics+PCB files in my github accordingly.
Also, a while ago I generated an "interactive BOM" using a pretty nifty plugin for KiCad - that can also be helpful:
https://htmlpreview.github.io/?https://github.com/ThorstenBr/Apple2Card/blob/main/Apple2Card/bom/ibom.html
I trief changing just R2 out and it still doesn't work with either a Harris or a Samsung labeled 82C55. Maybe these chips are fake but I could have sworn that they worked in some other card, maybe the Apple II Arduino card. That one has a completely different arrangement of resistors on it though.
danii4.7k.jpeg
You could try adding a second 4k7 to R2 in parallel (on the bottom side) as a test. The value of R2 may be as low as 2K without exceeding the 82c55's current capability. If that also doesn't help then it's cleary NOT related to the pull-down resistor - but an issue with something else.
(I have genuine Harris 82c55s, an entire tube from Digikey. They are date coded "99" instead of "97", but they indeed look very different: label printed in brownish color, the Harris "H" looks different, the label has three much longer lines of numbers, and "MALAYSIA", the country of manufacturing. So, quite different look to yours. They are from a different batch, they might have changed the label and logo, of course - but it could also indicate your suspicion might be right...)
I will try another 4.7k resistor and see if that changes anything. I'm currently working as a contractor at a major semiconductor firm, so I can tell you labeling chages a lot. One of the software systems I work on is for tracking labeling. However, I also know there are a ton of fake chips out there and I don't remember where I bought this one from. I know I bought at least one 82C55 from a reputable place like DigiKey or Mouser, I just don't remember which one, whether it was one of the Samsung that also don't seem to work or this Harris one. I know it wasn't the Oki chips, as those all seem to work and are all labeled basically the same.
Well, I just tried adding a 2nd 4.7k on the back... and what do ou know... It works.
So I am thinking maybe replace the 4.7k with a 2.2?
OK, 2.2K seems to do the trick, and the Samsung parts work with that too.
I'm down to one stubborn card that still doesn't work. I did make the R2 -> 2.2K mod because the last 82C55 I have on hand is a Samsung. That got it back to almost working... The 328P , 28C64B, 74HCT32, 74HCT00 and the 82C55 have all been tested in another card and are working... Oh, the Micro SD card is known good too.
Anyway, from a cold poer on it comes up with the "press return" prompt and then locks up. Sometimes it dumps to monitor at an addrsss around $904. Pressing ctrl-apple-reset gets the prompt again, and I can try to select the images, then it just sits there. The light at D9 LED blinks but nothing ever happens. Occasionally it will crash to the monitor again.
I've tried swapping out the chips. I've checked the resistors with a meter and they all seem to be correct. I've tried re-flowing all the solder joints... But no bueno...
Anyone gor any ideas? I'm thinking maybe I need to test out my new hot air rework station and take off the SD card slots and try again. This was the very first Dan ][ card I ever tried to build and I had a lot of "fun" soldering those. I've had a LOT of practice since then. I've built 16 working Dan ][ cards and 4 "Murmulators" which also each have a Micro SD socket.
I think the Atmega is running since the D9 LED occasionally flashes at times I'd expect it to. I think the 82C55 and ROM are all working and everything ahead of that is talking like it should...
I'm stumped on this one.
It's probably an open or bridge on one of the SD card pins. Reflow the solder by covering the pads with flux and then touching the iron to the pads to remelt them and use surface tension of the solder covered by the flux to adhere the solder to the pads and prevent the solder from bridging between pins.
Make sure JP5 is closed.
I don't have JP5 closed on any of my other cards. I put a jumper on JP5 to close it and it doesn't seem to make any difference.
Another thing I tried was hooking up a full size SD card slot to J6 SPI and it also doesn't seem to make any difference. Earlier I had removed the Micro SD socket I had originally installed and soldered in a new one. That didn't seem to matter either. I'm now doubting the socket the 328P is in, sometimes it seems a little loose. I'm going to check it with a meter for continuity on all the pins tomorrow. I might just cut it out and solder in a new one.
I know it works with JP5 open as well, but it’s better to keep it closed, unless you are programming the ATMEGA through the ICSP port while it is plugged in to the Appe II. This jumper shorts D7, which is there to prevent your little ICSP programmer from trying to power the entire Apple II and burning up. Having the jumper open causes a voltage drop on the Vcc of the ATMEGA because of D7 and it might cause issues.
Well, that is good to know. I will install a jumper on the JP5 on all of my cards then.
So now for a different Dan ][ related topic...
Software/disk images. I've been using the same 9 32MB ProDOS volumes for every Micro SD card I've made up.
BLKDEV01.PO - Total Replay v5.0-alpha.4.po
BLKDEV02.PO - Total Replay II Instant Replay.po
BLKDEV03.PO - games.po
BLKDEV04.PO - h32mb.po
BLKDEV05.PO - o32mb.po
BLKDEV06.PO - cffa3k.po
BLKDEV07.PO - booti.po
BLKDEV08.PO - floppy-emu.po
BLKDEV09.PO - smartport.po
That's good and all, but there is a ton of duplication on those, especially 6-9 which look like the only major difference is the splash screen.
What I'd love to have is a version of those with a Dan ][ splash screen. What I would really also love to put together would be all of the Dan ][ related utilities which are currently on separate .po images onto one image. I realize that maybe not all of them make sense to have mouned on the Dan ][ becase some of the utilities may not be able to be run from the Dan ][ if they are programming the Dan ][. But all the other stuff like the Internet related things and stuff would be cool to have there.
Coming up with a selection of disk images which eliminates most of the duplication would leave a lot of room for putting together other collections of ProDOS compatible software that might be useful. And also make up some empty 32M ProDOS volumes for people to store their working files on. I'm assuming right now that most people will use a Micro SD in slot 2 which works in the raw mode.
I don't know if anyone is familiar with how those splash screens are made. I guess I need to look into the disk images and see if I can figure it out.
OK, so looking at those 32MB ProDOS voumes it looks like there is a file called GAMES.SYSTEM or H32MB.SYSTEM or whatever that actually draws the splash screen and then displays a message. One of the disk images says something about being assembled with Merlin. I wonder where to find the source code for that. The web site mentioned www.apple-2.com goes to a parking space.
Thinking about what it does, it really wouldn't be all that hard to write someting new from scratch.
Well, I wrote an assembler program that draws a low res screen and waits for a keypress similar to the .SYSTEM files on those 32MB ProDOS images. That was pretty easy... but even back in the day I was more of a DOS 3.3 guy, so I'm not sure how to change the file type to FF as I gather they need to be I copied the BIN file onto one of the ProDOS volumes and I can run it OK.
JP5 has another purpose as well.
The output digital voltage levels from some SD cards may be marginal because the SD cards are powered at 3.3-3.6 V. By dropping the voltage on the ATMEGA328P (to about 4.2 V), it lowers the high threshold for the ATMEGA328P, but the output of the ATMEGA328P is still sufficient for low/hi voltages to the 82C55A (which typically use NMOS levels anyways).
So if one was having a problem reading/writing a SD card, you could remove the jumper and see if it works.
Good point! I guess I missed that one.
Pages