Jump to content
Have You Played Atari Today?
2600|5200|7800|Lynx|Jaguar|Forums|Store  

Kroko

Members
 View Profile  See their activity

  • Content Count

    429

  • Joined

  • Last visited

Posts posted by Kroko

  1. Posted

    I don't know the game very well' date=' but It runs without problems on my PAL TV :-)[/quote']

    Now that it seems to be confirmed to work' date=' I could quite easily also convert the PAL version. If there is any demand.[/quote']

     

    DEMAND :D

     

    On the one hand, this is not the most colorfull game, but where

    colors are used, the NTSC version looks especially bad on PAL TVs if you ask me ....

  5. Posted

    Ok. When is the release date for Krokodile Cart for the atari 2600? Is Krokodile cart going to come out before christmas of 2004?

     

    Hi bluetriforce,

     

    I am still working on it. I need to make a second prototype and do a lot of testing before it can be released ... I have no idea when this is finished ...

    Depends on how many problems there are.

     

    Kroko

  6. Posted

    How about making the layout etc. public? Maybe after your initial run has sold out?

     

    Yes, I will do that. I have always planned this as an open source project.

     

    I will finish the development, sell a few devices to get back prototyping costs and publish the stuff afterwards.

     

    Maybe if there is enough demand we can order a bigger amount of chips and PCBs which would reduce costs :)

     

    Kroko

  7. Posted

    Hi ajaxlex !

     

    I am not planning this. The chips have to be programmed and the system has to be tested somehow. If you solder this thing together on your own, then nobody would know why it is not working (just in case ...)

     

    Do you have an idea, how a kit could be sold ? How could we solve this problem. Who would "repair" them , in case people are not able to build them properly ? I guess that would have to be me :?

     

    Kroko

  8. Posted

    Thanks again for the great feedback, Andrew :-)

     

    The power supply is needed for programming. With the current

    design, it is not possible to program it reliably with the VCS

    switched on. But it needs power from somewhere.

     

    I would have had to cut of the databus and the lower 7 address-lines from

    the VCS for reliable in system programming. That would have required a different CPLD version which is more expensive and doesn't exist in a PLCC package.

     

    For the next version, I will put a jumper on the board. I will test if the cart will run without external supply if it is connected to the VCS supply. There are two things that have to be tested: Maximum current drawn from the VCS and also the "startup time" of the microcontroller. I hope it will start fast enough. So maybe, there will be a version that runs without external power (at least if used as a standard cart which is not reprogrammed).

     

    But then you would have to remove it from the VCS everytime you reprogram it, because otherwise the cartridge powers the VCS during programming and that will not work.

     

    Well the decision is difficult, we will see .....

     

    The price will most likely be in the 100 Euro region. I still don't know, how much a label a box and a manual would cost if someone made them for me. And the design is not yet completed. This is why I cannot be more accurate at this time.

  9. Posted

    Hey thanks for the cool names ....don't stop thinking :D

     

    I will fix a few bugs in the commander software before I send out

    the second prototype. It needs to be a tester in the "NTSC" world

    because I need some NTSC feedback ...

     

    I don't accept preorders, because it is still a bit too early for that.

    The next prototype run will still be only 2 or 3 devices....

     

    I will keep you informed ...

     

    Kroko

  10. Posted

    I have ordered 3 prototype PCBs. I sent the first to Andrew.

    One I will keep myself to go on with the development.

     

    I will wait until I get the first feedback from Andrew how

    well it works. But if it works, then there is still one prototype

    I could send to whoever is willing to pay for an unfinished

    product and is willing to do some testing.

     

    It will NOT support RAM and it comes without label and

    manual. (a real prototype :-))

     

    I still need a cool name or you will be in great danger that

    it will be called Krokodile Cart ...

     

    Kroko.

  11. Posted

    Just a thought, but maybe you could have the addresses

    $3C and $3D select the bank for the first 2k, and $3E and $3F select the scheme for the second 2k, which is normally hardwired.  

     

    Ah... let me think ... you want to extend it so that both banks can be either RAM or ROM. Adam had mentioned something like this on the stella list, though it was a bit different. I will see if there is enough space in the CPLD to make it that complicated. But in general I like your idea :-)

  13. Posted

    Why the  symbol? I was only clarifying my reference.

    Ups ... Is this symbol negative ? I thought it looks if somebody is thinking about something. No offense :-)

     

    The 2600 is generally fairly tolerant of doing bad things to it

    Yes, thats also what I heard, also because everything is NMOS

     

    Hmm, that's not a good sign. ... Have you tried this in many 2600s?

    Yes, I have quite a few :-) But I will send a few prototypes to other people to test it a bit better. I really think I made some other mistake.

    However I will check this again.

    Yes, but are you going to supply simple fixed carts using the same format?

    Well I didn't think about that. But in theory, I could make a write protected device without serial transmission capabilities and fixed bs scheme. Wouldn't be much less expensive though .....

    I thought more in the direction of impressive effects than in homebrew cart releases ....

     

    The Cablecart  

    The Brewcart  

     

    Hey those 2 are cool ! Thank you

  14. Posted

    Ok, in this case I do not support SC :ponder:

     

    Yes, the device needs its own power supply at the moment. It will not run without it. Do you have an idea how to change this ? I have never calculated how much power it would consume .... maybe the VCS supply is sufficient. How much current could I draw from the VCS ?

     

    11 Mhz is the clock for the CPLD and the 11 MHz is because the AVR UART is only accurate enough for 115KBPS serial transmission @11.0592 MHz. I tried also 7.something MHz, but that was not the kind of resolution I was aiming for in the CPLD timing. 11 seemed to be perfect to exactly time the validity of the data on the databus for best and worst case conditions of the 6507 bus specification.

     

    The BS scheme is stored in the controllers EEPROM. So the device remembers what was programmed. It does not have to be connected to a computer to run. But it is intended to, because people should download their brand new stella programs every 15 minutes to check how they look like on the real thing ;-)

     

    I need I/O Lines to program the Flash Perom. I made a compromise between CPLD resources and Controller resources. The 19 Adresslines have to be controlled during programming. The CPLD will only drive A7 upward (limited space ....:-(). A0 to A7 have to be controlled by the µC during flash programming. That are 8 Lines. Another 8 lines for the databus interfacing. (I didnt want to latch the data - not enough space on the CPLD.....)

     

    Another 4 Lines for the BS scheme transmission (one time transmisson to a CPLD register was not stable enough..for some strange reasons I have never found out). And a few control lines for Address latching ......2 lines for RxD TxD ... well infact they are almost all used.

     

    During normal operation, everything is tri-stated, when the cart is not selected. During programming, there are TTL levels on the databus and addresslines A0 to A6.

     

    Well I will listen to your advise and recommend to unplug the device from the VCS before it is programmed, although i never had problems with this. Have you made different experience with your devices ?

     

    Yes, the RAM is too big, because I am too stupid to figure out what I am doing wrong. Currently I need a CS2 input on the SRAM. The /OE /CE driving of the RAM simply doesn't work and I am too lazy to find out why. I want to finish this now and don't change the design details unless there is a real big problem or cost advantage. Maybe the chips are defective or its a timing issue - I don't know. I have developed the device with a 128K SRAM and now I will keep it. But you are right I could save 2 Euros with a smaller chip. Hehe ... but that will give us the possibility to use 128K of RAM for VCS PROGRAMMING !!!

     

    The real expensive parts are the CPLD, the µC and of course the PCB.

     

    The MAX232 + caps is less than 1 Euro at my supplier. But I will most defintely have a look at this MC1489. I was searching for a cheap device without caps but actually didn't find any. Thank you :-)

     

    Can you give me a hint how you have started to program FPGAs. Did you buy some Evaluation Board ? And how do you handle these SMD parts. I think they are a pain. I am very impressed by your CC2. I would be so happy if I could learn to do something like this, before I am 100 years old ...

     

    Kroko

  15. Posted

    Hey Tim,

     

    It looks like UK WOMBLE is going to send me enough cart cases

    for the next step of the development. It would of course have to be betatested by a few people before it can be released.

     

    Well, if people really like it, then I might have to buy some new cart cases which would cost about 5$ each. I guess people would like it more to pay

    3 additional $$ and get a complete device .....

     

    >As for a name how about "Sweet Mother of Christ the Cart I've been >waiting for."

     

    Hehe, thanks ... but that won't fit on the label ..... :|

     

    -Kroko

  17. Posted

    USB needs a dedicated USB chip which is more expensive and more complicated. In addition an eprom is needed to store device info. I am not aware of any non-smd parts, and I am not going to use SMDs. The driver stuff wouldn't be the problem because they are shipped with the chips. I thought about it, but decided to make it standard serial.

     

    SC is superchip. I am not aware that there are any license issues with that. I can not support arcadia because the CPLD is too small. I used the biggest non-smd version which is 108 macrocells. As far as I know, the license was necessary because of the audio transmission and because Chad used their code. Superchip is very easy to implement.

     

    I have never heard of "Bankzilla", what is that ?

     

    The power supply is external. There is no connection to the VCS 5V supply. Only ground is connected. You can download files even if the VCS

    is switched on. But the risk of transmission errors is higher and I recommend to switch the VCS power off, when a new ROM is transmitted.

    You will have to do that anyway, to start the game :-) The device has ISP headers, so the firmware can be upgraded via ISP programming, but not via serial download. The DC power has to be connected any time, its the only power source available. But a standard 6V dc supply is all you need.

     

    The serial port runs at 115kbps. "FuKung 512K demo" needs 1 minute for transmission and programming. An average 8K ROM-File about 1 second :-)

     

    Cost reducing is hard. I would if i had any more ideas. Maybe you can help, Chad. I need a µC which I am able to programm in assembler (AVR). It needs an UART, lots of IO-Lines at has to run @11MHz. I chose the cheapest part available here (non-smd). The SRAM is the only SRAM i can handle. I tried other parts but they didn't work as I wanted so I will use the 128K chip. The CPLD is almost too small, so i can't choose a smaller one. Serial transmission is done by the max232. Very cheap solution, can't make it cheaper ..... Do you have any ideas to reduce costs ? I would be very interested.

     

    E7 is not used by a lot of games. I could support it, but space on the CPLD is limited. I like the idea, to support a BS scheme with 128K of RAM available for the programmer. If I will not support the RAM then most likely I will support E7.

     

    Serial transmission from the VCS cart to the PC ? Sounds very cool ... I will think about that :-) What would you want to do with that feature ?

  20. Posted

    Unfortunately I have no idea how to support USB. If you really need USB, there are a few serial to USB converters out there ....

     

    And please don't compare it to the CC its simply very different from both CC1 and CC2. Its more or less a development cart for the programmers.

     

    Of course you can also use it just for fun :D

  21. Posted

    This is the first version of my flash-cartridge prototype:

     

    flashproto1_0.jpg

     

    Is supports 1K,2K and 4K, F8, F8SC,F6, F6SC, F4, F4SC, FA, E0 and finally 3F up to 512K

     

    The rom images are downloaded from the PC with a serial cable.

     

    Before I can realease this device, I would like to get a bit of feedback on a few subjects:

     

    :arrow: I need a name for this device :-) Any suggestions are welcome ....

     

    :arrow: I don't have a cheap source for empty atari cartridge cases. Is anybody able to help me with getting cheap empty cases ? I thought about one million copies of ET...

    or so :D

     

    :arrow: I am a terrible artist (most engineers are ...hehe). Would you like to help me to design a nice label ? What is the label material and who can print them ?

     

    :arrow: There is a bit of empty space on the CPLD. I have also working VHDL Code for CV,FE,F0 and E7 (Rom only) bankswitching. But I don't want to actually program it into the CPLD. I would like to use the empty space a bit differently: We have a 128K SRAM chip. I would like this RAM to be available for programming. In theory I could enhance the 3F bankswitching to support RAM. We could call it 3F+. 128K/256 that would be 256 blocks of RAM each with a size of 512Bytes which would be mapped into the first 1K of the cartridges address space. (512 bytes write port and 512 bytes read port). A write to 3F selects a 2K Rom slice into the first 2K of the cartridge. A write to 3E could select a 512Byte RAM slice into the first 1K of the cartridge. So the first 1K could be either ROM or RAM, dependent on what address you were last writing to (3E or 3F). What do you programmers think about this. Would this RAM-Slice size be good, or should I make it smaller, lets say 256 slices of 256Byte .....

     

    :arrow: And finally, how much money would you be willing to pay for this ? I have to compare this number with the material price and decide if it make sense to continue. At the moment I am not sure what to do. Either build a few units myself or releasing the whole thing open source and somebody else can build a larger number of them.

     

    Kroko 8)

  23. Posted

    Thanks for your answers.

     

    I think he has licensed it. But what I actually try to find out is,

    if I would need to buy a license if I wanted to support the AR

    bankswitching in my flashcart. I would not want to use any

    existing code or design, just the specs.

     

    I guess not, but i want to make sure before I start looking

    at AR bankswitching because it seems to be very complex

    and i don't want to spend too much time if i am not allowed

    to use it for free.

  24. Posted

    I remember that there was some kind of licence problem with the

    cuttle cart.

     

    Was this licence only necessary for the data transmission part, or

    is it also necessary to get a licence, if just the AR bankswitching scheme is used, but not the data transmission.

AtariAge
Forums
Store
General
Follow AtariAge

Copyright ©1998-2021 AtariAge

×
×
  • Create New...