Would you like to react to this message? Create an account in a few clicks or log in to continue.

You are not connected. Please login or register

Rom-component question

2 posters

Go down  Message [Page 1 of 1]

1Rom-component question Empty Rom-component question Tue Feb 23, 2021 2:46 pm

Mistral



Hi,
I am trying to replicate the Gigatron
https://gigatron.io/
schematic page 50
https://www.budgetronics.eu/data/mediablocks/Gigatron-manual.pdf
and came up to this point

Rom-component question Romv1_10

This eprom needs some data of course and I looked here :
https://github.com/kervinck/gigatron-rom/blob/master/ROMv1.rom
I downloaded it because it doesn't seem to show its raw contents on Github. Opened it with Notepad++ but only saw yibberish.

Rom-component question Romv1_11

In the assumption that the Simulide rom package can only handle hex-numbers I spent the next two hours looking for ways to open this rom-file and/or if a hex-file from this project already existed somewhere. Didnt want to install unknown programs and didn't find a solution. In the end, in Simulide I clicked on the eprom, clicked load data, navigated to the rom-file and clicked open. Surprise, it opened and at least gave response to the input address lines. Rightclicked again, clicked save data and a data-file was generated which I could see with notepad


Rom-component question 2romv110

My question, is this the exact representation in decimals of the original rom-file and so will it work the same? In other words, is the Simulide data-file exactly the same as the Github rom-file? Or are these just random numbers in the data-file caused by the file 'transformation' ? 

Sorry about the extensive post, but can't describe it much shorter. Also hope that other users can get some usefull information from this or maybe even better can answer this thing I don't understand.

2Rom-component question Empty Re: Rom-component question Tue Feb 23, 2021 7:02 pm

Fizik_S

Fizik_S

The original ROM content and the text representation in SimulIDE will be the same. In text form, the numbers are written with a sign (the highest bit of a binary number in ROM is " 1")

3Rom-component question Empty Re: Rom-component question Tue Feb 23, 2021 8:08 pm

Mistral



Thank you Fizik_S, that's good news. I don't expect to be fully able to replicate this because, among others, for the for me rather complicated output signals but that's oke since along the way I am learning more, even if the endresult is not reached. With your explanation I am now more confident to go to the next step. Want to draw the schematics page by page and combine them (as much as possible) at the end. Opened a new github repository for it and named it Tron TTL and it contains one simu-file at the moment. 

https://github.com/Alectus/Tron-TTL

4Rom-component question Empty Re: Rom-component question Tue Feb 23, 2021 8:52 pm

Fizik_S

Fizik_S

I finished the circuit of a 4-bit Nibbler computer ( https://www.bigmessowires.com/nibbler/ ). It worked in Simulide, but not completely. Not all programs run, and there is almost no reaction to pressing the buttons. I don't know how long it will take to fully debug. In this computer, the author took into account the features of some chips that I do not even know how to implement in the circuit. I don't want to redo the chip models... The screen saver of the program appears on the indicator and the program polls the buttons indefinitely and does not respond to their pressing. We'll have to debug it step by step.

Rom-component question Nibble10

5Rom-component question Empty Re: Rom-component question Wed Feb 24, 2021 2:44 pm

Mistral



It looks very nice and I will certainly give it a try and who knows maybe I find something and will let you know.
There is still something I don't understand with the eprom from the openings post. As described I have two files now, a rom-file and a data-file that was generated by simulide from that rom-file. Both work the same. I didn't test all addresses of course but a few. Like these first 6 in the eprom. Oh, I should mention Vuikie at tweakers told me how to use hex plugins in Notepad so I can read the rom file now. This is my comparison between rom and data file

rom-file * rom dump | data file * data dump   | output eprom in simulide

00 00               ....€ | 20 20        .0               | 0000 0000 0000 0000

18 80                .À.. | 20 30         -32744      | 1000 0000 0001 1000

18 C0               Ö..ÿ | 2c 2d        -16360       | 1100 0000 0001 1000



Programmer calculator shows:

-32744 = 8018h = 1000 0000 0001 1000

-16360 = co18h = 1100 0000 0001 1000

Because of big endian this eprom output fits precisely with the rom-file and with the decimal representation in the ram-file.

However, I can't figure out where the 20 20 - 20 30 - 2c 2d in the data file come from and what their relation is with the rest?

And what kind of characters is the rom file dumping anyway, where do these come from, how are they generated? Believe me, I searched myself for quite some bit but probably with the wrong search words.

Ps, @Fizik_S,
Would you like to share your 4 bit cpu simu-file or you rather develop it some more?

6Rom-component question Empty Re: Rom-component question Wed Feb 24, 2021 5:39 pm

Fizik_S

Fizik_S

The Gigatron computer does not have a processor. Its functions are performed by the AT27C1024 ROM along with a simple ALU. The ROM stores micro-commands that control the operation of the ALU, registers, and RAM. I.e., the ROM is the control device of this computer. I tried to change the combination scheme on the ROM using SimulIDE and I succeeded. Many radio amateurs replaced the hexadecimal decoder for the seven-segment indicator with a ROM chip with the corresponding firmware. It turned out much easier than using standard chips. So is Gigatron. The contents of the ROM are described in detail on the forums and on the official website. Under Proteus, there are Gigatron simulation files. You can find them here: http://www.nedopc.org/forum/viewtopic.php?f=68&t=20128&start=45#p156722 This scheme can be implemented in SimulIDE, because the output of information in it occurs on a standard two-line indicator, the model of which is in SimulIDE.

7Rom-component question Empty Re: Rom-component question Wed Feb 24, 2021 8:35 pm

Mistral



Thank you. Good to have some more background. 
A lot about programming and thats not for me. 
But a very good órganised schematic compared to the original.
This one has higher resolution.
http://www.nedopc.org/forum/viewtopic.php?f=68&t=20128&start=45#p156712

http://www.nedopc.org/forum/download/file.php?id=7960

Sponsored content



Back to top  Message [Page 1 of 1]

Permissions in this forum:
You cannot reply to topics in this forum