"Let's make GD ROM emulation happen" Facebook group.

That works in many cases, but the VGA rendering really needs it's own code path. It works automatically as long as the proper Sega setup code has been used and VGA hasn't purposely been disabled, but if anyone is freehanding and that part the actual code might need patching. So it works for most games, but fails for quite a few demos and homebuilt stuff that was made before people had put marcus findings into proper libraries.

Something that might be fun/useful if you have some spare flash/rom is to notify the user that the IP.BIN has been patched by also patching in a custom text-image about it below the "licensed by sega" logo in the usual manner.

 

Only a handful of DC games doesn't support VGA at all. Most games that don't support VGA can be forced into it by setting the ip.bin 62de byte to ascii character '1'. This should be *fairly* easy to do using DCIO. Some games won't work even with this fix though.

 

Good ideas there, Storm.

Yeah, it should be possible to patch pretty much anything on-the-fly once it's running.
Good to know about the homebrew stuff though - I thought there might be some stuff that has VGA disabled.

Looks like either way, I'll have to get the scandoubler working for the HDMI board, so it will make it compatible with 99.9% of modern TVs.

OzOnE.

 

When it comes to NTSC-U/J and Homebrew games, all of the ones I know of that aren't capable of VGA use or are capable of using the 240p mode so that isn't -too- horrible, although it's not ideal for VGA monitor users.
However, I do know some PAL games are trapped in 480i limbo and I don't know if using NTSC versions presents a difficulty to folks in those regions or not.

 

I've been out of the scene for a while and came back with a dream of finding out if someone had modified a pc slot drive to use as a slot GDR drive. No luck on that, but I find there's a SD card mod or a sd card reader for serial port that you can buy from a guy in China. Drawback, I hear neither is flawless.

Then I find that Deunan was/is working on a SD Card to emulate the GD-Rom. Drawback, no more info and it may never be released.

Now I find out about DCIO... someone, somewhere... give me some good news.

 

DCIO seems like it's on a good road!

Is it good news enought for you?

 

Hey Ozone, what do you think about adding scanlines to that hdmi output? Possible?

 

Yes. But how long before the end of the road? I think I've got limited time to get this done. I think I mentioned I was surprised at all the good work people have done with the DC while I was MIA. Now the problem is there are so many mods I want to do right away. I've scoured the internet, especially Youtube to find out what's going on with the scene and everyone has their own opinion on which way to do things the best way... I can't seem to find any straight and clear information.

 

Create a thread to ask nicely, and not hijack this one, and you'll have straight answers. Though if you think cybdyn and Ozone are too slow for you, learn FPGAs and start your own project!

@Fandango, if it's only dimming/zero-ing every other line, it should be quite easy to implement if he can do scan doubler already.

FG

 

I don't believe I was being rude, I really, really want to "make GD ROM emulation happen". What I was talking about was my desire for the topic of this thread, so I don't see how I was hijacking anything. No need for a new thread when in this very thread is the subject I'm wondering about. Not saying anyone is too slow, I really appreciate the work that people do for the beloved Dreamcast. Am I in a hurry?, yes.... but for reasons other than just wanting it now. If I thought I could handle starting a project like this, I would in a second. If there were any way I could help them, I would.
I apologize.

 

@ HowlingFury, The only part I though was rude was the "in a hurry" part. I genuinely wanted to suggest to start a new thread if you have questions about mods, as people here will help you fast and well, but you should ask in a new thread so this one doesn't derail. I don't mean you're derailing it right now. Sorry if I sounded rude.

But anyhow, news ozone?

 

i.m waiting for ARM (from OzOnE or anywhere) i hope i'll got it to the end of this month.
but i have got test board w/ similar ARM and i,m trying discover/compile/debug code how work w/ usb host.
it's not fast - as i'm started work w/ ARM mcu 3..4 month ago))) and every time i face w/ new problems: like i need wait parts, or i ordered Ulink2 for KEIL (it's IDE for ARM), in KEIL is not much examples for usb host... so i try used ATMEL Studio 6 ... but this doesnt work w/ ULINK (what? damn..grrrrr). so i need order J-Link programmer/debbuger. before i'm waiting - i try bring code to KEIL.

i think we'll see something good about DCIO in this year)))
if someone has new ideas, You're always welcome here (or in new thread)))

 

A new thread would be nice. Clicking on "something facebook something petition" makes me feel dirty every time.

 

Ditto to that . . . maybe a new thread with a roll-up of the steps and progress to date in the first post?

While I strongly believe that folks should take it on themselves to read everything in a thread and grok it without hand-holding, it's kind of intense for newcomers to the topic to meet 700+ posts in nearly 40 pages . . .

 

I guess a good idea would be for cybdyn/Ozone to make a new thread, with some background infos in the first posts, once the first DCIO prototypes are assembled and being tested, instead of updating us here. Seems like a nice state for a new thread, going from theoretical to practical!

 

NOOOOO!! DON'T DITCH MY THREAD!!!!

Kidding. A new thread would be good. Much of the beginning of this thread is just old speculation and arguing how my group on Facebook wouldn't do anything. 440 members and a few prototypes later I think it's time to let this thread die.

 

This is a project I've wanted to do for a long time! I'm really impressed with the progress so far.

Cybdyn's current solution uses an FPGA along with a microcontroller, right?
I think it could be done with only a microcontroller. Some of the high-speed bits would need to be in assembly, but other than that, it could be coded in C, which is much easier to work with.

I've used STM's ARM microcontrollers in the past, and they're really nice. Lots of features and peripherals.
This particular one would likely do the job:
http://www.st.com/web/en/catalog/mmc/FM141/SC1169/SS1577/LN1035/PF253031

Max clock speed is 168MHz, and it has an SD card controller and plenty of RAM. The MCU's DMA controller can be used to transfer from the SD card to RAM (probably much faster than 1.8 MB/s, too), though the MCU would need to process transferring the data from RAM to the Dreamcast. Should definitely be possible, though!
A hardware datasheet I have says the Dreamcast DMA transfers happen at 13.3MB/s max. So that's one byte every 75ns, but since it's a 16 bit bus, we have 150ns to set up each word transfer. If the MCU is clocked at 166MHz, that's one instruction every 6ns. So we get 25 instructions to essentially increment a RAM address and read a couple bytes to a GPIO port. Should definitely be doable.

In 3-4 weeks I'll be getting some boards fabbed so I can actually try it, so I'll report back then. I doubt the hardware will be the most difficult part, though; especially after reading this thread, it seems like getting the software side right will be most difficult. I will definitely be trying to find all the help I can get when I get to that part. Gotta keep looking over those spec sheets until I understand all of it.
Unfortunately, I don't think any of the links for OzOnE's old source code still work, or I'd be looking at that. Does anybody know if cybdyn has released anything of the sort, too?

I don't mean to create "competition", so to speak, but if I do get it working as well, it's not like more options could hurt anybody, right? Plus, as much as I just want a GDD replacement, I really enjoy a project, so I still wanna try to do it myself!

 

I'll start a new thread if I can get in on the BETA.

 

Mario - I'm interested in to see your device in action.

about software - there are files w/ information about SPI protocol, and you can find source code of DC emuLator for PC.

I doubt about such way - use of just ARM , maybe to use small cpld to interface DC and mcu.
but i dont know, maybe you have more practice, and you know more.

of course, i wanna reduce cost (and count of parts) in future, but i hope, this last board that i made - it should work.

 

Thanks, cybdyn. I'll have time in about a month to make a board layout and get a few made, so I can start programming. Until then, not much from me.

I do have the various GD drive documents and I've been going through them. I need to look at emulator source code, too.
Still have more to learn about the Dreamcast's GDD protocol, but I'll get there eventually. I have a logic analyzer I can hook up, too, so that should help. But surely somebody's done this before, right? Where can one find that kind of thing?

I doubt I know more than you, but I may have a different skillset. While I don't know FPGAs, I do know C quite well and have made many microcontroller projects.
With a fast enough processor it's always possible, the question is whether this one will do the job.