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

http://hackaday.com/2012/10/17/gd-rom-drive-emulated-to-use-sd-cards-instead/

Someone is getting somewhere to make it on Hackaday.

 

^I didn't see any new information on that hackaday link. It just looked like an informational posting about Deunan's device, with the same information we all learned from Deunan's blog in Aug/September. Did I miss something?

 

The board is now revision 2 with most if not all the issues solved. Final steps are to finalize design and go to kickstarter for production.

 

excited

 

I didn't see anything about kickstarter or any different info from Deunan's blog post about his prototype from back around September 3rd. In fact, at that time he even explicitly said he would not use kickstarter.

The September update is still the latest post on his blog.
http://dknute.livejournal.com/

I'm not saying you're wrong, I'm just confused because I don't see any new info other than what has already been talked about on Deunan's blog. Even the picture on hackaday is from his blog - which has already been mentioned in this thread.

 

Its just nice its on HaD, now more people will support and flood the sites.

 

I've already gotten several new members to the Facebook group due to the HaD feature.

 

Isaw and read it yeah

 

Has the project died now? No news from members who are working on it or from other places documenting progress...

 

it goes slow it's true, but i still have no reasons for make it faster))))

dont lose hope... i work in Altium Designer under such emu boards (ps1, dc)... it was my fault, i need made prototype first, then fix bugs and work under FW. cause even on this stage it can do useful things like fast SD/HDD connetion for "gd-rom call redirection method".

 

I'd offer to help fund this project right now but I just bought a Sega DS-16 and now I'm broke ATM. LOL!

 

i think you alone cant change things globaly...

generaly i just wanna said i still working under such devices, so your hope is still alive...

 

Hi i had a reply from nec

Here goes,

Dear Sir,

This is the NEC Global Gateway Webmaster in Japan.
We received your inquiry via NEC-UK website.
Thank you for your inquiry.

Unfortunately,we can not find out spu-3200 on our data base.

We are not sure that your sega dreamcast is used NEC’s spu-3200,
but it seems NEC only supply this part as the electronic material
for sega dreamcast.
In these cases, we can not support anything for you.
So, could you please contact to sega.

Thank you for your kind understanding.

Best Regards,
NEC Global Gateway Website Webmaster

 

@bart - were you looking for pinout details for the SPU-3200, or just a replacement?

I'm just wondering because it wouldn't be too difficult to work out the pinouts for the laser.
I've got most of them done already, just need to find the correct order for a few signals.

btw, I'm still SLOWLY working on the GD Emu too, and still trying to figure out a way of debugging the boot issue.
I'm looking at getting the direct Compact Flash / HDD connection working rather than the FPGA as CF / HDD will be easier and cheaper for people to use.

Unfortunately we're in the process of looking for another house, so I keep having to hide all the DC stuff away for viewings etc.

Do you like my DC "dev" box? :friendly_wink: (wow, the iPhone 3 camera is crap. lol)...
http://i45.tinypic.com/fp77fp.jpg
http://i46.tinypic.com/kap5bc.jpg
http://i49.tinypic.com/2e16592.jpg
http://i48.tinypic.com/34xqc76.jpg

OzOnE.

 

My work on it has unfortunately been shelved until I get more free time (coursework and exams are crazy). Also, prototype boards cost money and I'm saving up for an extended vacation this summer, so my budget is tighter now than it used to be. Expect to hear something from me around the beginning of February.

 

I've said it before and I'll say it again.

I'm available for anyone who needs a beta tester for their creation. As long as it is mostly P&P (at lest for the bus connector) I'm game. If software needs burning to CD, BIOS flashing, serial SD adapter needed, I have these things to use.

 

I am after new replacement lasers if possible i mean you can get a ols new kss210a but seems hard to find a spu-3200.

If anyone know where to buy tested working lasers i am happy to buy them.

 

I've had a quick look around for the SPU-3200 - they're available on eBay or Alibaba (quite cheap too)?

I finished the pinouts for the original though (Samsung type GD drive)...
These pinouts probably been posted somewhere before, but I wanted to double-check the circuit on my DC (VA1 PAL).


Dreamcast Laser pinout (SPU-3200 16 pins)...

Has a similar pinout to (some) 17-pin pickups, but the DC 16-pin one doesn't have a separate "RF" output pin (usually pin 1).
So pin 1 on the DC starts with the "VR" signal instead.

Apparently, there is a 16-pin and a 17-pin version of the SF-P101N pickup and this can be used on the DC too (if the first pin on the 17-pin version is chopped off)??

The RF chip (IC1) on the Samsung type Dreamcast GD board is usually marked "Yamaha 915", and actually appears to be a TA2140FN (RF amp IC).

The service manual for the Samsung SC-140 PC CD-ROM drive has a very similar layout to the DC GD.
I found the pinouts below using the SC-140 diagram as a reference (although the SC-140 uses a 17-pin pickup)...

Pin 1. (VR / Variable Resistor) just has a 100R resistor down to ground.
Pin 2. (PD / PhotoDiode) goes directly to pin 8 on IC1.
Pin 3. (LD / LaserDiode) comes from the collector of transistor Q1.
Pin 4. (C Photocell) goes via R202 (100K) to pin 4 of IC1.
Pin 5. (B Photocell) goes via R203 (100K) to pin 5 of IC1.
Pin 6. (D Photocell) goes via R204 (100K) to pin 5 of IC1.
Pin 7. (A Photocell) goes via R201 (100K) to pin 4 of IC1.
Pin 8. (E Photocell) goes via R206 (47K) to pin 7 of IC1.
Pin 9. (F Photocell) goes via R205 (47K) to pin 6 of IC1.
Pin 10. (VCC) = +5V from output of 78M05 reg (IC10).
Pin 11. (VCIN) = +2V1 from pin 12 of IC1.
Pin 12. (GND)
Pin 13. FOCUS COIL Minus
Pin 14. TRACKING COIL Minus
Pin 15. TRACKING COIL Plus
Pin 16. FOCUS COIL Plus


All pinouts are confirmed to match the expected circuit etc...

VCC on the GD board is supplied by a 78M05 +5V reg (IC10, voltage confirmed).
The "VCIN" voltage is confirmed. On my DC, this was around 2.07V

The photocell outputs match the datasheet for the TA2140FN (IC1), and the resistors / connections make sense.

So, photocells A and C each connect to a 100K resistor. The junction of the two resistors then connects to IC1.
(same for photocells B and D, but on a different IC1 pin).

(Photocells E and F just connect via the 47K resistors to IC1.)

The focus / tracking coil connections go to a BA5986FM driver chip, and match the expected signals.

The laser diode itself is driven by Q1 on the underside of the GD board.
This is a similar circuit to the SC-140 one.


So, I'll try hooking this up to a different pickup.

I want to try making a Slot-loading DC using a Wii drive mech. :friendly_wink:
I also have an old 6-CD changer from a Ford Cougar (Mercury), which would make a fun project.

OzOnE.
Ahh, @sonicdude - I still haven't had any IDE adapters made, but if you're up for a bit of soldering, you could try adding an IDE cable to you DC?...
http://www.dc-swat.ru/forum/thread-1902-post-23518.html#pid23518

The code isn't yet booting images either, but if you fancy messing with the current code I can re-upload it? (Filedropper links are dead now)
Do you do any C coding at all? Or know any methods for debugging some code on the DC?

 

I'm up for putting the IDE cable on. My unit is a NTSC-U VA0. I have no clue on working with coding or else I'd jump on that wagon and get to work since I have everything needed minus a C programmer. I know nothing about code. Just hardware side of things. As far as I can tell, using the direct method over the FPGA should be better. Not sure but the drives used may need to be formatted with something weird to work properly. (Like how the Xbox has a weird format on it...) The best method I can think of at this point will be either a bootloader image to go on serial SD adapter, CD, or burned to BIOS if the chip has been swapped. That or else a whole new BIOS with the proper protocols built in.

And just for the record, if the code can be made to use practically ANY IDE device, good. I just learned of Magneto Optical drives and would like to use a 1.3 GB model. Looks like a standard IDE internal 3.5" Zip drive but the disks can hold upwards of 1.3 GB data. Be perfect for me since these things last for god knows how long. They're a hybrid of FDD, CD ROM, and HDD. They can also be used just like a HDD being re-written over multiple times. Only problem is they're slow to write to since the process takes at least 2 to 3 passes...

 

Yep, the coding could be a bit of a problem if you've not done any C stuff before...

I'm not much of a C programmer myself, but I can understand most code and modify it OK.
When I started the direct-connection project a few months ago, it was the very first time I'd done ANY Dreamcast coding at all.

For the actual dev environment itself (under Windoze), I just used the R4 Dev ISO...
http://dchelp.dcemulation.org/downloads/DDEr4.rar

There are a few guides online for setting it up, but you don't really need to use the Dev++ IDE...
http://www.dcemu.co.uk/vbulletin/threads/97142-DC-Dev-R4-Dev-C-Guide

With the R4 ISO installed, you should be able to compile things straight from the Cygwin window.

I don't think there's much extra setting up that needs to be done?
Try installing it anyway, then you should be able to have a play with the examples.
These are normally under C:\cygwin\usr\local\dc\kos\kos\examples

(You really need to be familiar with the basic Linux commands under Cygwin to navigate the directories etc.)

Or, what I normally do is to edit the code using Notepad++, then only use the Cygwin window to run the actual compiler.
(I'm guessing Dev++ is similar to Notepad++, but probably allows you to compile the code with a single button-click?)

And here's a backup of my CF card / IDE redirection code (again, still not booting past Sega license!)...
http://www.multiupload.nl/KPN3MM05K9

Just to be clear, the above code is for the direct connection method (ie. IDE cable soldered to G1 port).
It's supposed to be used with the CF adapter or HDD set to SLAVE, so the GD drive can be left plugged in to get the DC to boot.

It also assumes you're using a SWAT SD card modded DC as well.
(It's possible to run the code from CD-R, but it would take FOREVER to try debugging if you have to keep writing a new CD-R or CD-RW each time!)

The CF card / HDD should have NO filesystem or formatting on it. It doesn't use FAT32 or anything like that at this stage, so Windows won't recognize it.

The code simply points directly to the LBA offset on the CF / HDD where your GD image (track03.iso) resides.
If your GD image is in .BIN format, then it will first need converting to ISO instead (it was just easier to do the sector calcs this way).

IIRC, you'll need to modify the code to tell it the LBA from where to read the image from CF / HDD.
(I used Winhex to copy track03.iso directly onto the CF card.)

The code has the basic redirection stuff in it, but I don't know if this is getting over-written in RAM after the image tries to boot?
tbh, there are lots of issues to sort out with this as I don't know how the hell to properly debug this stuff on the DC hardware itself?

If there was a simple way to hook up the DC to the PC to get lots of debug info, then we'd have a much better chance.
I've probably mentioned this many times, but just in case anyone thought the code was anywhere near complete. lol

Of course, you'd normally want to use the DC's serial port for simple debugging info, but this is taken up by the SWAT SD mod.
So, in some ways maybe it's better to boot the code from a CD-R. I wouldn't recommended it though, unless you're very sadistic. :wink-new:

OK, so there's the basic process. As you can tell, I hit a brick wall in the end as I can't figure out how to debug it.

There may be a way to still use debug output on the serial port while the "SWAT" SD card mod is still connected?
The idea is that the serial output to the PC is only enabled when the /CS pin is HIGH (inactive).
This would require an update from SWAT to get it to work though.

Yep, I know, it's all a bit of a mess. :dejection:

As I say, I'm still learning so if you can find more experienced DC coders who fancy hooking up the IDE cable,
they'd probably be able to debug it much easier and add more advanced features.

btw, I'm working on the assumption that Deunan will finish his board soon.
I don't really mind which method works as long as it works well, and isn't stupid expensive. :friendly_wink:

If anyone wants to give the cable + coding a whirl (even just to learn some C stuff on the DC), I might be able to give you some help with the compiler set up.

OzOnE.
P.S. Magneto-optical is a nice idea, but you might as well use CF cards these days.
I'd imagine MO drives / disks are still available, but they're probably getting rare by now?