New update:
http://brennanthl.wordpress.com/2012/10/26/vbsd-update-10262012/

Sorry about that e5frog, I didn’t mean to get snappy, it was just a bad day.

MineStorm wrote:
Chill out. We do this for fun, right ?

I do know what you mean about board revisions. Waiting for them is excruciating (because the fast turn-a-round service is SO expensive).

I recently dumped a box full. God knows what all that lot cost 🙁

In the past I had good experiences with Itead Studio for turnaround time and price, but the problem is they are sold out depending on when you try to order. I’m hoping in early November they will open up their ordering again so I can send out a board. I’m almost done making a VB Emulator board that I can use to debug the cartridge.

blitter wrote:
Take your time, Vaughnabe13. I’m particularly excited about this board and the potential for it (or a future revision) to allow for additional space in the expansion RAM area for bigger games. My current project is being developed for the FlashBoy and is taking quite a long time, so don’t feel too pressured from me anyway. 🙂

P.S. Are you the same Vaughnabe13 on the MTBS3D forums? *waves*

Yes I am, but I am not very active over there. Recently I was interested in the Oculus Rift so I was following the progress in that original monster thread.

[Vaughanabe13]

e5frog wrote:
What, are you spending your own money!!?? 😉

If you’re short on cash perhaps asking for donations alternatively some kind of kickstarter project or maybe pre paid orders would help.

Of course it’s a lot of work if you haven’t done it before, that’s always the case. Not just the hardware which can take a long long time to CAD when you’ve found the parts you need but all the software as well.

If you weren’t making one I’m sure Richard would have made one by now as he already have most parts of such a project done with FlashBoy and also his Vectrex cart with microSD card reader. I don’t think the market is that big so you probably need all of it yourself if you want to make something from the invested time and money.

Regarding the SRAM issue, already in the 70’s (maybe earlier) they used a battery and two diodes, it should still work… Diode prevents current to flow into the battery, a current limiting resistor could be added to prevent smoke and melting things if there’s a short…

You need a second diode for board VCC to block the battery from powering the whole board.

VCC of SRAM —-|<---- board VCC VCC of SRAM ----|<---- battery + GND of SRAM ------------------------ battery - I was considering using something similar in my Fairchild carts but I decided to try out FeRAM instead which worked out pretty well although they are quite expensive. I saw a RAM cartridge (game copier) for Commodore 64 that worked the same way, holding the memory with a battery. This should hold the SRAM powered for a LOOONG time (if it's a 5V SRAM) http://se.farnell.com/productimages/farnell/standard/42245229.jpg

So what can we do to ramp up your motivation? Not that I have the money right now to get one of these so on my account you can work on it a few more months. 😉

I was thinking if it would be possible to use the CAD:ed connector by one of our members and then try to find the female end of a PCMCIA connector (only the metalpieces that are in the connector) to make proper connectors. The type you have previously linked to worked most of the time on my veroboard built eprom cartridge but it would be nice to have something original looking that works all the time, even if you happen to bump the cart.

With all due respect, I think the wrong thing to do to “ramp up my motivation” is to minimize my problems and condescend me by telling me how a battery and diode works like I’m 5. I realize you are only trying to help, but I need you to understand I am not some amateur who has never built a complex digital board before. I’m an engineer, I do this for a living. I also don’t appreciate the subtle guilt tripping by telling me “Minestorm would have had it done by now”. Especially since I’ve been very vocal from the beginning that I didn’t want to make money on this project and I didn’t want Minestorm to stop any work he was doing. In fact, he can go ahead and make an SD cart right now if he wants. So can you. It can only help the VB community, and I’ve already said I don’t care about making money. In fact, I probably won’t even make back the money I spent prototyping.

The motivation problem comes from the fact that every time I run into a major problem, it usually takes weeks to resolve. Whether it’s a board issue that requires drafting a new board, or a programming issue that I don’t have the right testing equipment to debug, waiting for a new part order, etc., I get grinded to a halt every couple weeks or so and then once I’m not working on the project it is more difficult to resume. It’s just the nature of the beast when one person is trying to interface with obsolete technology from 17 years ago to do things it wasn’t meant to do. IMO the only mistake I have made on this project so far is to dangle a carrot in front of PVB. I did it because I needed information that PVB members already had, but it created expectations. The smart thing would have been to not announce anything, but I had to in order to gauge interest and see if there really was any kind of demand for something like this. If it wasn’t for that, I would have been wasting money on a project that has no future, and I needed to know there is a place for it alongside the flashboy. I am still not convinced of that, but it is what it is.

• This reply was modified 12 years, 1 month ago by Vaughanabe13.
• This reply was modified 12 years, 1 month ago by Vaughanabe13.

The biggest problems I’m having right now is not actually with the cart itself but how to properly test it. I can’t just solder wires onto the cart because the pins are too fine pitch (too close together) and all the components are surface mount, so I can’t just swap parts out at will as if they were legos. Right now I need to figure out a comprehensive test strategy, and I think the best way to do that is to build a daughter board that I can use as a “Virtual Boy emulator” and test signals. The board would plug into the VBSD cartridge and perform test reads and writes to verify operation and allow me to debug further. I don’t think this will take me too long to make, and I was planning on doing it anyway so that I can make ROM and RAM dumps of real VB carts.

Another decision I’m having to make right now is how to deal with the SRAM chip, since it requires power to maintain data. That makes it very hard to debug because I have to debug it live in the system. I may decide to change over to a Flash chip, which solves a lot of my problems and makes the cart cheaper as well.

Folks, I don’t think most of you have a clue how complex this project is, except for DogP and Richard. The end result is easy to imagine, but there are thousands of intermediate steps in between. I’ve already been through 3 board spins and invested hundreds of dollars out of my own pocket into this, so it’s not like I’m not committed. Also, I told Richard a looonnng time ago that he shouldn’t let me stop him from making a FlashBoy follow up. I’m just saying, this stuff is not easy, and I work a full time job and have a family and other hobbies… Anyway, here is the latest update:

http://brennanthl.wordpress.com/2012/10/22/vbsd-update-10222012/

L___E___T wrote:
What does FWIW mean?
For what I want(ed)?
For what it’s worth?

I’d be prepared to pay that, plus margin, plus shipping, assuming it come in a box with all the essentials like the flashboy did.

Hopefully lack of donor parts will keep costs down.

FWIW = for what it’s worth. I don’t know about final costs yet, I was just saying it won’t be any cheaper than that because if I sell these I want to make sure I’m at least not losing money. If I have to solder each board by hand that’s probably 1-2 hours per board manual labor as well. But I’m not sure about any of that yet, I don’t want to get ahead of myself.

e5frog wrote:
Looks nice, we’re all hoping for an affordable price and no need for updates. 😉

The cost to make this board fully assembled was about $60, FWIW.

New blog update! With pics!

http://brennanthl.wordpress.com/2012/05/02/vbsd-update-522012/

mbuchman wrote:
No clock to the FPGA? And no platform flash to configure the FPGA (or are you planning on the PIC to do that)?? and no JTAG to debug the FPGA?

Why not use a CPLD rather than FPGA? Don’t have to load those so it would be up and running right away.

No clock needed for what I’m going to do. Although I still have the capability for a 32MHz clock through a control pin (PIC) if I ever need one. I have the PIC handling all loading on power up, and my intent is to have the bitstreams stored on the microsd card so they can be upgraded or changed as necessary. I may decide to simplify the design once I have it working – this is a prototype board, after all.

The PIC can do JTAG if I need it to, but I’m planning on doing the majority of my debugging using a test bench.

thunderstruck wrote:
It sound like you compiled the GH-Midi Player. If you press A, B, Left Trigger or Right Trigger it plays different Video-Game themes. The displayed number is the note count. As the number changes the music should be played. Obviously you have to turn up your sound to hear the music.

Did you try to start the demo in RealityBoy or on Hardware?

All I did was compile the default file/project that was loaded from the pack. So yeah, I guess it was the midi player. I verified my computer had sound working but I couldn’t get any sound through either realityboy or mednafen…

e5frog wrote:
Looks promising, making all those footprints that aren’t in the library already is a real pain.

I wonder if there’s some central deposit for those things so you can upload and get other people’s stuff?

You should perhaps try and find some cheaper PCB manufacturer, I order from China when I build carts for the Fairchild Channel F and prototyping is done. Recent purchase was for 14 PCBs, plain double sided (basically cheapest alternative):

1) FRCJ01 2.28
2layers, 0.062″, FR4-TG130, 1oz, Lead Free HAL, GREEN SOLDER MASK, WHITE SILK SCREEN, Complex Shape, 3.50″ x 3.15″ Special Notes: Shipped As Individual Boards , Ship To Sweden
QTY: 14pcs USD11.817/pcs; Subtotal:USD165.44
Lead Time: 5 Business Day + 3 Business Days Shipping

I understand a four layer 0.032″ board will be a lot more expensive I just wanted to point out that there could be cheaper alternatives.

I work at a PCB factory in Sweden, we are however not cheap, it’s not impossible though I might be able to squeeze something like this in (one or two boards) at no other cost than the shipping price if there are other boards made with the same configuration (thickness, copper thickness, layers, material).

In my experience the shipping costs for various parts are what tends to make things more expensive, having a single supplier for all parts can be more economic than ordering from here and there even if the individual parts are cheaper.

Gideon is having trouble sourcing microSD card connectors in the same type that he has been using for the 1541U-II. Make sure you choose components that are available in the near future as well if you intend to do more than one batch (I guess it depends on how many is sold). 😉

Thanks for the tips. Luckily I planned ahead in order to make my PCB costs cheaper – just 2 layers, believe it or not! Hard to tell by looking at that mess.

I usually order my prototypes from BatchPCB. Their fab house does a good enough job and I can order just a few boards for the lowest cost I’ve seen. I don’t want to order 10+ boards for a prototype. But I’ll keep your advice in mind for the future.

Here’s my first take on the Rev3 PCB. I will most likely end up making several/many changes to this before I order the next prototype, but this gives you an idea how massively complex this project is. This board was 100% hand routed on two layers from scratch. All SMD packages were also created and measured by me.

Attachments:

1. 

   board.png
   

What is the demo project supposed to do, exactly? It compiled fine and it runs fine but I don’t hear any sound, and all I see is the background graphic and a number that counts up.

RunnerPack wrote:
As I said in the comment on your blog post, it’s great to have you back and working on VB stuff! :thumpup:

The new design looks great! I hope you can get costs down as much as possible, because I really want one!

The biggest factor when it comes to cost is actually the PCB itself. If I can order the PCB in semi-large batches I will get a huge discount per board, and that will drive the cost way down. My goal is to make this similarly priced to the FlashBoy. My rule is that “if I wouldn’t buy it myself, it’s too expensive”. It’s definitely something that I will keep in my head as I work on the design. But first my goal is just to get it working, obviously.

WOW, this is incredibly easy to use. This is a fantastic idea and I’ll definitely use it, since I had trouble getting an IDE and compiler set up before.

e5frog wrote:
Sounds like it will be expensive, do you have any aim regarding the cost?

Sounds to me a Spartan-3 should be enough. Can’t you download an entire VIC-20 emulation in one of those? You’d need to know how of course, don’t have those skills myself, made a 8×8 multi color LED battleship game with sampled sound effects some years ago that’s about it.

Spartan-3’s require a fourth power plane at 2.5V for all the VCCAUX connections. I didn’t have enough board room or patience for 4 power planes and another regulator. The cost difference between 3 and 3A isn’t too bad and I agree that a Spartan 3A is total overkill but I’ve looked into other options already.

The cost won’t be too bad except for the SRAM, because 32M parallel SRAM’s are expensive parts. I could save 90% of the cost of the SRAM by switching to an SDRAM but that adds a lot of complexity because of random access schemes (to interface with the VB) and row/column refreshing and clock generation. Another option is to downgrade the SRAM to 16M. The part I’m using has a pin-compatible 16M version, so I may decide to use 16M to save $10 or so, and then implement the flash save RAM and VBSD menu in the FPGA in memory blocks.

Again, this is a prototype and after I get it working I might decide to switch out the RAM and use a cheaper part to save cost.

Hey guys, just wanted to let you know I’m not dead! It’s been a long time but I have not given up on the project. Here is my latest update.

http://brennanthl.wordpress.com/2012/03/28/vbsd-update-03282012/

And Engadget is right there behind you, making a stupid headache joke at your expense.

http://www.engadget.com/2011/06/15/diy-virtual-boy-arcade-cabinet-costs-only-a-headache-to-play/

The 3DS looks like a child’s toy compared to the PSVita. The 3DS is to the original GameBoy what the PSVita is to the 3DS. That being said, I think it is definitely possible for the two consoles to co-exist and I don’t think the PSVita is going to destroy the 3DS or anything like that. Nintendo will still have their first-party franchises (Mario, Zelda, Donkey Kong, Smash Brothers, etc. etc.) and the games will still be good.

Also, we can’t forget that the original PSP was supposed to destroy the original DS, but ultimately Sony made some bad decisions with the PSP and the introductino of the DSLite basically destroyed the PSP. I would like to think Sony will not make the same mistakes they did with the PSP but who knows…

All that said, for my $250 I would be much more interested in a PSVita than a 3DS. I haven’t bought a 3DS because the launch lineup was terrible and I wanted to see what Sony had to offer. I’m extremely excited for a portable Uncharted and Little Big Planet (among others) and the PSVita hardware looks second to none.