Tuesday, 10 November 2015

Like riding a bicycle?

It's been a long time between posts, for several reasons. Babies, closing the office... resulting in not having the time nor the inclination to tackle engineering-related extra-curricular activities, especially those requiring significant effort!

So... an update? Unfortunately not related to NGPACE, but my first foray back into FPGA development for quite some time. I haven't been working with FPGA's at all in the last few years until very recently, and now I've been contracted to work on a design incorporating a rather nice FPGA (in fact the exact same device I am using for the NGPACE motherboard) that we are bringing up right now; a design that is perfect for retro video game emulation! ;)

I discovered only a few days ago that a colleague involved in the project at work owns a Tempest cabinet! That got me curious and after some research I re-discovered that Jeroen Domburg, a.k.a. Sprite, implemented Black Widow in an FPGA sometime around 2012, (very similar hardware to Tempest) although he interfaced that to a vector monitor.

Turns out that Tempest, unlike Black Widow, uses the Atari Mathbox hardware, which I don't have the time to implement right now. Rather, I thought I'd get Black Widow up and running on our new board, just for kicks. I used the same crude rasterising/decay technique that I did for Asteroids several years ago, although I added colour! It gives surprisingly good results, more than adequate for a proof-of-concept demonstration.



Maybe now I'm on a roll I'll be able to get back to NGPACE...

Thursday, 2 April 2015

Vaporware rendered in glorious 3D

Here's a very, very preliminary 3D rendering of the flash cartridge PCB.


The large euro connector at the bottom mates with the system adapter boards we designed and prototyped last year, allowing the cart to be used in both AES and MVS systems. The connectors down the right (near) side include serial debug connector, JTAG in/out chain, external power and cartridge PCB interconnect cable. On the left (far) side is the SD card slot.

The prototype is intended to be programmed stand-alone from the SD card before being inserted into the motherboard. A complete cartridge comprises 2 PCB's and the interconnect cable enables both to be programmed from the one SD card image.

Early days yet and features are still subject to change.

Wednesday, 1 April 2015

Laying it all out

Some pretty exciting (for me, anyway) news - I've been informed that layout is now complete on the flash cartridge!

I can assure you that this is NOT an April fools joke; at least if it is, I've been fooled as well!

So what does that mean from here? Well there's the tedious but necessary review of firstly the schematic (including checking FPGA pin assignments) and then the PCB, including footprints, layout and physical factors. This is a task ideally performed by someone other than the engineer that put it all together; namely, me. It's not fun at all, but ultimately a lot better than having to re-spin and manufacture a 2nd prototype run.

That aside, work-wise it probably couldn't have come at a worse time. I can't see us being in a position to be able to proceed with prototype manufacture for a few months yet, but hopefully no longer than that. In some ways the delay may encourage more rigorous review. Years ago with our first homebrew design we could only afford a single prototype run and we were so paranoid we reviewed the design to death; the result was a perfectly working board with zero patches!

Until then, I'll get stuck into the comprehensive review, and also enlist my former co-conspirator to review the top level design. Realistically it would probably take me at least a month anyway, given my work and family commitments, so perhaps work will sort itself out in time.

I'll post some eye candy - a 3D rendering of the PCB - in the next few days.

Thursday, 19 March 2015

Mid March Update

It's been a fortnight since the last update, and I'm pleased to say that there has been more progress again since then. I can't say how much longer it will take at this stage, but from appearances I'd estimate that approximately 70% of the layout is complete now. I'm secretly hoping that the closer it nears completion, the more David will be eager to complete it... and that this will be the last blog update before it is done.

Once layout is done I'll post a 3D rendering of the PCBA.

Thursday, 5 March 2015

Feb Update

I'm pleased to be able to say that my learned colleague is progressing well with layout of the flash cartridge prototype. Most of the component & connector placement has been finalised with routing for traces not connected to FPGA I/O pins nearing completion. Reportedly there's still a few weeks work remaining, based on his current time allocated to the project.

Aside from the above, there's not a lot else happening on the project. I've taken the FPGA HDL and software about as far as practical without actual hardware, so that's on hold, though it will probably be cracked open again when it's time to review the final flash cartridge schematic before production.

I've taken the opportunity to snap up a few cheap AES cartridges - for testing the main project of course. My latest acquisition was Art Of Fighting (1), which I was particularly interested in because the source code was found for it on a former developer's hard drive and is freely available for download on the net.

More when it happens...

Thursday, 15 January 2015

2015 and beyond

Part of my New Year resolutions is to be more productive with my time this year. I think 2015 is crunch time for this project, and if I don't get a move-on, someone is going to beat me to it. Whilst my spare time is going to be truly limited in the next 6-12 months, it's all about making the most of what little I do have, and not just whittling away the hours surfing the net.

Tune in here same time in 2016 to read me lamenting about this year and proclaiming the same resolutions next year.

Layout is continuing with my colleague in the process of morphing his analyzer PCB design (which was complete) into the flash cartridge. No ETA to report as he can only work on it in when his spare time and his motivation happen to coincide with one another. You get what you pay for! But I am happy to report at least that progress is being made, and the project isn't just languishing on the back-burner for months at a time as it has done so in the past.

I also managed to acquire Metal Slug 4 (MVS) - purely for this project of course. It had nothing to do with the fact that Metal Slug is my favourite franchise on the Neo Geo, if not any arcade platform. I should also resolve to actually play these games more and not just collect and emulate them!

Pretty much every reader of this blog has requested specifications of the flash cartridge, so at some point as layout nears completion I'll post those, along with a description of the 'bitmap mode' feature.

When layout is complete I should be able to post a 3D rendering of the PCB. There's simply nothing more exciting than being able to see vapourware in glorious full-colour 3D!