Français:OGPN25,fr

Newsletter Archives

December 15th 2007 Open Graphics Project Newsletter

• December 15th 2007 Open Graphics Project Newsletter
• Interesting List Quotes:
• Progress and developments
  • Documentation
  • Conexant Datasheets Lost
  • VGA Progress
• OGD Architecture
  • Texture Compression
  • Source Code
• Keeping an eye on the competition
  • Project VGA
• The Open Hardware Foundation
  • Mini FAQ for the Open Graphics Mailing list.

(Updated May 16 2009 about OHF and Linux Fund)

Interesting List Quotes:

Terry:

Chip companies hamstringing Open Hardware (projects) would be exactly as rational as Home Depot spreading FUD about home-improvement do-it-yourselfers (i.e. NOT).

Progress and developments

Documentation

Kenneth added a new page documenting the road to VGA in the Wiki. This is combined with the VGA Code Sprint for during the end of the year break.
http://wiki.opengraphics.org/tiki-index.php?page=VGARoadmap

With the new schematics the documentation was given attention:
Joseph:

I have added most of the components to the components page - anything we are missing? (...) A picture of the chip with the pinouts named might be quite handy. I tried using a table (..)

Terry had a look and then added connector and chip images with even more detail.
http://wiki.opengraphics.org/tiki-index.php?page=OGD1+components+guide

This is an easy project that visitors can help with because the main qualification is that you can read english. Post to the list if you want to help out.

Conexant Datasheets Lost

With the announcement of the schematic and the request to help out and look for bugs, Attila noticed:

not all data sheets are available online, namely for the SIL1178 and CX25874 are hidden somewhere.

Howard:

I have these. I not sure if I should allow them to be downloaded by everyone though. While these data sheets were free and downloadable from the manufacturer at some point in the past, this is no longer the case. He explained why he had chosen the chip: This was the only chip I could find that encoded SECAM and NTSC/PAL as well, that had, at the time (..) downloadable specs.

Joseph:

Has anyone actually asked SI/Conexant whether they mind if we post these for our members and prospective customers?

Luc:

maybe OHF would have more success if the foundation do this approach for its own as well?

Sebastian noted that Conexant have dealt with Open Source groups before:

At the beginning of the Prism54 SoftMAC driver project, I tried several times to contact them politely

Howard:

The CX25874 is almost functionally identical to the CX25870 and CX25898 . You can download the spec for the CX25898 here: http://www.icbank.com/icbank_data/online_seminar/conexant_060517/Aquila24_DS_102759A.pdf and the spec for the CX25870 here: http://arm.cirrus.com/files/schematics/edb9307/3_DATASHEETS/cx25870.pdf

If you look at the "internal Registers" section in the CX25898 document, the differences with the CX25874 chip are shaded. Comparing the two specs, I can tell you that the differences are inconsequential. I will try and contact Conexant this week and ask for permission to make the spec available. There seems to be no good reason why the CX25878 is widely available and the CX25874 isn't.

Some google work turned up a marketing document but not anything good enough for programming the chip. http://doc.chipfind.ru/conexant/cx25874.htm

Timothy:

If, in the end, they decide that we cannot share the datasheet, we have a couple of options.

One is to not populate the chip and implement an alternative. Hypothetically, we could do all of the modulation for a composite TV signals in the digital domain in the FPGA and send it through the VGA DAC.

Another option is to extract the essential facts from the material we have and publish our own documentation. This is perfectly legal, because you cannot copyright facts, we would have acquired the chips themselves through legal means, and since the datasheets were once publically published, they can't expect to remove every copy of the document that people had downloaded before they took it down or prevent people from talking about it. In short, we could have a Wiki page on it with all the info one needs to use it.

Terry started documenting the chips and adding drawings with inkscape:

Certainly it would be in the best interest of anyone on the project to collect copies all of the available datasheets for themselves. I think the best we can do is to create some documentation of our own for the chips we are using — which is something I've been working on. But it's obviously going to be a tiny fraction of what's in the full datasheets.

He also noted:

the datasheet link for the Spartan 3 family is broken. (...) it'll have to be found again.

Dieter:

Given that Chip companies are in the business of selling chips, not in the business of selling data sheets. The data sheet is essential to use the chip. Without a data sheet, there is no reason to buy the chip. It is reasonable for a web site to provide a link to the chip company's official copy of a data sheet and also provide a local copy in case the official copy is unavailable for some reason. (server down, DNS problem, accidentially deleted, etc.) I've seen this link plus local copy done on various websites for various works.

Timothy explained Traversal could not help here:

we'd be a nice big lawsuit target (...) Right or wrong, we wouldn't be able to defend ourselves.

Attila:

IMHO we should collect all datasheets of the parts we use and put them onto a website. Although most of them are freely available online it always takes some time to look for them.

He then announced that he had decided to provide a mirror of all the freely available datasheets: http://attila.kinali.ch/ogp/datasheets/

Because Attila is caching them, and if a vendor wishes to remove something, they should contact him.

VGA Progress

A couple of volunteers asked where the project was and how they could help. Timothy explained:

We should do the VGA controller. We have certain parts done and just need to finish up others.(...) The VGA controller will need a lot of assembly language programming. Can you help with that?

Actually, the first thing we need is an assembler for the nanocontroller. Patrick had done an assembler for the video controller in Perl.

Petter:

there is already an assembler for HQ in the OGP repository under tools/oga1hq. (It's written in C and uses a Bison parser.)

OGD Architecture

Texture Compression

Timothy:

Texture compression and z-buffer compression are things I want to stay away from for a while, until we can be sure to come up with something that isn't covered by a patent. For the most part, anything we've decided to use so far is either patent-free or not patentable, and while the patent system is crap for overturning crap patents, we at least have the moral high-ground. There just aren't that many ways of doing 3D graphics, and most of what we have is basically derived from first-principles and then tweaked to make it OpenGL compatible.

Source Code

Nova started an example that several helped grow into a quite basic tutorial on how to write HDL and test cases. This led into a small library of a few of the 7400 series chips.
http://wiki.opengraphics.org/tiki-index.php?page=HDL_Resources


If you want to practice with some basic code you are welcome to add to the list. However Timothy observed that it would not suit everyone as 'The 74-series examples are only useful if you have past experience using them' . He cautioned 'some people will get confused and try to do tristate and open-collector for logic that goes inside of an FPGA, which you can't do.'


Michael questioned about the license:

I'll go with whatever we want to go, this wasn't exactly rocket science. MIT, GPL, did we settle for something?

Terry:

Because these are so simple, I would highly recommend using a non-copyleft license (or even assignment to the public domain). You probably couldn't enforce a copyleft anyway, and it wouldn't be worth the trouble. They're kind of like teaching examples or exercises.

Michael:

Works for me, public domain it is. It is really neat, though. A great newbie project. ;-)

Keeping an eye on the competition

Project VGA

A small group of developers have started work on another graphics board and this was noted on the list. Terry commented regarding an article in the media that had not grasped the differences between the projects:

their concept is more of a simple framebuffer VGA card, not a 3D accelerated card. (...) The world's big enough for two so highly-differentiated open hardware development boards. So all in all, I think it's a lot more pluses than minuses.

Michael from Project VGA posted to the list and explained his intentions:

The goal of this project isn't to compete with anything, we just want to get something working. After we've gained experience with this (and helped more people getting their hands on a usable platform) and support VGA, we're going for driver development and 2D basics. After that, well, hopefully we can merge with OGP and go for the bigger stuff.

The Open Hardware Foundation

www.openhardwarefoundation.org is not accepting donations at this time. Donations to the Linux Fund help the Open Graphics Project build developer boards. www.linuxfund.org/projects/ogd1/

Mini FAQ for the Open Graphics Mailing list.

Where is the list?
Gmane.comp.graphics.opengraphics
How can I get it?
Using your favourite newsreader point it toward news.gmane.org and subscribe to comp.graphics.opengraphics
Which newsreader can I use?
There are many. Specialist software such as Pan (Linux & Windows) or Gravity (Windows) are well known, but there are many others, and often your favourite email software can mostly function as a newsreader.
Can anyone post?
Yes, though the list is moderated. Suggestions for this newsletter are welcome and are made through the mailing lists.