[Open-graphics] So what *ARE* we here for?
Gregory G Carter
gcarter at aesgi.com
Wed Dec 1 00:05:10 EST 2004
Randy Williams wrote:
>Quoting Steve Baker <sjbaker1 at airmail.net>:
>>This email is principally directed to the non-TS people here.
>>What is proposed by TS is essentially that TS do all of the hardware
>>design in secret and that they don't release the resulting designs
>>(well - they *might* - but "No guarantee" is as good as "No" in this
>No. That is not correct, even internally Tech Source isn't throwing the idea of
>opening the FPGA code away. The general consensus is that Mgmt. hasn't weighed
>in on the issue so we don't REALLY know what the folks upstairs will do.
>However, it is very possible that after a short period of time the code will be
>made available. The key point here, and something that you miss, is that Tech
>Source must see a profit motive in this effort. The closed source FPGA code is
>a way to do this.
Why can't we buy open FPGA code?
I buy open source code right now?
Why can't I buy open hardware?
>>It's clear to me that this situation has two major problems:
>>1) In the end, we OpenSource people will be unable to work on the
>> hardware design. It's always going to be TS's product. Our
>> job will be to help them out by writing drivers that'll never
>> work with anyone else's card.
>I'm not sure you've been following this list very well. I'm not aware of anyone
>specifically saying that the hardware design won't be involved across many
>people. AFAIK it will, and has already, include(d) a number of folks. And
>although there would be a Tech Source person actually doing the input, that
>doesn't take away from the work done by others so far. Of course our drivers
>won't work with anyone else. This is not new, nor is it something that should
>be a surprise. Nvidia's drivers don't work with ATI's cards, nor would we
>expect to see that here. Unless I've missed too many posts...
I agree with this.
>>2) The goal to have an Open graphics card is busted. All we have
>> is a closed hardware card with Open drivers. This isn't such a
>> wonderful thing because there are plenty of low-performance or
>> 2D-only cards with OpenSourced drivers.
>Again, I'm not sure you are reading this project very accurately. The simple
>fact of having totally OPEN drivers is the EXACT reason why this whole thing
>got started in the first place. While that advance may not interest you there
>is really no current product that has anything close to this. No, its not
>perfect, but it is a whole LOT more than what we have now.
I am not quite sure I understand.
If I need to buy a computer, what prevents me from buying it if I also
know the internal layout of the chipsets?
Same with a video card.
I know the internals of the card so that means I will steal it and
cannot buy it?
That is absurd.
There are IP laws that would protect others from copying it already in
place in the USA if someone wanted to make a shameless direct copy as well.
Orange computers tried to do that with old Apple ]['s and got sued out
Honestly this mightier than thou attitude by Hardware manufacturers
really pisses me off when I buy a product and cannot use it with Linux.
As more idiots realize software licenses are going away, and LABOR is
what that value of software is and where the money goes, software is
gradually becomming less important.
As Stallman has said many times, the software is free however the labor
>>That being the case, what would happen if we just ignored TS and
>>started our own group?
>>1) We might have a problem with getting access to VHDL/Verilog
>> FPGA design tools - and we'd have to hand-build boards to test
>> with. I have no idea how serious that issue is.
>Well, from what I hear some of the tools that we are considering are
>$20,000/license/user. These are on the cheap end. I'm also told that the
>Xilinx product is another answer and that is $1,000 to $2,000/license/user.
>While that's not much to Tech Source that is a LOT to a single person to spend,
>although not out of the realm of possibility.
>Hand building boards?? Outside of college, I've never seen it done with a
>production part. While there may be some residual value for experimental
>purposes, no production house (that I'm familiar with) will accept a hand-built
>board as a prototype. Even our experimental boards have professionally built
>PCB's to start with.
We may need the help of a University Institution.
After, that is where the web browser started.
Why not a video card?
>>2) TS tell us that if the hardware were free/open then other H/W companies
>> would rush at the chance to produce a clone - and that they
>> could under-price TS in the process. That's their ENTIRE reason
>> to keep the design secret.
>No, that's not correct either. It is one of the most compelling reasons, but
>not the only reason. Without a profit motive, that makes business sense, there
>is no reason to invest in this effort. Tech Source has already demonstrated a
>HUGE amount of good faith on this effort and invested in a large amount of
>hours of some very expensive people's time just to get this far. Not to
>mention all of the other folks who've already donated bandwidth and server
Again, this is preposterous.
There are laws to prevent this in the current copyright and ip law in
U cannot outright copy a product and sell it, it is illegal to do that
in Europe and also in the US.
I can cite many sources that HAVE tried to do this, and they ended up
withdrawing product and paying huge damages.
This rhetoric is the same argument to keep operating systems secret.
The only reason is primarily greed, and to disenfranchise as many people
as possibly in pursuit of scientific knowledge and advancement of the
human race for the benefit of the few super wealthy.
Whether I know what register A does or doesn't do isn't going to impact
my decision to buy the card.
As many have alredy pointed out, very few facilities exist to
manufacture these sorts of cards and as I have pointed out through
precedence, IP laws protect products for outright copying.
Do not insult peoples intelligence by continuing this line of thought.
We already proved it wrong with software, now it is time to move on to
It is a natural progression don't you think?
>> Well, dunno about you - but this seems like a GOOD thing for us
>> (albeit a bad thing for TS). Multiple vendors to choose from with
>> competition driving down the price.
>> It's possible that TS are wrong about this point - but if they are
>> then they have no reason not to open up the hardware design.
>See above comment. We are fine with competition, it's better for everybody, but
>there is a reason why some large firms (Sony, Sharp for example) are no longer
>registering patents. If you want to compete with them, then you spend the R&D
>to get there.
Having a open standard spec and then having everyone innovate around the
design protected by existing copyright and IP laws should provide
everyone opportunity to make a buck.
May the best organization win!!
>>3) We would have to design everything ourselves. That seems daunting
>> (and it is!) - but since TS appear to have essentially zero 3D
>> graphics expertise - and several people here have demonstrated
>> quite good knowledge and unbounded enthusiasm to do it - we may
>> not be any worse off. We'd have free control over the design and
>> could make it completely reprogrammable from the get-go. That
>> way, the design would be utterly Free/free and Open.
>Again, from where I sit, as a simple IT guy, this community has full control
>over the specs as they sit now. Take the TV-Out discussion for instance. If
>you are passionate about a feature, take it to the mat and see what happens.
Ra ra. Here here.
>>4) Some people here will leave the project if the hardware isn't Open
>> and non-TS people are unable to work on it. Several of us have
>> been talking about this off-list. If TS stay under their present
>> terms - then many of our experts (me included) will cetainly leave.
>> I immodestly submit that the group will be considerably worse off
>> as a result.
>That's going to happen anyway, even if this thing becomes what we all are
>working hard for. If you feel that you can do a better job under different
>circumstance, then feel free to do so. That doesn't change the viability of
>the project just because some folks leave it.
>>That being the case, do we *really* want to deal with TS at all?
>>I think not.
>Just a few reasons why it is a good idea to *really* want to deal with Tech
>Source (and yes, I'm a TS'er and this is my opinion).
>1) Who is going to spend the $$$ for the thousands of parts that these cards
>will need to be built?
Who is going to spend the billions to construct a OS Kernel?
>2) Who has the relationships with the Fab houses, PCB shops and parts suppliers
>to get this project deployed in a timely manner?
How can it possibly succeed there is no market and there is no
distributor and now one to support it.
>3) Who has very skilled people, who have spent years in the graphics business
>that already have working relationships with the large corporate vendors that
>will buy this product in quantity?
Only a company with PhD computer scientists with decades of experience
could possibly build a modern OS kernel.
Anything else is a copy of the UNIX intellectual property domain and
therefore illegal and UnAmerican!
>4) Who will invest the time/tools/engineering environment to get the project
>built to begin with?
No single individual could possibly possess the equipment and research
facilities to design and coordinate a modern OS kernel.
>5) Who will orchestrate the marketing program, the visibility at major trade
>shows, the advertising in needed trade magazines and promote the product
Linux will never succeed. It is not supported, it has now marketing
department and no one with the ability to promote it.
>While you obviously disagree with the direction this project is taking, and you
>certainly raise some interesting points, there a HUGE list of benefits come
>with what we are trying to do.
>Now that I've ranted, have I missed the mark?
>>---------------------------- Steve Baker -------------------------
>>HomeEmail: <sjbaker1 at airmail.net> WorkEmail: <sjbaker at link.com>
>>HomePage : http://www.sjbaker.org
>>Projects : http://plib.sf.net http://tuxaqfh.sf.net
>> http://tuxkart.sf.net http://prettypoly.sf.net
>>-----BEGIN GEEK CODE BLOCK-----
>>GCS d-- s:+ a+ C++++$ UL+++$ P--- L++++$ E--- W+++ N o+ K? w--- !O M-
>>V-- PS++ PE- Y-- PGP-- t+ 5 X R+++ tv b++ DI++ D G+ e++ h--(-) r+++ y++++
>>-----END GEEK CODE BLOCK-----
>>Open-graphics mailing list
>>Open-graphics at duskglow.com
>>List service provided by Duskglow Consulting, LLC (www.duskglow.com)
>Open-graphics mailing list
>Open-graphics at duskglow.com
>List service provided by Duskglow Consulting, LLC (www.duskglow.com)
More information about the Open-graphics