Open ATI R600/700 3D Graphics For Christmas?

[fhuberts]

what??

What "tricked" us with the 6xx was that the internal docs contained information on a lot of functionality we are not using in any drivers today, so the amount of information we had to cut out was much larger than in previous generations so it took a few attempts to find the right balance. The sheer complexity of the chip doesn't help either, since every review involves more people and more time than previous generations.

Why bother putting in the effort of keeping the information in sync with the features you're actually using? This also implies keeping back those that work off of the released documentation, which I'm actually very worried about. That can be considered anti competitive behaviour.

I can understand competition reasons but you could also say that it is must easier to just release it all and focus on innovation and not worry about competition. This is the new paradigm anyway: innovate or die.

Let the open developers come up with new and interesting way of using your hardware... :-)

Also, in your next architecture, just split out the encumbered parts into separate blocks and do not release info about them. Release everything for the rest :-)

[mityukov]

Since the DRM debate ruturns like a bumerang I'll answer that. It is cheaper to include DRM in all models even if some of those cards don't use it/need it. In fact it might have end up that this DRMless card would cost more then the one with DRM.

That's the answer from one of those DRM topics out there If you know a bit about economy and scale effect this should be quite logical.

That's something that I expected to hear back.

However, according to my 2nd part... What's more costly: all that time/money/risks, spent on thorough examinations of specs on DRM-contaning cards.. Or, producing of only few "Engineer" devices, and passing them to the driver devs without a need to hide _anything_?

So, just a kind of work-around to current problems..

[chaos386]

Why bother putting in the effort of keeping the information in sync with the features you're actually using? This also implies keeping back those that work off of the released documentation, which I'm actually very worried about. That can be considered anti competitive behaviour.

I can understand competition reasons but you could also say that it is must easier to just release it all and focus on innovation and not worry about competition. This is the new paradigm anyway: innovate or die.

Let the open developers come up with new and interesting way of using your hardware... :-)

Also, in your next architecture, just split out the encumbered parts into separate blocks and do not release info about them. Release everything for the rest :-)

I think what Bridgman meant by "functionality we are not using" was that there were features proposed during the design phase that didn't make it into the final hardware, not that there is some "secret" functionality on the chip, just waiting to be enabled by a driver.

Releasing documentation for parts of the hardware that don't exist or are non-functional would be misleading and would just hamper development efforts.

[energyman]

producing that 'enigneer' devices would be much, much more expensive. You need a new mask (thousands of dollars), wafers, you need several of them because the first couple with be defective. Time in the foundry, designing the masks, debugging the masks. You are asking for designing a new chip!

Just ask TSMC what a single wafer costs.
And then you have to add packaging and the creation of the cards...

do you want to spent ten thousands of dollars on special 'drm free' cards?

[energyman]

I think what Bridgman meant by "functionality we are not using" was that there were features proposed during the design phase that didn't make it into the final hardware, not that there is some "secret" functionality on the chip, just waiting to be enabled by a driver.

Releasing documentation for parts of the hardware that don't exist or are non-functional would be misleading and would just hamper development efforts.

or functionality later shown broken, too slow, not stable, too energy hungry when the same could be done more efficiently otherwise .... ,

[nanonyme]

producing that 'enigneer' devices would be much, much more expensive. You need a new mask (thousands of dollars), wafers, you need several of them because the first couple with be defective. Time in the foundry, designing the masks, debugging the masks. You are asking for designing a new chip!

Just ask TSMC what a single wafer costs.
And then you have to add packaging and the creation of the cards...

do you want to spent ten thousands of dollars on special 'drm free' cards?

Offtopic: I think the people who made Direct Rendering Manager and Digital Rights Management have the same short form should be shot. Twice. ^^

[elanthis]

Offtopic: I think the people who made Direct Rendering Manager and Digital Rights Management have the same short form should be shot. Twice. ^^

Toss is up there with the well-known genius who came up with the term "Free Software." No end of confusion that could've been avoided if a little thought had been applied.

[duby229]

Toss is up there with the well-known genius who came up with the term "Free Software." No end of confusion that could've been avoided if a little thought had been applied.

From what I understood, the term was deliberate... Stallman I believe coined it to promote freedom... (checked 1983) I personally prefer the term open source sofware rather than free software.

[mityukov]

producing that 'enigneer' devices would be much, much more expensive. You need a new mask (thousands of dollars), wafers, you need several of them because the first couple with be defective. Time in the foundry, designing the masks, debugging the masks. You are asking for designing a new chip!

Just ask TSMC what a single wafer costs.
And then you have to add packaging and the creation of the cards...

do you want to spent ten thousands of dollars on special 'drm free' cards?

Two questions:

1. Isn't it possible to disable DRM stuff in a kind of "firmware" (I mean, are pipleplines, shaders and so on, really perform some DRM natively on their own, or their only the "tools", used by a separate tricky program)?

2. If it costed only $ 10 000... It think it would be far more affordable than spending so many time (probably, paid to some people, starting from techwriters, and ending with lawers) on all that "adjusting of the document"... Really.

[Forge]

I do not like DRM and would happily sign up for a DRM-removed card.

I do not like endless rants about the usefulness of DRM in a thread about something completely different. This thread is about R600/R700 open source drivers, NOT about the DRM garbage that has delayed the project.

I would also like to voice my opinion that bridgman should push the powers-that-be to make DRM less obtrusive in future GPU designs, FWIW.