[Dawgmatix]

I wanted to know if anyone knows why drivers for Powervr chips arent widely available. Take for example the case of the Nokia N800 (and the 770 before that I believe), the device shipped with a 3D chip in it, but for the entire lifetime of the product a 3D driver has not been made available. Painful as that is, it doesnt affect me directly.

Ditto for the beagleboard. Where I have read multiple posts about the lack of drivers. Again this doesnt affect me but thinking of all that underutilised hardware ... is a sad thought.

However I am an openpandora early buyer, and it seems now there are some questions about whether this device will have drivers for the powervr chip. So anyone knows why this is the state of affairs ?

[tongue_of_colicab]

Im not sure, but isnt this a case of $$$? The HTC tynII I believe also had a 3d chip (not from povervr though I believe) and for a long time that didnt come with a driver either. The whole story behind that was I believe that you can buy the chip but that doesnt automatically buy you the driver. You have to fork out extra cash if you want that.

[Dawgmatix]

I find it extremely hard to believe that Nokia doesnt have mountains of cash sitting in a bank.

[argor]

Originally Posted by Dawgmatix

I wanted to know if anyone knows why drivers for Powervr chips arent widely available. Take for example the case of the Nokia N800 (and the 770 before that I believe), the device shipped with a 3D chip in it, but for the entire lifetime of the product a 3D driver has not been made available. Painful as that is, it doesnt affect me directly.

see http://www.internettablettalk.com/fo...hlight=powervr

Ditto for the beagleboard. Where I have read multiple posts about the lack of drivers

. beagleboard will have the drivers nou there are beta drivers out there
http://groups.google.com/group/beagl...af5abb79383525

Currently, there is a PowerVR SDK currently in private beta. It is
running under Linux and the evil skull demo was running under Linux.
Programming the PowerVR outside of Linux would be very difficult due
to a lack of available drivers and PowerVR hardware documentation.
The 3D SDK/drivers for Linux will be released before the end of the
year. You can get started coding for OpenGLES 1.1/2.0 using
Imagination's SDK that utilizes emulation on a PC:

However I am an openpandora early buyer, and it seems now there are some questions about whether this device will have drivers for the powervr chip. So anyone knows why this is the state of affairs ?

the Pandora people they have the drivers http://www.gp32x.com/board/index.php...dpost&p=672351

[Dawgmatix]

Hmm the nokia 810 link basically says no drivers.

The beagleboard link says drivers sometime next year.

The openpandora link says that they dont know when it will be out. (the current kernel tree doesnt have the driver).

[argor]

Originally Posted by Dawgmatix

(the current kernel tree doesnt have the driver).

no they did say they will not put the close source drivers in the public kernel tree
till before the pandora go out you do nou that are 2 kernel trees
still they have them

[TheArchitect]

Originally Posted by tongue_of_colicab

Im not sure, but isnt this a case of $$$? The HTC tynII I believe also had a 3d chip (not from povervr though I believe) and for a long time that didnt come with a driver either. The whole story behind that was I believe that you can buy the chip but that doesnt automatically buy you the driver. You have to fork out extra cash if you want that.

You also have some tricky IP issues with Linux drivers, one false move an all you driver source could end up in the public domain thanks to the GPL.

IP providers often impose (and quite rightly) tight controls on licensees of driver source. This is particularly purtinent with shader based systems due to the shader compiler being hosted on platform. These often contain a lot of "secret sauce".

BTW - I think thats a Qualcomm based device in the HTC which means AMD/ATI core. It was an OMAP in the original version I think (which you could over clock!). Not strictly relevant, but interesting non the less

[Dawgmatix]

Thanks for the info - I never thought about IP issues being involved.

Just hoping to get my hands on an opengl ES 2 capable device one way or the other, so that I can do some development. I guess I will just have to get by using the es 2 simulator that Imagination Technologies ships.

[TheArchitect]

Originally Posted by Dawgmatix

Thanks for the info - I never thought about IP issues being involved.

Just hoping to get my hands on an opengl ES 2 capable device one way or the other, so that I can do some development. I guess I will just have to get by using the es 2 simulator that Imagination Technologies ships.

Most of the major manufacturers and IP providers have developer programs. Try tapping one of them for a development board. There always on the look out for demo content etc.

You might have to sign a nasty NDA, but at least you'll get a real deal platform.

[tangey]

I would assume you'd have to look in the direction of TI for drivers, as they are the supplier of the Omap chip in the pandora that has "IMG inside" (listens for the jingle)

[Ailuros]

Dumb question: does the Nokia N800 contain MBX or SGX?

[TheArchitect]

Originally Posted by Ailuros

Dumb question: does the Nokia N800 contain MBX or SGX?

I think N800 is a PXA250 or similar Xscale chip isn't it? The N810 was the first OMAP based device and I think its a 2420, which would mean MBX and VGP. Same as the N93 (the camcordery one?) and N95.

[TheArchitect]

Originally Posted by tangey

I would assume you'd have to look in the direction of TI for drivers, as they are the supplier of the Omap chip in the pandora that has "IMG inside" (listens for the jingle)

Just be prepared to give away your first born to get at them

Pandora is not really an OEM device, its more of an "Enthusiast" dev, board, so it is actually a bit slack of them not to have secured a method of releasing the drivers.

Mind you the consequences of an IP breach maybe more than they could bare...

Would be in PowerVR's favour though to "do a deal" to give them a source of cheap, readily available development system like that (can't imagine OMAP dev. kits are cheap through a disty - thats if you can get them at all).

[tangey]

Originally Posted by Ailuros

Dumb question: does the Nokia N800 contain MBX or SGX?

According to this article

http://www.linuxdevices.com/articles/AT9561669149.html

The N800 is Omap2420 (hence full MBX)

The N810 uses it too, but clocks it quicker

Nokia has commited to Omap3 (SGX) for the next one:-
http://www.linuxdevices.com/news/NS9093153240.html

[argor]

Originally Posted by TheArchitect

Just be prepared to give away your first born to get at them

Pandora is not really an OEM device, its more of an "Enthusiast" dev, board, so it is actually a bit slack of them not to have secured a method of releasing the drivers.

the people behind the pandora have stated many time the drivers will be releasing
when pandora will come out

[Ailuros]

Originally Posted by tangey

According to this article

http://www.linuxdevices.com/articles/AT9561669149.html

The N800 is Omap2420 (hence full MBX)

The N810 uses it too, but clocks it quicker

Nokia has commited to Omap3 (SGX) for the next one:-
http://www.linuxdevices.com/news/NS9093153240.html

Thought so; I was wondering since the original poster was asking for SGX drivers.

Originally Posted by TheArchitect

Just be prepared to give away your first born to get at them

Pandora is not really an OEM device, its more of an "Enthusiast" dev, board, so it is actually a bit slack of them not to have secured a method of releasing the drivers.

Mind you the consequences of an IP breach maybe more than they could bare...

Would be in PowerVR's favour though to "do a deal" to give them a source of cheap, readily available development system like that (can't imagine OMAP dev. kits are cheap through a disty - thats if you can get them at all).

It's times more easier for PowerVR to release their reference drivers for MBX/SGX on their homesite. I don't see a reason why they shouldn't anyway.

[Laurent06]

Originally Posted by TheArchitect

(can't imagine OMAP dev. kits are cheap through a disty - thats if you can get them at all).

The BeagleBoard (link) is a nice low-priced ($150) dev board based on OMAP3530 (Cortex-A8, C64x+, SGX). Beta linux SGX drivers have been demonstrated (OpenGL ES 1.1 and 2.0).

[Laurent06]

Originally Posted by Ailuros

It's times more easier for PowerVR to release their reference drivers for MBX/SGX on their homesite. I don't see a reason why they shouldn't anyway.

My understanding is that Imagination policy for drivers is that the chip maker is responsible for driver distribution (see question 3 here).

[TheArchitect]

Originally Posted by Laurent06

The BeagleBoard (link) is a nice low-priced ($150) dev board based on OMAP3530 (Cortex-A8, C64x+, SGX). Beta linux SGX drivers have been demonstrated (OpenGL ES 1.1 and 2.0).

Neat! Thats a hell of a lot better than the $1495 for the official TI eval board.

Says on the Beagleboard.org site about Cortex A8 being superscalar... I'm not sure thats true.

[arjan de lumens]

Originally Posted by TheArchitect

Says on the Beagleboard.org site about Cortex A8 being superscalar... I'm not sure thats true.

Actually, it is. 2 instructions per clock, in-order.

[Ailuros]

Originally Posted by Laurent06

My understanding is that Imagination policy for drivers is that the chip maker is responsible for driver distribution (see question 3 here).

Which doesn't mean that it cannot or will not change.

[silent_guy]

Originally Posted by Ailuros

Which doesn't mean that it cannot or will not change.

I can't think of a single piece of silicon IP where the original vendor is responsible for the distribution of drivers. Each chip and, even more, each system/handset has sufficient differences that it's completely not practical to do so. This is especially so for kernel level drivers. Addresses can be different, memory access policies can be different, one core may have some bug fixes that the other one doesn't, in the IP itself or in the logic that surrounds it. Etc.

And no system provider is going to allow third party to release a kernel mode driver without extensive local testing.

If even in the highly standardized regular notebook space, official GPU drivers don't support those of a notebook, it's just not going to happen for the embedded market.

[RussSchultz]

Originally Posted by Laurent06

The BeagleBoard (link) is a nice low-priced ($150) dev board based on OMAP3530 (Cortex-A8, C64x+, SGX). Beta linux SGX drivers have been demonstrated (OpenGL ES 1.1 and 2.0).

Are these drivers linux desktop integration (i.e. integrated with X, the window manager, cairo, etc.) or are they simply 'your single embedded app can take over the entire screen and go to town'?

[tangey]

Originally Posted by RussSchultz

Are these drivers linux desktop integration (i.e. integrated with X, the window manager, cairo, etc.) or are they simply 'your single embedded app can take over the entire screen and go to town'?

Note sure how accurate this document is but according to:-

http://jkridner.s3.amazonaws.com/esc/ESC-341_Dompe.pdf

"Texas Instruments plans to provide open source Linux kernel drivers for the SGX along with binary only libraries toward the end of 2008 that can be used with BeagleBoard."

[RussSchultz]

Well, what I've seen so far is that when an IP provider offers "opengles drivers for linux" what they really mean are a library that is appropriate to have a single standalone application take control of the graphics hardware and command it through OpenGL-ES and not desktop integration/acceleration. No X, Cairo, OpenVG, etc. support. Just a raw, raw driver.

Of course, when you think about it, while the embedded IP may be relatively standard, the LCD/display controller hardware is almost always homegrown or comes from one of a bundle of providers. Then there's the video accelerator to factor in (which likely isn't from the same IP provider) and there's a whole passle of work to do to integrate everything together in terms of framebuffer organization and data structures.

But probably the real reason there aren't any "drivers" (i.e. like we have on our PCs) for the SGX (or other vendors) is because only recently (with the advent of the iPhone) has ARM and the rest of the SOC vendors gotten serious (with their Ubuntu mobile/netbook initiative) about having an accelerated portable desktop. Previously, the use case was more of a set top box, where a single user driver that controlled the entire screen was completely adequate.