@billt

Quote:

billt wrote: @ReverseGTR It'd be nice to see the open-source part from AMD themselves... I don't know who that Christopher guy is.

If you actually look on AMD's website for the most up to date or new drivers and apps you would find that most of them ensure that their hardware is supported on Linux.

Take what they are now doing with the Turion for example: http://www.amd.com/us-en/Processors/TechnicalResources/0,,30_182_871_13349,00.html

If a business making steps of not only releasing drivers and apps for an open source OS but also announcing that they will release documentation to anyone wanting to take a look at the code is not enough, what do you think is? Now I understand actually carrying out a promise is the only way for someone to truely earn a merit for it. I am not trying to preach to a choir how great open source but rather if a company makes actual steps towards supporting not only when it comes to software but also hardware you can't say they are just doing it to save R&D costs.

I do not know who Chris Blizzard is either beyond someone who went to th Red Hat Summit and was one of the first to report on AMD announcing steps to increase their support for Linux and eventually other OSes. As for I, I know I can't do anything to contribute to the latest drivers working on Amiga OS4 but I figure annoucing the possibility may help a tiny bit. As for what benefit this has for me right now, I can finally look forward to eventually running the latest Linux with XGL on my notebook.

@ReverseGTR

Quote:

If you actually look on AMD's website for the most up to date or new drivers and apps you would find that most of them ensure that their hardware is supported on Linux.

Supporting linux is not necessarily the same as open-sourcing your graphics driver. open-sourcing a CPU thing is not the same as open-sourcing a graphics driver. The AMD article in the OP talked about new and improved ATI drivers for Linux, but did not say one word about it being open-source. The blog from Chris whatzizname did, but I don't equate him to being ATI or AMD saying the same thing, I have no idea who he is or how he knows what he says, so I think a grain of salt is acceptable. I'd rather hear the Open-Source ATI/Radeon drivers quotes direct from AMD, not from bloggers I've never heard of before.

I've seen this headling at digg and other places, but would like to see a link to a press release from AMD before I believe it. I'd like it to be true, but I'll wait until I see it from more official sources before I throw my hands in the air and yell Halleluyea. Feel free to point me to such press releases related to ATI Radeon products. Turion articles are OT enough to end up in /dev/null. AMD was always more open than ATI, which may be good for us and/or open-source. But seeing how resistant "old ATI" was to letting go of any docs, it seems like patience is a good thing to keep around for a while even regarding "new ATI". Just opening the docs will be a great thing which I very much hope comes to be.

_________________
All glory to the Hypnotoad!

@billt
From what i understood, they will make some kind of reference driver and open up the specs, so that the OSS community can make a fully capable OSS version.
This was posted on the ati site itself, so it is definitely more than a rumour.

The current catalyst driver will afaik probably not be opened up though..

Last edited by Tomas on 06-Sep-2007 at 07:04 PM.

@billt

Chris is a senior developer at Red Hat, though he spends most of his time working on OLPC now. The XO machine is based on an AMD chip, so he's had plenty of opportunity to get close to the AMD Linux people.

In terms of the truth of the announcement - I was sitting five metres away from the AMD rep when he announced it at the Linux kernel summit yesterday, and I'd been hearing it for some time from various sources (I was one of the contributors to the reverse engineered driver basic 2D driver for the current ATI cards). The intention is clearly to provide the specs to interested developers. There's always the potential for it to get screwed up somehow, but I've no reason to doubt AMD's sincerity at the moment.

@Hans

Quote:

except all the newer cards that have AGP don't support AGP2x voltages

AGP bus speed doesn't really have anything to do with voltage (except that >2x requires lower voltage support)

AGP Version Voltage Possible Multipliers
AGP 1.0 3.3 volts 1x, 2x
AGP 2.0 1.5 volts 1x, 2x, 4x
AGP 3.0 0.8 volts 4x, 8x

Here you can see more info

@ReverseGTR

Mark Shuttleworth (leader of the Ubuntu project)'s talk on recent AMD's announcements. Refer to
http://www.phoronix.com/scan.php?page=article&item=827&num=1

ATI/AMD's New Open-Source Strategy Explained
http://www.phoronix.com/scan.php?page=article&item=826&num=1

Last edited by Hammer on 07-Sep-2007 at 10:53 AM.
Last edited by Hammer on 07-Sep-2007 at 10:51 AM.

_________________
Amiga 1200 (rev 1D1, KS 3.2, PiStorm32/RPi CM4/Emu68)
Amiga 500 (rev 6A, ECS, KS 3.2, PiStorm/RPi 4B/Emu68)
Ryzen 9 7950X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB

@billt

I have a riser design here, but it's for a riser with 2xPCI and 1xPCIe, so it has two bridges on it. Since this is primarily for the µA1 I kinda put it on hold until the µA1 can handle bridge chips (which it doesn't at the moment).

No point in making a product with no potential market...

Of course, I could whip out the oldest PCI->PCIe version of the layout and finish that. But again, is there a market? The chips cost me 20 USD, add another 20 for the PCB and 20 for the mounting and you're looking at something that (imho) 99.99% of AW.net's users will just ignore anyway...

_________________
This weeks pet peeve:
Using "voltage" instead of "potential", which leads to inventing new words like "amperage" instead of "current" (I, measured in A) or possible "charge" (amperehours, Ah or Coulomb, C). Sometimes I don't even know what people mean.

@Jupp3

Quote:

Jupp3 wrote: @Hans Quote: except all the newer cards that have AGP don't support AGP2x voltages AGP bus speed doesn't really have anything to do with voltage (except that >2x requires lower voltage support) AGP Version Voltage Possible Multipliers AGP 1.0 3.3 volts 1x, 2x AGP 2.0 1.5 volts 1x, 2x, 4x AGP 3.0 0.8 volts 4x, 8x Here you can see more info

That's exaclty what I was talking about. All the newer cards are 8x only (+ maybe 4x, but 0.8V only) so plugging them in to an Amigaone is likely to damage something.

Hans

_________________
Join the Kea Campus - upgrade your skills; support my work; enjoy the Amiga corner.
https://keasigmadelta.com/ - see more of my work

@olegil

Indeed, there is no sensible market for anything Amiga these days. If it's not worth doing, what about posting the design like Dennis did for his FPGA thing?

_________________
All glory to the Hypnotoad!

@billt

Well, it's not that there's no market for Amiga stuff, it's just that since my target motherboard didn't support my board I basically halted at 90%, awaiting better software and/or another board to target.

And I still have some investments in this project (like a few hundred dollars worth of chips), so I'm not giving up just yet. I need to see a market to finish, but it doesn't have to be very big. It's just that targetting a board towards a system that cannot use it isn't much point

Any mini-ITX could theoretically benefit from this riser, but it's not cheap enough for most PC users... We're talking one PCI split into two PCI and a PCIe, after all. So it's not going to be very fast for graphics, but it'll prolong the life of your Amiga by a bit

_________________
This weeks pet peeve:
Using "voltage" instead of "potential", which leads to inventing new words like "amperage" instead of "current" (I, measured in A) or possible "charge" (amperehours, Ah or Coulomb, C). Sometimes I don't even know what people mean.

@ReverseGTR

More information on their open sourcing is available here. It sounds like they might be sharing hardware documentation with just Novell's Suse team who will be responsible for the open-source driver.

Hans

_________________
Join the Kea Campus - upgrade your skills; support my work; enjoy the Amiga corner.
https://keasigmadelta.com/ - see more of my work

Link to some of the released hardware docs for the curious. It looks like it's just the 2D control stuff at the moment.

Hans

_________________
Join the Kea Campus - upgrade your skills; support my work; enjoy the Amiga corner.
https://keasigmadelta.com/ - see more of my work

Just a quick update: a basic 2D driver is available now. The license appears to be very liberal (i.e., no GPL). This could well be used as a basis for an Amiga driver.

BTW, what extactly is git? It looks like yet another CVS/subversion like tool.

Hans

_________________
Join the Kea Campus - upgrade your skills; support my work; enjoy the Amiga corner.
https://keasigmadelta.com/ - see more of my work

@Hans

Quote:

BTW, what extactly is git? It looks like yet another CVS/subversion like tool.

That's all it is. Torvalds made it for the Linux Kernel after some problem with bitkeeper which they used before that.

_________________
All glory to the Hypnotoad!

Could a driver be written for the Mediator boards or would the Mediator even work with AMD cards?

_________________
Amiga - still alive & kickin'!