Toutes mes réponses sur les forums
-
new version (march 2012) compared with org. mediator-driver
full color pointer
faster drawline
faster shapeshifter gfx
hw: a4000d (060/90) – mediator4000di – radeon9250
Ratte:
PIII 700 512MB + Radeon PCI 256MB
Amithlon PowerFB vs. OpenPCI Radeon
Same system same software only different drivers.
Update:
writepixelarray8: (7782 powerfb) 2978 -> 8200
drawelipse: (36454 powerfb) 35675 -> 39426
draw: (13964 powerfb) 13225 -> 49349
writechunkypixel: (7780 powerfb) 2980 -> 8179
writepixelline8: (86848 powerfb) 75950 -> 90690
bitmapscaling: (250 powerfb) 117 -> 265
Its now faster in all p96-tests compared to the powerfb-driver.
There is a new feature, the mousepointer is a alphachannel-hardwaresprite with noninverting 3rd color.
Amithlonfix completed.
If the radeon is the first gfx-card in the system, than it has to disable the « lite » chipset-emulation from amithlon (bootmenu & dosshell).
But this had to be done only for the very first gfx-card in the system!
Disabling has to be done by the following UAE-Traps:
FindCard / UAECALL #16
SetGC / UAECALL #21
and depending on D0-state in
SetDisplay / UAECALL #26
PIII 700 512MB + Radeon PCI 256MB
Amithlon PowerFB vs. OpenPCI Radeon
Same system same software only different drivers.
Update:
Driver (framebuffer-mode) has been successfull tested on
– Mediator (4000di)
– Prometheus
– Amithlon
Searching for GRex-Testers …
Ratte a écrit :
Hi Benjamin,
could you please send a mail to:
computer_hobby(at)web(dot)de
I got no responce to the mails I send to you.
(spamfilter?)
Thanx in advance.
André
Benjamin?
Could you contact me?
I think, my emails are rated as spam at your mailaccount.
Thanx.
André
Hi Benjamin,
could you please send a mail to:
computer_hobby(at)web(dot)de
I got no responce to the mails I send to you.
(spamfilter?)
Thanx in advance.
André
TitanMKD a écrit :
You can contact me by Email: bvernoux @ wanadoo.fr
I used this email-contact on 13.07. …
… provider dead or maximum of inbox reached ?
Hi Benjamin,
nice to see, that you are still arround.
If you remember, we allready stand in contact in 2009 and you are so kind to produce your beta4 for the mediator4000di on request.
Our OpenPCI-Radeon driver is in a working beta stage.
It works on the mediator and the prometheus (co-op with « mastatabs » tobias seiler).
G-Rex and PowerPCI/Amithlon are not tested at the moment.
On beta4 you removed the DMA-detection … thats a good solution, because the radeon-driver should support a lot of different memorysetups and chiptypes.
Radeon 7xxx-9xxx / 16 – 256 MB.
It could be nice if you add a api for « giving » dma-memory from the openpci-gfx-driver (here radeon-openpci) to openpci.
The radeon-driver has much code for correct memory-detection for all important radeon-cards and best should be to take 256kb from end of this calculated memory and transfer the result to your library as dma-memory.
How does dma work on other bridges with beta4, can i add dma-memory on the same way?
Or can i make some « allocdma »-mem check?
result = 0 -> adddma mem from radeon
result <> 0 -> dmamem added by pci-system and bypass radeon adddma mem
With the first « real » openpci gfx-driver and a little modified beta4 openpci.lib, i would advice to bump the version to 2.2 (3.0) …
Btw. can i have a a backup of your homepage.
If you like, we can host it on a1k.org.
br
Andre
For your information, latest OpenPCI is V2.1 beta4 from 2009
http://www.a1k.org/forum/showpost.php?p=288822&postcount=32
(remove the .pdf from the filename .. attachments on a1k.org are sometimes mystery)
Why we used OpenPCI is easy .. supporting Mediator, Prometheus, GRex & Amithlon with only one code.
And .. OpenPCI is coooool stuff!
10 sujets de 1 à 10 (sur un total de 10)