Callisto Technical Discussion

[esbr]

Hi All

Just installed Callisto Build 02 & the driver seems not to be loaded at all.

No indication at boot that the Callisto drivers is loaded.

 

My graphic card is an Ati Radeon 9600 (1002x4150)

Is there something to change in IOPCIFamily or not? Do i miss something?

 

Thanks for your help

 

Type: VGA-Compatible Controller

Bus: PCI

VRAM (Total): 256 MB

Vendor: ATI (0x1002)

Device ID: 0x4150

Revision ID: 0x0000

Displays:

Display:

Resolution: 1280 x 1024

Depth: 32-bit Color

Core Image: Not Supported

Main Display: Yes

Mirror: Off

Online: Yes

Quartz Extreme: Not Supported

[XanthraX]

Works 1/2 for me (ATI Radeon 9550, 256 M VRAM, AGP).

Can change resolution and refresh rate, but after every restart, my monitor shows a garbled image, that can be restored by click in the display icon on the menu-bar -> Display Preferences -> and chose the refresh between 75 and 85 (hard to see it, but no impossible, because the image is double).

Even the QE and CI shows "Supported" in the System Profiler, in Dashboard I have no ripples and no flip widgets. And no mose tearing to.

And not the last thing, when switching into a fullscreen 3D application (like iPhoto -> Play Slideshow) and back to desktop, my screen is verry garbled and unreadable. I must reboot with the buton on the computer case.

Thanks for this stage of the driver, I wrote these, only to inform you what are my problems.

Is something wrong on my install and settings?

[rjtd]

Greets.

Unfortunatly I cannot try the new Alpha2 driver because I don't have my laptop with me.

But the ones who do have a 9700 mobility or a laptop, can you confirm me that now 1280x768 works or can be choosen?

Thanks

[autoy]

Is something wrong on my install and settings?

 

Have you linked ATI9700GA.plugin to omni's framebuffer? See http://omni.starchaser.org/?faq question #11

[kua4ke4]

hi guys does beta 2 fixs the white screen bug + stripes changing resolution? and working 1/4 time? like karku4 and others

karku4? Is that me?

 

 

Works 1/2 for me (ATI Radeon 9550, 256 M VRAM, AGP).

Can change resolution and refresh rate, but after every restart, my monitor shows a garbled image, that can be restored by click in the display icon on the menu-bar -> Display Preferences -> and chose the refresh between 75 and 85 (hard to see it, but no impossible, because the image is double).

Even the QE and CI shows "Supported" in the System Profiler, in Dashboard I have no ripples and no flip widgets. And no mose tearing to.

And not the last thing, when switching into a fullscreen 3D application (like iPhoto -> Play Slideshow) and back to desktop, my screen is verry garbled and unreadable. I must reboot with the buton on the computer case.

Thanks for this stage of the driver, I wrote these, only to inform you what are my problems.

Is something wrong on my install and settings?

Well, I got the same situation as yours. But I think Callisto did noting for the fullscreen application, Apple did

 

And another thing, when the s-video cable connected with my tv, osx stop booting, with the switch -x -v I found "stv" in the Callisto loading log. (that happened in Alpha001 and it should be same in alpha002 because nothing about it in the changlog )

 

Sorry for my poor English again

[Tek_No]

I really hope Omni is also reading this thread...

 

Release 2 of the driver is going the correct way: although the display is still not usable on my 4c66 (radeon 9200 - HP Pavilion ZT3010EA) I can actually "see" the desktop and even start "System Preferences" etc...

 

Here's the info detected by the driver at startup:

 

May 8 13:56:33 localhost kernel[0]: Callisto: omni's ATI Driver&FrameBuffer, build 002, 07-May-2006

May 8 13:56:33 localhost kernel[0]: Callisto: Model Radeon Mobility 9000 Lf (rs300)

May 8 13:56:33 localhost kernel[0]: Callisto: MMIO range@0x18 (90400000) mapped to kernel virtual address 0x1ed15000

May 8 13:56:33 localhost kernel[0]: Callisto: 32 MB DDR SGRAM found on 64 bit wide bus

May 8 13:56:33 localhost kernel[0]: Callisto: VRAM range@0x10 (98000000:02000000)

May 8 13:56:33 localhost kernel[0]: Callisto: Found ROM @0xc0000

May 8 13:56:33 localhost kernel[0]: Callisto: Legacy ATI BIOS detected

May 8 13:56:33 localhost kernel[0]: Callisto: PLL values detected from Legacy BIOS: ref_clk=2700, ref_div=12, xclk=22000, min_freq=20000, max_freq=35000

May 8 13:56:33 localhost kernel[0]: Callisto: Found LVDS port, previously missing from the BIOS table, adding...

May 8 13:56:33 localhost kernel[0]: Callisto: LCD DDC info table found in BIOS!

May 8 13:56:33 localhost kernel[0]: Callisto: Port[0] DDC: DDC_CRT2, DAC: TVDAC, TMDS: Unknown, Connector: Proprietary

May 8 13:56:33 localhost kernel[0]: Callisto: Port[1] DDC: DDC_VGA, DAC: Primary, TMDS: Unknown, Connector: CRT

May 8 13:56:33 localhost kernel[0]: Callisto: Panel ID string: SEC

May 8 13:56:33 localhost kernel[0]: Callisto: Panel Size from BIOS: 1280x800

May 8 13:56:33 localhost kernel[0]: Callisto: (BIOS) FP: Hdisp=1280, Hblank=160, Hover_plus=48, HsyncWidth=32

May 8 13:56:33 localhost kernel[0]: Callisto: (BIOS) FP: Vdisp=800, Vblank=23, Vover_plus=2, VsyncWidth=6

May 8 13:56:33 localhost kernel[0]: Callisto: (BIOS) FP: Pixel Clock=71000

May 8 13:56:33 localhost kernel[0]: Callisto: Non-DDC laptop panel detected

May 8 13:56:33 localhost kernel[0]: Callisto: Found Monitor: LCD

May 8 13:56:33 localhost kernel[0]: Callisto: done with Radeon Mobility 9000 Lf

 

 

Panel-ID detected is CORRECT (is the exact same driver output as the BeOS Radeon driver is generating)!

 

Main problem seems to be the same as on BeOS some time ago: some wrong timing parameters: I altered the BeOS driver to use the correct pixel clock for my laptop and the screen was correct after this:

 

pixel-clock currently used by Omni's driver = 71000

 

Should be: 70875 (as being listed by Windows app PowerStrip)

 

Is there any display parameter file (see folder /System/Library/Displays/Overrides ) that is being used by Omni's driver?

 

Maybe Omni could release a driver that is reading a small (text???) file in which we could enter the value for our pixel clock so that we can test this this way...

 

As already said: Omni is going the correct way and what is happening is exactly the same as what was happening with the BeOS Radeon driver some time ago... I hope he is reading this

 

Take care,

Tek_No

[moseschrist]

Radeon Mobility 7500:

 

Chipset Model: Radeon Mobility 7500

Type: Display

Bus: AGP

VRAM (Total): 32 MB

Vendor: ATI (0x1002)

Device ID: 0x4c57

Revision ID: 0x0000

Displays:

Color LCD:

Resolution: 1024 x 768 @ 60 Hz

Depth: 32-bit Color

Core Image: Not Supported

Main Display: Yes

Mirror: Off

Online: Yes

Quartz Extreme: Not Supported

 

 

 

i also added the AppleI386PCI.kext to get AGP bus working.

resolution change works great.

 

thanks omni ! great driver release.

keep up the great work.

[cyclonefr]

Xanthrax : it's normal that's the common problem of omni's driver, this should be fixed (i hope) in futher versions. i do have the same problem

Kua4ke4 yes it was u sorry for the spelling but u should have written Quake4

[thedopefishlives]

Omni, you rock my world. My chipset is correctly detected, as is my laptop display panel. It's still impossible to change resolutions, though; no more black screen, but the image goes all to horizontal lines and is split in half - if I'm lucky. If I'm unlucky, the screen goes to garbled black and white. It also still only detects 16M of video RAM.

 

EDIT: First of all, my chipset is a Radeon Xpress 200M (0x5955). I'm just glad that Omni, in his grace, actually decided to support us - one of the first developers I know who does

 

Secondly, I think I've found the issue. Comparing the log of Callisto to the log of my Linux kernel's radeon framebuffer driver, it appears that Callisto's calculation of the pixel clock is off by an order of magnitude (71250 for Callisto vs. 7125 for radeonfb). Either they're measuring in different units, in which case the problem lies elsewhere, or they're measuring in the same units, in which case the pixel clock for Callisto is off on some chips. I'm attaching the radeonfb and Callisto logs for your perusal.

callisto.txt

radeonfb.txt

[Dark_Elf]

Great work Omni

It recognises my Mobility X1400 but still changing resolution doesn't work (I get black screen). I will wait patiently for a further release

[domino]

- ATI 9800 Pro (OEM)

- 4e48

- Samsung Magic SyncMaster CPG21 (T)

 

 Chipset Model:	Radeon 9800 Pro NH
 Type:	Display
 Bus:	PCI
 VRAM (Total):	128 MB
 Vendor:	ATI (0x1002)
 Device ID:	0x4e48
 Revision ID:	0x0000
 Displays:
Display:
 Resolution:	1600 x 1200 @ 75 Hz
 Depth:	32-bit Color
 Core Image:	Supported
 Main Display:	Yes
 Mirror:	Off
 Online:	Yes
 Quartz Extreme:	Supported

 

I get 1600x1200@75, with QE/CI. At first I got 2 screen split. Changed the ref rate to 60, then 75. Looking good.

 

edit1: WOW! I checked sys profiler again and it now says QE/CI supported! Don't know what's up. ??

 

edit2: I can change res and refresh rate. Went from 1600x1200 to 1280x1024 and back. Changed refresh from 85, 75, to 60. My max refresh at 1600x1200 is 75. Max refresh at 1280x1024 is 85.

 

edit 3: screen distortion after reboot

 

thanks omni

Edited May 8, 2006 by domino

[cyclonefr]

- ATI 9800 Pro (OEM)

- 4e48

- Samsung Magic SyncMaster CPG21 (T)

 

 Chipset Model:	Radeon 9800 Pro NH
 Type:	Display
 Bus:	PCI
 VRAM (Total):	128 MB
 Vendor:	ATI (0x1002)
 Device ID:	0x4e48
 Revision ID:	0x0000
 Displays:
Display:
 Resolution:	1600 x 1200 @ 75 Hz
 Depth:	32-bit Color
 Core Image:	Supported
 Main Display:	Yes
 Mirror:	Off
 Online:	Yes
 Quartz Extreme:	Supported

 

I get 1600x1200@75, with QE/CI. At first I got 2 screen split. Changed the ref rate to 60, then 75. Looking good.

 

edit: WOW! I checked sys profiler again and it now says QE/CI supported! Don't know what's up. ??

 

reboot like twice in a row, and you'll see the screen gets sometimes white, sometimes it works, sometimes is distorted as you said... isn't it ?

[greenman100]

Radeon 9700 NE:

 

Chipset Model: Radeon 9700 NE

Type: Display

Bus: PCI

VRAM (Total): 128 MB

Vendor: ATI (0x1002)

Device ID: 0x4e45

Revision ID: 0x0000

Displays:

Display:

Resolution: 1600 x 1200 @ 60 Hz

Depth: 32-bit Color

Core Image: Supported

Main Display: Yes

Mirror: Off

Online: Yes

Quartz Extreme: Supported

 

 

Callisto: omni's ATI Driver&FrameBuffer, build 002, 07-May-2006

Callisto: Model Radeon 9700 NE (r350)

Callisto: MMIO range@0x18 (f9000000) mapped to kernel virtual address 0x2ac69000

Callisto: 128 MB DDR SGRAM found on 128 bit wide bus

Callisto: VRAM range@0x10 (e8000000:08000000)

Callisto: Found ROM @0xc0000

Callisto: Legacy ATI BIOS detected

Callisto: PLL values detected from Legacy BIOS: ref_clk=2700, ref_div=12, xclk=27000, min_freq=20000, max_freq=40000

Callisto: Port[0] DDC: DDC_DVI, DAC: TVDAC, TMDS: Internal, Connector: DVI-I

Callisto: Port[1] DDC: DDC_VGA, DAC: Primary, TMDS: Unknown, Connector: CRT

Callisto: Connection on port 1 not detected, but there is connection on port 2 (swapping 2->1)

Callisto: Found Monitor: CRT

Callisto: done with Radeon 9700 NE

 

 

 

 

 

 

Switchres / modding 9700 kexts.

 

Sometimes takes a couple reboots/res changes to go. Minor mouse tearing.

[domino]

reboot like twice in a row, and you'll see the screen gets sometimes white, sometimes it works, sometimes is distorted as you said... isn't it ?

you're right. Sorry I missed that part of the the discussion on this very long thread .

[krowten]

reboot like twice in a row, and you'll see the screen gets sometimes white, sometimes it works, sometimes is distorted as you said... isn't it ?

 

you're right. Sorry I missed that part of the the discussion on this very long thread .

 

Same result... wait 003

[munky]

havent had a chance to check it out yet. anyone had success with an X1800 PCI-E ?

 

aside from that, great work omni (though I doubt you read this site anymore... )

[XanthraX]

Xanthrax : it's normal that's the common problem of omni's driver, this should be fixed (i hope) in futher versions. i do have the same problem

Kua4ke4 yes it was u sorry for the spelling but u should have written Quake4

 

Thanks. I installed and loaded SwitchResX and all gone fine (QE and CI). The resolution go fine after a reboot, no garbled screen. I wait paciently the new versions. In my previous post, I only want to give information, not criticize. In this step, I am more than satisfied by the capabilities of this version.

 

Ty!

[stmkjp]

rjtd, as i already pm'd u, alpha 2 driver shows up to 1024x768 only. but at least i can change refresh rate which i heard that changing refresh rate of LCD is not important as of CRT's though.

I think W3N users should wait till alpha 3 releases.

 

BTW, anyone tried keynote or other OpenGL application with full screen?

hope the driver also resolved mouse scrambling problem.

 

Greets.

Unfortunatly I cannot try the new Alpha2 driver because I don't have my laptop with me.

But the ones who do have a 9700 mobility or a laptop, can you confirm me that now 1280x768 works or can be choosen?

Thanks

[viperjp]

havent had a chance to check it out yet. anyone had success with an X1800 PCI-E ?

 

aside from that, great work omni (though I doubt you read this site anymore... )

 

X1800 XT PCI-E here but no success yet. When booting without -v I get the "restart your computer" screen. When using -v I see the graphics card gets recognized properly (X1800 XT) but it hangs up at the moment of loading loginwindow.app.

[cyclonefr]

BTW, anyone tried keynote or other OpenGL application with full screen?

hope the driver also resolved mouse scrambling problem.

no mouse scrambling = mouse tearing

[jpsolares]

I think for x300se o x300 the id will 0x5b60.

[stmkjp]

oh cyclonefr, i meant the mouse pointer scrambled to noise-like horizontal lines after keynote presentation issue, not mouse tearing problem.

I alredy heard that mouse tearing still remains if qe&ci enabled.

 

if u tried keynote ( or WOW), then please let me know the result.

 

no mouse scrambling = mouse tearing

[cyclonefr]

oh cyclonefr, i meant the mouse pointer scrambled to noise-like horizontal lines after keynote presentation issue, not mouse tearing problem.

I alredy heard that mouse tearing still remains if qe&ci enabled.

 

if u tried keynote ( or WOW), then please let me know the result.

yes i know, mouse scrambling is the same problem as mouse tearing! comes back when the video try to re-display the mouse pointer.

The only way to avoid this, is if "we" find a solution to not display the mouse cursor and then use mouse locator instead.

see How to remove tearing topic in the genious bar, for futher information.

 

PS: WOW never scrambled the mouse cause doesnt unload the mouse cursor.

[stmkjp]

ok. i found this on omni's site...

------------------------------------------------

Tush Radeon X300 PCIe 0x5b60 LCD

Comment: Works perfectly.There is mouse tearing, but the cursor doesnt fall to pieces after OpenGL games.

[cyclonefr]

just a lie, the cursor still scrambles since i ve just explained u. U can try and see.

This man must have ran WOW like a noob.