Headers for cached page www.cygnus.com/~raeburn/ct5xx.html:
HTTP/1.1 200 OK
Date: Tue, 09 Nov 1999 02:52:56 GMT
Server: Apache/1.3.4 (Unix)
Last-Modified: Sun, 20 Oct 1996 04:32:51 GMT
Connection: close
Content-Type: text/html
Chips & Technologies 655xx video chips
SPONSORED LINKS

Chips & Technologies 655xx video chips

This page is maintained by Ken Raeburn.

This page is (perpetually) under construction. Additional info and/or web pointers are most welcome! Especially for the chips I don't know much about.

Disclaimer: My maintenance of this document has nothing to do with Chips & Technologies, XFree86, any laptop manufacturer, or any other official organization. I'm just putting up what info I know, in the hope that someone else may have to do a bit less searching. Some of it might even be accurate.


Table of contents:

Features

The 655xx series chips are SVGA video controller chips for flat panel displays and CRTs. They also provide some level of CGA, MDA, EGA, and Hercules compatibility, and various accelerator features. They are designed with various features for reducing power consumption and optimizing display quality.

The 65540 and 65545 have programmable clock generator circuits, and the ability to map the full range of video memory at once, instead of using a pair of small banks like most non-accelerated PC video cards do.

The 65545 also has hardware cursor/pop-up support, and a bitblt engine.

The HiQVideo chips -- 65550 and 65554 -- include some new multimedia support, including video capture and display with scaling. A press release from April on the Zoomed Video Initiative mentions this. The 65550 is a HiQV32 chip, and the 65554 is a HiQV64 chip; the numbers apparently refer to the width of the interface between video chip and frame buffer memory. I don't know if the 65554 is available yet.

Systems

These chips are used in several laptop systems including: They are also used in ISA and PCI cards available from a number of companies; some are listed in the contact section later.

X support

XFree86 3.1.2 does include minimal support for the CHIPS 65520, 65530, 65540, and 65545, worked on by Jon Block (block@frc.com), Mike Hollick (hollick@graphics.cis.upenn.edu), and Regis Cridlig (cridlig@dmi.ens.fr). It recognizes these chipsets and will use them as simple, generic SVGA display drivers.

Regis put up a driver for XFree86 3.1 on Linux here, but I haven't checked whether it's got anything that didn't go into the next XFree86 release.

I have a modified version of the XFree86 3.1.1 driver here which will enable the linear addressing mode of the 6554x if you specify

	Options "linear"
in your XF86Config file.

Using the programmable clock, we should be able to generate a wide range of clocks. According to Nelson Minar, the beta-test (or was it demo?) version of the XInside commercial accelerated X server supports at least some higher clock frequencies, up to 65MHz; it's possible their final product might not support this chipset though. I'm working on programming the clock generator to support higher frequencies than the default 25.175MHz and 28.322MHz.

The 65545 also has hardware cursor support and a bitblt engine.

A group of people in Japan, Europe and Australia have been working on supporting these chips under X. The server, and diffs from XFree86 3.1.2, are available for anonymous ftp from ftp://ftp.biol.tsukuba.ac.jp/pub/X/XFree86/alpha/CT65545plus. Their work is more recent than any of the others above. David Bateman and Egbert Eich have done work on 65550 support; that code, and X server binaries, can be found at ftp://ftp.rp.csiro.au/people/dbateman/ct65550.

Bugs

Nelson and I have seen problems where, occasionally, if X is running when the system is suspended, you can't un-suspend it. Or maybe you can, and it just wedges up right away. In any case, sometimes it just dies. (At least, I only see it occasionally; I'm not sure about Nelson.) It doesn't happen if you switch to a virtual console in text mode before suspending, and it doesn't happen under the XInside server. If I can figure out what flag they might be setting differently, I'll see if I can fix it. But I don't even know how to reliably reproduce it at the moment.

There are synchronization problems of some sort at higher resolutions. Regis and Nelson have seen them. Since I use only my lower-resolution flat-panel display, I haven't, so I can't describe them well. (Nelson: Does the XInside server have this problem too?)

XF86Config data

Nelson has a NEC Versa P laptop with a 800x600 256-color LCD screen. Check his web pages for the configuration file he's using with some luck on the LCD and a Sony Multiscan 15sf. Above 704x600 Nelson runs into the synchronization problems, but 704x600 he seems to find reliable.

My Versa M/75C is 640x480 with 64K colors. Here are the "Monitor" section settings I use for the LCD display in 8bpp mode. I don't often use an external monitor.

    Bandwidth	25.2
    HorizSync	30-64
    VertRefresh	50-100

    Mode "640x480-28"
        DotClock	28.322
        HTimings	640 648 688 784
        VTimings	480 488 491 521
    EndMode

Windows/NT/OS2 support

I don't run Windows myself (except sometimes to play with the Dilbert Screensaver), nor NT nor OS/2, so I can't tell you anything about how well these drivers work.

Some Windows, Windows NT, and OS/2 drivers can also be picked up via the web pages at CHIPS.

This info was also sent to me a little while ago. (The email address doesn't appear to work any more.)


    From: Michael Wild <Michael@iee.org>
    Subject: Found : CT655xx drivers for NT
    To: raeburn@raeburn.org
    Date: Tue, 2 Apr 1996 20:55:58 +0100

    I recently spent some while tracking down an NT driver for my Acer
    laptop, which has a CT655xx vid chip. Along the way I encountered
    your excellent page. I noticed you don't offer help on NT, but I
    thought other visitors might like to know the drivers can be found
    at :

    <http://www.pc.ibm.com/cgi-bin/listftp?dir=34&sorttype=date>

    The file you need is :

    vf701nt1.exe Video Drv Windows NT - TP 701
	 (57066 bytes, 10-31-95) 

    This is for NT 3.51. The main advantages over the "generic" driver
    that comes with NT are that it is somewhat faster and supports 256
    colours.

I don't know if the package from IBM just includes a copy of the CHIPS driver, or what... Pier Borra reports that the driver works for his Toshiba 400 laptop as well. However, Jimmy C. Yang points out that this web site is intended for support of IBM's Butterfly laptops, so Chips' web site may be more desirable.

Getting Technical Data

Data sheets for these chips and video BIOS code are available from Chips and Technologies, Inc. Call them (see the contact info below) and ask for product literature.

Some of the data sheets are also available as PDF files via CHIPS' web site and ftp site. If you use ftp directly, instead of a web browser, don't forget to set "binary" mode. (The CHIPS ftp server, whatever it is, doesn't implement the "SYST" request. Many UNIX ftp client will default to ASCII mode -- which tends to trash binaries -- unless the "SYST" request gets a response indicating the server is a UNIX box.)

You'll need a PDF viewer such as xpdf or Adobe Acrobat to view these files. Unfortunately, some of the PDF files look like they were scanned in from hardcopy and converted. So the text is stored as semi-legible raw bits rather than characters, you get smudges, and you can see where the holes were in the original hardcopy.

... And the 65548 PDF file, when I looked at it, appeared to be "encrypted", which means you have to use Adobe's software to look at it. And, it means you're SOL if you're using Linux or NetBSD. I'll keep calling and bugging them for paper copies, thank you very much.

The latest versions I know of are:

part(s) publication number rev rev date notes
65510 DS161.1 0.7 (prelim) October 1992
65520/530 DS152.1 0.7 (prelim) November 1992 available on CHIPS web site
65525 DS163 1.0 (prelim) February 1993 also on web site
65535 DS165.1 2.0 March 1994
65540/545 DS170.2 1.2 October 1995 errata sheets available
65548 DS176.1 1.0 (prelim) August 1995 also on web site
65550 DS177.1 0.7 December 1995 HiQV32
65554 API32 0.1 December 1995 HiQV64 "Advance Product Information"
OC65530 UG113.2? 2.0 June 1993 VGA BIOS reference guide; on web site

Bjoern Kriews <bkr@drdhh.dialup.uni-hamburg.de> reports that CHIPS' German distributor, Rein Components, also has datasheets available.

There is also at least one Video BIOS data sheet available from CHIPS through their ftp site.

Chip availability: The earlier chips (through 65530?) are being phased out; I don't know if they're being used in new boards or systems these days. (Yet the latest datasheets I could get on them were still marked "preliminary"....)

The '545 and '548 datasheets, at least, have some mistakes and omissions. In case you don't get the associated errata sheets for the '540 that I got, here's a quick summary:

There are other minor problems I've run across too, but haven't summarized yet. The mailing list archives mention everything.

Some of the Toshiba laptops use a 65546 chip, which I understand is made exclusively for Toshiba. CHIPS won't give out the technical data on this chip; you'll have to contact Toshiba. I called Toshiba America, and talked to their tech support and laptop development groups. Each told me to talk to the other. One other source I got said he thought it was probably a 65545 with a different pinout, but he wasn't sure.

Other network resources

I've got a mailing list "ct5xx-hackers@cygnus.com" set up.

Some other web pages available:

The code being worked on in Japan can be found at this ftp site.

Also try various newsgroups: comp.windows.x.i386unix, comp.sys.laptops, and some of the Linux newsgroups.

Contact info

Phone numbers without leading "+" are US phone numbers; the "800" numbers may not be reachable from outside the US, so I've included non-800 numbers as well when I've got them.

For chip info:

Chips and Technologies, Inc.
2950 Zanker Road, San Jose, California 95134
Phone: 800-323-4477 or 408-434-0600. Fax: 408-526-2275.
http://www.chips.com/
Rein Components (D-Nettetal)
Phone +49 2153 733-91. Fax -575
ct5xx-hackers-request@cygnus.com
Laptop manufacturers:
Toshiba America (wasn't helpful to me, but I didn't try all that hard)
tech support: 800-999-4273
corporate office: 714-583-3000 (ask for laptop development group)
Video board manufacturers:
Added value of Tustin CA
1-800-439-7073
Aved Display Technologies
14192 Chambers Road, Tustin CA 92680.
Phone: 714-573-5000.
aveddisplay@aveddisplay.com
http://www.aveddisplay.com/
Dolch Computer Systems
3178 Laurelview Court, Fremont, CA 94538.
Phone: 510-661-2220. Fax: 510-490-2360.
http://www.dolch.com/
Earth Computer Technologies
Mintronix Corp.
Camarillo, CA
Phone: 805-482-1298. Fax: 805-987-8088.
mintronix@interramp.com
All their video boards are C&T-based.
Sage Inc
Santa Clara, CA.
Voice: 408-748-0500?, 408-748-8541. Fax: 408-748-8542.
sage@netlynx.com
http://www.netlynx.com/~sage/
Here's the availability of video boards last I checked. Note that not all of this is current, and some of it may be based on second- or third-hand reports. If you've got an update or addition for this table, please let me know.
company and product name date info updated and source 65545 65548 65550
Added value ISA PCI
Aved ? ?
Dolch "Viper" boards 12 July 1996 by phone ISA PCI not planned
Earth Computer Technologies "Earth Vision/ISA" 30 Jun 1996 from web page ISA
Mintronix 12 July 1996 talked to Derek Nguyen PCI PCI (9/96)
Sage Inc 4 Jul 1996 from web page ISA PCI (rumored)

This page was inspired by Nelson's versa-linux page, and the fact that I usually see one or two requests for info each time I go look into comp.windows.x.i386unix (which is about every month or two at most).

Thanks to the people who have contributed information used in this page, major or minor, including: Nathan Bond, Pier Borra (pier7@netcom.com), Regis Cridlig, Rob Howard (grasebyus@aol.com), Stuart Kreitman (skk@benelux.pa.dec.com), Bjoern Kriews, Nelson Minar, Derek Nguyen at Mintronix, Rich Paul (linguist@cyberspy.com), Shrikant (full name??) at CHIPS (shrikant@chips.com), Daniel Vachon (dvac@bighorn.dr.att.com), Michael Wild, Jimmy C. Yang (yangcx@expert.cc.purdue.edu).


Free Speech
Author: Ken Raeburn, raeburn@raeburn.org
Last updated: 7 July 1996