Linux-Development-Apps Digest #410, Volume #6 Tue, 7 Mar 00 19:13:13 EST
Contents:
Re: I can't stand this X anymore! (David T. Blake)
Re: Socket problem ("Ron Weigel")
Re: which ODBC or other API for Linux (Walter Rawdanik)
How can I map kernel space to user space.. (Kevin Shao)
Broadcast Socket ("Ron Weigel")
Re: How to safely interrupt select() ? ("Maxim Udushlivy")
Re: I can't stand this X anymore! (Donovan Rebbechi)
Re: How to safely interrupt select() ? (Mikko Rauhala)
Re: I can't stand this X anymore! (Jan Wielemaker)
multicast router (Ahmed Said)
Re: Interrupt Service Routines (Richard Bonomo)
Re: [Q] Idling with XLib ([EMAIL PROTECTED])
[Q] TickCount equivalent? ([EMAIL PROTECTED])
Re: Socket problem (Andreas Rottmann)
Re: I can't stand this X anymore! (David T. Blake)
Re: I can't stand this X anymore! (Steve Mading)
Re: I can't stand this X anymore! (Christopher Wong)
Re: I can't stand this X anymore! (Christopher Wong)
Re: I can't stand this X anymore! (Donovan Rebbechi)
Re: I can't stand this X anymore! (Christopher Wong)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (David T. Blake)
Crossposted-To: comp.os.linux.x,comp.os.linux.advocacy
Subject: Re: I can't stand this X anymore!
Date: 7 Mar 2000 18:47:15 GMT
Reply-To: [EMAIL PROTECTED]
Eduard Bloch <[EMAIL PROTECTED]> wrote:
> Am 6 Mar 2000 01:30:50 GMT schrieb/wrote/a écrit
> Donovan Rebbechi <[EMAIL PROTECTED]> in comp.os.linux.development.apps:
>
> >The only inadequacy with X is that the font system doesn't support
> >anti-aliasing, ie "font smoothing".
>
> Can we expect any changes in Xfree86-4 or the next major release? I am
> still surprised about the quality of font presentation when working via
> telnet (Putty) from a windows box.
xfstt is available for all non-RedHat X systems as a drop in
replacement for the X font server. There is lots on it on
the web, including several web sites.
There is also Freetype which is a library that provides true
type support for many applications. Of cource, Apple is now
preparing to halt all TrueType font use in linux due to
the patents they hold on TrueType fonts.
Metafonts from TeX are actually better in theory (and more
computationally advanced) than TrueType fonts, but they require
more computational power to render. However, xdvi viewers
tend to look pretty sweet while displaying text in X.
It will come around. If Apple were to licence TrueType patents to
XFree86 than it would come around a lot faster.
--
Dave Blake
[EMAIL PROTECTED]
------------------------------
From: "Ron Weigel" <[EMAIL PROTECTED]>
Subject: Re: Socket problem
Date: Tue, 7 Mar 2000 12:24:09 -0700
Andreas Rottmann wrote in message <[EMAIL PROTECTED]>...
>Andreas Rottmann <[EMAIL PROTECTED]> writes:
>
>Solved! I was reading in the strings via read - which of course read
>pending '\0' charcters too, (more strings than one in a read() call).
>I just give these strings out too and everything is ok.
You have discovered half of the problem. Messages received by a socket can
be combined, so you may receive more than one at a time. The other half of
the problem is splitting of messages. Once messages reach a certain size,
they can be split into multiple packets and received in pieces. Your code
needs to account for both cases. In your case (using '\0' as a delimiter
for ascii strings) if a messge is received with no terminator, it must be
combined with the next message. Try sending very long strings (>2048) to
test for this problem.
Ron
------------------------------
From: Walter Rawdanik <[EMAIL PROTECTED]>
Subject: Re: which ODBC or other API for Linux
Date: Tue, 07 Mar 2000 13:36:11 -0600
Use DBLib that comes with Sybase ASE for Linux. C based and will work with
MS SQL .
You might want to check out source code for X based SQL query tool which
uses DBlib ..
http://www.megsinet.net/~agatka/
Walter
[EMAIL PROTECTED] wrote:
> I looked at some ODBC solutions a while back and found nothing that
> I felt I could use for a number of reasons. I'm interesting in
> looking again, as there seem to now be quite a number of different
> options. However, as before, it sees the documentation is either
> very incomplete on them, or it assumes more knowledge about databases
> and ODBC concepts than I have.
>
> I need to run a client on Linux that accesses data on a Microsoft SQL
> server running on NT. Installing anything on the NT side is unlikely
> to be an option. On the Linux side it should be just a shared library.
>
> I'm not committed to any library. In other words, I'm not porting
> some program that expects to use some particular API. I will be
> writing the code in C so I can adapt to some API's. However, I do
> NOT want any library that runs its own event loop. The event loop
> will be in my code only. It is also preferrable to have asyncronous
> access (e.g. so my code doesn't get blocked in some function call
> while waiting on the database).
>
> None of the documentations I have looked at cover these areas. Now
> there are many more, and it would take quite a long amount of time to
> go digging deep into all of them to discover exactly how they work;
> time that I do not have.
>
> Ideally I would like a standardized protocol for basic database access
> (would not necessarily be capable of advanced proprietary features, but
> I won't be using those anyway).
>
> --
> | Phil Howard - KA9WGN | for headlines that | Just say no to absurd patents |
> | [EMAIL PROTECTED] | really matter: | Boycott Amazon.Com (AMZN) |
> | Dallas - Texas - USA | linuxhomepage.com | Shop http://bn.com/ instead |
------------------------------
From: Kevin Shao <[EMAIL PROTECTED]>
Subject: How can I map kernel space to user space..
Date: Tue, 07 Mar 2000 15:02:45 -0500
Is there any way to map kernel space (which is allocated in the driver)
to user space, or vice versa, so that both driver and the application
can simultaneously access to the same buffer ?
Kevin
------------------------------
From: "Ron Weigel" <[EMAIL PROTECTED]>
Subject: Broadcast Socket
Date: Tue, 7 Mar 2000 13:31:52 -0700
The following code run on NT but get an error 101 - (Network Unreachable)
when executing the sendto on the broadcast socket.
Any ideas?
Ron
#ifdef WIN32
#include <winsock.h>
#define GetSocketError WSAGetLastError
typedef int socklen_t;
#else
#include <unistd.h>
#include <errno.h>
#include <fcntl.h>
#include <unistd.h>
#include <netdb.h>
#include <sys/ioctl.h>
#include <sys/socket.h>
#include <netinet/in.h>
#include <arpa/inet.h>
#define GetSocketError() (errno)
#define SOCKET_ERROR -1
typedef int SOCKET;
typedef struct sockaddr SOCKADDR;
typedef struct sockaddr_in SOCKADDR_IN;
#define FALSE 0
#define TRUE 1
#endif
#include <iostream.h>
#include <stdio.h>
#include <string.h>
int broadcast(void)
{
int err;
SOCKET s;
int sendport = 4000;
SOCKET r;
int receiveport = 4001;
// setup send socket
{
// create socket
if ((s = socket(PF_INET, SOCK_DGRAM, 0)) == SOCKET_ERROR)
{
err = GetSocketError();
return err;
}
// set non-blocking
#ifdef WIN32
unsigned long nonblocking = FALSE;
if (ioctlsocket(s, FIONBIO, &nonblocking) == SOCKET_ERROR)
{
err = GetSocketError();
return err;
}
#else
int flags = fcntl(s, F_GETFL, 0);
if (flags >= 0)
{
flags &= ~O_NONBLOCK;
if (fcntl(s, F_SETFL, flags) < 0)
return errno;
}
#endif
// set broadcast
int optval = TRUE;
if (setsockopt(s, SOL_SOCKET, SO_BROADCAST, reinterpret_cast<const
char*>(&optval), sizeof(optval)) == SOCKET_ERROR)
{
err = GetSocketError();
return err;
}
}
// setup receive socket
{
// create socket
if ((r = socket(PF_INET, SOCK_STREAM, 0)) == SOCKET_ERROR)
{
err = GetSocketError();
return err;
}
// set non-blocking
#ifdef WIN32
unsigned long nonblocking = FALSE;
if (ioctlsocket(s, FIONBIO, &nonblocking) == SOCKET_ERROR)
{
err = GetSocketError();
return err;
}
#else
int flags = fcntl(r, F_GETFL, 0);
if (flags >= 0)
{
flags &= ~O_NONBLOCK;
if (fcntl(r, F_SETFL, flags) < 0)
return errno;
}
#endif
// set reuse address
int optval = TRUE;
if (setsockopt(r, SOL_SOCKET, SO_REUSEADDR, reinterpret_cast<const
char*>(&optval), sizeof(optval)) == SOCKET_ERROR)
{
err = GetSocketError();
return err;
}
SOCKADDR_IN sockAddr;
memset(&sockAddr, 0, sizeof(sockAddr));
sockAddr.sin_family = AF_INET;
sockAddr.sin_addr.s_addr = htonl(INADDR_ANY);
sockAddr.sin_port = htons((u_short)receiveport);
// bind to receive port
if (bind(r, reinterpret_cast<SOCKADDR*>(&sockAddr), sizeof(sockAddr)) ==
SOCKET_ERROR)
{
err = GetSocketError();
return err;
}
int backlog = 5;
if (listen(r, backlog) == SOCKET_ERROR)
{
err = GetSocketError();
return err;
}
}
char sendbuf[100];
sprintf(sendbuf, "%d test 0 0", receiveport);
cerr << "Send: " << sendbuf << endl;
{
SOCKADDR_IN sockAddr;
memset(&sockAddr, 0, sizeof(sockAddr));
sockAddr.sin_family = AF_INET;
sockAddr.sin_addr.s_addr = htonl(INADDR_BROADCAST);
sockAddr.sin_port = htons(sendport);
int flags = 0;
int bytes = sendto(s, reinterpret_cast<const char*>(&sendbuf),
strlen(sendbuf)+1, flags,
reinterpret_cast<SOCKADDR*>(&sockAddr), sizeof(sockAddr));
if (bytes == SOCKET_ERROR)
{
err = GetSocketError();
return err;
}
}
SOCKET c;
{
SOCKADDR_IN sockAddr;
socklen_t len = sizeof(sockAddr);
memset(&sockAddr, 0, sizeof(sockAddr));
if ((c = accept(r, reinterpret_cast<SOCKADDR*>(&sockAddr), &len)) ==
SOCKET_ERROR)
{
err = GetSocketError();
return err;
}
}
char receivebuf[1024];
{
int flags = 0;
int bytes = recv(c, receivebuf, sizeof(receivebuf), flags);
if (bytes == SOCKET_ERROR)
{
err = GetSocketError();
return err;
}
}
cerr << "Receive: " << receivebuf << endl;
return 0;
}
int main(int argc, char* argv[])
{
#ifdef WIN32
WSADATA wsaData;
WORD wVersionRequested = MAKEWORD(1, 1);
if (WSAStartup(wVersionRequested, &wsaData) != 0)
return -1;
#endif
int err = broadcast();
if (err != 0)
cerr << "Error " << strerror(err) << endl;
#ifdef WIN32
WSACleanup();
#endif
return 0;
}
------------------------------
From: "Maxim Udushlivy" <[EMAIL PROTECTED]>
Subject: Re: How to safely interrupt select() ?
Date: Tue, 7 Mar 2000 23:37:16 +0300
Reading from stdin is unacceptable, beacause this is a GUI program. I am
newcomer in Linux, and I see that there is no analogue of Win32
WaitForMultipleObjects function, which can wait both on files and some other
manually resettable events. Or may be I'm reading old documentation...
"fred smith" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Why not put:
> fileno(stdin)
>
> as one of the file descriptors being watched by select() ?? Then whenever
> it returns you can easily tell if the user hit a key or if some other
> event occurred.
>
> Fred
------------------------------
From: [EMAIL PROTECTED] (Donovan Rebbechi)
Crossposted-To: comp.os.linux.x,comp.os.linux.advocacy
Subject: Re: I can't stand this X anymore!
Date: 7 Mar 2000 20:37:37 GMT
On 7 Mar 2000 18:47:15 GMT, David T. Blake wrote:
>xfstt is available for all non-RedHat X systems as a drop in
>replacement for the X font server. There is lots on it on
>the web, including several web sites.
Are you sure you don't mean xfsft ? xfstt is only truetype AFAIK.
>type support for many applications. Of cource, Apple is now
>preparing to halt all TrueType font use in linux due to
>the patents they hold on TrueType fonts.
References ? AFAIK, the patents exist, but no threats have been made.
I don't see why Apple would be out to get Linux, especially given
that they were actively supporting MkLinux some time back.
>Metafonts from TeX are actually better in theory
... and in practice. Computer Modern blows anything available in TrueType
out of the water.
>It will come around. If Apple were to licence TrueType patents to
>XFree86 than it would come around a lot faster.
It already has come around. xfsft which is essentially xfs patched to
use freetype is in the 3.9x releases of XFree.
--
Donovan
------------------------------
From: [EMAIL PROTECTED] (Mikko Rauhala)
Subject: Re: How to safely interrupt select() ?
Date: 7 Mar 2000 21:26:47 GMT
Reply-To: [EMAIL PROTECTED]
On Tue, 7 Mar 2000 23:37:16 +0300, Maxim Udushlivy <[EMAIL PROTECTED]>
wrote:
>Reading from stdin is unacceptable, beacause this is a GUI program. I am
>newcomer in Linux, and I see that there is no analogue of Win32
>WaitForMultipleObjects function, which can wait both on files and some other
>manually resettable events. Or may be I'm reading old documentation...
Well, at least Gtk+ provides facilities to monitor file descriptors
as well as GUI events. I don't know about other toolkits as such but
would find them lacking if they didn't. See
<URL:http://www.gtk.org/tutorial/gtk_tut-17.html#ss17.2>
--
Mikko Rauhala - [EMAIL PROTECTED] - http://www.iki.fi/mjr/
------------------------------
From: [EMAIL PROTECTED] (Jan Wielemaker)
Crossposted-To: comp.os.linux.x,comp.os.linux.advocacy
Subject: Re: I can't stand this X anymore!
Date: 7 Mar 2000 21:04:10 GMT
In article <[EMAIL PROTECTED]>,
Donovan Rebbechi wrote:
>
>On Tue, 7 Mar 2000, Lyle R. Taylor wrote:
>
>> I would not say that lack of font smoothing is the only inadequacy of X.
>> It's weakest link appears to me to be the whole font system.
>
>I'd agree that the lack of a unified print/display system is a problem, but
>not necessarily a problem with X.
I think you're right that it is not an X-problem. X should remain
what it is: a remote protocol to handle screen, mouse and keyboard. This
approach allows for a layered API, which is the best option to allow for
distributed development and improvement.
Printers are another beasts and can properly be handled by the GUI library
at hand. The standard on Unix is the printer language (PostScript), which
at least allows me to deliver a document in printable format.
Out-of-the-box and widely available support for TrueType (or other
high-quality scalable fonts) would be nice though.
Regards --- Jan
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Jan Wielemaker Author of SWI-Prolog and the XPCE GUI library
SWI, University of Amsterdam http://www.swi.psy.uva.nl/projects/SWI-Prolog/
E-mail: [EMAIL PROTECTED] http://www.swi.psy.uva.nl/projects/xpce/
------------------------------
Date: Tue, 7 Mar 2000 23:53:51 +0200
From: Ahmed Said <[EMAIL PROTECTED]>
Subject: multicast router
hi all,
how do i configure linux to be a multicast router ?
thanks in advance.
=================================================================================
Testing can prove the presence of bugs, but never their
absence.
Edsger Dijkstra
------------------------------
From: Richard Bonomo <[EMAIL PROTECTED]>
Subject: Re: Interrupt Service Routines
Date: Tue, 07 Mar 2000 16:10:34 -0600
Atle wrote:
>
> Richard Bonomo wrote:
> >
> > From the feedback I am getting, I am developing the impression that
> > under Linux you can not do this, and I that I must write use
> > a device driver if I wish to employ the interrupts at all.
> >
> > Is this your understanding?
> Uh - let's put it this way: I haven't thought about it. If I have to
> think about it, I would say, well, I guess it would work.
> But here's the good news: What you are trying to do has already been
> done!
> And even if you wanted to do it to gain insight in the workings, you
> should get this package, because if was developed for a class of
> students who wanted to examine the parallel port.
>
> Feel free to drop me an email, and I will scan the archives to see if I
> can find it ...
>
> Atle
OK. Will e-mail direct.
Rich
--
************************************************
Richard Bonomo
UW Space Astronomy Laboratory
ph: (608) 263-4683 telefacsimile: (608) 263-0361
SAL-related email: [EMAIL PROTECTED]
all other email: [EMAIL PROTECTED]
web page URL: http://www.cae.wisc.edu/~bonomo
************************************************
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: [Q] Idling with XLib
Date: Tue, 07 Mar 2000 22:11:34 GMT
In article <[EMAIL PROTECTED]>,
[EMAIL PROTECTED] wrote:
> Dave Lawrence <[EMAIL PROTECTED]> writes:
>
<Snip>
>
> Qt should be a more portable framework. There are other cross-platform
> toolkits (but I don't remember their names - you could try searching
on
> freshmeat with the keyword "cross" or something like that).
>
> --
> /Kinkie
>
> Se sulla scatola c'e` scritto "Per windows 95 e superiori", dovrebbe
> funzionare sotto Linux, vero?
>
Thanx for the feedback guys.
-Ian
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: [EMAIL PROTECTED]
Subject: [Q] TickCount equivalent?
Date: Tue, 07 Mar 2000 22:18:36 GMT
I'm looking for a equivalent to GetTickCount() in XLib. The closest I
was able to find was getrusage() but I don't think that is what I want.
-Ian
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: Andreas Rottmann <[EMAIL PROTECTED]>
Subject: Re: Socket problem
Date: 07 Mar 2000 22:49:14 +0100
"Ron Weigel" <[EMAIL PROTECTED]> writes:
> Andreas Rottmann wrote in message <[EMAIL PROTECTED]>...
> >Andreas Rottmann <[EMAIL PROTECTED]> writes:
> >
> >Solved! I was reading in the strings via read - which of course read
> >pending '\0' charcters too, (more strings than one in a read() call).
> >I just give these strings out too and everything is ok.
>
>
> You have discovered half of the problem. Messages received by a socket can
> be combined, so you may receive more than one at a time. The other half of
> the problem is splitting of messages. Once messages reach a certain size,
> they can be split into multiple packets and received in pieces. Your code
> needs to account for both cases. In your case (using '\0' as a delimiter
> for ascii strings) if a messge is received with no terminator, it must be
> combined with the next message. Try sending very long strings (>2048) to
> test for this problem.
>
Thanks, I already realized this.
Andy
--
Andreas Rottmann (Dru@ICQ, 54523380@ICQ)
Pfeilgasse 4-6/725, A-1080 Wien, Austria, Europe
http://www.altern.org/arot/
mailto:[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (David T. Blake)
Crossposted-To: comp.os.linux.x,comp.os.linux.advocacy
Subject: Re: I can't stand this X anymore!
Date: 7 Mar 2000 22:35:34 GMT
Reply-To: [EMAIL PROTECTED]
Donovan Rebbechi <[EMAIL PROTECTED]> wrote:
> On 7 Mar 2000 18:47:15 GMT, David T. Blake wrote:
>
> >xfstt is available for all non-RedHat X systems as a drop in
> >replacement for the X font server. There is lots on it on
> >the web, including several web sites.
>
> Are you sure you don't mean xfsft ? xfstt is only truetype AFAIK.
For sure I did.
> >type support for many applications. Of cource, Apple is now
> >preparing to halt all TrueType font use in linux due to
> >the patents they hold on TrueType fonts.
>
> References ?
www.freetype.org
"
We are now in contact with Apple's legal department to discuss
the TrueType patents issues with them. However, we'll be unable
to comment on the current discussion until they take an official
position. This could take some time, so don't expect anything too
soon.
Even though we're unable to tell for now whether FreeType
violates them, this could mean that the free use of the library
could be illegal in the US or other countries like Japan, be it
in commercial or free projects.
For a more detailed description of the issues at stake, have a
look at our TrueType patents pages.
"
> AFAIK, the patents exist, but no threats have been made. I
> don't see why Apple would be out to get Linux, especially given
> that they were actively supporting MkLinux some time back.
There is an enormous difference between making hardware
specs available and supporting linux. Seen quicktime players
for linux lately ??
To add to this Microsoft currently pays a licensing fee
to Apple for use of TrueType fonts in Windows. It is also
highly unlikely that Freetype is legal, and that SOME
licensing would have to be arranged (or else Apple would lose
royalties from others that pay it to use TrueType).
The Freetype developers seem to think they have reverse
engineered the patents by only using the published specs -
but I think they are clearly wrong. Just because something
is a published spec does NOT mean the published spec is not
patented (RSA comes to mind).
Think of all the computer use that depends on TrueType fonts,
and the licensing that must be getting paid to Apple. Adobe uses
them in everything. Microsoft uses them in everything. Virtually
every Windows application that renders fonts uses the algorithm.
This is not chump change for Apple to throw out to the free
software community.
> >Metafonts from TeX are actually better in theory
>
> ... and in practice. Computer Modern blows anything available
> ... in TrueType out of the water.
But they are computationally more complex, and the rendering
takes more CPU cycles and code. There was a nice interview
with Knuth about this - TreuType fonts are at least in part
derivative of metafonts. It'd be interesting to see
metafonts become popular if use of freetype became illegal.
Knuth interview:
http://www.advogato.org/article/28.html
> >It will come around. If Apple were to licence TrueType patents
> >to XFree86 than it would come around a lot faster.
>
> It already has come around. xfsft which is essentially xfs
> patched to use freetype is in the 3.9x releases of XFree.
Which is almost certainly illegal, the fallout of which we
can only just barely currently grasp. What will Apple decide
to do ? Give up the royalties, or give up the patents ? I see
almost no way they are biased to give up the royalties. Patents
have been violated, and patent holders do not benefit from
giving the patents away after someone violates them.
--
Dave Blake
[EMAIL PROTECTED]
------------------------------
From: Steve Mading <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.x,comp.os.linux.advocacy
Subject: Re: I can't stand this X anymore!
Date: 7 Mar 2000 23:12:20 GMT
In comp.os.linux.advocacy Jan Wielemaker <[EMAIL PROTECTED]> wrote:
: I think you're right that it is not an X-problem. X should remain
: what it is: a remote protocol to handle screen, mouse and keyboard.
IMO it should be abstracted to "user interactive input/output devices"
instead of specifically iterating just the three "screeen, mouse, and
keyboard". Why? Because I'd like to see the ability to add remote X
support for things like sound and whatever other devices might come
our way some day. The fact that X is restricted to only the three
main devices is one of its few drawbacks. It was meant to be a wholly
remotable user interface, but it ignores some interface devices.
And printers too, because of the fact that the fonts where the user
sits (X server) might not be the same as where the program sits.
--
-- ------------------------------------------------------------------
Steven L. Mading at BioMagResBank (BMRB). UW-Madison
Programmer/Analyst/(acting SysAdmin) mailto:[EMAIL PROTECTED]
B1108C, Biochem Addition / 433 Babcock Dr / Madison, WI 53706-1544
------------------------------
From: [EMAIL PROTECTED] (Christopher Wong)
Crossposted-To: comp.os.linux.x,comp.os.linux.advocacy
Subject: Re: I can't stand this X anymore!
Reply-To: [EMAIL PROTECTED]
Date: Tue, 07 Mar 2000 23:31:02 GMT
On 7 Mar 2000 07:05:05 GMT, Boudewijn Rempt <[EMAIL PROTECTED]> wrote:
>
>Your comments were spot on, except for one thing: the TrueType font
>format provides for hints that are especially helpful for rendering
>on small resolution displays. Not all TrueType fonts use those, and
>a lot of those who do are autohinted by the font editing program, but
>the Monotype fonts Microsoft uses are superbly hinted. I wonder, would
>it be acceptable to take the PostScript fonts included with XFree and
>try to improve them? I've got access to some good font design software,
>myself.
I did mention hinting in my post that you responded to. I doubt if
improving the Type 1 fonts in XFree86 will help. As I said earlier, the
Type 1 rasterizer that comes with XFree86 is crap. You can try
contrasting what xpdf or Ghostview can do with what Adobe Acrobat (which
has its own rasterizer) can do. Even with anti-aliasing turned off,
acroread looks far better.
Chris
------------------------------
From: [EMAIL PROTECTED] (Christopher Wong)
Crossposted-To: comp.os.linux.x,comp.os.linux.advocacy
Subject: Re: I can't stand this X anymore!
Reply-To: [EMAIL PROTECTED]
Date: Tue, 07 Mar 2000 23:46:18 GMT
On 7 Mar 2000 05:54:29 GMT, Donovan Rebbechi <[EMAIL PROTECTED]> wrote:
>On Tue, 07 Mar 2000 04:28:06 GMT, Christopher Wong wrote:
>>On 6 Mar 2000 21:30:08 GMT, Donovan Rebbechi <[EMAIL PROTECTED]> wrote:
>
>>I don't think it would be easy to make such an assertion without some
>>screen shots. Fortunately, there are some here:
>>
>> http://www.dcs.ed.ac.uk/home/jec/programs/xfsft/renderers.html
>>
>>TT fonts look great, but Type 1 fonts look awful.
>
>No, one TTfont looks great. ( Monotype Times New Roman ). The page you
>cite says, and I quote:
>
> The rendering of a TrueType font depends crucially on the
> quality of the font. Those samples use Monotype Times New
> Roman, as included by Microsoft in their `core fonts'. Don't
> expect to find many TrueType fonts that display that well!
I was addressing your point which had the phrase "in practice". In
practice, TrueType on X is far superior to Type 1 on X, as that screen
shot amply shows. Microsoft has quite a selection of good fonts in their
free package: Andale Mono, Trebuchet MS, Georgia, Verdana, Comic Sans,
Arial, Arial Black, Impact, Times New Roman and Courier New. These 10 font
families are all well-hinted and professionally designed. It is not like
I need anything more. I don't care much about what is generally out
there, but what is specifically attainable on my Linux machine. Nothing
beats TT at this point.
>It's true that the MS fonts are very good. However, once you start going
>outside the realm of these special fonts designed especially for onscreen
>viewing / web browsing, you will find that the difference is less visible.
>In other words, if you want to load up a gazillion fonts, it doesn't really
>matter whether you go with TT or Type1.
In practice: if I "load up a gazillion fonts", there are two
scenarios. If I go with TT, I get excellent fonts most of the time and
ugly fonts some of the time. With Type 1, I get ugly fonts all of the
time. I think it does matter which one I would go with.
>>If you know how to get great-looking Type1 fonts, please share the
>>info and screen shots.
>
>See my font HOWTO. If you want a decent Type1 collection, get the
>bitstream 500 font CD. Or convert your TrueType to Type1 ( I bet you wont
>notice much difference ).
I would need more than a decent Type 1 font. I need a decent Type 1
rasterizer: the one that comes with XFree86 is crap. Thanks to TrueType,
I know what well-hinted, scalable fonts can look like. While I do not
have a TT to Type 1 converter, I doubt if I can get even one Type 1 font
that looks comparable to my TrueType fonts. I'll try your font RPM,
nevertheless (thanks). It's worth looking at.
Chris
------------------------------
From: [EMAIL PROTECTED] (Donovan Rebbechi)
Crossposted-To: comp.os.linux.x,comp.os.linux.advocacy
Subject: Re: I can't stand this X anymore!
Date: 7 Mar 2000 23:47:50 GMT
On Tue, 07 Mar 2000 23:31:02 GMT, Christopher Wong wrote:
>On 7 Mar 2000 07:05:05 GMT, Boudewijn Rempt <[EMAIL PROTECTED]> wrote:
>I did mention hinting in my post that you responded to. I doubt if
>improving the Type 1 fonts in XFree86 will help. As I said earlier, the
>Type 1 rasterizer that comes with XFree86 is crap. You can try
However, freetype are working on a Type1 rasteriser, which will probably
eventually replace the one currently used.
--
Donovan
------------------------------
From: [EMAIL PROTECTED] (Christopher Wong)
Crossposted-To: comp.os.linux.x,comp.os.linux.advocacy
Subject: Re: I can't stand this X anymore!
Reply-To: [EMAIL PROTECTED]
Date: Tue, 07 Mar 2000 23:48:05 GMT
On 7 Mar 2000 05:55:38 GMT, Donovan Rebbechi <[EMAIL PROTECTED]> wrote:
>On Tue, 07 Mar 2000 04:28:06 GMT, Christopher Wong wrote:
>
>>info and screen shots. That way, we can get a unified display and print
>>font system (Ghostscript does a horrible job with TT fonts). Without
>
>If your printer groks postscript, or some other language that can handle
>embedded fonts, ghostscript does not render the fonts. Your printer does.
If I throw out a bundle of extra cash to get a Postscript-capable
printer, then I get Type 1 fonts that look awful onscreen but great on
paper, and TrueType fonts that look great onscreen but awful on paper
(if at all).
Chris
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list (and comp.os.linux.development.apps) via:
Internet: [EMAIL PROTECTED]
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Development-Apps Digest
******************************
