Subject: Re: 21 inch monitor.... which one?
From: "Brian Hogan"
Date: 11 Jan 1997 04:51:24 GMT
Novak,
Thanks for the good evaluation of the Mitsubishi 91TXM. I've tried one
ViewSonic P815 and three Hitachi Elite 802's. I completely agree with
everything you said about the ViewSonic! My screen had darker areas, the
geometry was far from perfect, the colors were faded, and the controls
sucked!
The three Hitachi's I've had have been much better than the P815. Controls
were good, colors were brighter than average, and the clarity was pretty
good, even at 1600x1200. I've had problems with "brightness bleeding" --
bright objects on a dark screen produce light shadows or hazing.
I've been considering sending my current Hitachi back again (to NECX) to
get a completely new monitor, and the Mitsubishi was at the top of the
list.
A few questions:
1. How clear is small text at 1600x1200? (better than the Panasonic, fuzzy,
etc.)
2. How is the convergence? (that's what the H Stat is for, to adjust the
convergence of the electron guns)
Please post any response back to this newsgroup.
Brian
novak@nuc.net wrote in article <32e8ea23.51828338@news.alt.net>...
>
> >Sorry to post something off-topic (or likly to be off-topic). I'm
> >considering buying a 21 inch monitor. Having a hard time finding recent
> >tests/comparisons I'm turning to people that are likely to have them.
> >I'm currently looking at these:
> >
> > Nanao (Eizo) F2-21
> > Viewsonic P815
> > Hitachi AccuVue 21 (Mod 4921 asymmetric grille (?))
> >
> >Anybody have those? Good, bad, neat, nice, fuzzy, bright, bright colors?
> >Buy, stay away, for the money..ok?
> >
> >In case you have any other brand/model that can do 1600x1280 at 74+ Hz
> >let me know.
> >Please reply via e-mail. Thanks, MIKE...
>
> I first ordered a Panasonic S21. It is basically the same monitor as
> the Viewsonic P810 (Panasonic makes many of Viewsonic's monitors).
> Similarly, the P815 is basically the same as the Panasonic P21. All
> of these monitors use the same tube; have the same OSD; weigh,
> measure, and spec almost exactly the same; and *basically* they have
> the same case (although it looks a little different cosmetically).
>
> Anyway, I sent the S21 back for a Mitsubishi 91TXM. I'm glad I did.
>
> I think the Mitsubishi is a better monitor in every way except for it
> being heavier and for it having the two support wire shadows being
> visible. The Mitsubishi 91TXM uses the new .28mm DiamondTron tube.
Subject: Re: 21 inch monitor.... which one?
From: novak@nuc.net (novak@nuc.net)
Date: Sat, 11 Jan 1997 06:33:43 GMT
>Wow, that was a long post. Thanks. I take it you're about as
>enthusiastic about your 91TXM as I am! I searched for over a year
>before I laid down my hard earned dollars for a 21". I haven't seen
>the Panasonic here in Canada, and I was wondering if I had missed
>something in view of the rave reviews of that monitor I've seen
>recently in mags. Now I feel a lot better.
>
>I can concur with just about all the things you've observed with your
>monitor. So I'll just snip most of your post and add little things.
>
>Firstly, I'll relay my experience.
>
>I've seen the 21TX (previous generation 91TXM with .30mm instead of
>28mm). It looked good but a blurry in small text because of the
>30mm. I've seen a demo of the Nokia 445xi side by side the
>Nanao/Eizo FX2-21. I'm sorry to say the Nokia was way out of its
>league. The Eizo was nice, 1600x1200@80hz, .25mm but it had a lot of
>Moire pattern and the colors were just so-so. Maybe if I had more
>time I could have changed the colors to look better. But it was the
>most expensive at $3900 (new price, used to be $4300) Canadian. I
>also looked at the top of the line Hitachi Accuvue 49?? on another
>occasion, not side by side the Nanao. It was $3700 and I would say
>really close to the Nanao. I believe the Nanao in fact used the same
>Hitachi tube. Both of these looked really sharp in text quality. All
>demos were done at 1600x1200 by the way. Oh yes, the Hitachi claimed
>.22mm HORIZONTAL pitch, which is actually .25mm dot pitch measured the
>conventional way.
Are you sure that it is .25mm and not .27 or .28mm?
Check out these interesting posts by Bill Nott:
_______________________________________
He's not totally right - many designs use electromechanical relays to
change some circuit components related to different scan rates, and
that's the "clicking" which we hear. There is really no reason to
activate the degauss upon mode switch, although a few designs
apparently do so. As Sam indicates, you should hear both the click
*and* the hum if it's degaussing. I have not evaluated any design
where the degauss activates with a mode change, *but* I have evaluated
many designs which used the relays (and some Compaq monitors *do*
click when the mode is changed, but not all of them). More and more
designs are moving to solid state switches, now, but we're still
likely to hear about "clicking" for the next few years.
Regards,
Bill Nott - Sr. Monitor Engr.
Hitachi NSA and Hitachi America represent two different Hitachi
development & Mfg sites in Japan. Given the need to choose, I'd
tend toward the Hitachi America unit, as it is designed & built
at the Omika works, where they concentrate more on quality and
reliability. Both groups use Hitachi CRT's, which all come from
the same (Mobaru works) design & mfg. location.
In a previous life, I had extensive dealings with both
operations - but things may have changed over tha last 5 or 6
years. As always, there will be individual variations in each
monitor which fall within the manufacturing specification tolerances,
but may be objectionable to some users, so I always advise to
try and actually test the very unit you plan to purchase, if at
all possible (I don't favor mail ordering monitors unless you're
willing to deal with the possibility of needing to return an
unfavorable unit, perhaps multiple times).
Regards
Bill Nott - Sr. Monitor Engr (for more yrs than I care to admit)
0.22/0.16 mm dot pitch, incidentally, is 0.28 measured the
CLASSICAL way - don't let Hitachi get away with this travesty!!
regards,
Bill
_______________________________________
>I settled on the Mitsubishi 91TXM. It was just $2300, almost half the
>price of the Nanao. I think it has better colors. Yes, it has the
>damping wires but you get used to that quickly and won't even see it.
>According to calculations, a true 1600x1200 21" monitor requires .25mm
>to display all the dots. But let me tell you I run mine at 1600x1200
>and the text looks great. I swear it looks like .25mm. It goes to
>show that it's hard to compare stripe pitch to dot pitch.
I agree. The Panasonic S21 has a .25mm shadow mask tube, and I think
the .28mm DiamondTron tube of the Mitsubishi 91TXM looks sharper and
has richer colors.
>I did have a problem though. I started out at 1280x1024@85HZ.
>Things looked great. When I tried 1600x1200, I just wasn't happy with
>the sharpness of the text. I was blaming it on the .28mm. But I
>decided that before I throw in the towel, I should try a BNC cable.
>The wholesaler told me it wouldn't make any difference. Well, let me
>tell you he was dead wrong. I paid $75 for the cable and it REALLY
>cleaned up the text at 1600x1200@75hz. I was amazed. So Mitsubishi,
>if you're listening, trash the vga cable you ship with the monitor and
>give us a decent cable please.
I'm going to get a BNC cable as well. Hopefully, it will be even
sharper like yours is when I do. Did you send off for the optional
serial cable to use their windows software? If so, how does it work?
>>distracting. Also, I had to turn up the brightness and contrast all
>>the way up (the video level was set highest so that wasn't an issue)
>>to get it bright enough to be about as bright as I'd ever want it (the
>>Mitsubishi can go beyond that).
>
>Yes, the Mitsubishi has the brightest display by far. The Hitachi and
>Nanao were dim.
Really... the Hitachi and Nanao were dim eh. I think the brightness
of the 91TXM is a big plus.
>>The Mitsubishi was much less blotchy (with darker areas) than the
>>Panasonic. However, the Mitsubishi can be noticeably brighter around
>>the edges. For example, looking at an empty "black" DOS screen with
>>the brightness turned up past 15% (10% is all I use since it is a very
>>bright monitor and I like empty DOS screens to look *black*), the
>>center looks blacker than much of the rest of the screen where it
>>looks "grayer\brighter" (as is the effect with an empty DOS screen
>>when the brightness is turned up). When displaying color the brighter
>
>On mine, the only time I can tell the edges (about 1") are brighter is
>when I have a screen saver running with mostly a black screen and the
>room's light is out. Then I would notice the black screeen is
>actually not totally black, the edges are brighter still. But I
>usually do my work with light in the room. :)
Looks like that effect may be common with the 91TXM. I agree that it
isn't really noticeable most of time, and to me, it isn't really
distracting like the blotchy areas of the S21 was. I noticed that one
of the reviews I linked to that post said that the 91TXM seemed more
evenly lit that the Trinitron monitor he compared it to so maybe the
Mitsubishi is really good in that area for a Trinitron type monitor.
>>Something I don't like about the Mitsubishi is that it displays the
>>OSD for 5 seconds (defaulted to the power saving on\off function) when
>>it doesn't receive a signal before powering down. That means that I
>>see the OSD every time I reboot the machine for a second or two when
>>there is no signal. I wish they had an option for turning that off.
>>Also, there is a brief moment when Win95 is loading that there was no
>>signal to my monitor and the OSD displays for a second or two there as
>>well so I end up seeing it twice during a reboot.
>
>I could be wrong, but the only time I seem to see this is when I drop
>power to the pc but not the monitor. I run Win95 and this doesn't
>happen to me. I'm using a millenium. Maybe it's card dependent?
It could be. The Verite is not really a traditional video card.
>>The viewing area of the Panasonic is slightly bigger than the
>>Mitsubishi with the Mitsubishi being around 19.7" viewable (AFAIK,
>>Sony's Trinitron tubes of this size are only 19.2" viewable, BTW) and
>>the Panasonic being around 20" viewable. However, I like that the
>>http://www.mitsubishi-display.com/
>
>Is this true? I thought the Mit was actually bigger in surface area
>because it was more square than the others. I don't have a Panasonic
>to measure though.
The Mitsubishi is flatter than the Panasonic, but the Panasonic can
display 20" while the Mitsubishi can only display up to 19.7". Not
really an issue in my book. The display size is practically the same
and about as big as any 21" monitor out there.
>One thing I'll add which to me was a REAL PLUS was the size of the
>Mitsubishi. It was under 20" in all three dimensions. I'm tight on
>space. The other monitors were all 3" deeper. I swear the Mit looks
>like a 17" until you see the front. I just looks all screen. Awesome!
>:)
Hehe. Yeah, it is pretty impressive looking. It may not take up a
lot of space, but it is a good 15 pounds heavier than the Panasonic
S21. The quality and sturdy construction of the Mitsubishi is really
evident, IMO.
Just to digress for a moment, I also have had one of Mitsubishi's top
VCRs for a couple years. When looking under the cover I found out
that it's entire chassis is made of machined aluminum and it uses
metal gears and other quality component throughout. It rewinds a 6
hour EP tape in less than one and a half minutes! The picture quality
is excellent and flawlessly starts recording to a tape on the same
frame I stopped it on (it is an editing deck). Mitsubishi's products
are excellent, IMO. I really feel they have a knack for building
quality equipment that is built to last (although the front button
cover of the VCR is too flimsy, overall it is an excellently made
piece of equipment, IMO).
>BTW, you've got the Verite card. How do you hook that up? Daul
>input? How does it work out? If this is one of those video bypass
>cards, does it degrade the signal?
The Verite is an all-in-one video card. It works like most other
video cards and no pass-through is used. I normally run at
1152x864x16bit@85hz in Win95, and the picture quality is very good.
When I see a 3D accelerated game on this big monitor I can hardly
contain myself. If you don't have a 3D accelerator and you like
games, buy a Voodoo or Verite board and some 3D accelerated games, you
won't regret it. The Verite isn't capable of 1600x1200 so a Voodoo
add-on board such as Orchid's Righteous 3D sounds like a better choice
for you. The BNC\D-Sub button would work perfectly for switching
between the Voodoo and your 2D card so that you wouldn't even have to
use the pass-through. Take care.
_ __ __
/ | / / ____ _ __ ____ _ / /__
/ |/ / / __ \ | / // __ `/ / //_/
/ /| / / /_/ / |/ // /_/ / / ,<
/_/ |_/ \____/|___/ \__,_/ /_/|_|
Subject: Re: Heap Error in AutoCAD R13c4
From: Joćo Salgueiro
Date: Fri, 10 Jan 1997 15:05:38 +0100
David Thomas wrote:
>
> Occassioanlly on some of the machines we are using at work (Pentiums
> with between 32 to 64 MB RAM under Windows 3.1) we get a Heap Error
> Message. At one time (I think with version 12) we made some changes in
> the config.sys file--EMS switch change. Any ideas on what is causing
> this? Possible solutions. All help appreciated!
>
> David Thomas
> jvthomas@welchlink.welch.jhu.edu
> david_thomas@fmc.com
Read this:
Product Support
Segmentation Violation and Heap Errors in R13 Windows
Summary: This document outlines each of the possible causes of these
errors, from the most common to the most remote, and offers
troubleshooting suggestions to eliminate the source of the error.
Product(s) Release(s) Platform(s)
AutoCAD R13 Win3.1x
Audience: For Public Distribution - Global
Keywords: AUTOCAD SWAP SEGMENTATION VIOLATION HEAP MEMORY ERROR 988
Related documents: No
Document: US-MHK-TD111988.DOC Revision: 4.1
Creation date: January 18, 1995 Date last revised: August 22, 1996
Expiry/review date: August 22, 1998
The tips, tricks, examples and suggestions outlined in Autodesk Product
Support technical documents are suggested for use at your own risk.
Document contents are subject to change without notice. Autodesk is not
responsible or liable for damage or events that may occur as a result of
following suggestions from any Autodesk Product Support technical
document.
Overview
Segmentation Violation and Heap error messages in the Windows version of
AutoCAD Release 13 are general error messages that usually appear as a
result of a system memory conflict or memory shortage. Segmentation
Violation and Heap Error messages may be caused by one or more of the
following:
· Permanent Swap File set at less than 64MB
· 32-Bit Disk access or 32-Bit File Access enabled
· Windows System Resources too low
· Memory or Driver Conflict
· Software error
Troubleshooting the errors
In most cases, these errors can be cleared by increasing the size of the
Permanent Swap File in Windows. It is not always possible to set the
Swap File to the size that is recommended for use with AutoCAD Release
13 without getting an error message from Windows, that Windows will not
use Swap File space beyond the recommended size. The section below,
"Setting Adequate Permanent Swap File Size in Windows", explains the
procedure for allowing Windows to recognize a larger Swap File.
Windows System Resources and 32-Bit Disk and File access are also
Windows settings that will affect the performance of AutoCAD. These
settings will also be discussed in the corresponding sections of this
paper below.
The next most likely source of the conflict is an incompatible display
or pointing device driver, or a conflict in memory. This paper also
describes how to perform a system 'clean boot' to diagnose a conflict of
this kind.
Finally, the Segmentation Violation error may be the result of a flaw in
the program. This is unlikely, and should be considered only when the
other sources of these errors have been tested. If the Segmentation
Violation error is the result of a software error, it can usually be
reproduced with a specific sequence of commands. Specific examples of
this are shown in the last section of this paper, "Other causes for
Unhandled Exception errors".
Setting an Adequate Permanent Swap File Size in Windows
Heap errors usually occur when AutoCAD R13 is launched, when large
drawings are opened, and during memory intensive operations, such as
panning, zooming, rendering, modeling, saving, or regens. If a Heap
error occurs, the first thing to check is the size and type of the Swap
File in Windows. The minimum recommended Swap File is 64,000 KB
Permanent. If you are planning to work with large or complex files,
(files that include TEXT, HATCH patterns, BLOCKs, Xrefs, 3D models or
rendering, for example), you might need to set your Swap File to a
larger value. To change the size or type of the Windows Swap File:
· From Program Manager, select the Control Panel icon
· In the Control Panel Window, select the 386 Enhanced icon
· In the Enhanced Dialog Box, select the Virtual Memory button
· In the Virtual Memory Dialog Box, select the Change button
This Windows Dialog Box will display the current, maximum and
recommended sizes for the Swap File. If the recommended size is equal or
larger than the Swap File that you wish to create, enter the new size in
the "New Size:" box. You will now need to restart Windows for the new
Swap File to be created.
If the recommended Swap File size is smaller than the size that you wish
to create, you will receive a warning from Windows. A dialog box will
appear with the message that Windows will not use more than the
recommended size. Autodesk Product Support recommends the following
steps to have Windows recognize a larger Swap File:
1. Defragment the hard disk. Windows will not allow a Permanent Swap
File to be larger than 1/2 the size of the largest block of contiguous
free space on the hard drive. Defragmenting and optimizing the disk will
re-order all (or most) of the available of the free disk space into a
larger block or blocks.
2. Make more disk space available. Move less frequently used files onto
another disk or drive, compress files, if possible, and/or acquire more
disk space.
3. Edit the system.ini file. The system.ini is located in the WINDOWS
directory. Edit the file using an ASCII text editor, such as Notepad.
Look for the following line in the [386 Enh] section of this file:
PageOverCommit=xThis line does not exist in the system.ini file by
default. It Is placed there by the user, or during the installation of a
Windows application.
NOTE: This line must appear exactly as shown above. the system.ini file
is case sensitive, so the P, O and C must be capitalized, and the rest
of the letter characters must be lowercase.
The PageOverCommit value is multiplied by the amount of memory in the
system, and the result, (-approx. 1MB), is used as the Recommended Swap
File size, provided that enough contiguous space exists on the hard
drive. Therefore, a system with 16 MB of RAM and a PageOverCommit value
of 6 will have a Recommended Swap File of ~98 MB.
The PageOverCommit value can be set between 1 and 20. However, as this
value gets larger, paging activity will increase proportionately, which
will slow down the system. AutoCAD requires a Swap File of 64 MB
minimum, but this value doesn't need to be higher unless Heap Errors and
Segmentation Violation errors are occurring due to memory intensive
operations.
4. Follow the procedure above to establish a Swap File. The recommended
Swap File size should now be at least 64,000 KB. You must restart
Windows to create the new Swap File.
32-Bit Disk access and 32-Bit File access
If you are using Windows 3.x or Windows for Workgroups 32-Bit disk
access and 32-Bit file access should be disabled. The enable/disable
check boxes for these two settings are located in the Virtual Memory
dialog box , (the same dialog box as for setting Swap File size.)
Windows System Resources
Test any software that you are running concurrently with AutoCAD to
determine the amount of Windows system resources it consumes. Compare
the system resources value while running this software to the system
resources value without this software loaded. Different types of
programs consume these resources in different ways. For example, it is
possible, even in a system with a large amount of RAM, to run one
program that assigns a small amount of information to all of the Windows
memory allocation pointers. There is a fixed number of these pointers,
(called SELECTORS), and their use determines the amount of available
system resources. If one program uses most of these pointers, resources
won't be available to be used by other programs.
To check the level of available resources at any time, select About
Program Manager from the "Help" pull-down menu in the Windows Program
Manager screen. This value should be a minimum of 60% percent on a
computer with 16M of RAM to effectively load and run AutoCAD. A value of
less than 30% is dangerously low, and will usually result in an error
message or a crash. To make more resources available, close other
Windows applications that may be running concurrently with AutoCAD.
Also, closing and reopening Windows will free any resources that have
been "orphaned" by applications that closed without releasing sectors
correctly.
Memory conflicts, and conflicts with drivers
One of the most common sources of a Segmentation Violation error is the
presence of an incompatible display, pointing or plotting driver, or the
presence of conflicting application in memory. Common suspects include:
· older versions of Wintab digitizer drivers
· older versions of mouse drivers
· older versions of, or incompatible video card display drivers
The following procedure describes how to perform a "clean boot" to
diagnose the presence of an incompatible driver or application.
1. If the problem resulted in a crash or lock-up, check the hard drive
for lost clusters using the DOS command Chkdsk/F or Scandisk, and delete
any temporary files that may have been left on the hard drive during the
crash or error message. (AutoCAD temporary files usually contain a
dollar sign in the file extension, .AC$, or look for .SWR for swap
reserve.)
2. Backup the config.sys and autoexec.bat files. (ex: Copy these files
to a filename with a different extension, such as config.bak and
autoexec.bak. When you are ready to switch back to normal settings, you
can copy these files back to their original filenames and reboot the
system.)
3. Remove all lines from the config.sys file except the device statement
for himem.sys and 'Files=40'. Remove all lines from the autoexec.bat
file except the Prompt and Path statements.
4. Make sure that you have completely exited out of Windows or any other
program that was in use. From the DOS prompt, re-boot the computer by
turning the computer off and then on again. Perform a hard boot (off/on
switch) as opposed to a soft boot because the computer
may be loading utilities that will prevent some of the changes made to
config.sys and autoexec.bat from taking place with just a soft boot.
5. Attempt to run AutoCAD for as long as it usually takes for the
problem to occur.
If AutoCAD now operates correctly, edit the config.sys and autoexec.bat
files, adding in one device driver or TSR command at a time that you
find in your back-up files. Reboot the computer after each addition and
start AutoCAD until the error reappears. When the error reappears, the
last item added is the likely cause of the problem. Contact the
manufacturer of the errant utility for a new version of the device
driver or TSR, or try reloading the device drivers or TSR programs in a
different order.
However, if AutoCAD continues to fail with the simplified config.sys,
continue with the following steps:
1. From Windows, go to "Windows Setup, Options". Re-configure Windows to
use only the generic VGA display driver/video.
2. Rename the configuration file, acad.cfg, to another filename (ex:
acadcfg.old) so that AutoCAD will start the configuration process over
from scratch. Reconfigure AutoCAD for the system pointing device and no
plotter. (After you are done testing AutoCAD, you may return to your
original configuration, by renaming acadcfg.old back to acad.cfg).
3. Run AutoCAD for as long as it usually takes for the problem to occur.
If AutoCAD now operates, one of the peripherals devices is causing the
error. Re-configure AutoCAD, adding one device each time until the
problem driver is identified. Contact the manufacturer of the driver to
see if a newer version is available.
If AutoCAD continues to work properly after adding all the previously
used drivers, this may indicate that the configuration file was
corrupted.
If AutoCAD still fails when Windows is configured only for a VGA display
driver, you should begin to suspect a system level malfunction or
corrupted program files. Run a virus checker on the system to see if a
virus could be causing the problem. Re-installing AutoCAD or any program
which is running inside or alongside AutoCAD (add-on, MS Windows, etc.)
may resolve the problem. If not, go into the CMOS setup (the procedure
for accessing CMOS varies with each computer brand) and disable any
secondary cache and ROM shadowing.
If AutoCAD still fails to operate, then the hardware reseller or
manufacturer may need to inspect the computer for faulty components. It
is sometimes useful to install and run AutoCAD on another system (if
available) to confirm the suspected hardware problem.
Other causes for Segmentation Violation or Heap errors
If a 'Segmentation Violation' error occurs as the result of a specific
sequence of commands, it is possible that the error message is occurring
due to a flaw in the program. In order to verify this, it is important
to eliminate all of the other circumstances under which this error
message might occur.
First, follow the procedure for running AutoCAD in a stripped system.
Also, configure the system for VGA display, in order to eliminate the
possibility of a display driver conflict. Next, follow the procedure for
establishing an adequate Permanent Swap File in Windows.
After the above steps have been taken, attempt to recreate the
occurrence of the error message under the simplest possible
circumstances. For example if you suspect that a specific sequence of
commands will produce the error, open a new drawing in Release 13, using
no prototype, and enter that command sequence only. If a 'Segmentation
Violation' error appears consistently using this sequence of commands,
it is likely that is a result of a defect in the program.
Your local Authorized Autodesk Dealer can be contacted to assist you in
resolving and reporting the defect. If you wish, you can report a
suspected defect directly to Autodesk. Do this by filling out a Bug
Report form, (supplied in the back of the Installation Guide), and
mailing it to the address shown on the form.
There are some known command sequences in R13c0 that will produce a
Segmentation Violation error. (See below for examples).
1. Draw a line; Select the line to activate GRIPS; Select Floating Model
Space from the View pull-down menu. A Segmentation Violation will occur.
2. A Segmentation Violation will occur if a Bhatch object is bounded by
objects that are part of a Group, (defined in the Group command) the
Group objects are placed on a frozen or locked layer, and the Bhatch
object is edited using Move, Rotate, etc.
NOTE: Autodesk Product Support strongly recommends that you update to
the latest incremental release of AutoCAD if you feel your problems are
due to a software bug. The above command sequences no longer generate
Segmentation Violation errors in R13c1 and later releases. Please
contact your Authorized AutoCAD Dealer to obtain the latest incremental
release of AutoCAD.
