True progressive scan PC DVD player-- MPACT2 -- WOW! Initial Impressions
Christy
Yesterday I finally got an MPACT2 based PC DVD player. I have been
looking forward to seeing this video board/MPEG2 decoder for the last two
months, ever since I spoke with one of the engineers on IRC.
What got me excited about this chipset is that it does true progressive
scan DVD decoding on DVDs that are encoded from film sources. It also
has a special 720x480x16bit native graphics mode for DVD playback so it
sounded like something worth waiting for and to get excited about.
You can get more info about the mpact chipset by visiting www.mpact.com
I ordered the DVD kit from http://www.digitalconnection.com/ for $355
including overnight shipping. The kit itself cost something like $309.
It comes with the Pioneer 2.6x DVD rom drive, which to my suprise was a
slotted (trayless/caddyless) drive like the slotted pioneer CD roms. It
just has a big long slot that you stick the DVD or CD in.
When I first got the kit I noticed that the heat sink on the power
regulator had fallen off, which wasn't a good first sign. Since we have
an electronics lab in our house we just glued it back on. I suspect this
was a problem in shipping so if you buy from digitalconnection just make
sure you check for any damage and return it if there are any problems.
Unlike other DVD decoder cards this one is an actual 8 Meg video card
with 3d accelerator. It supports pretty decent modes and refresh rates
up to something like 1600x1200 at 75 Hz.
Instead of installing this in my desktop machine I decided it was time to
build a home theater PC to put in my projection room. Just for the
record, I use an Electrohome ECP 4100 and an IEV Turboscan T-1500
linedoubler on an 84" Screen. My DVD Player is a Sony DVP S7000 (april
manufacture) and my LD is a Pioner CLD 99.
So how does a PC stack up? PC-TV convergence here I come...
I went down to Pixel USA (a great store in Campbell California for low
prices and reasonably knowledgeable people) and picked up a Pentium II
333 and an ABIT BX6 motherboard and 128 Meg of PC100 Ram and a 6.3 Gig
hard disk. All this set me back $1050.
Since Pixel USA didn't have a good wireless keyboard (avoid the Logitech
as it only has 6 feet of range) and I didn't like the choices of cases I
went to Fry's for those things. I found an excelent wireless keyboard
and mouse stick (kind of like a thinkpad controler) and a good case.
These cost me another $150. BTW I am very impressed with one of the
lines of cases that you can get at Frys.
So all in all I spent $1500 on this whole kit, which is less than a
Pioner Elite DVD 09. According to the documentation, this board and
drive will run on a machine as low as a Pentium 166. By going with a
less insane PC you can probably put together a PC DVD player for around
$800, which is less than a Sony DVP S7000.
The reason I wanted more was so I could play Unreal and Quake2 on my
projector, but I digress.
Anyhow I built this PC installed the card/DVD Rom and windows 98.
I was impressed with how easy it was to install the software for the
MPACT card. Unlike other vendors where the installer tells you to go
fiddle with the control panel, this one installed fully automatically
when I inserted the CD.
Once I was satasfied that the machine was working I hooked it up to my
projector and fired it up. I put the Terminator 2 DVD on it and launched
the entertainment center application.
Needless to say, I was impressed. The picture revealed more horizontal
detail than my Sony S7000 through the line doubler. And better yet there
was no hint of line doubler type artifacting. The picture was just
goregous. It looked better than watching through a Faroudja.
I played the opening to Austin powers where you see Austin march around
the street with people in polka dotted and striped clothing. This scene
is an artifact extravaganza for line-doublers, other PC-DVD players.
With the MPACT2 board these parterns were STABLE with no artifacting
artifacting what-so-ever. The only thing that struck me as odd was that
the scanlines were a little more visible than I was used to when playing
DVDs through the line doubler in 16:9 mode.
So after fiddling with the controls a little more I found the problem.
The player was doing 4:3 down conversion of the 16:9 disk. So this blew
my mind. The 4:3 down conversion on this player looked as good as 16:9
playback through my Sony DVD player using the line doubler.
I switched it to 16:9 and played it again. Now it was beautiful and the
most filmlike that I've ever seen on a home theater system. It was
amazing to watch the opening of Terminator2. It was even more amazing to
watch the approach to the Skynet building after Sara shoots the
scientest. The diagonal lines of the building caused artifacting on
every other system I have ever tried it on--even a Faroudja line
quadroupler. Here it was stable and this player actually revealed the
way the building was actually lit and this made it clear why that scene
causes so many other systems fits when it tried to play it back. So when
I tried Austin powers again it was perfect too.
My partner watched this with me and we both felt that we saw a level of
fine detail and definition that we had just never seen before on our
system or any system.
In 16:9 mode I could sit about a screen width away and I didn't see scan
lines or any kind of video processor artifacting.
So basically this $1500 PC beats a $10,000 Faroudja. And in some ways it
gives a better picture than their quadroupler, although it obviously
doesn't display a line quadroupled image. The really exciting part of
this is that this board is only second generation. I can't wait to see
what third and fourth gen produce.
I will say this thing looks better than the Phillips DVX8000 which uses a
propritary video board that offers a 720x480 video mode but it uses
digital domain line doubling.
I put in Video Essentials for a moment and the Snell and Willcox zone
plate looked the best I had ever seen it. I need to do more testing on
DVDs from interlaced sources and I'll report on that. As it was, I was
up until five in the morning playing with this thing.
Some other things:
The MPAC2 board comes with coax digital out and it played AC3 through my
B&K decoder. Since it was late and our housemate was sleeping I couldn't
really put the sound through its paces yet but the fact it has a built in
digital out is promising. It also has a stereo out for folks that don't
have a digital processor. If you have a sound card you can hook the
player up to it and use its audio out if you wish. Personally I like
using the built in outputs.
The only thing that caused me any conern was that this player plays in
720x480x16 bit rather than 24 bit. However the results I saw were
amazing. I will delve into this with further testing. I haven't figured
out all the quirks of the software and hardware yet.
The MPACT2 software allows you to set region code at install time just
like the encore and it allows you to change it nine times. Now all we
need is a region hack for this board.
For anyone considering a PC-DVD player: forget everything else if you can
live with a new video card. I've tried a Real Magic Hollywood card,
Encore Dxr2, Cinemaster and they all look like garbage compared to this
card. The other cards have noticable artifacts in their line
doubling/scaling algorithms.
For what it is worth, I am in no way affiliated with Chromatic Research
(the makers of the chipset) or Digital Connection.
Cheers
--
Christy
-- Replace nospam with aurinia in order to reply
Frank Stearns
What was the horizontal resolution from the Snell and Willcox zone plate?
Christy wrote:
> I put in Video Essentials for a moment and the Snell and Willcox zone
> plate looked the best I had ever seen it. I need to do more testing on
> DVDs from interlaced sources and I'll report on that. As it was, I was
> up until five in the morning playing with this thing.
>
Brian Nelson
I wanted to know if you noticed that the picture on the television
appeared to be cropped a little on both sides like the creative labs
dxr2 board does.
Christy
In article <35842b6f...@news.fast.net>, ppr...@fast.net says...
> Forgot to ask - this won't do DTS will it? If so, I'm buying mine
> next week!
I don't know.. However since this is more of a general purpose processor,
I think there is a decent chance that a software upgrade may allow for
DTS. The best thing to do is to ask chromatic themselves.
William Hendershott
how did you hook up the output of the player to your projector? Does it have
an RGB output?
Christy
In article <357F115C...@accesscomm.net>, fste...@accesscomm.net
says...
> What was the horizontal resolution from the Snell and Willcox zone plate?
The resolution exceeds the range that the S&W zone was intended to
measure. In otherwords I see distinct lines all the way to the end of
the resolution measuring area.
One thing I did look at more closely tonight is the difference between
DVDs from true interlace sources and DVDs from film. Unlike most film
DVDs, DVDs from interlace sources must be processed through a line
doubling algorithm designed to handle interlaced video. This can prove
quite difficult to do. Consequently the performace of the MPACT board on
these sources is not as good as through my linedoubler. Oh well. I'll
take films looking amazing for now and live with the doubler for my
interlaced DVDs and Laserdiscs.
Christy
In article <357F299D...@ucdavis.edu>, bane...@ucdavis.edu says...
> I wanted to know if you noticed that the picture on the television
> appeared to be cropped a little on both sides like the creative labs
> dxr2 board does.
Are you talking about the TV out or the monitor out?
If you are talking about the TV out, please remember that TVs overscan.
A test with Video Essentials confirms this. if this is the case then
this is normal behavior and the MPACT2 will act the same as the Encore in
this regard. If you wish to eliminate the cropping you (or a qualified
tech) will need to open up your TV and adjust it to reduce overscan.
This might not be a good idea on all TVs.
If it does it on the monitor out then that would be quite a surprise to
me.
Christy
In article <357F3B4B...@maps.com>, wmhe...@maps.com says...
> how did you hook up the output of the player to your projector? Does it have
> an RGB output?
Since this is a computer it has a VGA out. Just buy an RBG cable and you
can hook it up to your projector.
Paul Prior MD
On Wed, 10 Jun 1998 14:26:48 -0700, ka...@nospam.com (Christy) wrote:
>Yesterday I finally got an MPACT2 based PC DVD player. I have been
>looking forward to seeing this video board/MPEG2 decoder for the last two
>months, ever since I spoke with one of the engineers on IRC.
... deleted ...
A very interesting and well written report. Please keep us updated!
I am very anxious to see what this progressive DVD format will bring
to us!
--
Paul Prior MD ppr...@fast.net Don't Blame Me...
PGY-IV Ob/Gyn I Voted For Bob Dole
TRHMC-Reading, PA
Paul Prior MD
Forgot to ask - this won't do DTS will it? If so, I'm buying mine
next week!
ale...@my-dejanews.com
christy:
Is there any way to "scale" this output (ie double it?). With an 8 or 9"
crt, I understand 480P still allows you to see scan lines
In article <MPG.fe899f74...@news.netscape.com>,
-----== Posted via Deja News, The Leader in Internet Discussion ==-----
http://www.dejanews.com/ Now offering spam-free web-based newsreading
Dean McManis
Christy,
Thanks. I have also been looking at PS-DVD ROM drives, and was planning
on upgrading my PC. I have a similar ECP FPTV/IEV setup from Hi Rez as
you, and I'm also in the Bay Area. So I'm probably going to jump on this
and try it out this week. It sounds wonderful. I don't mean to appear
like a copycat, but it seems whenever I am researching a new HT product,
you post a review after you purchased exactly what I'm looking for.
What kind of wireless keyboard did you get?
Thanks for the info,
-Dean.
Christy
In article <355692...@inreach.com>, de...@inreach.com says...
> Christy,
>
> Thanks. I have also been looking at PS-DVD ROM drives, and was planning
> on upgrading my PC. I have a similar ECP FPTV/IEV setup from Hi Rez as
> you, and I'm also in the Bay Area. So I'm probably going to jump on this
> and try it out this week. It sounds wonderful. I don't mean to appear
> like a copycat, but it seems whenever I am researching a new HT product,
> you post a review after you purchased exactly what I'm looking for.
I'm glad my posts help someone out there :)
> What kind of wireless keyboard did you get?
It is the PC-Concepts Pro-Series Surfboard. I found it at Frys. It is a
totally black keyboard with a small (think padish) joystick on the right
with three buttons around it. It sells for $79.95.
Dale Adams
In article <MPG.fe8e1c61...@nntp.best.com>, ka...@nospam.com
(Christy) wrote:
Thanks for the info on this new processor.
I'm a bit curious as to what the MPACT2 processor actually does with true
interlaced sources. Does it just combine fields (which should be very
obvious due to the motion artifacts) or does it perform some sort of
motion artifacts correction? Some cheap line doublers simply interpolate
a new field from the current one on every frame, thus significantly
reducing vertical resolution (i.e., a very soft looking picture). Can you
tell if the MPACT2 does this?
I know that without knowledge of the algorithms the MPACT2 uses it may be
difficult to tell what it's really doing. However, I would think you
could easily tell if it was doing nothing or if it was always reducing
vertical resolution by interpolating a new field. Any observations here?
You might also want to try the image montage from Video Essentials. It
consists of a lot of really good source material for evvaluating imaging
systems and should give the MPACT2 processor a good workout. In
particular, there are a lot of tricky transitions from 24 FPS film sources
to interlaced video sources. There are also transitions between 24 FPS
film sequences which have apparently been edited in the video domain, as
the 3:2 sequencing is imperfect. What does the MPACT2 do when you
encounter these transitions? (Again, I realize it may be difficult to tell
what it's really doing, but your subjective evaluation would be very
interesting).
I'm also a bit surprised that you couldn't tell that you only had 16
bits/pixel. This could be due to either the source material you were
viewing or some level of dithering in the MPACT2. Try a source like a
concert video (the Fleetwood Mac disc works pretty well here, and is a
good test for interlaced video artifacts as well) and look at the gradual
shadings around lights. There should be obvious banding unless some form
of dithering is being done.
Does the MPACT2 processor only work at 16 bits/pixel (bpp) or is there a
true 24 bpp mode (perhaps with a memory upgrade)? I'd have to say that my
interest in the processor would be severely limited if it could only do 16
bpp. Personally, I'd want to see at least 8 bpp, and preferably 10 bpp.
- Dale Adams
DVDO, Inc.
Francis Li
I hope people aren't confused by the message- I was for a while. Most
people will be using the composite and S-Video outputs of the card, which
are NTSC/PAL interlaced signals...
To see the "progressive scan" output, you need to watch it on your monitor
or hook the VGA output to a converter/projecter.
Do you mean that your projecter can handle the non-standard 720x480 full
screen resolution without any problems?
I have the Leadtek Winfast 3D S800 card which is also an MPACT2 based board.
I also agree that the DVD output to the monitor is very sharp and clear.
But I have to admit I'm a little disappointed with the overall performance
and software package of the card. I'll post my initial impressions later...
Francis
Andy KONG
From the URL provided by Christy, I believe she is referring to the
3DFusion DVD card. I haven't tried this card at all but I had tried
the Leadtek Winfast 3D S800 briefly today. In fact I delibrately go
to see the demo of this card after reading Christy's post on MPact2
chipset. I must say this is a very good valude for money 2D/3D/MPEG
card, but I was not blown away but its DVD decoding performance.
But before I jump to the conclusion, can someone confirm that the
Leadtek Winfast 3D S800 is identical (or very similar) to the 3DFusion
is terms of chipset used, features, software, and quality of MPEG
decoding?
Andy KONG (Remove "." in "netvigator.com" to response with email)
Christy
In article <dale-11069...@banzai.accesscom.com>, da...@dvdo.com
says...
That's exactly what I did and above.
> consists of a lot of really good source material for evvaluating imaging
> systems and should give the MPACT2 processor a good workout. In
> particular, there are a lot of tricky transitions from 24 FPS film sources
> to interlaced video sources. There are also transitions between 24 FPS
Can you tell me exactly where these transitions occur. I did notice that
some material looked really gorgeous and some had interlace caused
artifacts. Like the opening shots of the guy running looked good. The
shots of the stadium looked crapy.
> film sequences which have apparently been edited in the video domain, as
> the 3:2 sequencing is imperfect. What does the MPACT2 do when you
> encounter these transitions? (Again, I realize it may be difficult to tell
> what it's really doing, but your subjective evaluation would be very
> interesting).
>
> I'm also a bit surprised that you couldn't tell that you only had 16
> bits/pixel. This could be due to either the source material you were
I'm surprised too. Honestly. I'll have to find someone at MPACT to talk
to about this. It seems to switch into this special video mode.
> viewing or some level of dithering in the MPACT2. Try a source like a
> concert video (the Fleetwood Mac disc works pretty well here, and is a
> good test for interlaced video artifacts as well) and look at the gradual
> shadings around lights. There should be obvious banding unless some form
> of dithering is being done.
>
> Does the MPACT2 processor only work at 16 bits/pixel (bpp) or is there a
> true 24 bpp mode (perhaps with a memory upgrade)? I'd have to say that my
> interest in the processor would be severely limited if it could only do 16
> bpp. Personally, I'd want to see at least 8 bpp, and preferably 10 bpp.
The thing has 8 meg of memory and it does 24 bpp up to 1280x1024 (which
is also a little wierd). I will have to look at it more closely and try
and see if this isn't just pilot error on my part.
Greg Rogers
> Yesterday I finally got an MPACT2 based PC DVD player.
Thank you for posting this great review. It certainly got my interest
running high.
I've been digging into this and can add a little more information on the
Mpact 2 Media Processor from Chromatic Research, which is the heart of the
PC DVD player from Digital Connection. (That's a mouthful.)
> What got me excited about this chipset is that it does true progressive
> scan DVD decoding on DVDs that are encoded from film sources.
Indeed it does. The Mpact 2's "full window" mode should be selected to
drive a front projector or an RGB multiscan monitor. Everything I say here
applies to this mode. (There are additional modes to put a DVD picture in
a window on a computer monitor that work differently than described here.)
The Mpact 2 decodes film-source MPEG video into 24P progressive frame
video and then outputs it in RGB format at 72P by repeating each frame
three times. That is the only RGB output rate provided for driving a
monitor full frame. Therefore, the monitor or projector must accept a 37.8
KHz horizontal scan rate. That is a bit more than the usual 31.5 KHz scan
rate used by line doublers. The RGB DACs are included on chip in the Mpact
2 media processor.
In this full frame mode, the Mpact outputs a standard 720x480 active pixel
picture in each progressive frame. It works with both 16:9 format DVDs and
4:3 format DVDs. It will also downconvert the 16:9 format to the 4:3
letterboxed format using a vertical downconversion filter. I am waiting
for more information on this filter's characteristics.
> The only thing that caused me any conern was that this player plays in
> 720x480x16 bit rather than 24 bit.
According to the Video Product Manager for the Mpact 2, the documentation
is incorrect on the pixel depth. He indicated the Mpact 2 RGB output, in
the 720x480 DVD mode, has 18 bit color - RGB(6,6,6). Further, he stated
that it was not 24 bit color - RGB(8,8,8) because of "memory
considerations".
The DVD player product is being shipped to me tomorrow for review. So I
will have more to say on its performance next week.
Greg
--
**************************************************************
* Greg Rogers http://www.cybertheater.com *
* CyberTheater(tm): The Internet Journal of Home Theater *
**************************************************************
Greg Rogers
> I'm a bit curious as to what the MPACT2 processor actually does with true
> interlaced sources. Does it just combine fields (which should be very
> obvious due to the motion artifacts) or does it perform some sort of
> motion artifacts correction? Some cheap line doublers simply interpolate
> a new field from the current one on every frame, thus significantly
> reducing vertical resolution (i.e., a very soft looking picture). Can you
> tell if the MPACT2 does this?
It uses a motion adaptive (not motion compensated) algorithm to
de-interlace the video fields. This means it dynamically selects between
different interpolation techniques depending on the amount of motion
detected in the picture. Most line doublers use some type of motion
adaptive algorithm for deinterlacing video sources (non-film sources).
There are more motion adaptive algorithms than there are de-interlacing
products, so you must carefully compare each product's performance on a
wide variety of video pictures to decide which works the best (or really
which is least bad).
I will be sent one for review tomorrow so I should have comments to add to
Christy's next week.
Cheong
I also had a Winfast S800. Usually I watch DVD under 1024X768 at 16bit colour.
(I see no difference in picture quality at either 16bit or 24bit. Some
specification from Leadtek says that it does true color color space conversion
at either 16bpp or 24bpp for the frame buffer)
The image quality is really fantastic. Under 1024X768 at 72Hz on my 17" monitor.
It just good like a picture, not a video screen. But quality of TV output is
quite poor.
Christy
In article <6loml1$o03$1...@ultra.sonic.net>, f...@sonic.net says...
> I hope people aren't confused by the message- I was for a while. Most
> people will be using the composite and S-Video outputs of the card, which
> are NTSC/PAL interlaced signals...
>
> To see the "progressive scan" output, you need to watch it on your monitor
> or hook the VGA output to a converter/projecter.
Sorry I wasn't more clear on that point. I kind of thought it would be
clear from the context that I was using a computer.
> Do you mean that your projecter can handle the non-standard 720x480 full
> screen resolution without any problems?
A CRT projector capable of doing SVGA (800x600) or better should handle
720x480 just fine. A projector capable of only VGA (640x480) may or may
not work depending on the details of the decoder.
> I have the Leadtek Winfast 3D S800 card which is also an MPACT2 based board.
> I also agree that the DVD output to the monitor is very sharp and clear.
> But I have to admit I'm a little disappointed with the overall performance
> and software package of the card. I'll post my initial impressions later...
Is the software package from MPACT or is it from someone else? I thought
the software from CFusion was OK. It's interface had its clumsy points
and it had no keyboard control that I could find which was a drag. But
it was pretty easy to use with a mouse.
Christy
In article <35842e2b...@news.netvigator.com>,
Andy...@net.vigator.com says...
> >people will be using the composite and S-Video outputs of the card, which
> >are NTSC/PAL interlaced signals...
> >
> >To see the "progressive scan" output, you need to watch it on your monitor
> >or hook the VGA output to a converter/projecter.
> >
> >screen resolution without any problems?
> >
> >I also agree that the DVD output to the monitor is very sharp and clear.
> >But I have to admit I'm a little disappointed with the overall performance
> >and software package of the card. I'll post my initial impressions later...
> >
>
> From the URL provided by Christy, I believe she is referring to the
> 3DFusion DVD card. I haven't tried this card at all but I had tried
> the Leadtek Winfast 3D S800 briefly today. In fact I delibrately go
> to see the demo of this card after reading Christy's post on MPact2
> chipset. I must say this is a very good valude for money 2D/3D/MPEG
> card, but I was not blown away but its DVD decoding performance.
When you demoed it, did you set it up for 16:9 mode and squish the
vertical height and play it in full screen mode?
This makes a big difference. Did you also demo it on a film DVD like T2
or Fifth Element. On DVDs of video, as I noted in a follow up, the
handling of real interlaced video is nothing special. It does about the
same as the Real Magic Hollywood card or the Encore in this respect.
I also noticed more pixel garbage when I ran it through a computer
monitor than through my projector. But I also see similar garbage when
running my line doubler or any other DVD player through that particular
monitor so it didn't worry me much.
What does worry me some is that the 720x480 video mode only appears to
exist in 16 bit mode. I still need to sort this issue out.
I do think for PC playback this card works better than the others. I
consider the Encore too problematic due to its analog overlay scheme and
how hard it is to eliminate color fringing.
Christy
In article <gregsr-1106...@u3-250.spiritone.com>,
gre...@SpiritOne.com says...
>
> running high.
I'm very excited about this toy, especially since something this good has
shown up this early. It means we'll only see even better stuff come from
them and others down the road.
> I've been digging into this and can add a little more information on the
> Mpact 2 Media Processor from Chromatic Research, which is the heart of the
> PC DVD player from Digital Connection. (That's a mouthful.)
>
> > What got me excited about this chipset is that it does true progressive
> > scan DVD decoding on DVDs that are encoded from film sources.
>
> Indeed it does. The Mpact 2's "full window" mode should be selected to
> drive a front projector or an RGB multiscan monitor. Everything I say here
> applies to this mode. (There are additional modes to put a DVD picture in
> a window on a computer monitor that work differently than described here.)
>
> The Mpact 2 decodes film-source MPEG video into 24P progressive frame
> video and then outputs it in RGB format at 72P by repeating each frame
> three times. That is the only RGB output rate provided for driving a
> monitor full frame. Therefore, the monitor or projector must accept a 37.8
> KHz horizontal scan rate. That is a bit more than the usual 31.5 KHz scan
> rate used by line doublers. The RGB DACs are included on chip in the Mpact
> 2 media processor.
My electroholme ECP4100 thinks the horizontal scan rate is 31.5. I
wonder if it is screwing up on this since there seems to be a bit off
quirkiness about the picture. By quirkiness I mean that it was shifted
sideways and needed some tweaking in order to position in correctly. I
wonder if I need to adjust something in the projector sync settings.
I know that multisync systems actually only sync to a set of ranges.
> In this full frame mode, the Mpact outputs a standard 720x480 active pixel
> picture in each progressive frame. It works with both 16:9 format DVDs and
> 4:3 format DVDs. It will also downconvert the 16:9 format to the 4:3
> letterboxed format using a vertical downconversion filter. I am waiting
> for more information on this filter's characteristics.
I will look forward to hearing your comments on 4:3 downconversion. I
for one was very impressed with it.
> > The only thing that caused me any conern was that this player plays in
> > 720x480x16 bit rather than 24 bit.
>
> According to the Video Product Manager for the Mpact 2, the documentation
> is incorrect on the pixel depth. He indicated the Mpact 2 RGB output, in
> the 720x480 DVD mode, has 18 bit color - RGB(6,6,6). Further, he stated
> that it was not 24 bit color - RGB(8,8,8) because of "memory
> considerations".
I don't understand memory considerations here because the card is an 8 MB
card. I wonder if there is some other strange issue like the size of the
processor's cache.
> The DVD player product is being shipped to me tomorrow for review. So I
> will have more to say on its performance next week.
I look forward to hearing your comments. I just hope I didn't overblow
it. The picture looks damn good but I admitedly haven't A-Bed it with a
Faroudja.
Dale Adams
In article <MPG.fe9dc56a...@news.netscape.com>, ka...@nospam.com
(Christy) wrote:
> In article <dale-11069...@banzai.accesscom.com>, da...@dvdo.com
> says...
> > You might also want to try the image montage from Video Essentials. It
>
> That's exactly what I did and above.
>
> > consists of a lot of really good source material for evvaluating imaging
> > systems and should give the MPACT2 processor a good workout. In
> > particular, there are a lot of tricky transitions from 24 FPS film sources
> > to interlaced video sources. There are also transitions between 24 FPS
>
> Can you tell me exactly where these transitions occur. I did notice that
> some material looked really gorgeous and some had interlace caused
> artifacts. Like the opening shots of the guy running looked good. The
> shots of the stadium looked crapy.
>
> > film sequences which have apparently been edited in the video domain, as
> > the 3:2 sequencing is imperfect. What does the MPACT2 do when you
> > encounter these transitions? (Again, I realize it may be difficult to tell
> > what it's really doing, but your subjective evaluation would be very
> > interesting).
You can pretty much tell which are film sources and which are video
sources by the production values in the scenes. I.e., if it looks like it
came from a movie, it probably did.
It's my guess that the video sequences were shot to create explicit test
cases. (Your example of the stadium scene is a good example of this. You
note that it doesn't look so hot on the MPACT2, which means it probably
isn't doing a very good job of deinterlacing on that scene.) For
instance, the scenes panning along a road under an overpass, along a
bridge with poles whizzing by, and pans and zooms of bridges are video.
Most (all?) shots with people (actors) in them are film. The opening
shots (like the guy running) are film.
One interesting transition occurs between two film sources where the 3:2
pulldown sequence is messed up. The transition from the soccer game in
the rain (a film source) to a couple walking under bleachers (also a film
source) was apparently edited in the video domain without regard to the
3:2 sequencing since that sequencing is disturbed during the scene
transition. (If you have a way to capture the fields and view them in
slow sequence you can see this.) This could be either a goof-up on Video
Essentials, or, more likely, is another explicit test case to see how a
video processor handles this sort of transition sequence.
Tommy Hwang
Sorry, but I am not familiar with Progressive scan... But
If my guess is right, it is just the elimination of flickering
or appeared elimination of flickering. Right? If so,
how would the PC-DVD compare with my current setup?
Toshiba DVD player connected to a Toshiba 65 inch TV
with doubl scanning to reduce/eliminate flickering? I use
component input here obviously. Thanks...
Mike
Progressive scan will display all scan lines in ONE pass,
line doubler receives an Interlaced signal (all odd lines
in one pass, all even lines in the next, so 2 passes to complete
a frame). Through an algorithm it creates one frame per
interlaced pass. So the line-doubled picture should be
softer, and the progressive picture should be sharper.
A progressive output from a DVD player should look
better than output from the best line doubler.
Steve Zipser
The fabulous Miranda Quartz-2 line doubler will line double in both
progressive scan AND interlaced mode.
Zip
--
Sunshine Stereo, Inc http://www.sunshinestereo.com
Tel: 305-757-9358 Fax: 305-757-1367
9535 Biscayne Blvd Miami Shores FL 33138
PASS Labs Carver Lightstar Miranda CODA Audible Illusions CEC
Camelot Technology Audio Logic Parasound Kinergetics Cabasse
Chiro Benz Micro Gallo Acoustics Dunlavy Audio NEAR NHT Jadis
Niles Zenith INTEQ Crystal Vision Straightwire Mordaunt Short ESP
Rega Vans Ever's Cleanlines Monster Cable ENTECH EAD Arcane Audio
Sunshine Stereo encourages all audiophiles to support their local
dealers. If you do not have a local dealer, we will gladly assist
you with all your audio and video needs! *** ENJOY THE MUSIC! ***
Francis Li
: When you demoed it, did you set it up for 16:9 mode and squish the
This is definitely a neat thing to do with anamorphic DVDs- I find that the
MPACT2 downconversion introduces quite a bit of blocky pixellization on
edge, sort of like aliasing.
: I also noticed more pixel garbage when I ran it through a computer
: monitor than through my projector. But I also see similar garbage when
: running my line doubler or any other DVD player through that particular
: monitor so it didn't worry me much.
I think it is because of the sharpness and clarity of displaying on a PC
monitor without analog overlay. On a TV, you simply turn the sharpness
down. You can't do that on your PC monitor! I think the solution to this
is to sit back a little further from the monitor.
: What does worry me some is that the 720x480 video mode only appears to
: exist in 16 bit mode. I still need to sort this issue out.
I find this to be a problem myself. I was watching Air Force One on my
monitor (anamorphic, using the monitor controls to squeeze it to the right
aspect ratio), and I could notice dithering and banding in the sky.
Although sitting back further from the monitor solves this a bit too..
: I do think for PC playback this card works better than the others. I
: consider the Encore too problematic due to its analog overlay scheme and
: how hard it is to eliminate color fringing.
Unfortunately, I like the Encore Dxr2 playback software better than
Entertainment Center. The Entertainment Center lacks features like time
search, bookmarking (which I actually find useful at times), and variable
speed fast-forward and rewind. My full review will come later, the time
spent giving the card a full test, including 3D benchmarks, etc.
Francis
Christy
In article <3580F7D7...@glink.net.hk>, sc...@glink.net.hk says...
> I also had a Winfast S800. Usually I watch DVD under 1024X768 at 16bit colour.
> (I see no difference in picture quality at either 16bit or 24bit. Some
> specification from Leadtek says that it does true color color space conversion
> at either 16bpp or 24bpp for the frame buffer)
> The image quality is really fantastic. Under 1024X768 at 72Hz on my 17" monitor.
> It just good like a picture, not a video screen. But quality of TV output is
> quite poor.
How did you play it full screen at 1024x768? I thought no matter what
mode you ran in it would switch to 720x480 18bbp mode.
Cliff Watson
Christy wrote in message ...
Christy,
You are correct. According to my RGB monitor's OSD, the resolution is
720x480 at H=39.3 kHz and V=72 Hz for DVD movies.
Non-DVD disk files (MPEG-1 and 2) will play back at the resolution set for
the monitor.
Thanks for the review. I received my 3DFusion card today and love it.
Thanks, Cliff
Dean McManis
Christy,
I got my new 300 Mhz Pentium II system from Pixel USA today.
I ordered the DVD/MPACT2 card combo yesterday and I'm eagerly waiting
for it to arrive. I couldn't find the cordless PC Concepts keyboard at
the Campbell Fry's Electronics store, and the PC Concept's website
didn't mention a cordless product, but I'll keep looking.
-Dean.
mcd...@my-dejanews.com
Wow is right! I got my AGP version of the 3D Fusion from Digital Connection
last week. To keep it in good company, I slipped it into an ATX machine with
a P-II 400 on a 100Mhz BX bus, 64 meg of 7ns ram, and a new Pioneer DVD
drive. My Sony 17 couldn't be happier, (it justs wants to be bigger)! The
software installation was fast and flawless, it didn't ask any questions that
I could screw up, and I've played with all of the functions without once
reading the manual. I've looked at quite a few MPEG-2 decoders and I can say
that the 3D Fusion wins HANDS DOWN! As for the money I saved, well.... I
spent it on the P-II 400! I'm not much of a game player, but I can envision a
lot of DVD titles on my credit card statement. It's almost painful to watch
regular TV anymore, much less a VHS tape, yuk! I can't wait for large screen
HDTV now! The appropriate audio compliment will be when I mate this video to
the audio of my 1500 watts of Haffler & Crest power, a pair of ESS's and my
JBL 18" Subwoofer! I often find myself paying more attention to the quality
of the picture than to the movie I'm watching. It's a real treat. Please
e-mail me with DVD titles of striking quality or effects.
Paul Prior MD
On Sat, 13 Jun 1998 08:09:47 GMT, Dean McManis <de...@inreach.com>
wrote:
Please post -your- impressions/experience when you get it up and
running! Many of us are very interested.
Christy
In article <355958...@inreach.com>, de...@inreach.com says...
> Christy,
>
> I got my new 300 Mhz Pentium II system from Pixel USA today.
> I ordered the DVD/MPACT2 card combo yesterday and I'm eagerly waiting
> for it to arrive. I couldn't find the cordless PC Concepts keyboard at
> the Campbell Fry's Electronics store, and the PC Concept's website
> didn't mention a cordless product, but I'll keep looking.
The exact name on the box is
The (pro series) surfboard
remote wireless keyboard
by PC concepts
they give a phone number 1-800-735-6071 on the box.
I would try other Frys in the area.
The Frys label reads
PC Concepts 61533
Surfboard/Pro Series 2|335
Remote wireless keyboard
Built-in pointing device 2291248
I think that last number is the SKU.
Christy
In article <6lti8m$o0p$1...@nnrp1.dejanews.com>, mcd...@my-dejanews.com
says...
> Wow is right! I got my AGP version of the 3D Fusion from Digital Connection
> last week. To keep it in good company, I slipped it into an ATX machine with
> a P-II 400 on a 100Mhz BX bus, 64 meg of 7ns ram, and a new Pioneer DVD
> drive. My Sony 17 couldn't be happier, (it justs wants to be bigger)! The
> software installation was fast and flawless, it didn't ask any questions that
> I could screw up, and I've played with all of the functions without once
> reading the manual. I've looked at quite a few MPEG-2 decoders and I can say
> that the 3D Fusion wins HANDS DOWN! As for the money I saved, well.... I
> spent it on the P-II 400! I'm not much of a game player, but I can envision a
> lot of DVD titles on my credit card statement. It's almost painful to watch
> regular TV anymore, much less a VHS tape, yuk! I can't wait for large screen
> HDTV now! The appropriate audio compliment will be when I mate this video to
> the audio of my 1500 watts of Haffler & Crest power, a pair of ESS's and my
> JBL 18" Subwoofer! I often find myself paying more attention to the quality
> of the picture than to the movie I'm watching. It's a real treat. Please
> e-mail me with DVD titles of striking quality or effects.
I'll post here just so others can see.
First of all I'm glad you like this machine. I'm very happy with mine.
Here is a list of some of the better DVD titles to try on your new toy.
Watch all these in 16:9 anamorphic mode and shrink down the vertical size
of your screen when you watch. If you haven't tried that, you're in for a
treat because the picture looks even better that way:)
Fifth Element
Terminator 2
In the Line of Fire
Desperate Measures
Outbreak
First Knight
Copycat
Das Boot
Air Force One
Christy
>
> > The only thing that caused me any conern was that this player plays in
> > 720x480x16 bit rather than 24 bit.
>
> According to the Video Product Manager for the Mpact 2, the documentation
> is incorrect on the pixel depth. He indicated the Mpact 2 RGB output, in
> the 720x480 DVD mode, has 18 bit color - RGB(6,6,6). Further, he stated
> that it was not 24 bit color - RGB(8,8,8) because of "memory
> considerations".
After getting some email from other folks with rave reviews of the MPACT2
in comparasion with a Faraoudja in their show room I thought some more
about this 18bit issue and I realized something.
MPEG2 does a DCT to convert the data in a macroblock from the spatial
domain to the frequency domain. After this it reduces the quantization
(eg color depth) of the data. In some frequency ranges it reduces it
more than others. I wonder how many of the frequency ranges (eg slots of
the DCTed macroblock) use 7 or 8 bit encoding vs 6 or fewer bits. Can
John DeGoof or some other MPEG2 expert please comment?
Afterall the human eye can only differentiate something like 64 gray
scales and much less in colors. It is most sensitive to greens and least
sensitive to reds.
So I think in this case since the limited bit depth of the MPACT2
actually interacts favorably with the way MPEG2 works. If this is the
case then it is a nice change since we all know how other things like the
sharpness control on a TV interacts poorly with MPEG2.
Cheers
Mark Hapner
Before everyone runs out to buy an MPACT based DVD adaptor you should
look at the new Holywood 3 from from Sigma Designs. It is likely to
provide superior picture quality at higher resolutions in true color.
Its available now (it just came out in the last few weeks).
I'm check with Sigma Designs on DTS support but I'm guessing its not
there. But for $120 you can throw it away when a version with DTS
comes out.
http://www.realmagic.com/hllywood.html
http://www.computeresp.com/prdct/976/222.html
For scan converted display of TV, s-video and composite sources at
resolutions up to 1600 x1200 in true color I'm using and ATI All in
Wonder Pro AGP 8 MB graphics adaptor with very nice 3D support which
is very close to Voodoo performance. It does an excellent job. Even
regular VHS tapes look like films (albeit a bit fuzzy) with no scan
lines even with your nose on the screen (I'm using a 1280 x 1024
display on a 90" screen).
There is simply no reason to even consider a line doubler from this
point on.
http://www.atitech.ca/products/pc/aiw_pro/index.html
http://www.computeresp.com/prdct/706/791.html
On Wed, 10 Jun 1998 14:26:48 -0700, ka...@nospam.com (Christy) wrote:
>Yesterday I finally got an MPACT2 based PC DVD player. I have been
>looking forward to seeing this video board/MPEG2 decoder for the last two
>months, ever since I spoke with one of the engineers on IRC.
>
>What got me excited about this chipset is that it does true progressive
>scan DVD decoding on DVDs that are encoded from film sources. It also
>has a special 720x480x16bit native graphics mode for DVD playback so it
>sounded like something worth waiting for and to get excited about.
>
>You can get more info about the mpact chipset by visiting www.mpact.com
>
>I ordered the DVD kit from http://www.digitalconnection.com/ for $355
>including overnight shipping. The kit itself cost something like $309.
>It comes with the Pioneer 2.6x DVD rom drive, which to my suprise was a
>slotted (trayless/caddyless) drive like the slotted pioneer CD roms. It
>just has a big long slot that you stick the DVD or CD in.
>
>When I first got the kit I noticed that the heat sink on the power
>regulator had fallen off, which wasn't a good first sign. Since we have
>an electronics lab in our house we just glued it back on. I suspect this
>was a problem in shipping so if you buy from digitalconnection just make
>sure you check for any damage and return it if there are any problems.
>
>Unlike other DVD decoder cards this one is an actual 8 Meg video card
>with 3d accelerator. It supports pretty decent modes and refresh rates
>up to something like 1600x1200 at 75 Hz.
>
>Instead of installing this in my desktop machine I decided it was time to
>build a home theater PC to put in my projection room. Just for the
>record, I use an Electrohome ECP 4100 and an IEV Turboscan T-1500
>linedoubler on an 84" Screen. My DVD Player is a Sony DVP S7000 (april
>manufacture) and my LD is a Pioner CLD 99.
>
>So how does a PC stack up? PC-TV convergence here I come...
>
>I went down to Pixel USA (a great store in Campbell California for low
>prices and reasonably knowledgeable people) and picked up a Pentium II
>333 and an ABIT BX6 motherboard and 128 Meg of PC100 Ram and a 6.3 Gig
>hard disk. All this set me back $1050.
>
>Since Pixel USA didn't have a good wireless keyboard (avoid the Logitech
>as it only has 6 feet of range) and I didn't like the choices of cases I
>went to Fry's for those things. I found an excelent wireless keyboard
>and mouse stick (kind of like a thinkpad controler) and a good case.
>These cost me another $150. BTW I am very impressed with one of the
>lines of cases that you can get at Frys.
>
>So all in all I spent $1500 on this whole kit, which is less than a
>Pioner Elite DVD 09. According to the documentation, this board and
>drive will run on a machine as low as a Pentium 166. By going with a
>less insane PC you can probably put together a PC DVD player for around
>$800, which is less than a Sony DVP S7000.
>
>The reason I wanted more was so I could play Unreal and Quake2 on my
>projector, but I digress.
>
>Anyhow I built this PC installed the card/DVD Rom and windows 98.
>I was impressed with how easy it was to install the software for the
>MPACT card. Unlike other vendors where the installer tells you to go
>fiddle with the control panel, this one installed fully automatically
>when I inserted the CD.
>
>Once I was satasfied that the machine was working I hooked it up to my
>projector and fired it up. I put the Terminator 2 DVD on it and launched
>the entertainment center application.
>
>Needless to say, I was impressed. The picture revealed more horizontal
>detail than my Sony S7000 through the line doubler. And better yet there
>was no hint of line doubler type artifacting. The picture was just
>goregous. It looked better than watching through a Faroudja.
>
>I played the opening to Austin powers where you see Austin march around
>the street with people in polka dotted and striped clothing. This scene
>is an artifact extravaganza for line-doublers, other PC-DVD players.
>With the MPACT2 board these parterns were STABLE with no artifacting
>artifacting what-so-ever. The only thing that struck me as odd was that
>the scanlines were a little more visible than I was used to when playing
>DVDs through the line doubler in 16:9 mode.
>
>So after fiddling with the controls a little more I found the problem.
>The player was doing 4:3 down conversion of the 16:9 disk. So this blew
>my mind. The 4:3 down conversion on this player looked as good as 16:9
>playback through my Sony DVD player using the line doubler.
>
>I switched it to 16:9 and played it again. Now it was beautiful and the
>most filmlike that I've ever seen on a home theater system. It was
>amazing to watch the opening of Terminator2. It was even more amazing to
>watch the approach to the Skynet building after Sara shoots the
>scientest. The diagonal lines of the building caused artifacting on
>every other system I have ever tried it on--even a Faroudja line
>quadroupler. Here it was stable and this player actually revealed the
>way the building was actually lit and this made it clear why that scene
>causes so many other systems fits when it tried to play it back. So when
>I tried Austin powers again it was perfect too.
>
>My partner watched this with me and we both felt that we saw a level of
>fine detail and definition that we had just never seen before on our
>system or any system.
>
>In 16:9 mode I could sit about a screen width away and I didn't see scan
>lines or any kind of video processor artifacting.
>
>So basically this $1500 PC beats a $10,000 Faroudja. And in some ways it
>gives a better picture than their quadroupler, although it obviously
>doesn't display a line quadroupled image. The really exciting part of
>this is that this board is only second generation. I can't wait to see
>what third and fourth gen produce.
>
>I will say this thing looks better than the Phillips DVX8000 which uses a
>propritary video board that offers a 720x480 video mode but it uses
>digital domain line doubling.
>
>I put in Video Essentials for a moment and the Snell and Willcox zone
>plate looked the best I had ever seen it. I need to do more testing on
>DVDs from interlaced sources and I'll report on that. As it was, I was
>up until five in the morning playing with this thing.
>
>Some other things:
>
>The MPAC2 board comes with coax digital out and it played AC3 through my
>B&K decoder. Since it was late and our housemate was sleeping I couldn't
>really put the sound through its paces yet but the fact it has a built in
>digital out is promising. It also has a stereo out for folks that don't
>have a digital processor. If you have a sound card you can hook the
>player up to it and use its audio out if you wish. Personally I like
>using the built in outputs.
>
>The only thing that caused me any conern was that this player plays in
>720x480x16 bit rather than 24 bit. However the results I saw were
>amazing. I will delve into this with further testing. I haven't figured
>out all the quirks of the software and hardware yet.
>The MPACT2 software allows you to set region code at install time just
>like the encore and it allows you to change it nine times. Now all we
>need is a region hack for this board.
>
>For anyone considering a PC-DVD player: forget everything else if you can
>live with a new video card. I've tried a Real Magic Hollywood card,
>Encore Dxr2, Cinemaster and they all look like garbage compared to this
>card. The other cards have noticable artifacts in their line
>doubling/scaling algorithms.
>
>For what it is worth, I am in no way affiliated with Chromatic Research
>(the makers of the chipset) or Digital Connection.
>
>Cheers
Mario Kalogjera
What type of outputs does this MPACT2 card have? What jacks/signal paths
(RGB, S-video, composite YUV)? Can it be bought separately?
Thanks for the answer,
Mario Kalogjera
mario.k...@usa.net
Christy wrote:
go...@hotmail.com
> > Unlike other DVD decoder cards this one is an actual 8 Meg video card
> > with 3d accelerator. It supports pretty decent modes and refresh rates
> > up to something like 1600x1200 at 75 Hz.
TRUCATED
> > So basically this $1500 PC beats a $10,000 Faroudja. And in some ways it
> > gives a better picture than their quadroupler, although it obviously
> > doesn't display a line quadroupled image.
If you say the output can be as high as 1600x1200, then there must be scan
upconversion done to the progressive line doubled signal which is essentially
640x480 VGA signal. The set-up should allow interpolator type function ie.
line 2.5X (800x600), line 3x (1024x768), line quadrupling (1280x920) and
beyond.
Try outputing the signal at 1024x768 which is line tripling at the highese
possible vertical refresh rate to get a stable (non flickering) ouput. The
scan lines should be filled in nicely.
Let me know whether that works. I'm also in the process of settinbg up a
killer PC based DVD source as well as PC based line doubler for thlaserdisc
source. For the latter, I am thinking of using the ATI all in wonder pros
video inputs and upscanning that to 1024x 768. have you tried that before?
what's the quality if you have?
If you could please copy me you reply post via email cos i get on the web only
once in a while. thanks
Dennis
go...@hotmail.com
Christy
In article <6lvcfk$2at$1...@nnrp1.dejanews.com>, go...@hotmail.com says...
>
>
> > > Unlike other DVD decoder cards this one is an actual 8 Meg video card
> > > with 3d accelerator. It supports pretty decent modes and refresh rates
> > > up to something like 1600x1200 at 75 Hz.
>
> TRUCATED
>
> > > So basically this $1500 PC beats a $10,000 Faroudja. And in some ways it
> > > gives a better picture than their quadroupler, although it obviously
> > > doesn't display a line quadroupled image.
>
> If you say the output can be as high as 1600x1200, then there must be scan
> upconversion done to the progressive line doubled signal which is essentially
> 640x480 VGA signal. The set-up should allow interpolator type function ie.
> line 2.5X (800x600), line 3x (1024x768), line quadrupling (1280x920) and
> beyond.
When I was talking about output up to 1600x1200 I was refering to its
function as a video card. Its best playback mode is fullscreen which is
only available in 720x480. You can play DVDs in a window but I don't
think you will get as good results. I haven't seen a good resampler yet
on a computer.
> Try outputing the signal at 1024x768 which is line tripling at the highese
> possible vertical refresh rate to get a stable (non flickering) ouput. The
> scan lines should be filled in nicely.
But this plays 720x480 progressive. Tripling of 480p is 1440p which
isn't practical on most displays or projectors.
> Let me know whether that works. I'm also in the process of settinbg up a
> killer PC based DVD source as well as PC based line doubler for thlaserdisc
> source. For the latter, I am thinking of using the ATI all in wonder pros
> video inputs and upscanning that to 1024x 768. have you tried that before?
> what's the quality if you have?
As I said, I haven't seen a good resampler (what you call an upscaler).
I have read somewhere that the all in wonder pro doesn't have all that
good video. The only decent DVD decoder that we have available right now
is the MPACT.
> If you could please copy me you reply post via email cos i get on the web only
> once in a while. thanks
Ok, done:)
Hope this helps...
r...@optivision.com
In article <MPG.fec7c68a...@nntp.best.com>,
ka...@nospam.com (Christy) wrote:
>
> In article <gregsr-1106...@u3-250.spiritone.com>,
> gre...@SpiritOne.com says...
> [stuff deleted]
> >
> > is incorrect on the pixel depth. He indicated the Mpact 2 RGB output, in
> > the 720x480 DVD mode, has 18 bit color - RGB(6,6,6). Further, he stated
> > that it was not 24 bit color - RGB(8,8,8) because of "memory
> > considerations".
>
> After getting some email from other folks with rave reviews of the MPACT2
> in comparasion with a Faraoudja in their show room I thought some more
> about this 18bit issue and I realized something.
>
> MPEG2 does a DCT to convert the data in a macroblock from the spatial
> domain to the frequency domain. After this it reduces the quantization
> (eg color depth) of the data. In some frequency ranges it reduces it
> more than others. I wonder how many of the frequency ranges (eg slots of
> the DCTed macroblock) use 7 or 8 bit encoding vs 6 or fewer bits. Can
> John DeGoof or some other MPEG2 expert please comment?
Quantization of DCT coefficients does not alter the "color depth" of the
image data. If you input 8-bit samples to the DCT/quant process, you get
8-bit output from the dequant/IDCT process. The only thing that quantization
does do is add errors to the reconstructed pixel levels. Here is an example:
First, we start with an 8 x 8 block of pixels (these values are from
some actual captured video):
106 99 99 98 102 110 113 122
159 147 131 120 113 112 117 118
201 198 183 171 155 136 127 116
212 210 207 203 191 183 161 121
226 214 213 208 209 208 158 87
221 218 212 214 222 188 108 57
216 218 214 216 199 128 63 56
215 210 215 213 154 79 53 57
The output of the DCT will be:
1263.750000 268.556481 -95.134045 32.805784 5.750000 0.060174 -0.372102
-0.439796
-119.462773 -168.535537 99.358267 -0.237354 -19.889084 7.193988 0.071848
0.357827
-184.827283 -0.415735 31.483126 -53.939000 35.789550 -8.143926 0.141243
-0.277785
-10.934237 -43.690853 -32.002795 46.252509 -13.675425 -7.670069 9.165051
-0.001160
0.250000 -31.855892 6.364687 -6.920115 -7.750000 17.037060 0.340239 -0.229596
6.299784 -5.577267 -13.613778 -7.738089 7.852199 -10.179133 0.039135 -0.409630
-0.021281 5.710864 -6.858757 0.097545 -0.482820 -0.490393 0.016874 -0.383167
0.388621 6.789808 0.245245 0.024789 -0.006823 -0.158927 0.596691 0.962161
If we were to run these coefficients through the IDCT, we would recover
the original pixel values with no errors (lossless). However, we didn't
get much compression and in fact, it looks like we ended up with more
data than what we started with. The first step in quantization is to
take the integer value of the coefficients. This gives us:
1264 269 -95 33 6 0 0 0
-119 -169 99 0 -20 7 0 0
-185 0 31 -54 36 -8 0 0
-11 -44 -32 46 -14 -8 9 0
0 -32 6 -7 -8 17 0 0
6 -6 -14 -8 8 -10 0 0
0 6 -7 0 0 0 0 0
0 7 0 0 0 0 1 1
Notice that quite a few coefficients end up with a zero value. Now we are
getting some compression, and the run-level Huffman encoding that's used
later will take advantage of the zero coefficients. Also note that most
of the zero coefficients are high frequency coefficients. This is why
the DCT was chosen as the basis for MPEG-2. It's taking advantage of the
fact that most blocks in an image don't have that much high frequency
information to begin with. Next, we quantize the coefficients with the
default intra quantization matrix and a quant scale of 2. The default
intra quantization matrix is:
8 16 19 22 26 27 29 34
16 16 22 24 27 29 34 37
19 22 26 27 29 34 34 38
22 22 26 27 29 34 37 40
22 26 27 29 32 35 40 48
26 27 29 32 35 40 48 58
26 27 29 34 38 46 56 69
27 29 35 38 46 56 69 83
Notice that the high frequency coefficients are more heavily quantized.
We are still taking advantage of image content, but we're also factoring
in the visual perception of high frequency information to get more
compression. Be aware that this matrix can be tailored for high bitrate
applications. A typical high bitrate matrix might be (John DeGroof,
don't try this on your Minerva system. It won't work correctly and
I would imagine it's not supported in the first place):
8 16 16 16 16 16 16 16
16 16 16 16 16 16 16 16
16 16 16 16 16 16 16 16
16 16 16 16 16 16 16 16
16 16 16 16 16 16 16 16
16 16 16 16 16 16 16 16
16 16 16 16 16 16 16 16
16 16 16 16 16 16 16 16
However, since DVD is bitrate limited the default matrix is almost
universally used. After the quantization process with the default
intra matrix and a quant scale of 2, we end with these coefficients:
158 67 -20 6 1 0 0 0
-30 -42 18 0 -3 1 0 0
-39 0 5 -8 5 -1 0 0
-2 -8 -5 7 -2 -1 1 0
0 -5 1 -1 -1 2 0 0
1 -1 -2 -1 1 -1 0 0
0 1 -1 0 0 0 0 0
0 1 0 0 0 0 0 0
Running these coefficients through the dequant/IDCT process results in the
following block of pixels:
106 99 99 98 102 110 113 122
159 147 131 120 113 112 117 118
201 198 183 172 155 136 127 116
212 209 207 203 191 183 161 121
226 214 213 208 209 209 158 87
221 218 212 214 223 188 108 57
216 218 214 216 199 128 63 56
215 210 215 213 154 79 53 57
If we compare the original block with the reconstructed block, we get an
error map that looks like this:
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 1 0 0 0 0
0 -1 0 0 0 0 0 0
0 0 0 0 0 1 0 0
0 0 0 0 1 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
Wow, not too shabby! Let's try again with a quant scale of 4. The quantized
coefficients will be:
1264 272 -95 33 0 0 0 0
-120 -168 99 0 -13 0 0 0
-180 0 26 -54 29 0 0 0
-11 -44 -26 40 -14 0 0 0
0 -26 0 0 0 17 0 0
0 0 -14 0 0 -20 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
Which gives us more zero coefficients (and more compression). The error
map will be:
1 3 -1 1 1 -1 1 -5
0 0 6 5 0 -1 0 2
-3 0 -2 -8 0 5 -3 5
3 5 -2 -6 5 1 -13 -7
1 -3 -3 5 0 -4 1 3
0 3 1 -2 -9 -5 8 4
-5 5 4 -4 -5 3 3 -8
2 -3 6 4 0 6 3 -5
At a quant scale of 4, the reconstructed block begins to degrade although
this would look just fine to the visual system since only a couple of
pixels have a significant error. If we keep the quant scale at or below 4,
then the reconstructed video will look pretty damn good. If we keep
the quant scale at or below 2, then the reconstructed video will be
essentially identical to the source and will be considered "transparent".
The whole point of variable bitrate (VBR) MPEG-2 encoding is to keep the
quant scale at minimum values (versus constant bitrate encoding where the
quant scale is used to control bitrate).
So, the bottom line is that when it's said that MPEG-2 encoding is "lossy",
the ONLY thing that is being lost is accuracy of reconstructed pixel
levels. There is absolutely NO loss in frame rate, pixel resolution,
color depth, or anything else. And when the accuracy of reconstructed
pixels is kept high by variable bitrate encoding (as demonstrated above),
MPEG-2 video quality can be essentially identical to the master from
which it is encoded. This is why DVD is such a fantastic format and
anyone who slams MPEG-2 compression just doesn't realize what it's
capable of.
If anyone would like the 'C' code that I used to generate the above
data, e-mail me.
> Afterall the human eye can only differentiate something like 64 gray
> scales and much less in colors. It is most sensitive to greens and least
> sensitive to reds.
>
> So I think in this case since the limited bit depth of the MPACT2
> actually interacts favorably with the way MPEG2 works. If this is the
> case then it is a nice change since we all know how other things like the
> sharpness control on a TV interacts poorly with MPEG2.
>
> Cheers
> --
> Christy
> -- Replace nospam with aurinia in order to reply
>
Ron K6MPG
Billy Newsom
Mark Hapner wrote:
> Before everyone runs out to buy an MPACT based DVD adaptor you should
> look at the new Holywood 3 from from Sigma Designs. It is likely to
> provide superior picture quality at higher resolutions in true color.
> Its available now (it just came out in the last few weeks).
>
> I'm check with Sigma Designs on DTS support but I'm guessing its not
> there. But for $120 you can throw it away when a version with DTS
> comes out.
>
> http://www.realmagic.com/hllywood.html
> http://www.computeresp.com/prdct/976/222.html
>
> For scan converted display of TV, s-video and composite sources at
> resolutions up to 1600 x1200 in true color I'm using and ATI All in
> Wonder Pro AGP 8 MB graphics adaptor with very nice 3D support which
> is very close to Voodoo performance. It does an excellent job. Even
Uh, what was all of that? Are you displaying teevee signals on your PC monitor
or computer screens on your teevee? You sort of lost me, at least.
> regular VHS tapes look like films (albeit a bit fuzzy) with no scan
> lines even with your nose on the screen (I'm using a 1280 x 1024
> display on a 90" screen).
A projector? What kind? I'm asking these questions because it seems like your
teevee projector is capable of displaying the same resolutions as PC monitors,
and I can't tell which way you're going, or why.
Anyhoo, tell me what you think about the ATI card -- I've heard it's a good
video card, but I thought it did MPEG-2. Anyway, what does all of this
conversion allow you to do? Record computer screens on video tape? Do you
capture/record teevee shows to your hard disk? If your projector is so awesome,
why buy a 17" monitor?
> There is simply no reason to even consider a line doubler from this
> point on.
Speaking of all of this converting, I've seen a product that will supposedly
blow the doors off any 2D or 3D digital comb filter -- although a signal from a
DVD player won't need one. It's called the Crystal Vision VPS-1.
> http://www.atitech.ca/products/pc/aiw_pro/index.html
> http://www.computeresp.com/prdct/706/791.html
>
> On Wed, 10 Jun 1998 14:26:48 -0700, ka...@nospam.com (Christy) wrote:
> >Yesterday I finally got an MPACT2 based PC DVD player. I have been
> >looking forward to seeing this video board/MPEG2 decoder for the last two
> >months, ever since I spoke with one of the engineers on IRC.
--
Billy Newsom, webmaster of The Motherboard HomeWorld
http://www.motherboards.org/
Cliff Watson
Mario,
The Mpact2 3DFusion DVD is a 2D/3D graphics card/DVD decoder combination
card.
The card has video signal paths for RGB, S-video, and Composite YUV.
S/PDIF for Dolby Digital coax.
Stereo mini plug for Dolby Pro Logic and/or SRS TruSurround 2-channel
analog.
The PCI version has a internal audio out to connect with a sound card and
the AGP version has three internal connectors; aux in, CD in, and audio out.
Yes, the 3DFusion card can be bought separately from Digital Connections.
I'm currently using the 3DFusion with a Hitachi GD-2000 DVD-ROM drive.
Best Regards, Cliff
Mario Kalogjera wrote in message <3581C682...@zg.tel.hr>...
Mark Hapner
The ATI does not contain an MPEG2 decoder; however, ATI provides a
software implemention that only supports the AGP version. It uses the
scaling and scan conversion hardware on the card. Since dedicated DVD
controllers are so cheap I would use one like the Hollywoood3.
I use a PC and high rez crt projector as a combination home theater,
super web TV and game machine.
Anyone thinking of buying a projector should be certain that it
supports atleast 1280x1024 resolution. In my opinion, that is the
minimum resolution needed for a PC/home theater. As I noted earlier,
inexpensive video display hardware now exists to display any video
source on such a system with surprising quality. As I'm typing this
I'm watching the world cup in a 3' window on the other side of the
screen.
Folks who are planning home theaters that are not considering driving
them with a PC are ignoring the major shift in PC convergence
technology that has occurred. They will likely be kicking themselves
for this error in the near future.
In particular, several of the projectors frequently discussed on this
alias are not suitable for use as a PC monitor. Folks should think
twice before buying one of these.
Those looking for a good wireless keyboard should consider the
VersaPoint unit. I've found it to be an excellent, omni-direction
keyboard.
http://necxdirect.necx.com:8002/cgi-bin/auth/ifilelnk_q?key=0000106014&nonce=guest_cesp
-- Mark Hapner
On Sun, 14 Jun 1998 09:30:42 -0500, Billy Newsom
<webm...@motherboards.org> wrote:
>Mark Hapner wrote:
>> Before everyone runs out to buy an MPACT based DVD adaptor you should
>> look at the new Holywood 3 from from Sigma Designs. It is likely to
>> provide superior picture quality at higher resolutions in true color.
>> Its available now (it just came out in the last few weeks).
>>
>> I'm check with Sigma Designs on DTS support but I'm guessing its not
>> there. But for $120 you can throw it away when a version with DTS
>> comes out.
>>
>> http://www.realmagic.com/hllywood.html
>> http://www.computeresp.com/prdct/976/222.html
>>
>> For scan converted display of TV, s-video and composite sources at
>> resolutions up to 1600 x1200 in true color I'm using and ATI All in
>> Wonder Pro AGP 8 MB graphics adaptor with very nice 3D support which
>> is very close to Voodoo performance. It does an excellent job. Even
>
>Uh, what was all of that? Are you displaying teevee signals on your PC monitor
>or computer screens on your teevee? You sort of lost me, at least.
>
>> regular VHS tapes look like films (albeit a bit fuzzy) with no scan
>> lines even with your nose on the screen (I'm using a 1280 x 1024
>> display on a 90" screen).
>
>A projector? What kind? I'm asking these questions because it seems like your
>teevee projector is capable of displaying the same resolutions as PC monitors,
>and I can't tell which way you're going, or why.
>
>Anyhoo, tell me what you think about the ATI card -- I've heard it's a good
>video card, but I thought it did MPEG-2. Anyway, what does all of this
>conversion allow you to do? Record computer screens on video tape? Do you
>capture/record teevee shows to your hard disk? If your projector is so awesome,
>why buy a 17" monitor?
>
>> There is simply no reason to even consider a line doubler from this
>> point on.
>
>Speaking of all of this converting, I've seen a product that will supposedly
>blow the doors off any 2D or 3D digital comb filter -- although a signal from a
>DVD player won't need one. It's called the Crystal Vision VPS-1.
>
Dean McManis
Christy,
Thanks for the keyboard info. I called the San Jose Fry's, and they said
that only Campbell and San Jose stores carry them, and that they were
both out of stock. But I had to go to the San Jose Fry's anyway to pick
up something else, and there was one left on the shelf!!
It looks like a good design. Light, compact, with good tactile feel.
Tomorrow, I get my Pioneer DVD/MPACT2 combo (hopefully) so I'll finally
get to put it all together, and try it out myself.
I can hardly wait.
-Dean.
go...@hotmail.com
> > If you say the output can be as high as 1600x1200, then there must be scan
> > upconversion done to the progressive line doubled signal which is
essentially
> > 640x480 VGA signal. The set-up should allow interpolator type function ie.
> > line 2.5X (800x600), line 3x (1024x768), line quadrupling (1280x920) and
> > beyond.
>
> When I was talking about output up to 1600x1200 I was refering to its
> function as a video card. Its best playback mode is fullscreen which is
> only available in 720x480. You can play DVDs in a window but I don't
> think you will get as good results. I haven't seen a good resampler yet
> on a computer.
I went to the MPact2 homepage and saw the specs which indicate that the chip
has scaling capabilities besides displaying the DVD native resolution. Is
that the "window" display you were referring to? Couldn't you set the display
at 1024x768 and size the window to the max? Would appreciate if you could
advise on picture quality when that is done.
>
> > Try outputing the signal at 1024x768 which is line tripling at the highese
> > possible vertical refresh rate to get a stable (non flickering) ouput. The
> > scan lines should be filled in nicely.
>
> But this plays 720x480 progressive. Tripling of 480p is 1440p which
> isn't practical on most displays or projectors.
When I say tripling, i'm referring to the trpling of interlaced 480lines, as
in line doubling means doubling of 480i to 480p. Therefore by line trpling i
mean 1024x768 progressive, or thereabouts. This is important to me as my
projector still show visible scan lines at 480p and blurring at 960p
(quadrupling) due to line overlapping. I have tried the Deuce Video Scaler
and the "sweet spot" or native resolution of my projector is line tripling or
720p. BTW, the deuce gave a smooth and stable picture but was low in
resolution and sharpness even compared to the Dwin LD2 (obsolete model)
> As I said, I haven't seen a good resampler (what you call an upscaler).
> I have read somewhere that the all in wonder pro doesn't have all that
> good video. The only decent DVD decoder that we have available right now
> is the MPACT.
Is this the only Mpact based card at the moment? Which is your view is the
best card to use to convert interlaced NTSC laserdisc signals to progressive
(preferrably at 1024x768) signals to be displayed on monitors or graphic
grade projectors?
Thanks
Dennis
Greg Rogers
In article <355BDE...@inreach.com>, Dean McManis <de...@inreach.com> wrote:
> Tomorrow, I get my Pioneer DVD/MPACT2 combo (hopefully) so I'll finally
> get to put it all together, and try it out myself.
>
> I can hardly wait.
Both I and my CyberTheater partner Bill Cruce, received separate units
yesterday. Neither of them will play region coded discs. We will contact
Digital Connection tomorrow and see if we can discover the source of the
problem. The installation is very simply and gives no indication of any
problems but the units simply refused to play region coded discs. They
play Video Essentials since it isn't region coded. If anyone else has seen
this problem and solved it, please let us know.
I've looked at the test patterns on VE and also the video sequences. But I
want to see some 16:9 film material before I say more.
Greg
--
**************************************************************
* Greg Rogers http://www.cybertheater.com *
* CyberTheater(tm): The Internet Journal of Home Theater *
**************************************************************
go...@hotmail.com
> The card has video signal paths for RGB, S-video, and Composite YUV.
>
Is that inputs or outputs? I need an s-video input so that i can use the card
as a line doubler or quadruapler.
Christy
In article <gregsr-1506...@u1-121.spiritone.com>,
gre...@SpiritOne.com says...
> In article <355BDE...@inreach.com>, Dean McManis <de...@inreach.com> wrote:
>
> > Tomorrow, I get my Pioneer DVD/MPACT2 combo (hopefully) so I'll finally
> > get to put it all together, and try it out myself.
> >
> > I can hardly wait.
>
> Both I and my CyberTheater partner Bill Cruce, received separate units
> yesterday. Neither of them will play region coded discs. We will contact
> Digital Connection tomorrow and see if we can discover the source of the
> problem. The installation is very simply and gives no indication of any
> problems but the units simply refused to play region coded discs. They
> play Video Essentials since it isn't region coded. If anyone else has seen
> this problem and solved it, please let us know.
Mine played region one discs just fine on Win98. Since the software
allows you to set region through the installer I wonder if there could be
a driver bug.
What OS and version are you running under?
Cliff Watson
go...@hotmail.com wrote in message <6m2l19$ksd$1...@nnrp1.dejanews.com>...
Sorry Dennis, Should have stated they are all outputs.
Cliff
joe...@nowhere.com
Now I remember why I asked you to marry me ;-]]]]
----joe
>On Wed, 10 Jun 1998 14:26:48 -0700, ka...@nospam.com (Christy) wrote:
>Yesterday I finally got an MPACT2 based PC DVD player. I have been
>looking forward to seeing this video board/MPEG2 decoder for the last two
>months, ever since I spoke with one of the engineers on IRC.
>
>What got me excited about this chipset is that it does true progressive
>scan DVD decoding on DVDs that are encoded from film sources. It also
>has a special 720x480x16bit native graphics mode for DVD playback so it
>sounded like something worth waiting for and to get excited about.
>
>You can get more info about the mpact chipset by visiting www.mpact.com
>
>I ordered the DVD kit from http://www.digitalconnection.com/ for $355
>including overnight shipping. The kit itself cost something like $309.
>It comes with the Pioneer 2.6x DVD rom drive, which to my suprise was a
>slotted (trayless/caddyless) drive like the slotted pioneer CD roms. It
>just has a big long slot that you stick the DVD or CD in.
>
>When I first got the kit I noticed that the heat sink on the power
>regulator had fallen off, which wasn't a good first sign. Since we have
>an electronics lab in our house we just glued it back on. I suspect this
>was a problem in shipping so if you buy from digitalconnection just make
>sure you check for any damage and return it if there are any problems.
>
>Unlike other DVD decoder cards this one is an actual 8 Meg video card
>with 3d accelerator. It supports pretty decent modes and refresh rates
>up to something like 1600x1200 at 75 Hz.
>
>So basically this $1500 PC beats a $10,000 Faroudja. And in some ways it
>gives a better picture than their quadroupler, although it obviously
joe...@nowhere.com
>Those looking for a good wireless keyboard should consider the
>VersaPoint unit. I've found it to be an excellent, omni-direction
>keyboard.
>http://necxdirect.necx.com:8002/cgi-bin/auth/ifilelnk_q?key=0000106014&nonce=guest_cesp
>
>-- Mark Hapner
Agreed. I have been using this wireless keyboard for months...it's
great and can be found for $70-$79.
---joe
Alen Koebel
Dale Adams wrote:
> One interesting transition occurs between two film sources where the 3:2
> pulldown sequence is messed up. The transition from the soccer game in
> the rain (a film source) to a couple walking under bleachers (also a film
> source) was apparently edited in the video domain without regard to the
> 3:2 sequencing since that sequencing is disturbed during the scene
> transition. (If you have a way to capture the fields and view them in
> slow sequence you can see this.) This could be either a goof-up on Video
> Essentials, or, more likely, is another explicit test case to see how a
> video processor handles this sort of transition sequence.
According to Joe Kane himself it is indeed an explicit test case for
testing how long it takes a line multplier to detect and lock-on to
the 3/2 pulldown sequence.
---------------------------------------------------------------------
Alen Koebel Video and Display Engineering Electrohome Ltd.
www.electrohome.com Disclaimer: You know the rest .....
"Non existat prandium gratuitum." - Fulminous Minor
Alen Koebel
Christy wrote:
> Afterall the human eye can only differentiate something like 64 gray
> scales and much less in colors. It is most sensitive to greens and least
> sensitive to reds.
Although this is often said, it's not really true, or at least is true
only for very specific circumstances. What is important for video
applications is the ability to distinguish gray shades as they are
smoothly increasing or decreasing in level across the image. That's
what you see when you look at someone'e face or a blue sky, for
example. That's where quantization errors will be most noticeable.
For these images, the difference between 6-bit linear quantization
(18 bit color) and 8-bit linear quantization (24 bit color) is very
obvious. Eight bit quantization is easily noticeable as discreet steps.
Ten bit is required in practise for a smooth image, especially if the
image is bright. That said, any level of quantization can be made to
look much better by dithering. A 2 x 2 ordered dither, for example,
can simulate about two bits of additional grayscale resolution, at
the expense of increased noise in the image and some blurring of edges.
Another way to improve the grayscale resolution is to transform to a
non-linear quantization. In this way, six bits can store the
equivalent of eight (or more) bits of linear grayscale.
Christy
In article <3585AF...@electro.com>, ako...@electro.com says...
> Christy wrote:
>
> > Afterall the human eye can only differentiate something like 64 gray
> > scales and much less in colors. It is most sensitive to greens and least
> > sensitive to reds.
>
> Although this is often said, it's not really true, or at least is true
> only for very specific circumstances. What is important for video
> applications is the ability to distinguish gray shades as they are
> smoothly increasing or decreasing in level across the image. That's
> what you see when you look at someone'e face or a blue sky, for
> example. That's where quantization errors will be most noticeable.
> For these images, the difference between 6-bit linear quantization
> (18 bit color) and 8-bit linear quantization (24 bit color) is very
> obvious. Eight bit quantization is easily noticeable as discreet steps.
> Ten bit is required in practise for a smooth image, especially if the
> image is bright. That said, any level of quantization can be made to
> look much better by dithering. A 2 x 2 ordered dither, for example,
> can simulate about two bits of additional grayscale resolution, at
> the expense of increased noise in the image and some blurring of edges.
> Another way to improve the grayscale resolution is to transform to a
> non-linear quantization. In this way, six bits can store the
> equivalent of eight (or more) bits of linear grayscale.
So you need ten bits for grayscale. Now how many bits do we need for
each color component in RGB space (or YUV if you prefer). Couldn't you
encode reds with harsher quantization than say green?
go...@hotmail.com
I am thinking of using the ATI all in wonder pro as the video processor
(deinterlace and line multiply) for external laserdisc source and the 3DFusion
DVD kit as the DVD source for output to a graphic grade CRT projector.
As the 3DFusion is itself a video accelerator, will it conflict with the ATI
card?
Thanks
Dennis
> Earlier I wrote:
>
> > His machine is running Win98 and mine Win95 B. Mine is a plain old Intel
> > 440LX motherboard with P-2 processor. The 3D Fusion PCI video board is the
> > ONLY board installed in the computer. There is one master hard drive on
> > the primary IDE channel and the Pioneer DVD player as a master on the
> > secondary IDE. Both players will only play non-region coded discs like
> > Video Essentials. We have reinstalled the software a zillion times.
>
> OK, the mystery appears to be solved. The Pioneer drives apparently are
> not compatible with the Intel Bus Master IDE Drivers. We both had these
> installed in our respective machines. After we removed them, the 3D Fusion
> cards finally appear to be operating normally. I'm exhausted after trying
> to make them work for 3 days, so testing on the big screen must wait for
> tomorrow night.
>
> Greg
>
> --
> **************************************************************
> * Greg Rogers http://www.cybertheater.com *
> * CyberTheater(tm): The Internet Journal of Home Theater *
> **************************************************************
>
go...@hotmail.com
> > I have seen DVD images produced by standalone products with the above
> > features projected by a graphic grade projector (Runco IDP980 ultra) and
when
> > viewed from a distance about twice the screen width, it approaches 90% of an
> > HDTV (1080i) video which I saw projected through a Sony G70 from the same
> > viewing distance.
>
> Either you are exaggerating or the Sony projector was not set up well.
> 1080i is much more than a 10% or so improvement over line multiplied DVD.
> That said, line multiplied DVD that has been matched to the display
> device (i.e., the number of lines being displayed is at the projector's
> "golden" frequency) gives a very pleasing image. I saw line quadrupled
> DVD (The Fifth element) on a Hughes/JVC ILA-12K just a couple of days
> ago. It was emminently watchable, even on a theater sized screen, but
> it could not really compare to the HDTV images that preceded it
> (scenes from Titanic and The Lost World, played on a D5 machine).
It wasn't a back-to-back A/B comparison like yours but it was only about 3
hours apart. It was my subjective impression of image quality from 2x screen
distance viewing. Any viewing distance closer would start to show the
difference considerably.
I saw both the euro DVB HDTV broadcast as well as the ATSC HDTV. Both were
transmitted through air from a station 10km away. The only complain I had was
slight flicker from the 2x viewing distance and slight twitter from 1x
viewing distance.The DVD set-up had a flicker free stable picture at >75Hz
refresh rate.
HDTV is definitely far superior and you can have a far larger viewing angle
before artifacts start to irk you. What i was saying was that I am only very
slightly less satisfied with the best DVD picture on a very good set-up
compared with a run-of-the-mill HDTV (broadcast as opposed to direct source
from Harddrive or similar device) viewed from 2x distance. I don't know if
the Sony G70 is a lesser performer compared to the Runco980 ultra.
But the future is exciting isn't it? Now to get the laser surgery on my eyes
to get 100% resolution from them!
Dennis
Greg Rogers
> I am thinking of using the ATI all in wonder pro as the video processor
> (deinterlace and line multiply) for external laserdisc source and the 3DFusion
> DVD kit as the DVD source for output to a graphic grade CRT projector.
>
> As the 3DFusion is itself a video accelerator, will it conflict with the ATI
> card?
Sorry Dennis. I'm not qualified to answer PC or Windows questions. Call
Digital Connections for these kinds of questions. Sorry.
Now its off to test this baby on some real video.
Brad Wilson
Mark Hapner wrote in message <35869242...@news.thegrid.net>...
>See my post on the RealMagic Hollywood3. It is superb.
Well, I bit the bullet based on your review (and the darn low price <g>) and
ordered a Hollywood 3. It will be here Friday. I'll make sure to post my
thoughts on it once I've dinked around with it for the weekend.
So, no go on the REALmagic Ventura, can't use the Mpact because I don't need
another video card, and the Dxr2 card performance left a LOT to be desired
(which really turned me off to analog overlay technology in the first
place). Hopefully this one will do the goods.
-- Brad Wilson, student of Objectivism -------- bradw<at>pobox<dot>com --
"I would remind you that extremism in the defence of liberty is no vice!
And let me remind you also that moderation in the pursuit of justice is
no virtue!" - Barry Goldwater
Mark Hapner
I have seen images from DVD s-video on both a Faroudja quad and the
ATI.
I have not seen images from cable TV on a Faroudja.
On Tue, 16 Jun 1998 14:25:09 -0700, ka...@nospam.com (Christy) wrote:
>In article <358738eb...@news.thegrid.net>, m...@thegrid.net says...
>> On Tue, 16 Jun 1998 04:51:16 GMT, go...@hotmail.com wrote:
>>
>> I have not seen external line doubled TV images so I can't make a
>> comparison with the ATI.
>>
>> I have seen the Toshiba RPTV with built in line doubler and the ATI
>> image seems superior.
>>
>> What I can say is that with my normally mediocre TCI cable feed, good
>> channels have a film like quality without any visible scan lines or
>> jaggies.
>>
>> Poorer quality channels sometimes have a fine herringbone pattern that
>> comes and goes.
>>
>> Color quality seems good. Overall I'm pleased with image quality.
>>
>> You also get a few nice features for free like TV in a window. Closed
>> caption scanning for key words which triggers a video capture. TV
>> magazine capture mode that stores a program as a set of stills. A
>> thumbnail screen that automatically updates. Scan conversion of
>> composite and s-video sources. Support for DBX TV sound. Support for
>> all the new TV features in Win 98, TV out which looks surprisingly
>> good, real time zoom, individual color, contrast, brightness and tint
>> controls for each source, etc.
>>
>> If you bought this from Faroudja it would probably cost $50K. From ATI
>> it costs $150 more than their regular grapics adaptor.
>
>I'm confused. You say you have not seen external line doubled TV images
>and yet you talk about the ATI as doing as well as a Faroudja. So have
>you seen the output of a Faroudja or not? Please clarify...
>
>Thanks
go...@hotmail.com
> I have been reading the posts on the 3DFusion decoder/display card with
> great interest. I am interested in better performance from my DVD player in
> my media room, which includes a CRT front projector and Sony 7000, among the
> requisite other stuff. I demo'd a Philips DVX8000 awhile back, and was
only
> partially impressed. I have been demo'ing line doublers and scalers
> (currently using demo 3rd gen. "Deuce" - still has decoder probs - supposedly
> will be built with a new decoder within 60 days... another story...), and
> none have really done it for me. IEV was an all around good performer, but
> not stellar, IMO.
I have seen a 2nd gen deuce vs a Dwin LD2. The deuce handled motion pretty
well and the picture at 1024x768 showed visible scan lines only when i was 5
ft away from screen. Picture at 72Hz refresh rate was very stable. The BIG
drawback compared to the Dwin (which is an obsolete model) was much lower
resolution and sharpness from the deuce. The color was also not as good (ie.
less natural). The one i saw had a Bt 848 chip i think. DO you know what chip
they will be putting into the 4th gen deuce.
In your opinion, does the ATI all in wonder pro's processing of laserdisc
signal come close to the deuce? If so, then i could live with that as an
interim solution until something better comes along. Otherwise what's a good
cheap interim solution. You see, i just bought a used projector that accepts
only RGB and so no processor, no picture.
Thanks
FRiC
In article <6m7ivk$4q1$1...@nntp0.detroit.mi.ameritech.net>,
Brad Wilson <this....@my.email.addr> wrote:
>Mark Hapner wrote in message <35869242...@news.thegrid.net>...
>
>>See my post on the RealMagic Hollywood3. It is superb.
>
>Well, I bit the bullet based on your review (and the darn low price <g>) and
>ordered a Hollywood 3. It will be here Friday. I'll make sure to post my
>thoughts on it once I've dinked around with it for the weekend.
So, what does the Hollywood 3 have over the Hollywood 2?
Mario Kalogjera
ki...@cmbn.rutgers.edu wrote:
>
> In article <MPG.fe8e1c61...@nntp.best.com>,
> ka...@nospam.com (Christy) wrote:
> > One thing I did look at more closely tonight is the difference between
> > DVDs from true interlace sources and DVDs from film. Unlike most film
> > DVDs, DVDs from interlace sources must be processed through a line
> > doubling algorithm designed to handle interlaced video. This can prove
> > quite difficult to do. Consequently the performace of the MPACT board on
> > these sources is not as good as through my linedoubler. Oh well. I'll
> > take films looking amazing for now and live with the doubler for my
> > interlaced DVDs and Laserdiscs.
> >
> Interesting. All video is stored on DVD in an interlaced format, and
> film sources have been through 3:2 pulldown to convert them to 29.97fps.
> Does the card do some sort of reverse telecine to get rid of the frames
> with mixed fields?
>
> Charles King
>
> -----== Posted via Deja News, The Leader in Internet Discussion ==-----
> http://www.dejanews.com/ Now offering spam-free web-based newsreading
--
This is completely wrong. DVD is stored in only possible format -
progressive frames. The fact that MPEG-2 frames are internally
interlaced if encoded from an interlaced source doesn't make DVD an
interlaced format. It only does the best it can to properly predict
intra-field changes thus MPEG approximation takes place only on even or
odd lines, horizontally, not mixing the two fields since it would cause
quite a mess.
Check your movies in frame-advance mode and then compare that to an
interlaced source. Since all DVD players can only advance full frames
(unlike VCRs) you will notice how movie frames advance cleanly on after
another, while frames of an interlaced source have every second line
displaced (when viewed on progressive displays) or shudders between the
two fields a frame consist of on TV (interlaced displays).
The problem here is that most todays players and DVD cards take that
film source progressive signal and interlace it for display, as
instructed by the DVD stream, i.e. the 3:2 pulldown is not performed at
transfer time, but rather in the player itself by the PAL/NTSC encoder
chip. MPACT-2 seems to put out original progressive signal from a film
source that the progressive scan TV can display unprocessed. I cannot
however predict what would happen with interlaced sources on a
progressive scan TV using MPACT-2, probably there would be a combing
effect present like on PC-monitors, unless the TV was also able to
produce interlaced display.
Regards
Mario Kalogjera
E-mail: mario.k...@usa.net
mario.k...@zg.tel.hr
myt...@usa.net
Member of: FX InterZone Media Group
www.geocities.com/hollywood/academy/5055/
fxint...@geocities.com
ki...@cmbn.rutgers.edu
In article <MPG.ff08838a...@news.netscape.com>,
ka...@nospam.com (Christy) wrote:
>
> So the player actually does the 3:2 pulldown.
Hah! Of course, I'd forgotten about that. All is now clear. Thanks
chri...@my-dejanews.com
In article <6m6659$jkm$1...@nnrp1.dejanews.com>,
chri...@my-dejanews.com wrote:
>
> In article <MPG.fe899f74...@news.netscape.com>,
> ka...@nospam.com (Christy) wrote:
> >
> > Yesterday I finally got an MPACT2 based PC DVD player. I have been
> > looking forward to seeing this video board/MPEG2 decoder for the last two
> > months, ever since I spoke with one of the engineers on IRC.
> >
> > What got me excited about this chipset is that it does true progressive
> > scan DVD decoding on DVDs that are encoded from film sources. It also
> > has a special 720x480x16bit native graphics mode for DVD playback so it
> > sounded like something worth waiting for and to get excited about.
> >
> > You can get more info about the mpact chipset by visiting www.mpact.com
> >
>
> I have been reading the posts on the 3DFusion decoder/display card with
> great interest. I am interested in better performance from my DVD player in
> my media room, which includes a CRT front projector and Sony 7000, among the
> requisite other stuff. I demo'd a Philips DVX8000 awhile back, and was
only
> partially impressed. I have been demo'ing line doublers and scalers
> (currently using demo 3rd gen. "Deuce" - still has decoder probs - supposedly
> will be built with a new decoder within 60 days... another story...), and
> none have really done it for me. IEV was an all around good performer, but
> today - am anxious to install and try in a 333mhz AGP PC. Christy, among a
> few others, also seem to be at least trying this board with their projectors,
> with good result, on film transfers off of DVD. At the cost of this board,
> if it performs as well as some say, to use it only on film transfer DVDs
> would still be a no brai>
<input type=
chri...@my-dejanews.com
In article <35869242...@news.thegrid.net>,
m...@thegrid.net (Mark Hapner) wrote:
>
> On Tue, 16 Jun 1998 10:51:03 -0400, "Brad Wilson"
> <this....@my.email.addr> wrote:
>
> See my post on the RealMagic Hollywood3. It is superb.
>
>
> >God help me, I want DVD! :-p I have a Creative Dxr2 card and it blows. I
> >want something that uses the feature connector. I don't want a whole new
> >video card (ala Mpact2) because I love the ATI All-in-Wonder Pro card. I
> >tried the REALmagic Ventura (not technically an end user product), and I
> >can't get the thing to work. I suppose I'm Cinemaster bound.
> >
> >What else is there to choose from? Does ATI have a DVD daughter card
> >available?
> >
> >
>
>
Just posted a minireview of the 3Dfusion/Mpact DVD I recieved yesterday ...
Mostly excellent - take a look. Also had to order a Hollywood3 just to
compare - should have that tommorow - may take a few days to get the time to
put it in and uninstall / reinstall drivers etc... - Ill let the group know
the results.
- Chris.
Alen Koebel
Well, you don't absolutely *need* 10 bits displayed, unless you are
super critical. You do, however, need 10 bits in the processing
stages prior to display in order to display 8 bits. With dithering,
8 bits is more than acceptable. Without dithering you can see the
steps in the grayscale on certain images, but not on most. Even so,
most people would still consider the image to be very good.
Regarding your question about color, I don't know the answer to that.
Maybe Greg can take a stab at it.
Christy
In article <6m8vjg$unv$1...@nnrp1.dejanews.com>, chri...@my-dejanews.com
says...
> In article <35869242...@news.thegrid.net>,
> m...@thegrid.net (Mark Hapner) wrote:
> >
> > On Tue, 16 Jun 1998 10:51:03 -0400, "Brad Wilson"
> > <this....@my.email.addr> wrote:
> >
> > See my post on the RealMagic Hollywood3. It is superb.
> >
> > ATI does have a hardware DVD upgrade but is an oem only product.
> >
> > >God help me, I want DVD! :-p I have a Creative Dxr2 card and it blows. I
> > >want something that uses the feature connector. I don't want a whole new
> > >video card (ala Mpact2) because I love the ATI All-in-Wonder Pro card. I
> > >tried the REALmagic Ventura (not technically an end user product), and I
> > >can't get the thing to work. I suppose I'm Cinemaster bound.
> > >
> > >What else is there to choose from? Does ATI have a DVD daughter card
> > >available?
> > >
> > >
> >
> >
>
> Just posted a minireview of the 3Dfusion/Mpact DVD I recieved yesterday ...
> Mostly excellent - take a look. Also had to order a Hollywood3 just to
> compare - should have that tommorow - may take a few days to get the time to
> put it in and uninstall / reinstall drivers etc... - Ill let the group know
> the results.
I don't mean to sound dumb but where did you post the minireview? If
here, what is the article header.
Thanks
Brad Wilson
> Just posted a minireview of the 3Dfusion/Mpact DVD I recieved yesterday
...
> Mostly excellent - take a look. Also had to order a Hollywood3 just to
>compare - should have that tommorow - may take a few days to get the time
to
>put it in and uninstall / reinstall drivers etc... - Ill let the group
know
>the results.
Turns out the Hollywood3 isn't really a product. Most companies selling a
"Hollywood3" are really selling you a REALmagic Ventura, which only works
with a very limited number of video cards. My ATI is supposedly on the
list, but I still couldn't get it to work. So I suppose I'm going to have
to get a Cinemaster or a Hollywood 2. Maybe the new Hollywood+ will be
good.
Chris Lewis
In article <6m8vjg$unv$1...@nnrp1.dejanews.com>,
chri...@my-dejanews.com wrote:
>In article <35869242...@news.thegrid.net>,
> m...@thegrid.net (Mark Hapner) wrote:
>>
>> On Tue, 16 Jun 1998 10:51:03 -0400, "Brad Wilson"
>> <this....@my.email.addr> wrote:
>>
>> See my post on the RealMagic Hollywood3. It is superb.
>>
>> ATI does have a hardware DVD upgrade but is an oem only product.
>>
>> >God help me, I want DVD! :-p I have a Creative Dxr2 card and it blows.
I
>> >want something that uses the feature connector. I don't want a whole
new
>> >video card (ala Mpact2) because I love the ATI All-in-Wonder Pro card.
I
>> >tried the REALmagic Ventura (not technically an end user product), and I
>> >can't get the thing to work. I suppose I'm Cinemaster bound.
>> >
>> >What else is there to choose from? Does ATI have a DVD daughter card
>> >available?
>> >
>> >
>>
>>
>
> Just posted a minireview of the 3Dfusion/Mpact DVD I recieved yesterday
....
> Mostly excellent - take a look. Also had to order a Hollywood3 just to
>compare - should have that tommorow - may take a few days to get the time
to
>put it in and uninstall / reinstall drivers etc... - Ill let the group
know
>the results.
>
> - Chris.
>
>-----== Posted via Deja News, The Leader in Internet Discussion ==-----
>http://www.dejanews.com/ Now offering spam-free web-based newsreading
Sorry All,
DejaNews seems to have about a 50% success rate with passing
my posts... Its getting quite annoying.
Anyway, that mini review I mentioned ? Here it is (3rd attempt).
After a fairly painless install, I put in the Fusion3D/Pioneer 2.6 IDE
DVD. A few minor snags, but no real big deal. Documentation is weak.
After figuring out (no help from docs.) that 800x600x16 is required for full
screen DVD playback, I was ready to see what the fuss is all about:
WOW is the right word. Absolutely the right word. Again, I am viewing thru
a CRT projector (100" 4:3) from about 15' away (also on 17" PC mon.)
Video Essentials film sequences, you know, the Ammish guys in purple shirts
with straw hats? --They look GREAT! (Not the Quaker/Ammish; the actual
video.) As pointed out elswhere, however, non 24fps (presumably) source
material can look rough. Very rough. But, an oddity I noticed is that non
24fps material varies widely in its quality; not just traditional hard-to
line-double stuff either. All kinds of stuff can look bad (Again, with
perfection as the benchmark). The video from this card, in any mode, looks
far better than the other MPEG decoders that I have seen, and in 24FPS,
looks better than any video Ive seen, other than HDTV demos at trade shows.
Its that good. As far as test patterns, all the still frames look
essentially (no pun intended) perfect. NO noise. NONE. The S/W moving
sphere looks quite bad, however (non 24fps I suppose...). The thin
green/purple vertical stripe pattern MAY look a little off; cant tell -
Haven't compared to S-Vid from Sony 7000 yet. Also, another quirk: Red
seems to give this thing more of a fit (at least in non 24fps) than other
colors. Check out Fleetwood Mac, "Goldust Woman" - Red spotlights on stage.
UGLY. Bands, Blocks, digipixamess all around them. Also, I noticed that
the intro to Batman Forever (credits flying in) also may not look quite
right - some are red also. This is not verified; again, need to check 7000
video as ref. This worries me a little as this SHOULD be 24fps, but bear
in mind that when you pop a movie disc in, alot of the material on the disc
is not (or may not be) 24fps, such as movie studio intros, trailers, FBI
type screens etc... I noticed the "FBI" screen looking terrible on some
discs, perfect on others. But when the actual movie starts, WOW. Again,
the possible exception (POSSIBLE) is the Batman intro - but when the credits
are over: perfect.
Here is a sidebar, which may should be the headline here, but My setup
will NOT play the Bond flick TN dies. It plays the MGM intro, will play the
"trailer", but will not display the Menu or play the movie properly - I get
severely choppy audio and super blocky / partial video. Plays fine in 7000.
All other discs I played worked perfectly. Have tried all PC / Setup /
Setting changes I can think of, no diff. Anybody else tried this flick ?
In conclusion, I pretty much have the same thing to say that Christy
said in her review. For the price, its a no brainer. Bear in mind I am
using this as a DVD player for a home theater, where I am VERY picky about
image quality. Some of the harsh words I had maybe shouldn't be so harsh
for those of you who just want to watch DVD on your PC screen. Its pretty
fair all the way around for that. My comparisons are based on trials (in
home) of several line doublers, scalers, and PC cards.
I couldn't stand it; I also ordered a Hollwood3 which should be here
today. I'll let the group know how it works when I get a chance to put it
in (there goes my clean driver installation...)
- I'll comment more on both boards as I have time.
- Chris.
Francis Li
I have my initial review of the Leadtek Winfast 3D S800 8MB PCI card. It is
also an Mpact 2 3DVD card. Nothing too new for people who have been
following this thread on DVD, but it also has some 2D and 3D benchmarks
along with screenshots.
http://www.sonic.net/~fli/movies/dvd/leadtek_winfast_3d_s800_8mb_pci.htm
Francis
Mark Hapner
The product I bought/reviewed is the Hollywood+ which uses analog
overlay and should work with most any graphics adaptor. I bought it
from NC Buys.
http://www.netsales.net/index.wcgi/netcent/prod/1184670-1
As you can see they call it a Hollywood3. Given Sigma Design's product
name confusion for this product I can imagine that this might be
happening. What evidence do you have that it has?
-- Mark Hapner
Greg Rogers
Christy (or anyone else with a 3D Fusion), please check Tomorrow Never
Dies on the 3D Fusion. Another reader sent me email that is wouldn't play.
I checked my copy and it does the same as his. Just makes a blocky mess on
screen.
I just finished my first 8 hours of evaluating the 3D Fusion. I've got
some interesting comparisons but its too late (1:30 am) to write them up
now. I'll write them up tomorrow evening and post later tomorrow night.
Erik Olson
In alt.video.dvd Alen Koebel <ako...@electro.com> wrote:
> Christy wrote:
>> Afterall the human eye can only differentiate something like 64 gray
>> scales and much less in colors. It is most sensitive to greens and least
>> sensitive to reds.
64 gray levels? I thought it was 100 levels of contrast which is different
than levels of gray?
And least sensitive to Red? Hmmm.
> Although this is often said, it's not really true, or at least is true
> only for very specific circumstances. What is important for video
Often said?
I thought the human eye was least sensitive to the color blue?
What about the blue only button on the pro Sony monitors?
This is for other reasons, but .....
erik olson
er...@best.com
Christy
In article <gregsr-1806...@u1-70.spiritone.com>,
gre...@SpiritOne.com says...
>
> Christy (or anyone else with a 3D Fusion), please check Tomorrow Never
> Dies on the 3D Fusion. Another reader sent me email that is wouldn't play.
> I checked my copy and it does the same as his. Just makes a blocky mess on
> screen.
>
> I just finished my first 8 hours of evaluating the 3D Fusion. I've got
> some interesting comparisons but its too late (1:30 am) to write them up
> now. I'll write them up tomorrow evening and post later tomorrow night.
I just tested TMD this morning and mine also fails right in the menu. I
tried other MGM discs to see if they were a problem and they weren't.
I wish I had my old version of T2 that didn't work on the encore to see
if this is the same problem. My guess is that this is a software problem
that can be fixed with an upgrade to entertainment center.
Christy
In article <6mbp23$9tf$1...@nntp1.ba.best.com>, er...@shell14.ba.best.com
says...
> In alt.video.dvd Alen Koebel <ako...@electro.com> wrote:
> > Christy wrote:
> >> Afterall the human eye can only differentiate something like 64 gray
> >> scales and much less in colors. It is most sensitive to greens and least
> >> sensitive to reds.
>
> 64 gray levels? I thought it was 100 levels of contrast which is different
> than levels of gray?
>
> And least sensitive to Red? Hmmm.
>
> > Although this is often said, it's not really true, or at least is true
> > only for very specific circumstances. What is important for video
>
> Often said?
> I thought the human eye was least sensitive to the color blue?
I thought it was red but my memory could be faulty. If it isn't red then
why does NTSC composite video bandwidth limit reds to 40 lines of
resolution and limits blue to 135 lines. I thought the design of NTSC
color was based on the limits of the human eye. NTSC is an analog form
of perceptual coding.
Cheers
Alen Koebel
Erik Olson wrote:
>
> In alt.video.dvd Alen Koebel <ako...@electro.com> wrote:
> > Christy wrote:
> >> Afterall the human eye can only differentiate something like 64 gray
> >> scales and much less in colors. It is most sensitive to greens and least
> >> sensitive to reds.
>
> 64 gray levels? I thought it was 100 levels of contrast which is different
> than levels of gray?
The human eye has a huge dynamic range due to the action of the iris
but can only sense light levels within a relatively limited range at any
particular fixed iris dilation. The figure of "100 levels" was probably
derived by testing people's ability to distinguish between the levels
when looking at large areas of two different grays side by side (that's
the "specific circumstance" I mentioned). For a video image the eye's
ability to distinguish gray levels in a smoothly varying ramp is more
important. The eye's ability to detect the edges formed by the
transitions between quantization levels comes into play. The eye can
distinguish in excess of 256 levels under these conditions.
Alen Koebel
Christy wrote:
>
> In article <6mbp23$9tf$1...@nntp1.ba.best.com>, er...@shell14.ba.best.com
> says...
> > > Christy wrote:
> > >> Afterall the human eye can only differentiate something like 64 gray
> > >> scales and much less in colors. It is most sensitive to greens and least
> > >> sensitive to reds.
> >
> > 64 gray levels? I thought it was 100 levels of contrast which is different
> > than levels of gray?
> >
> >
> > > Although this is often said, it's not really true, or at least is true
> > > only for very specific circumstances. What is important for video
> >
> > Often said?
> > I thought the human eye was least sensitive to the color blue?
>
> I thought it was red but my memory could be faulty. If it isn't red then
> why does NTSC composite video bandwidth limit reds to 40 lines of
> resolution and limits blue to 135 lines. I thought the design of NTSC
> color was based on the limits of the human eye. NTSC is an analog form
> of perceptual coding.
NTSC limits spatial resolution, not grayscale (or color) resolution/
quantization (the so called z-axis).
Chris Lewis
In article <6mbbru$s...@dfw-ixnews6.ix.netcom.com>,
Chris Lewis <cl...@ix.netcom.com> wrote:
>In article <6m8vjg$unv$1...@nnrp1.dejanews.com>,
> chri...@my-dejanews.com wrote:
>
>>In article <35869242...@news.thegrid.net>,
>> m...@thegrid.net (Mark Hapner) wrote:
>>>
>>> On Tue, 16 Jun 1998 10:51:03 -0400, "Brad Wilson"
>>> <this....@my.email.addr> wrote:
>>>
>>> See my post on the RealMagic Hollywood3. It is superb.
>>>
>>> ATI does have a hardware DVD upgrade but is an oem only product.
>>>
>>> >God help me, I want DVD! :-p I have a Creative Dxr2 card and it
blows.
>I
>>> >want something that uses the feature connector. I don't want a whole
>new
>>> >video card (ala Mpact2) because I love the ATI All-in-Wonder Pro card.
>I
>>> >tried the REALmagic Ventura (not technically an end user product), and
I
>>> >can't get the thing to work. I suppose I'm Cinemaster bound.
>>> >
>>> >What else is there to choose from? Does ATI have a DVD daughter card
>>> >available?
>>> >
>>> >
>>>
>>>
>>
>> Just posted a minireview of the 3Dfusion/Mpact DVD I recieved yesterday
>.....
Ive been busy, busy. I removed/uninstalled 3Dfusion, Re-installed my
ATI AIWPro VGA card, and added the Hollywood3 (PCI mpeg2 decoder - overlay).
Installation was painless, documentation (while never technical enough) is
much more complete than Fusion's. Basically, what I wanted to do was watch
some DVD in the same (native) Rez/Refresh setup (on my projector and PC mon)
that the Fusion did: 720X480 @72hz. Since this card can scale, and is an
overlay, whatever windows rez. you have selected is also what you watch DVD
in, even full screen (not the case w/Fusion). So, the ATI card has a
720X480
mode (maybe most cards do now) so thats what I set up. I also specified a
72hz refresh. I assumed this would bypass the scaling portion of the chip
and give me similar results to the Fusion-generated video. After some
slight
sofware glitches in making 720X480 work (have to uncheck the "keep aspect
ratio" box for some reason, as well as recalibrate the setup/borders) full
screen, I was ready to view.
OUTSTANDING. I didn't think it would be, but it was; just like Mark
said. Bear in mind I have had one evening with it (albeit a long/late one)
however.
Could it be any better than Fusion I thought? I think it may be.
Here
is why (so far). Check intro to Batman Forever on Fusion; where the Warner
Bros. logo morphs into the Batman symbol. I thought on the fusion that
during this sequence, the neon-blue drop shadow around the bat symbol was a
little pixelated. A LITTLE pixelated. Im picking knits, but I did notice
it the other night, and its one of the notes I made. I quickly checked this
on the Hollywood, after I was convinced it was a contender, and I see NONE
of this. I was impressed, because it was so slight on the Fusion, I thought
I was probably being way to critical. Maybe not, because its gone on this
board. At this point, I wanted to see what it could do on non-24fps
material
(where the Fusion really falls flat, IMO). Its far superior to fusion.
Fleetwood Mac was the worst looking thing I have seen from anything that is
supposed to be "high quality" on the Fusion, but on the Hollywood3, it
looked, I think, about as good, if not better in some ways than my Sony 7000
through an
IEV-class doubler. The only thing I could detect that was not positive is
that the non 24fps material MAY, again MAY, be a little soft. Bear in mind
that Fleetwood Mac is probably the softest DVD I've seen to begin with
anyway.
This softness, if its there at all, is very slight. Its NOT in 24fps
material,
I do know that. Its difficult to quantify differences in "softness",
because I cant quickly compare 2 setups (certainly not Fusion and
Hollywood,
because I don't have 2 computers at home).
I feel that the majority of non24fps material I saw, (Vid. Essentials video
portions, F.Mac) Looked better on the whole than an analog connection to an
IEV-class unit.
One of my few complaints with the IEV is that it was slightly soft (to my
eye), But
most of the units in its class are also. The hollywood3 is very much a
capable performer
on non-24fps material. Not perfect, but far better than the Fusion, IMO.
With all of this behind me, I was ready to see what the scaling could
do, since Ive
never been happy with 640X480 (doubler) because the scan lines in brighter
portions of the
screen are still fairly visible from my viewing distance. I like 800x600
or 1024x768
(at least for my Proj. and my viewing dist.). So knowing that base
performance was definately
there, I bumped up rez. to 800x600 (still 72hz). The shocker? Still looks
amazing. Of course
its better, in that scan lines totally disapear! I tried and tried to find
something that I could
throw at it to make it hiccup. Couldn't do it. Bear in mind, however,
that I am not sure what
kind of material is "hard" for scaling technology. It is relatively new,
and I would assume that
high motion, jagged edges, fine detail, etc... and combinations of these
would cause problems, as
they do for doublers. But I think that would be incorrect, as I think
these cause problems for the
de-interalcing portions of doublers, and with the Hollwood, the de
interlacing (of non 24fps) is very
good / excellent (so far). So maybe scaling is "easy" as long as you feed
it a solid signal to scale.
Anyway, this is getting long, but I wanted to get some info out.
In summary:
A) Excellent 720x480 at 24fps
B) Apparently no penalty for up-scaling (at least to 800x600)
C) Very Good 30fps (again, any rez.)
D) Better sofware app. than Fusion, IMO, because of hotkeys.
E) Lots of technical tweaks avail. in driver (even though no documentaion)
F) Can keep your (better-performing/existing) VGA card
G) DIRT cheap ($120)
- This is very preliminary, but I think I will go with Hollywood3
based on what I've seen so far. "So far" being the key.
BTW, It will play TN dies perfectly ;) - Chris.
krom...@my-dejanews.com
In article <MPG.ff3296e3...@news.netscape.com>,
ka...@nospam.com (Christy) wrote:
>
> In article <6mbp23$9tf$1...@nntp1.ba.best.com>, er...@shell14.ba.best.com
> says...
> > In alt.video.dvd Alen Koebel <ako...@electro.com> wrote:
> > > Christy wrote:
> > >> Afterall the human eye can only differentiate something like 64 gray
> > >> scales and much less in colors. It is most sensitive to greens and least
> > >> sensitive to reds.
> >
> > 64 gray levels? I thought it was 100 levels of contrast which is different
> > than levels of gray?
> >
> > And least sensitive to Red? Hmmm.
> >
> > > Although this is often said, it's not really true, or at least is true
> > > only for very specific circumstances. What is important for video
> >
> > Often said?
> > I thought the human eye was least sensitive to the color blue?
>
> I thought it was red but my memory could be faulty. If it isn't red then
> why does NTSC composite video bandwidth limit reds to 40 lines of
> resolution and limits blue to 135 lines. I thought the design of NTSC
> color was based on the limits of the human eye. NTSC is an analog form
> of perceptual coding.
>
> --
> Christy
> -- Replace nospam with aurinia in order to reply
>
Hmm, are you sure you don't have that backwards? (135 to red, 40 to blue)
The Human eye is definately far less sensitive to blue than red. In fact,
the red sensitivity is not much below green, but blue is far below both
of them.
Andy K.
Vincent Trinh
Brad,
The new Hollywood+ will display 24frames/second on your vga monitor (if the
source was 24fps like most movies), therefor, the quality on your vga
monitor will be amazing - alot better than any other decoder I've seen (I
haven't seen the MPACT2 decoder yet though). On TV it will obviously be 60
fields/second still. The only "disadvantage" of the hollywood+ is that
there there is no simultaneous display on tv and vga.
--
Vincent.
Chris Lewis
In article <6meg72$3...@sjx-ixn11.ix.netcom.com>,
Chris Lewis <cl...@ix.netcom.com> wrote:
2 Quick notes to add to the above: This afternoon I was doing some
more viewing and noticed a lipsync problem with Hollwood3 - very obvious.
Stopped disc and restarted and it seemed to go away. Hmmm....
Also, I noticed VERY slight "interfenece" in the vga output from time to
time; rf noise probably - not visible on proj from more than 4'. Seems to
come and go, like something else (in my house / pc) may be causing it. As
picky as I am though, its not a concern; a shielded loopthru cable or a
ferrite choke on this one would prob. banish all traces of RF - its very
minor. - Chris
misc...@my-dejanews.com
Hi Greg, this is Kei at Digital Connection, and I'm sorry was a little cranky
today when we spoke. Four days of PC EXPO and a six hour flight home with no
sleep after packing all the equipment, and then speaking to you 4 hours later
was just plain bad timing. I hope you had more success with Chromatic on the
scan rate issue. To answer the new James Bond TND movie problem, I was
alerted of the problem by a user of another vendor's card, and contacted
Chromatic last week. Rand from Chromatic did admit to the problem stating
that Chromatic has fixed the problem, but the Mediaware now has to be
updated. He did mention that there is a patch which is much smaller, and
depending on the size of the patch, we will either put it on our site for
download or ship it on another disk as soon as it arrives. He also mentioned
a work-around for this problem by using DirectShow based DVD player app.
Apparently the DVD Player which comes with Win98 can play these movies. He
also mentioned player by InterActual Network, which is bundled on most recent
Warner DVD Videos, but I don't see any of these stop gap measures to be a
full solution. We will update all future drivers, and ship new drivers for
customers who have already purchased the 3DFusion.
In article <gregsr-1806...@u1-70.spiritone.com>,
gre...@SpiritOne.com (Greg Rogers) wrote:
>
>
> Christy (or anyone else with a 3D Fusion), please check Tomorrow Never
> Dies on the 3D Fusion. Another reader sent me email that is wouldn't play.
> I checked my copy and it does the same as his. Just makes a blocky mess on
> screen.
>
> I just finished my first 8 hours of evaluating the 3D Fusion. I've got
> some interesting comparisons but its too late (1:30 am) to write them up
> now. I'll write them up tomorrow evening and post later tomorrow night.
>
> Greg
>
> --
> **************************************************************
> * Greg Rogers http://www.cybertheater.com *
> * CyberTheater(tm): The Internet Journal of Home Theater *
> **************************************************************
>
Christy
In article <6meslt$m...@sjx-ixn5.ix.netcom.com>, cl...@ix.netcom.com
says...
[stuff deleted]
So where is a good place to order a Hollywood3 card? I'd like to take a
look at one too since it will run side by side with the 3Dfusion card.
Thanks
Mark Hapner
Chris Lewis
In article <MPG.ff509864...@nntp.best.com>,
ka...@nospam.com (Christy) wrote:
>In article <6meslt$m...@sjx-ixn5.ix.netcom.com>, cl...@ix.netcom.com
>says...
>[stuff deleted]
>
>So where is a good place to order a Hollywood3 card? I'd like to take a
>look at one too since it will run side by side with the 3Dfusion card.
>
>Thanks
>--
>Christy
> -- Replace nospam with aurinia in order to reply
Christy,
I don't think the 2 cards can run simultaneously (in the same
machine, anyway) ? At least not under Win95. If you know an easy way, I'd
like to hear about it, or if you are familiar with Win98, I understand it is
more flexible and more Multimedia friendly (esp DVD) - Will pick up a copy
tommorow. Can the 2 cards exist in the same machine under Win98 ?
- Chris.
Christy
In article <6mj5o5$9...@dfw-ixnews9.ix.netcom.com>, cl...@ix.netcom.com
says...
[stuff deleted]
>
>
> I don't think the 2 cards can run simultaneously (in the same
> machine, anyway) ? At least not under Win95. If you know an easy way, I'd
> like to hear about it, or if you are familiar with Win98, I understand it is
> more flexible and more Multimedia friendly (esp DVD) - Will pick up a copy
> tommorow. Can the 2 cards exist in the same machine under Win98 ?
> - Chris.
I'm not sure why you are saying this. AFIAK, like the Hollywood1, the
Hollywood3 is an add in card with a video bypass cable and it is designed
to work in conjunction with your primary video card, not replace it. So
this should work on both win95 and win98. I'm running 98 so it won't
matter even if on the remote chance that they changed the hollywood3 to a
full fledged video card like the MPACT board.
Even if the hollywood3 doesn't do better on film dvd's, the fact it does
better on interlaced DVDs will make it a worthwhile addition to the
machine.
BTW, I believe that some video cards like the matrox millinium II will
let you run with two identical cards on a win95 machine.
Cheers
Chris Lewis
In article <MPG.ff7190ac...@nntp.best.com>,
ka...@nospam.com (Christy) wrote:
Maybe we got signals crossed, but when you said you wanted to add a Hollywood3 to your
system, which I assume now has a 3Dfusion VGA/MPEG card already in it, I meant that I don't
think the 2 MPEG portions of the cards can exist simultaneously. I realize that one card is a
VGA card and the other (hollywood) is designed to work with any vga via a pass thru, but
remember that the Fusion is not an ordinary VGA, in that the embedded MPEG decoder is addressed
by drivers windows sets up to use every time you want to play an MPG clip or DVD. How are you
going to tell windows WHICH mpeg decoder you want to use each time you fire up a disc? How can
you have windows load 2 sets of drivers for 2 of the same type cards ? In other words, if you
had a VGA card with a sound card chipset built onto it (they exist), would you also be able to
pop in a Sound Blaster board and make both the embedded sound card on the VGA AND the SB work
simultaneously ? or at least designate which one to use until you specify a change ? I may
not be making myself clear, but you are right; the Hollywood is a card which should work with
any VGA; but the MPEG decoder part of the fusion is where the conflict is going to be, I think.
It is possible that the MPEG and other required drivers are only loaded when you run the DVD
application, which in the case of these 2 cards are different. If that is the case, it just
might work. But I know in win98 there is "built in" support for DVD and MPEG, so it won't/can't
be that simple there. I have 3 optical drives in my machine; CD-Rom (scsi),CD-R (scsi) and DVD
ROM (IDE). Its easy to pick one of these or the other to do something, they are "F" "G", and
"H". Will you be able to tell windows you have 2 (two) MPEG cards, "A" and "B" ?
I would take out my ATI and put the fusion back in to try it, but I think I'll get a mess.
- BTW, the RF interference on the Hwood3, as super slight as it is, is only at
800x600 @72hz. Any other rez/refresh doesn't cause this AT ALL. The oscillation of the card(s)
at this combination of frequencies must cause this slight interference. Also have noticed a
slight jitter on Vid. Essentials still frames (lasts about 1-2 seconds after still is accessed).
Most noticable on grid patterns or others with many Horiz. lines. When the frame does lock up,
there seems to be a slight "blurring" of certain horizontal lines on the frame. Only some
frames do this, and I haven't seen it at all on any video material, except that at the beginning
of Austin Powers, the text "typing" across the screen (over the van in the desert) seems to
jitter (whole screen may jitter) slightly for about 1-2 seconds, every time I play it. - Just
some critical observations for any who care and are comparing... Still working on all the
settings in the driver setup window to see if that may change this at all. I think it is only a
"still" mode thing...
- Chris.
it...@emeraldnet.net
Depending on the driver support for the MPACT card you may be able to
disable the MPEG driver if it is separate from the VGA driver (which it
probably is). Look in Device Manager and see if you can find it. Or call the
card manufacturer.
-Aaron
Chris Lewis wrote in message <6mkmg5$3...@dfw-ixnews4.ix.netcom.com>...
Christy
In article <6mkmg5$3...@dfw-ixnews4.ix.netcom.com>, cl...@ix.netcom.com
says...
[stuff deleted]
>
>
> system, which I assume now has a 3Dfusion VGA/MPEG card already in it, I meant that I don't
> think the 2 MPEG portions of the cards can exist simultaneously. I realize that one card is a
> VGA card and the other (hollywood) is designed to work with any vga via a pass thru, but
> remember that the Fusion is not an ordinary VGA, in that the embedded MPEG decoder is addressed
> by drivers windows sets up to use every time you want to play an MPG clip or DVD. How are you
> going to tell windows WHICH mpeg decoder you want to use each time you fire up a disc? How can
> you have windows load 2 sets of drivers for 2 of the same type cards ? In other words, if you
> had a VGA card with a sound card chipset built onto it (they exist), would you also be able to
> pop in a Sound Blaster board and make both the embedded sound card on the VGA AND the SB work
> simultaneously ? or at least designate which one to use until you specify a change ? I may
> not be making myself clear, but you are right; the Hollywood is a card which should work with
> any VGA; but the MPEG decoder part of the fusion is where the conflict is going to be, I think.
> It is possible that the MPEG and other required drivers are only loaded when you run the DVD
The heart of the MPACT2 DVD decoder is specialized software that works
with the MPACT2 entertainment center. And if I run the entertainment
center it shouldn't care that a hollywood card is installed too since the
entertainment center will talk to the MPACT2 drivers. Similarly for the
Hollywood.
Win98 AFIAK has no concept of a default DVD decoder and even if it did I
doubt that either of these products would use it instead of looking for
the drivers for their card. There is no reason that I know of that would
prevent me from installing both sets of drivers.
Windows even lets me have a serial mouse and a port mouse hooked up at
the same time and runs both drivers. Win98 supports dual video cards.
The only sort of conflict that I know of is with direct3d accelerators
because by default they go to one of your accelerators and you have to
use a utility to switch it.
I'm getting the card on tuesday or wednesday so I'll see first hand what
the issues are with this.
Mario Kalogjera
interlaced and a separate non-interlaced DAC (that's why Encore's progressive output is in fact a scan-doubled interlaced signal). It seems it
is only commercially acceptable to employ some simple switching between the two, but simultaneous interlaced and progressive streaming is
either impossible or to expensive to do. Am I right here, anyone?
Mario Kalogjera
mario.k...@usa.net
Chris Lewis
In article <6mkmg5$3...@dfw-ixnews4.ix.netcom.com>,
Chris Lewis <cl...@ix.netcom.com> wrote:
The 3Dfusion and the Hollywood + WILL setup in the same machine. I installed the 3dfusion
back into machine (removed ATI card) and re-installed all sofware. It works, as Christy
suspected. I am, however somewhat surprised. I know win98 has a DVD app. in the accessories
group, and Im not sure how that would work with 2 mpeg decoders (I am win95.) Anyway, it works,
so I wont ask questions.
Some quick observations:
1) The RF interference patterns/noise is worse with the 3Dfusion as the VGA card. Again, it is
only at 800x600 @72 hz. If you vary the Rez. or refresh, it dissapears (on either vid. card).
On the Fusion, you have to go at least 10hz up or down on the refresh rate to erradicate the
noise. ATI works better as the VGA as far as this little problem is concerned.
2) The Fusion has no 720x480 windows rez. (only for "DVD Mode" full screen playback), so the
Hollywood cannot genlock to that rez/refresh combination in order to display native DVD rez.
With this in mind, at 800x600 the output of Hollywood will go through the scaler (again, it
works amazingly well, however).
3) I have (so far) found it difficult to banish all of the vertical line "noise" through the
horizontal adjusment items in the driver(s) with the Fusion as the VGA. Varying resolutions and
refresh rates for the Holly to genlock to can alter this significantly, but have not found a
combination that works as well as most of the combos would using the ATI as the VGA. Not sure
why this is; maybe ATI has a more "stable" output of some kind. Again, ATI is the better VGA
for the Holly to mate with here.
4) Still frames on Vid. Essentials look sharper (horizontally) in most cases On the fusion.
I still think that in actual playback, there may be little or no difference between the 2.
Interested to see what others will say on this.
5) These cards simply have a different set of "artifacts" than traditional line doublers do (at
least on the progressive mode). You have to look for things that are usually no problem at all
on line doublers that may or may not be fine on these decoders - my observations so far, anyway.
6) My inital comments that the pink/green vertical stripe pattern didn't look "quite right" on
the Fusion is correct; It doesnt. (looks ok on Holly).
7) Initial comments on certain 24fps artifacting on Fusion also seem to be right; notice less
of this on Hollly, except for "jitter" in some cases, which I never saw on the Fusion.
- Let me go kill some more time with these things - Hope this is interesting to
somebody.
- Chris.
Christy
In article <6mn44d$q...@dfw-ixnews8.ix.netcom.com>, cl...@ix.netcom.com
says...
[stuff deleted]
> 2) The Fusion has no 720x480 windows rez. (only for "DVD Mode" full screen playback), so the
> Hollywood cannot genlock to that rez/refresh combination in order to display native DVD rez.
> With this in mind, at 800x600 the output of Hollywood will go through the scaler (again, it
> works amazingly well, however).
The drivers for some reason do not allow you to normally select 720x480,
however I did manage to trick it. Here's how how I did it in win98. I
would expect it to work on win95 too but I haven't verified it.
1 - Set the computer to 640x480 mode and close all other applications.
2 - Leave the control panel browser open and launch the entertainment
center.
3 - Once it is completed launching hit ctrl-alt-del and when the task
list window appears select the entertainment center and hit the "end
task" button.
4 - It may say that the app isn't responding. If it does then you tell
it yes you want to terminate it. The screen may look a little funny but
don't worry.
5 - now open up the control pannel and you'll see that it is set to
720x480.
6 - Slide the slider to another resolution and slide it back to 720x480
and hit Apply or OK.
7 - Now hit ctrl-alt-del and choose shut down, or choose shutdown from
the start menu. The system may not let you shut down because of the way
we terminated entertainment center so you may have to hit reset.
8 - once you reboot the system will be happily running in 720x480.
That's it. Sheesh, seems real lame that they couldn't have just given me
720x480 as one of the valid graphics modes. After all some people
(myself included) own widescreen monitors.
Chris Lewis
In article <MPG.ff8c8149...@nntp.best.com>,
ka...@nospam.com (Christy) wrote:
>In article <6mn44d$q...@dfw-ixnews8.ix.netcom.com>, cl...@ix.netcom.com
>says...
>[stuff deleted]
>
>> 2) The Fusion has no 720x480 windows rez. (only for "DVD Mode" full screen playback), so the
>> Hollywood cannot genlock to that rez/refresh combination in order to display native DVD rez.
>> With this in mind, at 800x600 the output of Hollywood will go through the scaler (again, it
>> works amazingly well, however).
>
>The drivers for some reason do not allow you to normally select 720x480,
>however I did manage to trick it. Here's how how I did it in win98. I
>would expect it to work on win95 too but I haven't verified it.
>
>1 - Set the computer to 640x480 mode and close all other applications.
>
>2 - Leave the control panel browser open and launch the entertainment
>center.
>
>3 - Once it is completed launching hit ctrl-alt-del and when the task
>list window appears select the entertainment center and hit the "end
>task" button.
>
>4 - It may say that the app isn't responding. If it does then you tell
>it yes you want to terminate it. The screen may look a little funny but
>don't worry.
>
>5 - now open up the control pannel and you'll see that it is set to
>720x480.
>
>6 - Slide the slider to another resolution and slide it back to 720x480
>and hit Apply or OK.
>
>7 - Now hit ctrl-alt-del and choose shut down, or choose shutdown from
>the start menu. The system may not let you shut down because of the way
>we terminated entertainment center so you may have to hit reset.
>
>8 - once you reboot the system will be happily running in 720x480.
>
>That's it. Sheesh, seems real lame that they couldn't have just given me
>720x480 as one of the valid graphics modes. After all some people
>(myself included) own widescreen monitors.
>
>Cheers
>--
>Christy
> -- Replace nospam with aurinia in order to reply
Clever of you; I will have to give that a try. I suspect that whenever you change resolutions
from the taskbar (or otherwise) for "normal" computing, you would have to go back to 720x480
through the same method you outlined. My ATI card simply has a 720 setting on the taskbar...
Nice. Anyway, A couple more observations, Hollywood vs Fusion
1) I may be repeating myself, but holly is (seems) a TAD softer than Fusion. I can only really
verify it by having them both in the machine. Its VERY close. They both measure out the same
on the horiz. rez pattern, but in film mode, Fusion seems a click sharper. It may be that the
scan line structure is still visible on Fusion, which may lend that slight artificial resolution
look, or it may have something to do with the contrast setting, seeming to look more detailed
when its actually the contrast, but Im not sure. Neither of these cards has a "normal" contrast
control; they both react differently than analog ones we are used to, and the Fusion card seems
maybe a little more "punchy", IMO. Also, both cards can easily be "overdriven" by a brightness
or contrast setting that is too high. If so, the top 2 or 3 levels of the gray scale may be
identical - be careful. On the sharpness issue, bear in mind that I am comparing 720x480 native
on Fusion to 800x600 (scaled) on Holly, but I already determined that any difference in detail
between 720 and 800 on the Holly is essentially nil.
2) The fusion may be "smoother" more often than the Holly. By smoother I mean that when a pan
is taking place, or a steadily moving object is in view, sometimes both cards can "jerk"
slightly, but Ive seen this with doublers as well. It typically is not repeatable if you
reverse scan and re-play. I am REALLY splitting hairs, but thats what makes if fun, right?
I can't remember the last time I spent this much time trying to see how bulletproof a $120
piece of Audio / Video equipment was... I'd say either of these things is a good match for an
$800 Sony 7000 and (at least) a $2000 doubler (in film mode). Thats progress. I cant wait for
the "next" one to come out...
I think Greg has done some detailed viewing with a more formal approach to quantifying
the performance (at least on the Fusion) - I look forward to reading his post on it.
More info as I see it.... - Chris.
misc...@my-dejanews.com
The Mpact 2 card is not separate from the MPEG-2 decoder, it is a media
processor based 2D/3D/DVD card with one chip controlling the whole shebang.
As such, you can elect not to install the Entertainment Center, although why
I don't know. To answer the other question, the 3DFusion DVD or for that
matter Mpact 2 based VGA cards do not support multiple VGA in Win98. That is
unless some engineers at Chromatic Research decided that they want to write
the driver for this, if it's possible.
In article <6mkoq1$65n$1...@news13.ispnews.com>,
<it...@emeraldnet.net> wrote:
>
> Depending on the driver support for the MPACT card you may be able to
> disable the MPEG driver if it is separate from the VGA driver (which it
> probably is). Look in Device Manager and see if you can find it. Or call the
> card manufacturer.
>
> -Aaron
>
> Chris Lewis wrote in message <6mkmg5$3...@dfw-ixnews4.ix.netcom.com>...
misc...@my-dejanews.com
Yes, it will since the Mpact 2 Media Processor does not support Multiple VGA
in Windows 98 systems. I'm not sure if the ATI can do so either, but this
capability has to be supported by the video card.
On Coffee Break
Kei
In article <6m76ek$dbd$1...@nnrp1.dejanews.com>,
go...@hotmail.com wrote:
>
> I am thinking of using the ATI all in wonder pro as the video processor
> (deinterlace and line multiply) for external laserdisc source and the 3DFusion
> DVD kit as the DVD source for output to a graphic grade CRT projector.
>
> As the 3DFusion is itself a video accelerator, will it conflict with the ATI
> card?
>
> Thanks
> Dennis
> > Earlier I wrote:
> >
> > > His machine is running Win98 and mine Win95 B. Mine is a plain old Intel
> > > 440LX motherboard with P-2 processor. The 3D Fusion PCI video board is the
> > > ONLY board installed in the computer. There is one master hard drive on
> > > the primary IDE channel and the Pioneer DVD player as a master on the
> > > secondary IDE. Both players will only play non-region coded discs like
> > > Video Essentials. We have reinstalled the software a zillion times.
> >
> > OK, the mystery appears to be solved. The Pioneer drives apparently are
> > not compatible with the Intel Bus Master IDE Drivers. We both had these
> > installed in our respective machines. After we removed them, the 3D Fusion
> > cards finally appear to be operating normally. I'm exhausted after trying
> > to make them work for 3 days, so testing on the big screen must wait for
> > tomorrow night.
> >
> > Greg
> >
> > --
> > **************************************************************
> > * Greg Rogers http://www.cybertheater.com *
> > * CyberTheater(tm): The Internet Journal of Home Theater *
> > **************************************************************
> >
>
Dean McManis
I just wanted to say thanks for everyone's input on the performance,
setup, and product comparisons of the Mpact2 3D card. I bought a new
Pentuim II/Mpact2/DVD combo right after Christy's initial post, but I
made the mistake of buying Windows '95 from Fry's Electronics to load on
the machine.
The card worked OK, but didn't play about half of the movies that I
tried, and would not work in full-screen mode. After working with
Digital Connection's tech support (who were both helpful and
knowledgable) I discovered that the Win'95 version sold to me was
several generations old, and needed several revision patches to work.
So I now have Win '98 coming and I'll post my impressions of the
card/DVD when I get it working.
-Dean.