VS7: Best way to break up large clip?

VS7: Best way to break up large clip?

Post by Adam Whi » Fri, 06 Jun 2003 11:58:48

Using VS7, with the 7.01 patch applied (W2k Pro SP3, if it matters)

I have a 4 hour MPG file, about 15 Gb size.  I want to make a DVD
using only, say, one hour from this clip.  I see two ways to proceed:
1:  use the mark-in and mark-out features in the Edit tab, then use
Create Disk under the Share tab (I'm keeping this simple for my
example, here)
2: mark the same in/out points on the Edit tab, then Extract the clip
to a new mpg file, then open a new project and work on the smaller

I tried the firsrt method, because I didn't want to wait for the
Extract operation and did not really see the need to split the 4Gb MPG
file.  But the rendering (Create Disk step) took a lot longer than it
normally does when I make one-hour DVDs.  Eventually I got an error
message that said "unable to create file."  I noticed under the
temporary files directory that there were a large number of VOB files
... enough to hold the entire MPG file, in fact.  This brings me to my
first question:  Does VS actually process the entire MPG clip even if
it only wants a small portion of it?

If the answer to that question is YES, as it appears to be ... then I
guess I know the answer to my 2nd question:  which way (of the two
methods I outlined above) is best to use for making a small movie from
within a larger clip -- assuming that I'll actually make several small
movies from the clip (movie one = minute 1 to minute 60, movie two =
61 - 120, etc.)?

Are there any other aspects to splitting the large MPG of which I need
to be aware?  I don't want to lose any quality from my original

Thanks for your advice / thoughts on this.


VS7: Best way to break up large clip?

Post by Jerry Jon » Sat, 07 Jun 2003 05:40:23


ULEAD VideoStudio 7 utilizes a feature called Smart Render.

The purpose of Smart Render is to give the user the ability to edit
files without re-rendering the entire file.

If one adds a title to a video segment, for example, Smart Render will
ensure that only the segment of video where the title is located is

If one adds a transition, Smart Render will re-render only the segment
where the transition is located.

To ensure that Smart Render works properly, it is necessary to make
sure that one's file attributes 'match' the VideoStudio project

For example, if one chooses the following project settings...

MPEG files
24 Bits, 720 x 480, 29.97 fps
Field Order A
(DVD-NTSC),   4 : 3
Video data rate: 8000 kbps
Audio data rate: 224 kbps
MPEG audio layer 2, 48 KHz, Stereo

...and one's clips match these project settings, then the Smart Render
feature should work.

If one's video clips have attributes that do not match one's project
settings, then re-rendering will occur.

Hope this helps,

Jerry Jones


1. Q: Better ways to assign color?

Hi, everybody!

I would like to get some help regarding the use of color in IRIS
Explorer.  Specifically, I would like to find out if there is a better
way of assigning colors based on the values of a variable in the data.
I must tell you in advance that I am a begginer in using Explorer, so
please bear with me on this.

Here is what I am trying to do:  I have about 3800 spheres that I need
to draw.  I want to assign different colors to those spheres based on
the value of a certain variable.  I can easily assign a specific color
to the whole 3800 spheres at once with cxGeoColorAdd() after I draw the
spheres with cxGeoSpheresDefine().  However, it seems that there is no
way to assign different colors to each of those spheres while calling
the cxGeoSpeheresDefine and cxGeoColorAdd only once or a few times.  

I can think of two approaches.  One approach is to group the spheres
based on the variable that will be used for color assignment and call
the cxGeoSpheresDefine and cxGeoColorAdd combo several times (the number
of calls will be the number of colors that I want to use).  This seems
to be a reasonable approach unless I also want to have the capability to
select and change the color-assignment variable among many variables
interactively, and regrouping data everytime the variable changes seems
to be overly expensive operation not to mention the extra storage

Second way I can think of is to define each sphere one by one with
accompanying cxGeoColorAdd call, which is what I am currently doing.
So, I end up calling these two function combo 3800 times, and it takes
quite a bit of time to render them (they show up though).  Also,
Explorer blows up in my face sometimes complaining that I exhausted the
shared memory if I am trying to do something with them, like changing
the radius of the spehres.  By the way, I am running this in SGI Indigo2
IMPACT with 128 Mb ram.

So, I am thinking there's got be a better way of doing this, but I have
no idea as to how to do it.  Have anyone graffled with this kind of
problem and found a way to handle it?  I would really appreciate any tip
or insight on this matter.  Thanks a lot in advance.


Department of Geography
The Ohio State University

2. Images disapper from 98 to 95

3. calibration..Screen to Printer How, best ways etc.

4. Should I buy?

5. SPEED--best ways to boost a PM6100?

6. Help!

7. Q: The best ways to convert QuickTime to FLI/FLC

8. Is real time raytracing possible?

9. SPEED--best ways to boost a PM6100AV?

10. Upgrading SDK from 7 to 8.1 breaks clipping

11. Clipping/Large scale models

12. Large POV Files (How Large is Large ?)