VideoStudio 5 is out.

[Dr Mordrid]

http://www.ulead.com/vs/runme.htm

Interesting note: VBR MPEG's. Sounds like the new Ligos kernel is in the mix?? Wonder when MSP6 will get it's update? Hmmmm.....

I'm DL'ing the demo tonight to check it out....

Dr. Mordrid


[This message has been edited by Dr Mordrid (edited 11 January 2001).]

[Hulk]

The Pentium 4 support looks interesting. I wonder when these instructions will be incorporated into MS Pro.

[Jerry Jones]

It looks like the hottest video editor in the beginner price range that I've seen so far. That variable bit rate encoding is a surprise.

Cutting edge stuff, for sure.

[johnpr98]

Seamless capture is a biggy (98 & ME)
Video filters
Simple PIP
Scene detection
Batch capture

johnpr98

http://www.johnpr98.com

[Vidgal]

As usual, my timing is off. Just got Videowave IV for xmas in order to have MPEG editing capability and quick and dirty MPEG capture/output for web stuff. A couple of questions (if anyone knows or finds out): Will Video Studio MPEG1-2 be better? and Would installing Video Studio 5 in any way effect my MS Pro 5.2? Can't mess with the old warhorse as he's done me right with me G400TV.

[Hulk]

Vidgal -

Can't help you with the other question but I have MS Pro 6.0 and VideoStudio 4.0 on the same system with no problems. I doubt that the MPEG I/II encoding of VS 5.0 will be anywhere near as good as that of TMPGenc so I wouldn't worry too much about that.

Johnpr98 -

Have you figured out a way to output to tape using the AIW TV out without the black border around the picture?

Or do you use your camera to output projects to tape?

[johnpr98]

Hulk

Media Player 7 is probably better for TV output than than it's predecessor.

I don't use TV playback from the AIW, In fact I don't use it for much of anything.

OK I use the very good DVD playback, the TV tuner is used very little at the moment, of course as a Video card it's fine.

I always output to my Sony TRV120 cam or use it's DV to analogue conversion facility to copy onto SVHS.

Regards

johnpr98

http://www.johnpr98.com

[Dr Mordrid]

Just finished playing with the VideoStudio5 demo for about 30 min. First impressions;

The interface is nicely integrated and fairly intuitive.

NO SVCD PRESET. The custom video size box is grayed out so you can't try making your own. I sure hope this isn't the case in the full version.

VideoCD quality is about halfway between that of MSP6 (old Ligos engine) and TMPGEnc. Still room for improvement.

The top MPEG-1 bitrate is only 2000 kbps. This makes no sense in light the max setting for MPEG-2. See below.

Maximum MPEG-2 bitrate is 14,648 kbps.

VCD format MPEG's are accepted by Nero and burned CD's play well.

Adaptece accepted them as well, but it crashed after adding the MSPro6 MPEG. Figures.

I had some problems with DV files rendered to full frame MPEG-2's. I'll have play a bit to see if this is a problem with some beta drivers I'm running or a problem with VS5 itself.

All renderings in TMPGEnc done with dead-flat stock settings, so the differences might be greater if they were tweaked.

The Advanced MPEG settings in VS5 do show up, most likely because I have them enabled in ulead32.ini for MSPro6. I haven't tried using them yet.

Dr. Mordrid


[This message has been edited by Dr Mordrid (edited 12 January 2001).]

[Jerry Jones]

I have to say I've been reading the following Ulead / Intel collaboration Web page:

http://www.ulead.com/vs/new1.htm

And my question is this: Would somebody with a Pentium 4 processor *please* report to us your transition render times????

I consider this Pentium 4 optimization to be a critical new issue for my next computer build decision.

I was formerly thinking about going AMD, but now I'm more persuaded to build my new system around a Pentium 4 given Ulead's obvious *direct* collaboration with Intel engineers on very specific program features.

I suspect this is going to be a *very* important issue as development moves forward on MediaStudio Pro 7 -

I mean - logically - we might see near *real time* performance from dual Pentium 4 processors on most transitions, filters and perhaps color correction by the end of next year.

[hapz]

Am I correct with my interpretation that Video Studio 5 does not allow analog capture (e.g. froma tape deck) but only from a digital source?

------------------
Harald

[Jerry Jones]

Analog capture is supported:

http://www.ulead.com/vs/capcards.htm

[Jerry Jones]

Again, if anybody has a Pentium 4 equipped computer, please test the transitions that Ulead has optimized in cooperation with Intel as explained here:

http://www.ulead.com/vs/new1.htm

I am *extremely* curious to know how many seconds it takes to render these 25 optimized transitions.

[Hulk]

Jerrold -

"Real Time" rendering is a tricky thing to define with MS Pro 6.0.

For example. In my MS Pro bench http://www.hyperactivemusic.com/MS%20Pro%206.0.htm if the project is rendered in 78 seconds (the project length) then you could assume it was rendered in real time. The only problem is that because some parts of the project are more cpu intensive during rendering (transitions, overlay effects, etc..) they will be rendered in less than real time while the "easier" parts are rendered faster than real time. Overall though, rendering is real time.

The problem is that if you are creating a project and checking various portions, what good is real time overall rendering when you only need to see a small portion. Or if you want to check the whole thing, if the cpu can't render the whole thing real time then you'll still have to wait for the preview. Granted that preview will only take the project time.

My test clip would probably need about 4500MHz PIII for overall real time rendering. But for actual real time rendering of the most complex part, who know? Maybe 2 or three times that amount.

Anyway. I think that dual 2GHz P4's would be very close to real time in most circumstances.

One more thing. I'll bet that P4 won't be any faster than a similarily clocked PIII (if that could go that high). The P4 "optimizations" thus far (Flask MPEG for one) only keep P4 as efficient as PIII clock for clock. That 20 stage pipeline that allows the high clock speeds has a high branch misprediction penality. I'm wagering that you could test VideoStudio 5 on a PIII 500 and basically assume that P5 1.5GHz will perform 3 times faster. The SSE2 optimizations are only good enough to make up for the enormous branch prediction penality of the P4.

As I have said, thus far I have seen NO benchmark where the P4 was more efficient clock for clock than a PIII.

[Jerry Jones]

You and I have probably read the same articles about the P4 and the skepticism about it...

But the point I'm trying to make is there's nothing like "real world experience."

I think that's the only way we're going to find out in the case of VideoStudio 5.

[Hulk]

I agree 100%. The only way we'll find out is by testing.

Even if my assumptions are correct. P4 will be the fastest chip for video editing simply because it will scale to the highest speeds.