First of all, you have never been "rightly" pissed off, none of your crude and vulgar rants here on the forum have ever been justified.

In case you are rendering intermediate AVI files, there is no flag in the header of an AVI file that specifically states whether it's interlaced or not. This is a limitation of the AVI spec, not Vegas. This information is contained with MPEG2 and AVCHD files, just not AVI files.

Secondly, anyone wanting smooth slow motion from a 24fps or 25fps source material is simply asking too much. The higher the temporal resolution of the source material, the better the slow motion. This is why one overcranks the camera for decent slo-mo. You have Dynapel's Slowmotion, USE IT.

Interesting...

I had to give this a try...



(still processing but should be ready shortly)

Perrone Ford...I just looked at your test. Is that 30P or 24P footage?

Yes I am rightly pissed off..

Is it a day that ends in "y" again already?

Rob

I use progressive in vegas all the time.. slows down great! Heck, I take 60fps clips & slow them down to 30 & they look AWESOME!

I'm betting you just don't really know what you're needed to do.

For high quality slow motion from progressive footage, you need "optical flow" which can be found in After Effects, Final Cut Pro, Premiere Pro, and a few other NLEs, just not in Vegas.

Optical flow synthesizes new frames. It is compute-intensive, but the result is worth it.

60i to 30p is a special case, where you convert the 60 fields per second to 60 progressive frames with interpolation.

I think it's even in iMovie now.
Then again all the programs you listed kind of "inhale" at mixing audio.

Bob.

Look at the footage I just posted. It's progressive, and it was done in Vegas. I typically use a different program, but Vegas does fine if you give it good source material.

It's 60p at 50%.

http://www.vimeo.com/5027327

Yeah, I hated making that video. Progressive slowmo sucks. Vegas is horrible.

Pfffft.
---------------------------------------------------------------
Perrone- Looked good to me.

Yea Brad... yours looked as bad as mine... LOL!

Shoot properly, and Vegas slo-mo looks just fine.

Perrone- What shutter was that at?

i80 degree... so for that shot, 1/120. I tend to shoot shutter angles so that it's consistent if/when I decided to change frame rates. On this shoot I used 60p and 24p.

"you need "optical flow""


is that an AE plugIn?

First of all, you have never been "rightly" pissed off, none of your crude and vulgar rants here on the forum have ever been justified.

((Whatever you reckon John... If you believe that NONE of my posts or points are valid, then I can only assume you have tunnelvision...
Im not the only one having these incessant problems that are yet to be fixed...
Crude? No, im just a lil too straightforward for some and I dont have time for PC'ness when it comes to my work ))

In case you are rendering intermediate AVI files, there is no flag in the header of an AVI file that specifically states whether it's interlaced or not.
((Really?? Wow.. have you tried xfering progressive scan footage and then xfer interlaced footage from tape... and insert BOTH of those DV AVI captured files from tape into a timeline.. ANY timeline... OK do that then hit the properties of said individual clip and tell me what you see in regard to field order information provided... ))

This is a limitation of the AVI spec, not Vegas.
((No its a limitation of BOTH... but Im not refering to the file format, im refering to Vegas' managemnt (i should say MISMANAGEMENT) of it.
All Vegas needs to do to manage this files properties correctly is to read the first frame or field within the file.
It must do this (read the file) ANYWAY when its drawing its SFK file, so there is no excuse for V not to be able to recognise the file which has been imported ))

This information is contained with MPEG2 and AVCHD files, just not AVI files.
((Im not refering to metatags in these transport streams. Im refering to how Vegas manages its Porgerssive media DIRECTLY))

Secondly, anyone wanting smooth slow motion from a 24fps or 25fps source material is simply asking too much.
((No, i think after 5 yrs of development, some PROGRESS should have been made by now... Weve seen FCP and CS implement new engines, there are cheap external app options (Dynapel for one) and fair enough their not perfect, nothing it, but at least ITS SOME PROGRESS.
My gripe isnt about the fact that i have to use an external app, in fact i really couldnt care less. my gripe is how Vegas CANNOT manage progressive scan FULL STOP, PERIOD.
All it can do is play it back and throw filters on it. It attempts to flip the field order when you hit reverse WHEN THERE IS NO FIELD ORDER... It duplicates frames when you slow a clip down be it to 99/8% to 5%, it doesnt matter, there IS NO interpolation engine within slowmotion, only within its deinterlacing. Hell, it doesnt even blend frames... ))

The higher the temporal resolution of the source material, the better the slow motion.

((I beg to differ... IMO a faster shutter speed makes a massive difference....
Ive written about this many moons ago in fact...
As youre wanting to emulate a faster framerate (for slower output), you must also emulate the shutter speed nuance as your slowdown affects the global image, imcluding motion blur...

Try this with traffic and you'll see what Im talkin about...
Shoot at 1/50th and slow to 50% and your visuals LOOK as though theyve been shot at 1/25th. Increase your shutter and that element of motion blur is non existant.
Fine detail of the vehicles IS visible.

Now I personally do not see how resolution has anything to do with how a moving shot is captured or processed (unless its scaled DOWN from a higher resolution of coure) when the CCD/CMOS itself has not changed its resolution during acquistion. In additon.
With slowmoito the biggest issue is blur.
The idea is to get rid of it during the recording phase.

heres a link to an example of slowmotion with a variety of shutters.
Available for 7 Days or 100 downloads.
http://www.yousendit.com/download/Y1RvK3BCSU9OQnhMWEE9PQ
this is from a 1080i acquired source from a canon A1. I would have prefered 1080f, however I knew id be using alot of slowmotion with this. This has been scaled to 720p for the master, then scaled again to 320x240 for this extract demo.

Youll see how Vegas can handle interlaced material slowed down to ANY variable rate and output to a lower res progressive (in this case from 1080i to 720p).. Theres no issue with that at all, there is however the fact that I cant output this same project to 1080f/p from within Vegas (looks crap) and I cant output to 1080i as that too looks crap.
So the point of this clip (going on a tengent here) is to show how much ive slowed down the footage using stupidly high shutters (the first clip is about 1/50 and 1/75 <notice the blur in the girls arms>, after the short black break, it jumps to 1/250 and 1/500 <notice the feet and hair detail in the closeups... hardly any blur (if that, and immaculate detail>) and as slow as i got it (30%) it looks incredible.
So hopefully this demonstrates the differences in shutter for those wanting to do slowmotion.
Its not about resolution IMO, as I get teh same results from DV...
Slightly off topic but shows what can be done when u know how each tool works with the other.
Again John, I hear your point, however I dont see how res has anything to do with it.
From my experience (there might be others out there like yourself who disagree I dunno), its all about frame rate and shutter speed as is demonstrated here))

This is why one overcranks the camera for decent slo-mo. You have Dynapel's Slowmotion, USE IT.

((I am. Sadly i have to render out to uncompressed as this install of Windows wont run HuffyYUV :(
The point however is that I shouldnt have to do this 4yrs on... when progress is being made on so many levels from so many different companies. And I DONT want to jump ship because even with these nuances of Vegas it IS a good tool.

It just doesnt drill holes in all the kinds of bits of wood i point it at...

oh one other thing, just so you all know, if you have your monitor refresh rates set to anything higher than the output hz (ie, if your GFX card is set to 60hz and your watching 25p or even 50p for that matter (or 60i material on 100hz LCD's), then you wont notice the stutter of the bad slowmotion or flipped field problems....

>The higher the temporal resolution of the source material, the better the slow motion.

"temporal resolution" = resolution in time, in other words how many frames or fields per second are captured. I think it's pretty non-controversial that the higher your frame rate, the better your slowed-down footage can look.

temporal resolution" = resolution in time

non controversial.. agreed, but thats not what were talkin about here..
Even IF the resolution was as high as it woudl be (lets say 720@50p, runing at 1/50th), slowed down the slowmotion WOULD STILL emulate that slower shutter speed, as Vegas DOES NOT use the redundant second progressive frame in the 50p transport ... I wish it did, then id shoot everything at 50p and drop it in a 25p timeline and not think about it....
Some people have claimed success with slowed down 50p, I am yet to see actual results.. but the point here is that not everyone has a 50p camera... and most likely those working in 1080 dont want to scale down if they dont have to... im yet to find a cam that shoots 1080@50p....

What youre refering to is pretty much overcranking or redundant frame capture (such as 50p) some might say its not redundant, but thats another thread.. if THATS what is being refered to as temporal resolution, then saying 1080i is a higher res than 720p, theoretically that argument can also backflip whereby TECHNICALLY 720p is temporally higher in res than 1080i... which again would be another thread...

Look, lets just agree that Vegas slowmotion with progressive footage sucks... im not the only one to say it or experience it... and if you dont agree with me, thats cool too... im not fussed or phased... i was simply hoping SOny would have put some R&D into it as opposed to wow factor efects which are al pretty but not as much of a necessity as format management

DJ,

Any chance that you could post 3-5 seconds of your original footage, along with a VEG file that slows that down and, when I render from it, will show me the lousy slow motion? In other words, I want to recreate your lousy slow motion. You should be able to use the Vegas "smart render feature" to cut off a several second chunk of video. The VEG file is absolutely essential as well.

I ask because I downloaded your footage from the YouSendIt link you provided earlier in this thread, and not only viewed it through my media player, but also then put it on the Vegas timeline and then served it out to a program that separates the fields so I could look and see what is going on within each progressive frame. While I don't disagree with many of your points about Vegas' handling of slo-mo and the improvements that have been made by other programs, I am also suspecting -- based on the sample footage you provided -- that you may have run up against a problem that I found several years ago, and which can be solved by using different Vegas settings.

So, I may be able to improve your video and at least partially solve your problem.

I also downloaded the sample given by another user, and one reason it looks so good is that Vegas is simply using the alternate fields, and then re-scaling from that. This is easy to demonstrate because, if you download that footage from YouTube (which I did), put it on the Vegas timeline, and then serve it out through an AVISynth script that separates the fields, you will see that as the girl's hand waves across her face, each field shows perfect video with no trace of either blending (which is how Vegas does slo-mo) or motion-estimation synthesis (which is how Twixtor, Dynapel, and many other "high-end" products do slo-mo). That approach always leaves tell-tale morphs and distortions that are easy to spot when looked at frame-by-frame.

Getting back to the point, the problem is that unless you do exactly 50% slo-mo, the program has to take some other approach. Also, with the field-to-frame approach I just described, while you get great temporal results, you do lose half the vertical resolution. Of course when you start with 1080i/p and are going to 720 or 480, most people won't catch this.

So, let me see if I can help ...

Optical flow is the way to go Boris optical flow is very good although I never got them to work through Vegas as a plug so I use it as a stand alone

I ask because I downloaded your footage from the YouSendIt link you provided earlier in this thread, and not only viewed it through my media player, but also then put it on the Vegas timeline and then served it out to a program that separates the fields so I could look and see what is going on within each progressive frame. While I don't disagree with many of your points about Vegas' handling of slo-mo and the improvements that have been made by other programs, I am also suspecting -- based on the sample footage you provided -- that you may have run up against a problem that I found several years ago, and which can be solved by using different Vegas settings.

((I appreciate yoru help L, however theres nothing wrong with the clip Ive uploaded.
The one i uploaded was to demonstrate shutter speed variances within the slowmotion process as someone mentioned resolution plays a big part in slowmotion, I added that its moreso shutter speed which affects perception of detail (ie less motion blur for slowmo'd material. I will however upload a clip and a Veg once i finish ths current edit im working on...))

So, I may be able to improve your video and at least partially solve your problem.

((that would be nice!! Ive tried everything over the last 5 or so years, prolly even longer actually... lol ))

I also downloaded the sample given by another user, and one reason it looks so good is that Vegas is simply using the alternate fields, and then re-scaling from that.
This is easy to demonstrate because, if you download that footage from YouTube (which I did), put it on the Vegas timeline, and then serve it out through an AVISynth script that separates the fields, you will see that as the girl's hand waves across her face, each field shows perfect video with no trace of either blending (which is how Vegas does slo-mo) or motion-estimation synthesis (which is how Twixtor, Dynapel, and many other "high-end" products do slo-mo). That approach always leaves tell-tale morphs and distortions that are easy to spot when looked at frame-by-frame.

((Is this the field split trick?? I recall reading about that several years ago, i just never had the time to play with it :( ))

Getting back to the point, the problem is that unless you do exactly 50% slo-mo, the program has to take some other approach. Also, with the field-to-frame approach I just described, while you get great temporal results, you do lose half the vertical resolution. Of course when you start with 1080i/p and are going to 720 or 480, most people won't catch this.

So, let me see if I can help ...

((I'll upload some media and a veg file when i get a chance!!))

thx again