"That's a complete switch to mine."

Not from what I've read and seen.

1) You are testing V12, we're not, any difference might be due to that alone.
2) Your results using V12 for 60i are the same as ours for V10, only 50i shows a difference.

"What can I do further?"

Start filling in forms.

Bob.

Bob, I kinda know I'm using VP12 and Nick isn't.

That's your point?

G

What can I do further?

Render the files directly from Bob's files, without nesting, so you see what the results should look like and/or to see if we have a discrepancy between the Vegas versions.

Also, repeat the 50i nested test from scratch, following my step-by-step test method to the letter. It's very easy to slip up, so it's worth double-checking.

Anyone else care to do this test in V12 so we have more than 1 sample?

"Bob, I kinda know I'm using VP12 and Nick isn't.

No. You said:
"That's a complete switch to mine."

That statement is open to interpretation.
In the context at the time it could be taken to mean that you got exactly the opposite result to Nick and me and yet going on what you previously showed you didn't.
Other people are reading this, they may not be aware of our off forum communication, they may not carefully read through all the posts and they may, in fact I think they are, drawing the wrong or different conclusion(s). We need to be very specific in our posts and ensure we create the least possible opportunity for our words to be interpreted.

Bob.

Thank you Bob and Nick.

OK . . .

All my testing is in VP12.

Bob to me:That statement is open to interpretation.

Not for me it isn't, but I shall explain further: Concerning combing, Nick sees the combing in 60i but I see the anomaly in 50i. That to me is a switch, and that's why I used the word "switch". Nothing more nothing less.

Nick to me: Also, repeat the 50i nested test from scratch

I understand, Nick, this was your response to my "what can I do further". If so I was asking what can add/suggest/complete? My next "task", if you guys couldn't think of anything further I could do, is to submit your methodologies and outcomes and my Video to SCS.

In terms of further testing, sure Nick I can do this, but what I did, in providing that video, was to simply open Bob's 2 Renders and screen capture the outcomes. I also had a SKYPE session with Bob and showed him this Combing too. I was happy just to test Bob's Render outcomes, I hadn't created anything Nick, so I hadn't made any slip-ups or changes to your methodology. I was purely visually testing the Renders.

If you - Nick and Bob - are in agreement, what I intend to do is to:

i) simply copy and paste your methods to repro this, to SCS

ii) a link to this thread which will explain the intricacies of this testing

iii) that video I made

Is there anything else I can add?

It would be real neat if both of you made available your own videos of this happening so that SCS has 2 other sources to compare, contrast and confirm against.

Great thread this and great work from the both of you!

Cheers

Grazie

Nick sees the combing in 60i

I don't see combing in either 50i or 60i. Combing isn't the problem.

It would be real neat if both of you made available your own videos of this happening

I'll see if I can do a screen capture video. Never done one before.

AH! Penny just dropped for me Nick . . you said: "But my 59.94p project behaviour is the same as yours Grazie.", which I interpreted as the same as my 50i, but you were emphasising that the 60i I got is the same as your 60i!!! - Got it, thanks.

Screen capturing is fun. I use Frederic Baumann's Screen Recorder - marvellous!

Grazie

Not the best video or audio quality but...

Watch in 720p on YouTube not here.

[url=

OK, I've just made significant progress with this.

I have established that the problem is not related to the frame rate at all. It is related to the amount of downscaling that is being done. If you downscale nested interlaced footage to less than 50% of its height (number of lines) then the problem starts.

1080-50i > 576-50i - No problem
1080-60i > 576-60i - No problem
1080-50i > 542-50i - No problem
1080-60i > 542-60i - No problem
1080-50i > 538-50i - Problem (progressive output)
1080-60i > 538-60i - Problem (progressive output)
1080-50i > 480-50i - Problem (progressive output)
1080-60i > 480-60i - Problem (progressive output)

Perhaps anyone who has filed this bug could add these findings to the ticket, along with a link to my video above.

Nick,

Absolutely brilliant!

Your YouTube video was excellent, and the steps you took there were 100% identical to the tests that I did several weeks ago when I submitted my trouble ticket to SCS (which, BTW, they have still failed to respond to after almost three weeks).

However, what makes me say "brilliant" is your discovery that the problem is related to sizing percentage and not to frame rate. While I've always known that PAL is a slightly different resolution than NTSC, all my work with these two formats has always focused on the problems related to changing frame rate. It never would have occurred to me that PAL is slightly more than 50% of the HD vertical resolution whereas NTSC is slightly less. That was a very clever insight.

I'm getting too old to remember all the things in your long posts and great tutorial about getting great video for YouTube and all the related work, but it seems to me that when we were all looking a hulu dolls and related videos that we were all suspecting that something was not quite consistent with how Vegas handles deinterlacing when scaling. While none of that previous work involved nesting, I have to wonder whether the subtle bug that is the subject of this thread might not have some further, even more subtle ramifications.

Thanks to you and Robert and "Grazie" for all your work on this. Now if Sony would get involved ... If anyone at SCS is reading, this is a very important issue.

Thanks John.

My findings could do with a little verification from others before we take them as gospel. I tested the 538 and 542 resolutions by rendering to uncompressed AVI. Perhaps you could run them through your frameserver>avisynth method?

And of course I've only tested up to V10. Grazie's test in V12 indicates the problem still exists, but we still have the additional combing irregularity in his 50i V12 test.

I'll take the thanks!

John, please explain, simply, so I can understand what your: " If anyone at SCS is reading, " . What it could really mean for SCS? John, let's get 'em totally connected to this activity - yes?

Grazie aka Graham.

Why should SCS care? My goodness, you must be making a joke: if a video editing program doesn't handle interlacing correctly then it is completely useless for its intended purpose. A bug like this, even though it only happens under certain circumstances, strikes at the heart of this product's basic purpose.

As to Nick's question about rendering to different formats, it turns out there is an interesting issue that I stumbled into back when I developed those Deshaker scripts:

PAL field reversal in Deshaker script

The essence of the problem is that PAL uncompressed reverse fields. This apparently is something done by design, and you will find all sorts of posts about it if you Google:

PAL uncompressed field reversal

Thus, any tests you do using uncompressed might possibly mask the real issue because of this underlying industry anamoly (yes, this is something that apparently happens because of convention adopted a long time ago). By contrast, the problem we're wrangling here is a real bug in Vegas.

Thanks John.

G

Thanks for the details. We were able to repro the issue. It only occurred when the nested project was being scaled down (for example an interlaced HD nested project in an interlaced SD project or project being rendered to interlaced SD).

It will be fixed in the first Vegas Pro 12 update.

For ForumAdmin:

That is very good news. Thank you! And thanks again to Nick, "Grazie," and Robert ("farss"). It was worth returning temporarily to the forum in order to get this fixed.

However, the problem with tech support (who STILL hasn't responded to my ticket from over three weeks ago) remains: how many other bugs that haven't received this sort of attention and in-depth testing and analysis continue to plague the product release after release? I don't want to sound un-grateful that this one is being addressed, but I do want to point out that an awful lot of other problems remain.

Thanks ForumAdmin! Nice to know the time spent testing this was worth the effort.

Great to see SCS picking up the cudgels and running with this.

I say this for 2 reasons:

1] The Nesting Issue will get fixed

2] SCS have rightly acknowledged and responded to the well thought through, applied testing procedures coming from evidenced based material.

Great work Guys - all round!

Grazie

Once nesting is fixed in Vegas 12 what chance do we have SCS will update older versions with this same problem?

Jerry K

Jerry K : Once nesting is fixed in Vegas 12 what chance do we have SCS will update older versions with this same problem?

Forum Admin : It will be fixed in the first Vegas Pro 12 update.

Until you ask SCS directly, I understand from FA that Nest fixes would be confined to VP12. What anybody says here, outside of SCS, could only be conjecture.

There has been good work, which has taken much time by individuals in pushing this to the forefront for SCS to get this rectified. As I said, I'm looking forward in seeing this resolved in VP12, as FA promises.

Grazie

BTW!!!!! - After a Gazillion posts I've made, I've only just noticed the very faint "Reply Box Enlarger" control option (seesh . . )

For the record, today I received the first and only response to my tech support submission regarding this problem.

Problem submitted: 10/01/2012 10:43 AM
First human reply: 11/02/2012 03:15 PM

Over one month.

This is for professional software that costs $599.95.

Unacceptable.

Well, very unacceptable is correct!
From Thread Subject: RE: Oh WOW! 5 weeks for support to answer!
Reply by: videoITguy Date: 11/1/2012 8:08:00 PM
Can anyone say what the support of SCS actually is? IS it a contractor or is the service from Bangladesh with accents and all? Is it high-school interns?

You may be surprised to learn what the tech support of some major organizations actually consists of. I can name ( but won't do it here) that many of you on this forum are dealing with at least two major players in third-party video support hardware whose support (?) person is a hobbyist who is in the PR section of the company who volunteered to be the support staff. That's it!

Professional support organizations (I was an IT project manager for several Fortune 500 enterprises) use a very disciplined support management system. They start with tier one which is self-directed rapid support through FAQ's and high-school interns. Escalation goes to tier two which is more disciplined technically trained staff with at least several years of experience if response becomes complex or untimely ( usually defined as more than 24 hours to 2 days) and finally tier three - gets you to supervisory level where the question must be answered or there is a chance that some heads will roll.


Be certain to validate whether this works differently in Build 394 of VegasPro12 as promised!!