There is always a very very slight color saturation loss when coding to mpeg2 with the Main Concept encoder which can be allowed for by jucing up the colors a very very slight amount. I noticed this in V4 and when switching to V5 it was reduced considerable. Rember that what I am speaking of is "splitting hairs".

JJK

VC - "When you set your color (colour) in Vegas are you referring to the video screen or the external (television) monitor?" - When I preview on an external monitor , from within Vegas, it is okay, the colours are fine and rich. When I make/author/create/burn a DVD through DVDA5, place this final DVD "platter" in a DVD player on another monitor and view/see/observe the same "good" files the colours then appear washed out - enough for me needing to recalibrate the TV. Before you say it is because you are using another TV .. well I've run other Vegas rendered and created files - both mpegs and WMV - and their colours are good - if not excellent.

JJK - Yes I'd accept "hair-splitting" but this is between "acceptable" and a coaster .. . honest! The weirdest thing is that on recalibration the colour quality returns.

I'm just wondering if the total size of the project, being above 4.4 gb - could be a factor? Other DVD stuff I've burnt recently has been well below this. I'm going to view the smaller sized projects just to see if there is any type of "super" compression going on that would lead to this type of de-colorization [ my word .. it isn't technical . . ]

Anybody think that this is a possibility?

Best regards,

Grazie

Could it be that the de-saturation is due to the player itself, rather than the DVD?

I've just done a test, putting a SMPTE colour bar on the timeline, followed by Yellow Porches, followed by a rendered MPEG of same.

The rendered yellow porches were consistently "off" by one, i.e.:
brightest yellow 235 235 16, but rendered, 235 234 15

The rendered SMPTE bars were similar e.g.:

Rendered:________________On the timeline:

White 235 235 235________ 235 235 235
Yellow__179 179 15________180 180 16
Cyan____16 179 179_________16 180 180
Green___15 179 179_________16 180 180
Magenta 180 16 180__________180 16 180
Red____179 16 180__________180 16 180
Blue_____16 16 180 __________ 16 16 180

Would this be enough to cause the de-saturation you're seeing?

Alistair

Okay, I'd accept this, if it wasn't also true that non-DVD authored mpegs, and WMV, play at an acceptable saturation .. get my gist here?

It's only "since" I've been authoring DVDs that this de-sat is happenning. The same player plays the "other" files beautifully .. no need to re-calibrate.

Thanks for the tips and I'll get around to doing your test too. This wont be in the next day or two though.

Thanks again,

Grazie

Well, Mr. Graham, you certainly have stumped me! (Which wouldn't take much.)

Just for the heck of it, though, you might want to try another DVD player as Alistair suggested... just to be sure.

J--

Use color correction and Set your Black and White points then Lower the Gamma before rendering to Mpeg 2.

John

A friend of mine put a whole series od test patterns onto a DVD. They were all checked on a waveform monitor prior to encoding. Play it back on a DVD player and check levels coming out of player and they're not the same. He was using a Pansonic STB DVD recorder and pretty much proved what I've seen happen using Vegas.
Record video that's well outside BC legal to VHS and play that back to my very dodgy monitor and the thing goes nuts, total loss of sync everywhere I'd expect it to becuase the levels are out of spec.
Now make a DVD of the same video, play it back to the same monior and not a single hiccup.
It seems that either mpeg-2 encoders or more likely DVD players are clamping the levels to within legal limits. I'd suspect that what's happening in Grazie's case is the video is outside legit levels and the TVs adjust for that. Play it back through the DVD player and the now legal levels look washed out.
Only way to really be sure is with a waveform monitor, see what you're recording to VHS or whats coming out of the VCR versus what's coming out of the DVD player. It maybe that the DVD is what's correct and all else is wrong.

farss, excellent analysis, thank you.

Grazie

The problem is the black levels are out of whack. There are a couple of reasons for this:

The first has to do with the camera. North American equipment is designed for an IRE of 7.5 and European and Japanese equipment for an IRE of 0. Most of us buy Japanese video cameras set up for an IRE of 0 and when we plug them into NTSC equipment, the black levels we see are incorrect (darker than it really is). While the DV equipment sold in NOrth America is set up for 0 IRE, the DVD players you buy add "setup" or 7.5 IRE unless they have a setting such as "black enhance" which is in effect changing it back to 0 IRE.
If the Mpeg 2 encoder is NOT altering the black level and you are using the same TV monitor if you compare the DV tape and the DVD you made from it they will look different because the DV is 0 IRE and the DVD is 7.5 IRE. The DVD is showing you the correct levels, but this is the first time you are actually seeing it. The ligting conditions in the video will determine how much of a difference you end up seeing. In addition to that the camera settings for exposure will make a very big difference as well. Much of the time the autoexposure overexposes the video so manual exposure is necessary more often than not. LCD monitors are usually way off and outputting to NTSC video monitors will be off by their very nature (7.5 IRE). If you have a Proc Amp and AC current you can hook it up between the camera and the monitor to correct the setup level while you shoot.

The second is the Mpeg 2 encoder itself. You will have to calibrate your monitor first (discussed many times elsewhere. See BillyBoy's site for a good method). Then you can tell if the Mpeg2 encoder is changing it. Sometimes the compression that takes place can fool you due to the loss in detail, but if you do it a few times you will be able to tell if the encoder is changing the black levels especially if you don't drop the bitrate below say 6000. From what I have seen, I believe the Vegas MC encoder changes it.

Another issue is when you capture transcoded VHS footage. Some capture cards "require" 0 IRE and yet others have a jumper for switching between 7.5 and 0 IRE.

John

craftech, thank you!

Lots to take on baord. I notice you refer to NTSC. I'm PAL, what differences should I be being careful with?

Grazie

In your case, if you are using a PAL monitor, which I assume you are, the DVD player "should" not be adding a setup of 7.5 to the video so that would leave the MC encoder. I would try using the color corrector, setting the black and white points with the eyedropper and lowering the gamma.
I would also check the capture card to make sure it is set up for and expecting 0 IRE.

John

Same here, our test were done in PAL so no setup issues at all and test sources were precision test pattern generators.
I'll throw another red herring into the fire too, DVD is 4:2:2 whereas DV (PAL) is 4:2:0, that may also have some bearing on it, don't know if there's also a color space conversion going on as well.
But also bear in mind Sony cameras record pretty hot video to start with.

craftech - I'm using a PAL monitor; you are falling back on the MC encoder as the "variable" yes? Meanin, I would do what? And why WOULD the MC Encoder be doing anything? PLease remeber I can play back MC encoded NON-DVD-authored files with no need for monitor calibration . .meaning MC encoded stuff works well, it is only when I go down the DVD-authoring route that things go south. Using the colour corrector seems a bit of a fundamental "sledghammer" to crack this wlnut? Am I being naive?

This I don't undersstand - "I would also check the capture card to make sure it is set up for and expecting 0 IRE." . . I've been capturing and Printing to Tape and getting very successful results. It is only when I go near DVD Authoring, things change.

Yes farss, apart from having me rolling about with the mixed metaphor { 'herrings & fire' !!!! } - "I'll throw another red herring into the fire too, DVD is 4:2:2 whereas DV (PAL) is 4:2:0, that may also have some bearing on it, don't know if there's also a color space conversion going on as well." . . .Canon camera here. Tends to be rather cooool . . . In any event, I don't understand [ my fault not yours ] or what would be the correction needed to improve my postion from what you say; your greater experience and deep understanding of these things has left me way, way behind.

Please advice,

Grazie

farss

DVD is 4:2:0 the same chromiance/Luminance sample rates as DV.

Grazie

I've only got DVD Archietct 1 and have not noticed any desaturation of colours but I haven't done any side by tests so there could be some differences.

You could try checking your original clip levels in Vegas using the scopes and importing some of your rendered mpeg and checking the levels of the mpeg.

martin

ps At work we use phillips set top DVD recorders that seem to desaturate colours a noticeable amount.

Grazie,
I also mentioned in another post of yours regarding this issue that possibly DVDA 1.0 is doing it. In terms of the capture card, read the sheet that came with it and see if it has a jumper. although I don't think that is the issue and it will generally be more of a problem capturing transcoded analog such as VHS.
Why do I have to color correct?
It may be the nature of what you shot which is making it more noticeable.
Try my suggestion with a loop render of color corrected sections that have the same lighting. In other words, split a representatively lit scene having the same camera exposure into smaller sections of say 10-15 seconds each or longer depending upon how much footage you have to play with. Color correct using the eyedroppers to set the black and white points. Number the clips to do gamma adjustments, starting with 950 and working your way down in increments of 50. I have gone as low as 750 with certain scenes.
Loop render, author, and burn it onto a DVD/RW and play it on a bunch of different DVD players, then see for yourself. If the contrast is nice and the colors are good, but you still want more color, do the same thing using the saturation slider and raise it VERY little at a time.

Now there IS another possibility here, and it goes back to that battle thread about video monitors a few months ago. My school of thought is the same as BillyBoy's in that a TV monitor is sometimes better than an expensive broadcast monitor for calibrating. "Proper" calibration is usually "darker" than preset levels in commercial televisions. Many videographers (particularly thise producing VHS) produce "washed-out" videos because they calibrate their expensive monitors properly, then the customer plays them on televisions with default settings and the videos look washed out. That's why, if you want to do this and get good feedback from customers, you need to try the videos (or loop experiments of them)on as many different types of commercial TVs as you can until you come up with your own "proper settings". Take your video to a store that has a bunch of TVs. Here in the states we have BestBuy and CircuitCity for such things.

John

Martin,
I'm fairly certain that you're wrong, I'll have to go do some research about DVD being 4:2:2, possibly the MC encoder is only 4:2:0 but all the specs I've seen on high end encoders indicate they work in 4:2:2 via SDI.

Grazie,
I'll drag out my calibration DVD and see what levels come out of my DVD player. In the interim here's a simple test. Look at your video on the histogram scopes. Is the luminance outside the range 16-235? If so then you're outside BC legal. Apply BC filter, Extremely Conservative template. Render out small section, PTT as you did before and encode to DVD. If my theory is correct you should see far less difference between the DVD and DV version.

This issue has been raised indirectly some time ago in discussions about the need to stay within BC legal limits. General comment from Sony was if going out to VHS or for broadcast you need to worry about it. If going to DVD ignore it. My own experience backs this up. Question is if going to DVD why isn't that an issue? Probably answer? The players are clamping the levels on composite output.

As an aside I'd agree with a comment somewhere in the discussion about what gets broadcast. One of the local stations tried pulling a sneaky by pushing their levels to 110% to make their content look more attractive. They got a serious rap over the nuckles from the regulator.

Gentlemen -Thank you. A very thoughful and intelligent assessment. I'll give the testing a go ASAP. This wont happen prior to the weekend.

BTW -Hello MArtin! Good to hear from you .. . SO you too experience such a thing too. Ah, I aint goin nutz!

Grazie

Well Martin is correct, DVD is 4:2:0. During my searches I did find this interesting article:
http://www.hometheaterhifi.com/volume_8_2/dvd-benchmark-special-report-chroma-bug-4-2001.html

Does demonstrate how complex the whole encoder/decoder issue is.

Well, nice article, and even more "confusing". So, what are the solutions? Knowing - even if I did "understand" what had been written - the position on 4:2:0 viz DVD with the to and from encoding 4:4:4 or 4:2:2, what next?

Many thanks Martin for bringing this to my attention and many many thanks farss for doing this latest research.

As a footnote: At what level does/should a typical videographer/editor - for example myself - need or have too get this complex? Should this level/complexity be "pushed-out" of the way by the Sonys and MainConcepts of the World? After all, I'm doing and making videos for clients using highly complex programmes and brilliantly created software, I have no comprehensive idea or deep understanding what on Earth is going on under the hood/bonnet of this vehicle that I drive everyday. And I'm making a living from these packages.

Bottom line: I want to be ble to make/monitor/PTT/DVD author and EXPECT that the quality to be be within acceptable margins throughout the whole process. This HAS been the case with everything I've done so far .. Capture miniDV footage>Edit and Preview on external monitor> create WMVs & MPEGSs and VCDs/SVCDs/ . . Now only when I author DVDs I have come to this enpass. Hmmm... How much knowledge do I OR am I expected to have to do this? - Tell yer what folks? I've yet to revert to my old U-Lead simple MovieFactory DVD authoring package? I'll give that a go with the same file . . . hopefully I'll get the same result. If I DON'T then . . arrggghh! ! !

Thanks again lads!

Grazie

At what level does/should a typical videographer/editor - for example myself - need or have too get this complex?

I know just what you mean (I think). As I've said in this forum before, I've never been the most "technical" person in world. I understand the basics enough to get the job done, and, at the risk of sounding bragish, the talent to do it well. If and when something in the process "breaks down," more often than not, I can find the answer to fix it. That doesn't mean I understood the problem, it just means that I found a fix for it. For me, it's like driving my car. I'm a good driver. I know my car's abilities and limitations (as well as my own) and I drive within those boundries. If and when the car "breaks down," I'm dependent on someone else to fix it. Other than filling it with gas or changing the oil and tires, I know nothing about cars. I don't think I have to be an automotive engineer to drive a car well. I feel the same way about making videos. I don't have to be (shouldn't have to be?) a software and/or video engineer to make decent videos.

I want to be ble to make/monitor/PTT/DVD author and EXPECT that the quality to be be within acceptable margins throughout the whole process. ... Now only when I author DVDs I have come to this enpass.

Again, I'm right there with you. Unlike so many of the people here, I don't have the talent/knowledge base to diagnose 90% of the problems people post here. That's why I don't reply to too many. Most of the problems I read about here, I'm not experiencing, so I have nothing to offer. If I could be there--if I could "see" what's happening--more often than not, that helps me to see what may be happening, or not happening, and offer a suggestion as how to fix it. Then, add to that mix the differences between PAL and NTSC issues...

I know you're frustrated, I would be, too, and my heart goes out to you, it really does. However, I think if you will use the week-end, if time permits, to clear your head, get away from the problem altogether, relax, and have some fun, when you readdress the issue on Monday and go over your entire process, hardware, software, media, media player(s), monitors, cables, etc., from head to tail, you may well find what the problem is.

J--

Hi Grazie

Hope you are well.

I'm not really sure how much you need to know, it seems to me ironically that the more professional hardware/ software (Avid/Media 100 in my personal experience) the less you need to know about such matters as they sort all of this out for you. But these are used by exactly the type of person who probably would know how to make all things legal etc. I use the term professional losely, those who work in broadcast environments.

If you are interested in learning more, I found it helped me understand the limitations of the DV format, have a look at Adam Wilt's site (type his name into google). He writes for an american magazine called DV, and manages to explain things in a relatively simple way without skimping on facts.

Hopefully Sony will sort out this problem in a future release.

Martin

You've indirectly highlighted a point I always try to make. One of the reasons movies shot on 35mm look so much better than DV is because it costs so damn much to shoot on 35mm. When it costs that much every minute the camera is rolling you take a lot more care over what's happening in front of the camera.
Same applies to high end editing systems, when it costs you serious money per hour just in depreciation you again find the 'cost' of getting the technical knowledge relatively smaller. You can make just as many stuff ups on any system at any pricepoint. I'll admit Vegas lets you easily do a lot of things that can get you into an ocean of grief if you don't know what you're doing and so in some respects you perhaps need a better understanding of the processes and limitations.
Then again knowledge isn't expensive, pretty well all of it can be had for free off the web these days. I've never been one to see any downside to having a deeper understanding of anything, the more I know about what's happening 'under the bonnet' the more I enjoy the ride.