Video keeps failing harding test - working with stock - HELP NEEDED!!!
by Lawrence Vaughan (l_vaughan)
on
Jun 27, 2008 at 1:41:59 pm
I have made a music promo, unfortuantely i have had it back from MTV twice now as it keeps failing the harding test (to test to see if it triggers epilietic siezures).
I am working with stock from the prelinger archives, which - to be honest - has been a nightmare to work with for various reasons.
The wierd thing is, the test has failed twice, both on different areas. I edited the video after it failed the first time, then it failed on a totally un-related area on the second test, which the first test didn't pick up.
I am getting "luminance flashes" between 0.5 and 2.2. Is there a way I can filter these out in the edit rather than get the video sent back again? The "luminance flshes" can not even be made out to the naked eye. I have changed the problem areas, then it picks up something else once I send it to MTV again.
Clearly something is happening when I export the video and get it transferred to Digibeta (same file, same edit - just changed slightly), OR - something is happening in Final Cut pro when I export to DV tape which is making this happen.
£60 a time for a digibeta transfer is becomming quite expensive. Especially if I get it sent back again!
Re: Video keeps failing harding test - working with stock - HELP NEEDED!!! by Rob Alexander on Jun 27, 2008 at 1:50:58 pm
Can you describe the shots that you're failing on. Often it's something that you don't necessarily think of as a flash - for example a close up of flickering flames.
Re: Video keeps failing harding test - working with stock - HELP NEEDED!!! by walter biscardi on Jun 27, 2008 at 1:57:31 pm
Are you getting rejected from MTV U.S. or MTV Europe?
I'm asking because if you're in Europe obviously you're editing in PAL and if you're delivering to MTV U.S. then you're converting the edit to NTSC.
This conversion process if not done by the proper hardware, can add millisecond flash frames at edit points. They're not really flash frames actually, they're kind of a split frame of the outgoing and incoming video at the edit point. The video frame essentially gets split in half or into 1/3's so it looks like a flash frame or a millisecond edit.
If you have a bright area of the incoming or outgoing image, this could account for the luminance flash. It's not really a flash, but the edit point being converted to NTSC. Just a thought.
Walter Biscardi, Jr.
Biscardi Creative Media HD and SD Production for Broadcast and Independent Productions.
Re: Video keeps failing harding test - working with stock - HELP NEEDED!!! by Ed Arsuffi on Jun 27, 2008 at 3:57:33 pm
In addition to comments prior. Also, look for repetitive patterns within a specified time gap. That would trigger the pass/fail measure. What was tested? Your DV tape, the Dbeta transfer or a file created from the Dbeta txfr?
If it was the Dbeta is the method of transfer to Dbeta from your DV tape a good transfer. Is that process introducing the image anomalies? i.e. From deck to deck how is the visual image being converted. That is Video out of the DV deck to video in on the DBeta deck. SDI, Firewire converter, analogue? An incorrect signal setup can introduce video levels that cause issues as well.
In any case you should be able to request the data from the harding software app log. It is time stamped and should tell you if it is Luminance based (like you stated) or spatial.
Re: Video keeps failing harding test - working with stock - HELP NEEDED!!! by Bob Flood on Jun 27, 2008 at 3:48:03 pm
whoa
I dont know wahts more bizarre, the fact that there is a test to see if your video is cut so fast it will trigger a seizure, or that MTV uses the test on the clips before they air them.
so does this mean that "MTV style editing" is no longer so fast it will cause a seizure? If so, what is? "Men In Black trailer" style editing?
there are some shows on food network that may fail that test
Re: Video keeps failing harding test - working with stock - HELP NEEDED!!! by Chris Borjis on Jun 27, 2008 at 5:07:52 pm
[Bob Flood]"
I dont know wahts more bizarre, the fact that there is a test to see if your video is cut so fast it will trigger a seizure, or that MTV uses the test on the clips before they air them."
no kiddin.
I'll have a seizure if I even see 2 seconds of the channel.
Re: Video keeps failing harding test - working with stock - HELP NEEDED!!! by Jason Porthouse on Jun 27, 2008 at 5:19:03 pm
Better brains than me have ventures to suggest that the Harding test is an absolute con... and I'd be suspicious if your failing twice in different places. Check that you're not having any transfer issues when you're laying off to DigiBeta. If you're confident that the tapes are identical in the area that failed on the second pass have a word with the testers.
Other than that, failures are often just about luminance extremes, and some can simply be cured by clamping the luminance - so if you've a flickering fire scene then taking that down so the flames are maybe 60% may well cure it.
The kit for the test is about £12k... I heard recently that there are plans for a software version. It's going to be... around 12k. It's a license to print money, and I'm not sure of the efficacy of the test at all in terms of stopping seizures. If someone came up with a plug-in that would mimic this test accurately then they'd make a mint - OK, you wouldn't be able to use it for 'official' clearance but you would at least be able to check before taking it for the 'proper' test...
Jason
_________________________________
Before you criticise a man, walk a mile in his shoes.
Then when you do criticise him, you'll be a mile away. And have his shoes.
Re: Video keeps failing harding test - working with stock - HELP NEEDED!!! by david bogie on Jun 27, 2008 at 6:36:23 pm
Mr Vaughn, you've posted this same question in at least two FCP forums, maybe others. You're getting a wide variety of interesting suggestions but they're dramatically different. If you get this sorted out, please go bck to the various forums and close your threads with the answer that worked for you. If anyone every searches for "Harding Test" they'll find some decent advice.
bogiesan
This is my standard sigfile so do not take it personally: "For crying out loud, read the freakin' manual."
Re: Video keeps failing harding test - working with stock - HELP NEEDED!!! by Rob Alexander on Jun 27, 2008 at 9:00:25 pm
I don't know whether it's fair to call the Harding a con. It's just testing to standards laid down by OFCOM - and these haven't changed in a long time. Problem is (and that's why so many more programmes are getting caught out) that the Harding's much better at testing to these standards than it's predecessor the GORDON box!
I'm afraid off the top of my head I can't remember the numbers but it detects an x percent luminance change over x percent of the frame - more than 6 of these in 2 seconds and you fail. A classic example is paparazzi at a premiere, all those flashes going off even though they don't white out the entire frame is enough to trigger the Harding, it's not flash frames or fast cutting necessarily. And as mentioned above repetitive patterns can trigger it too.
The chances of a fit being caused by any of these may be remote but these are litigious times and the broadcasters are protecting their arses.
Re: Video keeps failing harding test - working with stock - HELP NEEDED!!! by Lawrence Vaughan on Jun 28, 2008 at 9:01:06 am
Wow! A massive response. Thanks guys, it is really really appreciated.
I will answer in order:
The shots are basically simple shots of some people in a street. Nothing from what I can tell that crazy. No flickering flames or anything like that.
It's getting rejected from MTV Europe, so no problems converting it to NTSC as of yet.
MTV will only perform the harding test from a digibeta tape. No other formats are supported (apparently). As far as I know, they are doing it from the tape itself not a dbeta txfr. However - I have no idea how the process works - I literally send them the tape. I will ask however - and let you know on that one.
In terms of the transfer itself, everything seems fine to me. Not entirely sure if they use analogue or firewire. However - I will be getting another transfer done, I might hold their tech hostage until he tells me everything I need to know!
I understand they, MTV and OFCOM are covering their arses. My girlfriend is a lawyer! I also understand these tests "have" to be done, it's just becoming an expensive method of trial and error (£60 per transfer).
In terms of a "fast edit" this apparently, is safe (providing it's not like a shot per frame). The test looks for flashes, patterns (such as moving lines, hypnotic circles etc). The video is fairly fast, but nothing that MTV or indeed anyone else would consider unsafe.
I'm having seizures from dealing with MTV, they are honestly the most unprofessional company I have EVER worked with. When contacting via e-mail If you want answers from them you are guaranteed a response of two, maybe three words of little help, often patronising.
When asked if there was any way of me testing if the video would fail I was greeted with a response which read "find a harding machine".
Personally, I feel the harding test is non a con, merely inconsistent. I'm not entirely sure of it's efficiency, I've never seen one working. However - I have read the repost, although fairly detailed in it's content there is something niggling away that isn't quite right. I am 90% confident that the transfers that I sent were identical. Something may have gone wrong during the trnsfer, I cannot rule that out. Worryingly, all of the local TV / broadcast / corporate video stations near me have never even heard of the harding test, yet they are broadcasting content nationally!! I think they are probably looking into it by now...
I understand I have posted this in two forums (apple forum), once I get everything sorted I will post my results and close the thread. I sincerely want to help anyone who might have this problem in the future. Posting in two forums - I just need help.
---
Any other response it greatly appreciated. My next step: Calling the tech who has done these transfers and ask him if they can test it once the transfer has been done (providing they have a harding machine). I will also ask for the video MTV currently have to be tested again. I want to see if the results are the same as last time. There is a chance that their harding machine could be faulty / inconsistent.
Re: Video keeps failing harding test - working with stock - HELP NEEDED!!! by Lawrence Vaughan on Jun 28, 2008 at 12:40:31 pm
No. Normal speed/ The first time it failed I had some footage of a car speeding down a street (on a totally different part of the video, so unrelated) Sped up X 500! Obviously, that had to be taken out.
Re: Video keeps failing harding test - working with stock - HELP NEEDED!!! by Ryan Mast on Jun 29, 2008 at 4:20:03 am
What about colors? Never used a Harding test before, but according to the info on their website (http://www.hardingfpa.com/), it tests for luminance AND color. Red and blue in particular. You could ask if they would send you a screenshot of the graph that it generates, to see if it's luminance or color that's causing your video to fail.
Re: Video keeps failing harding test - working with stock - HELP NEEDED!!! by Lawrence Vaughan on Jun 29, 2008 at 9:04:04 am
In the report it breaks down when and where these instances happen. It gives timecode and the value (as far as I know from 0.1 to 5). It gives red, spatial and luminance in a column. Mine is just a problem of luminance.
Blue, strangely - isn't mentioned.
