82 TVBEurope QC Forum
when considering a QC strategy. Particularly unique to loudness compliance is that it is often best conducted separate from the rest of the QC process. And that good loudness compliance systems can not only fi nd problems, but can fi x them, too. Singhal: Three of the key issues for developing a successful fi le-based QC strategy are, fi rst, having a fi rm understanding of one’s workfl ow and the various issues that are encountered at different stages, so that proper tools are selected. For example, for customers with high Adaptive Bit Rate workfl ow, a QC tool with best ABR functionality may be a better choice than a general QC option. Secondly, having a good knowledge of the workfl ow metrics – content volume, fi le characteristics, types of QC issues to look for, expected QC performance – in order to properly estimate the software needs and to maximise the value of the investment. And fi nally, proper development of QC analysis ‘templates’ or use of pre-defi ned templates based on established specifi cations, with proper operator training to review and act on the results. Westlake: As with most technical solutions, you often get what you pay for. Carefully analyse your needs and think about how quality control can fi t into your workfl ow as seamlessly as possible. Consider at what points in your workfl ow you should verify your fi les. A system that integrates well with transcoders, MAM, editing systems, playout systems and storage will save much money and heartache in the long run. Reliability and future- proofi ng need to be considered as standards and formats change. You need a system that will keep up with future requirements. If you have end- clients, what are they using for QC? Can you share content test plans or templates? Don’t be taken in by a pretty interface.
Is QC automation always more effi cient?
Mehring: QC automation is more effi cient as long as it can be built into an automated workfl ow that includes human intervention – but only for exceptions. This is known in manufacturing environments as ‘autonomation’. This is part of the reason why at Snell we liken a typical media company’s workfl ow to a media factory. Understanding workfl ows as sophisticated assembly lines helps media companies gain massive effi ciencies through autonomation and other strategies. Singhal: With proper planning and smart insertion of QC automation in a fi le-based workfl ow, a QC tool can increase the overall
Karl Mehring Snell
Steve Nunney Hamlet Video International
Adam Schadle Video Clarity
Vikas Singhal Venera Technologies
operational effi ciency. No QC software is 100 per cent perfect, but it can provide invaluable assistance at various stages of busy fi le-based operations. Some of our customers say that they have been able to achieve 80 per cent automation with Pulsar, which we believe is signifi cant for most organisations. Vasudev: Yes it is more effi cient, as human QC is very time consuming and, hence, costly. QC automation allows operators to focus their attention only on those fi les fl agged as errors. Aside from the effi ciency aspects, automated QC tools are the only realistic method of validating metadata, and their ability to ‘look inside’ the fi le makes them able to identify issues that are easily missed using human QC alone.
Are there any downsides to auto QC?
Begent: Auto QC is good at detecting technical problems such as metadata, bitrate, GOP structure, syntax errors and other parameters which cannot be found by manual QC. However, over reliance can be dangerous. It will only fi nd those items you have set it to test for and should be used as a tool in conjunction with human QC. It cannot replace human contextual perception – for example, is that red hue a video error or a deliberate artistic effect, are the correct two football teams shown playing in the video, and is the content appropriate for family viewing?
Mehring: We see downsides only when the planning phase has not been followed, which leads to too many false positives. That said, these are easy to overcome with the right choice of system and equipment. Nunney: Auto QC can provide so much data it can be overwhelming for users, when all that is needed is to know if it’s a ‘go’ or ‘no go’ situation. Samad: Yes. The problem with auto QC is that it can be diffi cult to alter as workfl ow requirements change. With this in mind, I would always suggest looking for tools and building workfl ows which have some degree of fl exibility. Schadle: Yes. Implementation of a new version either in the fi le processing device or the auto QC device might introduce a change that could cause a problem or stop the workfl ow. It is possible that the auto QC process either does not work with the device/network path change that was made or if the testing solution itself has a new version. This should be assessed outside the workfl ow before it is placed into the critical
“Auto QC can provide so much data it can be overwhelming for users, when all that is needed is to know if it’s a ‘go’ or ‘no go’ situation” Steve Nunney, Hamlet Video International
path of output production demanded by the asset delivery cycle. Vasudev: So long as you give consideration to what you are trying to achieve with an automated QC strategy, there should be no downside whatsoever. Automated QC is a critical element for any organisation implementing a fi le-based workfl ow and is the only way of achieving cost effective and repeatable results. Westlake: No. Auto QC is a useful tool to ensure that video and audio content reaches the consumer in a form that will play, and that it will look as good as was intended. Where there are regulations – such as those for fl ashing images – auto QC is a valuable tool in preventing companies from inadvertently breaking the law.
Is faster-than-realtime always needed?
Davenport: In time-critical business such as news, single-fi le throughput and latency are critical and priority will be placed on processing speed.
www.tvbeurope.com September 2014
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84 |
Page 85 |
Page 86 |
Page 87 |
Page 88 |
Page 89 |
Page 90 |
Page 91 |
Page 92