Understanding HD & 3G-SDI Video In HD

I'm giving a talk at Broadcast Video Expo next week on HD infrastructure (10Gig Ethernet, OM3 fibre & 3G Video) and I was pleased to see several manufacturers are doing similair. The link is to Tektronix's 3G poster - all good stuff.

The great 3G cable shoot-out

Thanks to Simon Hillman for tabulating them in a way that makes trends clear. The link in the title is the directory with the A1 plots.

The tests based on a pathological signal are more extensive and test for six cable types - three HD and three SD. Since we were trying to spot trends due to cable length we feel this is the most informative of the sets of data. Simon re-did a subset using only the three HD types and 1080/50P colour bars to test for jitter which (as you'd expect) doesn't vary to any degree with length but you can see the relative damage barrels and U-Links do.

The conclusions that spring out are;

• At 3G using coax specified for HD 60m seems to be the workable cable length before attenuation becomes an issue and the eye closes below 400mV.
  
  
• SD coax goes about half the distance - this seems counter-intuitive as most SD coax has a notional analogue bandwidth (+/- 6dBs) of 360Mhz - three octave less than HD coax. Clearly the signal recovery in the WFM8300 is at play.
  
  
• The variation between the best (most expensive) and worst cable at HD before the signal becomes sub-optimal (i.e. worse than 3dBs attenuation) is less than 10m with Belden 1694 coming out on top.
  
  
• The 8300 was still able to recover a signal at 150m with Belden but only 120m with the Draka DC DVC13C. At these length the mean time between corrupt video frames would be unacceptable.
  
  
• The Condufil 1694-equivelent tracks the more expensive Belden cable very well.

As mentioned this was really a test of run-lengths for a practical guide to cabling TV facilities. In the bulk of the tests we used the correct BNC crimp connectors and the proper tools for the brands of BNCs (attached by an experienced wireman) - We did try and provoke jitter by mixing up connectors with cable but it seemed to make scant difference. It does seem for 3G HD video the newer style 4.5Ghz are to be preferred over the original HD-type cable.

In the end I suspect that these results will represent the best possible world as Tektronix gear is known to drive a coax line optimally and has excellent return loss on its inputs. Other manufactures are less so and if our experience with 1.48G 4:2:2 HD is anything to go by the massive variation in the quality of line-drivers and receivers will make these results meaningless. Nobody (particularly in these hard economic times) builds a facility with only Sony and Tektronix equipment.

Many thanks to Tom & Lee at Tek for the loan of the equipment and advice and Simon and Graham at Bryant for providing the various cable types, ends & tools. Please note all original information and test results are the property and copyright of root6

3G cable 'shoot-out'

Three gigabit HD-SDi video has been around for a couple of years now but I've yet to see a comprehensive review of different cable types and how they handle the newer 1080p50 signals. Bryant Broadcast contacted me a couple of months ago and asked if I could conduct a test of the six cable types they recommend for HD work. So, I've borrow a brand new Tek WFM8300 (launched at IBC, not that I'd have known!) and have six drums of cable and all the BNC ends and appropriate crimp tools at the ready! I'll be measuring jitter, noise and the overall state of the eye pattern at 10m intervals from 200m down to 10m and then when we have a good feel for the various cable types we'll turn our attention to the damage that jackfields, u-links and BNC couplers do to the signal.
The image above is a screen capture of the machine with 2m of Belden 1694 cable - rated at 4.5Ghz!

3G HD-SDi standards

Yesterday I was doing a day of training - Root6's Video101 (you can grab my slides here) was what I was doing, but the day started off with a presentation with my old pal Lee Ballinger of Tektronix. He went through SMPTE372 and the now sixty(!) transport formats it covers. A 3G payload can carry many variations of Y Cb Cr / RGB / XYZ colour, framerates etc. You can even send two 1.48G HD-SDi streams down one side of a 3G connection - this is being refered to as SMPTE 292B (an extension on the original HD-SDi spec).
One of the things I'd not realised was inter-link timing discrepancy - it can be a max. of 40nS (not long!).
Anyway - Lee's presentation will be available as a video on Root6's site when Mark pulls his finger out and edits/encodes it. Whilst there you can check out me banging on about 10-gig ethernet.

Tektronix - why they really are STILL the best! pt.1

I often say that customers only ever avoid buying Tek because they can't afford them - nobody ever says "...I need a Hamlet/VideoTek/Leader etc." because it's the best solution, rather it's cheaper than Tek. However - in the last year they've had the WVR/WFM 5000 range which puts pay to even that.

Anyway - last week I went to their Vision '09 day - a kind of pre-NAB training day for users and dealers. They ran four sessions.

Upgrades to WVR/WFM-series 'scopes

• ANC data - the HD-SDi ancilliary data stream can now encompass user-defined type. So, you could make a new ANC-type called whatever you like and as long as you know the pre-amble values you can have the Session Data page watch for those packets. Very useful if you wanted to individually mark the output of encoders and then check to see which machine made an HD-SDi stream. I suggested it to Root6's Dev team for Content Agent.
  
• Dolby information - On the Audio Session tile you can now see the values for the DolbyE metadata - Dialogue Normalisation, Dynamic range and Downmix. You can also check the the guardband figure to see if the embedded DolbyE data is locked to video.
  
• Black/Frozen - this is a new parameter in the video QC section - previously you could check for extended periods of silence on the audio but now the instrument can check for frozen or extended periods of black - it's window'able as well so you can check if the lower-third financial ticker (for example) has frozen despite the studio video continuing.
  
• Infinite Persistence - you can turn any display into a storage 'scope - useful for checking for those extreme events that occur rarely.
  
• AFD - a few more vertical-interval signals (although they're now rolled in with the ANC data) can be detected - widescreen flag, V-chip, etc.
  

Overall a superb set of features which come for free (just download the updater) - WFMs can do it off the USB socket on the front, WVR machines over the network. This will all be available after NAB.

WFM7100 screen grabs

I'm currently prep'ing some training notes for video, audio & QC test and measurement. I've often said it, but Tektronix really are the best-of-breed for television signal monitoring. People only buy Hamlet, Harris, Videotek etc. because they won't stretch to a Tek.
So, here are some screen grabs - handsome!

Tektronix WVR7120 audio pinouts

Why do Tek lock their PDFs?!

Tektronix WVR7100 & NTSC gamut in HD

Talking about NTSC colour gamut on an HD recording seems a nonsense (how could something in 709 colour space have any relation to composite legality?) - however, since Sky take delivery on HD-Cam (and increasingly SR) they insist that all recordings also conform to a composite colour space. For those recordings in 29.97 and 59.94 it's an NTSC colour space they need to also conform to. This seems a bit daft to me - knobble your nice HD production so we don't have to worry about it. Surely the best way would be to transmit the HD and then stick a legaliser across the down-converted output rather than making the HD master conform to a small colour space so that down-conversion is easier. Talk about casting your pearls before swine!

Anyhow - one of our customers uses Tektronix WVR7100 rasteriser waveform monitors and has been QC'ing for delivery to Sky and noticed that with an NTSC master (either SD or HD) the Tek consistently throws composite gamut errors. The second screen-grab shows that the legaliser feeding the Tek has had the colour gamut wound right down (see how cropped the two diagonals on the arrow-head are) yet the 'scope continues to flag errors. It even does this with monochrome material!
So - I have to get on to Tek. The problem isn't there on 24, 25 or 50i material, just NTSC-derived stuff.

Upgrading a Tek WFM7100

I demo Tektronix monitoring products and today I had to upgrade our loaner WFM7100. I'd done software on these chaps before and you can either do it over the web interface or via a USB stick on the front panel. In the case of this version 3.0 update (from the previous v1.25) you have to replace the front control panel, update the boot loader over the network and then use the USB stick to upgrade the instrement.

It's quite a change - it's made it into a WVR7100!
It seems a lot more stable and quicker. Also - it's not a touch-screen any more - it has the same front panel button set as a WVR (but arranged around the display). Using the Java app to control it you can't tell it apart from a WVR. Amazingly Tek ship this upgrade kit for free - given how well made the new front panel is I imagine they are taking a hit of hundreds of pounds per customer update.

I was tickled to see the attached label on the USB stick - you might kill your machine by doing the upgrade and if you do we'll bill you to fix it! I can report the update was entirely successful!
Actually - Tek upgrade procedures are always entirely straighforward and the instructions never leave anything to chance.

Another cool feature is that (being a dual-link HD-SDi machine) it now supports dual live inputs (at YUV 4:2:2, not RGB 4:4:4!) so you can be monitoring two video signals at once!

The PAL vertical interval

Our very own Kevin King asked me about the vertical interval in a PAL signal - the above diagram is excellent and has all you need to know. It's taken from a Tektronix training manual from the late eighties by Margaret Craig called Television Measurements - PAL Systems.
Essentially you should bear in mind the following;

• Active video starts at line 23 - actually line 23 is a half line so the first 26 u-sec is blank - people who don't know about video often notice this!


• The other half line is the first half of line 623 (end of field 2) - the reason for the two half-lines is to give the line-scan circuit the best chance to make it back to the same point when it starts the new field.


• There is no half-line at the start of field-2 or the end of field-1


• Field 1 and Field 2 both end with five equalising pulses and then start the next field with five broad pulses, to be followed by more eq pulses. These are the only lines not to have a colour burst - that's what Bruch Blanking does on older SPGs.
  

The reason for a lot of this is down to the stability of phase-locked loops constructed from valves rather than transistors - if you have (potentially) unstable oscillators for every waveform (line-drive, field-drive, sub-carrier etc.) then you have to take every effort to make sure nothings changes phase too quickly - hence the half lines, the broad and equalising pulses etc.

ITU Rec 601 vs Rec 709 colourspace

Every superhero knows that in transitioning from standard definition television to high def we've adopted a different matrixing function for component to/from RGB conversion. The numbers for (old-skool) Rec 601 are thus;

Y = 0.299R +0.587G +0.114B
Cb = 0.564(B-Y) + 350mV
Cr = 0.713(R-Y) + 350mV

And the new kids on the block (Rec 709);

Y = 0.213R +0.715G +0.072B
Cb = 0.539(B-Y) + 350mV
Cr = 0.635(R-Y) + 350mV

So, not only has the weighting of the colours that make up the luminance path changed but the weighting of the colour difference signals is different. I've heard varying accounts of why they felt the change was necessary - I think it's probably to do with cameras and telecines (now be entirely CCD-based as opposed to the ubiquity of tubes when 601 was being formulated) and display devices (are we going to be able to buy a tube'd monitor by the end of this year?!). The new values better reflect the tri-stimulus nature of human vision and are less bound by the very noisy response of the blue-tube in image acquisition devices of yester-year.

However, one of the upshots of this is that digital devices that can receive an SD/HD-SDi bitstream have to be able to switch in the appropriate matrix. If that isn't the case then you'd notice a green cast on pictures if you switched between standards (going from HD to SD) or a magenta error going the other way. In the case of a monitor you'd have to re-calibrate the white point to D65.

The reason this has cropped up is that a facility (where I've just started to offer them colour calibration advice) has noticed that a monitor that was lined up correctly for HD working is showing the wrong colourimetry when being sent an SD feed. It's gone green (and not with envy! - oh, and that isn't the facility in case you're wondering!). It's a JVC DTV1700 series monitor which (although a cheapie at <£2k) has an EBU-phosphored tube (so you can calibrate it to 6500k at the white point). It looks like JVC's input card doesn't do the matrix switch. So, I'm wondering what other monitors do - I was sure the Sony BVM-D range did (but those monitors started in the mid-teen thousands of pounds). Any comments from people who've hit this before? As an aside the image (right - click it!) is from a very good Tektronix poster entitled Understanding Colors and Gamut - I have many copies (along with the equally exciting Understanding High Definition Video!) - give me a yell if you want one.

Tektronix WFM7100

I've just my hands our new Tek WFM7100 demo unit and after one quick firmware update (to v. 1.23, fact fans!) I've been monkeying about with it ('cause I'll be demo'ing it post IBC). First thoughts are that it's very sluggish - screen update on waveforms is fine but navigation is slow. It feels like i's running on top of something (Windows Embedded?!) and that seems to be born out by the fact that when you hit it over the Java interface the menu on the front panel screen remains with the last selection you made and the remote display changes as per selections you make on the browser - almost like a remote desktop.
Anyhow - feature wise it is pretty much a WVR7100 with some nice extras. The biggest step is the screen grab facilities. In the same way that video or audio 'events' (gamut, level etc.) can be placed in the log, close the GPI, honk an SNMP alarm or just indicate on the front panel you can now have an error condition dump a screen-grab to a USB thumb drive. Now you can imagine exporting the XML log with still-frames of the offending video - excellent for monitoring the state of a cable head-end (for example). Another addition is having a picture o/p SVGA as well as the instrement's display out. We often install the WVR series rasterisers into machines rooms and DA the feed to several places (including for the operator in the machine area) - but since this has a built-in insterment display it would be advantageous to have just the picture (on a 15¨ TFT panel) thus avoiding the cost of an HD monitor.
Unlike glass-tubed traditional 'scopes (and even some other rasterised models - specifically the Videoteks) the 7100 generates the graticules in the same digital space where the HD/SDi stream is demultiplex'ed - the upshot of this is that they are ultimately accurate - something no other 'scope can lay claim to (unless they've just been calibrated!). Having spent time doing a head-to-head with the other manufacturers (Videotek, Omnitek and Hamlet) I conclude that the only reason folks don't buy Tek is the price. In terms of accuracy, ease of use and feature set the Tectonix are way out in front. Their automated QC features are pretty impressive and make delivery reports very easy to prepare.

Dual-link HD SDi is dead (and other things!)

I had a great meeting with Lee Ballinger - a sales engineer at Tektronix who looks after Root6. He periodically updates me on new products and gets me up to speed for when I have to demo them. He is a great guy - an old school engineer who is very much grounded in test and measurement. The WFM7100 series is the new range of portable rasterised video 'scopes and looks set to really clean up - they have addresses the (very few) weaknesses of the WVR range and have a killer product. I'll blog a bit more about it when I've had a day with our new demo unit. Lee also told me about a couple of interesting developments that he's been party to in other Tek divisions;

• Sony are set to kill off the dual-link HD-SDi interface - as a three-gig interconnect it has always been seen as ungainly and they aren't going to launch any more products that use it - a ten gig multi-mode fibre (straight to the back of the VTR/Telecine/etc.) is just around the corner. This will mean that those stick-in-the-mud engineers who refuse to take fibre seriously (or do it half-heartedly) will have to pull their fingers out.


• Quad-core 4.4Ghz Intel chips - he's seen production samples in quantity and they are coming soon - Tek's new seven gig 'scope wasn't quick enough and they had to rush a set of prototypes to Intel to allow testing of those new microprocessors!

Tektronix & automated MPEG test, H264 and VC-1

I went to a Tektronix training seminar this morning and really got a lot out of it. The first session was led by a chap called Dr Mark Nicholls who amongst other things set me straight in my perception of the current cream of MPEG4-like codecs;

• MPEG4 implementations (DivX, XVID, WM9 etc.) score over previous DCT-schemas (MPEG2 principally) in that they allow forward and reverse reference to distinct macro-blocks rather than discrete frames. That is where the increased compressibility comes from and explains some of the weird effects you see in corrupt DivX files!
• More recent implementations (H.264 & VC-1) allow for variable size macroblocks as well - in effect the compression can track the detail better and really allocated data-space where things need it.
• No broadcasters currently implement PCR timing recovery in the transport stream - I wonder if this explains why the MS-DVR format files my homebrewed PVR produces are always 12-frames wrong sync-wise (i.e. there is no way of precisely timing each GOP)? You get this effect if you try and play the files outside the MediaPortal app or if you import them into VirtualDub.

The rest of the session was devoted to Cerify - a very clever system for automated QC of LAN/SAN-based media - Rupert had a look at the API and is convinced it would be a cinch to make Artbox drive it. It really was a very impressive demo and I imagine it will be the kind of thing that lots of broadcasters would buy into in coming years. A kind of n x virtual-WVR7100s continually hoovering up new media and analyzing it for all video/audio and transmission parameters (including perceived parameter - measurement of compression PSNR for example).
Anyhow - it's probably copyright, but all the training notes are here.

Why Tektronics are still the best in their field

At Root6 we sell lots of bits of video kit and for video test and measurement Tektronics are still the best. Their rasterised video test sets are superb and one of the killer features is the ability to treat the thing like a webserver for both control and logging. You hit the device with a browser which gets served a Java app that allows you full access to the instrement. Recently we've had real trouble with Tiger (Mac OS-X v.10.4) and the WVR7100 - after much fiddling I got this very complete reply from Tek's engineering group;

Thank you for contacting Tektronix technical support.

Part of the problem here is that there is no official Sun JVM for the MacOS platform, instead, Apple wrote their own version which looks much nicer and is mostly compliant... mostly. We usually recommend version 1.4.2 on Windows computers (since v1.5 is still a little buggy) but it's hard to say what version would be best for a Mac.

What I get when I try to run the WVR remote user interface depends on which browser I use. Internet Explorer simply gives me a gray screen and never does anything - that's fine, even Microsoft has given up on that browser. Safari seems to get close but at the very last second gives me a Java exception when trying to access a cookie so it looks like there's something subtly different about the way Safari handles cookies that prevents it from working with the WVR. FireFox worked perfectly, and until I can figure out why Safari choked my recommendation would be to download FireFox and try accessing the WVR with it. You should be able to get FireFox at www.mozilla.org.

If you still have trouble with FireFox then let me know and I'll work with you to resolve this issue. Also, if you still have trouble it would be helpful for me to know what errors (if any) you receive when trying to access the remote user interface, what version of Java you're running and what version of the OS you're running. If all else fails we might have better luck with a stand-alone version of the Java application which you can download from the Tektronix website.

One last thing... make sure you have cookies enabled, the WVR remote user interface needs them to function properly.

I hope you find the above information useful and please feel free to contact me with any further questions you might have.

So there you go - GET FIREFOX!

Tektronix video site is a superb resource not only for their products but for all manner of info on test signals etc.