Sony BVM-E250 OLED broadcast monitor

I had an hour or so to set up a new Sony OLED grade-1 next to a Vutrix HD Pro-24 (their grade-1 LCD display which I like) - the Sony is twice as expensive as the Vutrix (£20k vs £10k). OLED as a technology is supposed to have a number of advantages over thin film transistor (TFT - the important bit of an LCD display. A few are;

• In a TFT layer the polarisation of light is twisted through 90 degrees to allow the pixel to be illuminated by the virtue of the fact that the front filter is 90 degrees offset from the rear. In effect the transistor stops the light from the fluorescent (or nowadays LED) source. Since the light source is a few millimeters behind the pixel there are chromatic distortions inherent and since thin-film transistors don't shut-off all the light when turned off there are black-level issues. These are the two best known problems with LCD monitors in film & TV grading.
• Thin-film transistors have a limit to how quickly they can be cycled - typ. 16ms at best (I know some manufacturers claim faster but it's smoke & mirrors). OLEDs can be cycled a lot quicker for better response.
• In a TFT display the place where the colour is made (the three sub-pixel RGB transistors) is physically separate from the light source - not so with OLEDs where the illuminating LED is also the colour-maker.

So - with deeper blacks and fewer chromatic problems you'd think OLEDs were the way forward. The only thing to consider is the life-span. The blue OLED elements have an estimated life of 10k hours (around a third of the backlight of an LCD). Also - the metameristic character of OLEDs is different and so colour-management tools will need to be upgraded (I just spent £7k on a new LCD photometer!).

I thought out of the box the pictures on the BVM were very good - close to the VuTrix I'd just calibrated to illuminant-D (6504K at 80Cd/m2 for peak white). Response seemed as good with much better blacks, particularly from different angles. The monitor's de-interlacer didn't seem as good as I'd have expected for video-shot material but camera pans etc looked better than the VuTrix.

Overall I was impressed, but not an extra £10k impressed!

UPNP has always been a bad idea!

UPNP is a protocol that allows an application to open up ports on a router so that incoming packets from the Internet get to the correct IP address on the LAN. It's typically used to allow the XBox360 to set up open ports through your router to allow multi-player gaming. If both XBoxes are behind NAT routers there is no way that unsolicited traffic from one can make it to the other (hey, I never wanted your bullets to hit me!). Skype suffers thus if both callers are behind NAT routers (i.e. in most cases; who has an internet-facing IP address on their machine nowadays?) - details here). More recent versions of Skype will make use of UPNP if it's on the router.
You won't be surprised to learn that it's a Microsoft technology and I've always encouraged people to disable it on their routers. Any piece of malware inside your network can open ports and invite any other nasties in. In the case of XBox there are about four ports you need to open up for the Live! service to work. Anyhow - it turns out that Linksys routers have a bug that allows UPNP activation on the WAN side - that's right, with the correctly formatted packets you can open ports through a Linksys router from the Internet. Using something like UPNP Port Mapper will allow you to scan Internet IP addresses and open ports on those routers.

The title link is to the article on The H.

Flame ain't all that

The facility that I'm building at the moment has a couple of Flame Premium suites - the very best version of Autodesk's TV and film compositing and finishing tool. Now I know very little about Autodesk products - I've never really worked in facilities where they were used. Flame, Smoke and the rest of them have been around for nearly twenty years and so you'd think they are what software people would describe as "mature products".
Anyway - a currently model Flame runs on an HP Z800 workstation with a Kona K3 card for video i/o (more about that in a minute!), an nVidia 4500 for desktop and their own proprietary fibre attached storage called "stone". They can typically handle two streams of 2k in realtime.
All very good, but everything is specified in a system - even down to the Eizo monitor you can use for the desktop display. This is because the SDi output for preview (which is not the output of the K3) is made by looping the 2nd DVI output of the graphics card into a daughter card that converts it to SDi. Consequentially you often see bits of the GUI on the HD video monitor. Now the reason for them specifying a certain model of GUI monitor becomes evident - they have to run the output at a rate near video for the DVI output to be convertible! This gets even more silly when you find out you have to use their own provided long DisplayPort cable to run it. We'd run in LC loose-tube fibre with DVI extenders which initially only worked whilst Linux was booting in text mode (i.e. before the X11 display subsystem could run). I had to throw in a Lindy EDID manager to fool Linux into thinking it really had the Eizo directly connected before we could run the desktop (and hence the Flame application).
This, along with a ton of little bits that the kit ships with; an 8-port ethernet hub, Lucid AES->analogue converters etc gives the impression they expect you to install the system on your dining room table rather than in a pro video facility. It really has the feel of a v.7 Avid from 1996! Cobbled together from third-party bits and only just running (everything on the hairy edge).
It reminds me of the Abekas DVEous (again, 1996) - one of the design engineers confided in me that if every bit of silicon on the video-processing board only performed to published spec then the system couldn't work. They relied on everything outperforming itself.

Mute the TV automatically!

I've been a fan of ComSkip (a PVR plugin that automatically detects/removes TV adverts from recorded MPEG2 transport streams) and I firmly believe that technology will allow us to 'tame' various media sources - advert blocking in Firefox makes the web a nicer place, for example. Anyway - this is great, what a project;

iPhone 4 battery fault? NO!

Yesterday my iPhone 4 starting showing all the signs of a failed battery - the back of the 'phone was hot and it could hold a charge for around three hours. Darn - replacement handset with all the attendant messing around (although it a lot easier than it ever was on Windows Mobile!).
Anyhow - Googling around brought up a few sites with folks suggesting the mail daemon can get stuck trying to synchronise with Exchange and merely turning off push and letting the 'phone entirely discharge (or doing the hard-reset method of holding down the home and power buttons until the 'phone has powered off and restarted) and then re-enabling push sorts it.

settings>mail,contacts,calendar>fetch new data>advanced

By jove - it did the job. I can only assume the handset burns through the battery by keeping a data connection open continually.


I haven't blogged much recently because I've been on holiday. Pics here

Making iPhone ringtones from MP3s

1. Start iTunes and find the song you want to convert. (It must be an MP3.)
2. Right-click the song and choose Get Info.
3. Click the Options tab.
4. Check the Start Time and Stop Time boxes, then enter times for each (no more than 30 seconds apart, the maximum length for a ringtone). I used 0:00 and 0:30, respectively, as "Spit It Out" has a perfect ascending lead-in.
5. Click OK, then right-click the song again and choose Create AAC Version. You should immediately see a new 30-second version of the song. You need to make your import settings are set for AAC (I normally leave them as MP3).
6. Drag that version out of iTunes and into the folder of your choice.
7. Delete the 30-second version from iTunes and undo the Start Time/Stop Time changes to the original.
8. Open the folder containing the 30-second AAC file you dragged out of iTunes, then change the file extension from .m4a to .m4r. Double-click it and it immediately gets added to iTunes' ringtone library.
9. Finally, sync your iPhone. When it's done, you can head into the settings and select your new ringtone.

Couple of gems; Spanish Flea and Hawaii 5-0

Two location-based things I'd like from my smartphone

1. Location aware mute; I'd love to have a silent setting that unset itself when you moved more that (say) 50m from your current location. I have often put my iPhone into silent mode for a meeting and then noticed later in the day that I've missed several call because I forgot to take it out of that mode. Also - time-based silent-mode. I'd like my 'phone to auto-silence every Sunday 10:30 - 12:30 'cause I'm in church (for example).
2. Location aware alarm; I've often missed a train stop because I've fallen asleep listening to a podcast or music. Why can't I set an alarm that rings in my earbuds when I get within five miles (say) of a location? Also - the ability to announce the time every ten minutes in the earbuds would be great.
   

Bitcoin

Unlike other digital currencies, Bitcoin avoids central authorities and issuers. Bitcoin uses a distributed database spread across nodes of a peer-to-peer network to journal transactions, and uses digital signatures and proof-of-work to provide basic security functions, such as ensuring that bitcoins can be spent only once per owner and only by the person who owns them.

My introduction to Bitcoin was via Steve Gibson's "Security Now" podcast; (check out episode 287 here). Along with the Wikipedia article my interest was peaked. Bitcoin seems to be a real, credible digital currency that is cryptographically secure. However - a currency needs to appeal to more than engineers and the economists are up in arms about it - read the very entertaining exchange here.

For the moment Bitcoins are trading for 1 BTC: 20$

Estimating the time a job takes

For the last eight years I have spent a lot of my time quoting for broadcast installation builds. From small single edit suites to £1.5m broadcast editing centers I think I have a good feel for how long most jobs take. I recently did a search of my arrogated purchase orders folder and discovered I've bought more than 500 equipment cabinets, nearly 1000 video jackfields, and just over 100 kilometers of bulk fibre optic cable - all in the last eight years!
Anyhow - I often get quotes thrown back at me by chief engineers who assume I'm trying to fiddle them and they will typically say;

You've quoted ten days for a wiremen to wire those audio patch panels back to krone blocks; I'm sure my guy could do it in seven.

Even as a chief engineer he's probably only built one or two big machine rooms in his time. He might think he has experience of wiring technical facilities, but I've hired and paid wiremen to do many hundreds of audio panels in recent years. Possibly more than all the panels every chief engineer in Soho will oversee this decade! I know how long things take.

Anyway - for the last five years I've done an audit at the end of each year to compare the number of wiremen and engineer days each job used against what I thought when I quoted. It turns out that I naturally underestimate the time required by around 25% - If I quote 100 wireman days it'll be nearer to 130 when all is said and done. This isn't down to me not knowing how long each part of the job takes - it's a function of wasted time;

• Client won't provide passes to everyone and so guys have to wait idle in rooms or go looking for the pass.
• Deliveries are late, guys are idle
• Customers change their mind, but not enough (or it isn't politic) to warrant a change order
• Faulty parts - we always have to replace those for free!
• Various other things you didn't expect.

So - I've made it my habit to still estimate the number of days as I think it should be and then add on the 25% extra. Consequently my quotes have been getting a lot more accurate over the last few of years. I also discovered the same effect with parts but in the other direction; I always over-estimate the amount of cable/connectors required by around 10%.

So far I haven't done anything about that!

Tektronix WVR5200

During a manic day I was fortunate enough to bump into my old Tektronix mucker Tom Perry who had a new WVR5200 in his rucksack and gave me a quick in-the-street demo! It seems like a much more complete instrument than the current WVR5000 'scope and I can't believe it won't cannibalize their WVR7000 and 8000-series business. The things that stood out for me are:

• Much improved four-tile monitoring with four inputs that can be configured as 4 x SDi (270M, 1.48G, or 3G - yes!) or 2 x dual-link. With four discrete inputs you can display all four at once.
• An SDi o/p that can be any of the inputs OR a test output with bars or pathological signal.
• Audio loudness - and up to 16 channels via embedded groups
• Full Java control (the 5000 lacked this)

So I think this is excellent - the only thing missing is physical layer measurements. Some of the better features are paid-for upgrades (license key) but at £4.5k this represents superb value.
I shall write more when I've had one in to evaluate.

Tricaster tally-light interface for EX3s

One of my favorite parts of the job is prototyping and making little interface units so that equipment from different manufacturers can talk to each other.

This gadget allows the Tricaster TXCD850 studio production system to light the tallies on Sony EX3 cameras; although the Tricaster has "wet"-style GPI outputs it can drive the 12V needed for the lights. So, simple buffer circuit with relays to drive the studio tallies;

There are some photos in a Facebook album - click the title link.

Extending Sony 8-pin camera remotes

How far can the camera control signals from any smaller Sony camera go? If you talk to Mr Sony it's around 50m (the longest cable they sell) - but as the pictures (below) suggest it's at least the length of a box of cat5e cable! The CCA5 cable they sell is north of £500 so I recommend you hot-foot it over to RS, the Hirose ends are part numbers 685-1166 and 685-1163 for the lady and the gentleman and by consulting my scrappy wiring notes (above) you can brew your own for a tiny fraction of the cost. You can also adapt the cable to send down existing structured cable routes (cat5e / cat6 / cat7).

The title-link is to the F23's maintenance manual; that SR field-recorder has every Sony standard interface on it and so you can find the pinouts for whatever you might be using on your EX3 for example.
Rather splendidly the DC supply that runs back from the camera to the remote is the unregulated feed and so even if you loose a few volts down your home-brewed cable the regulator in the RM-B150 won't care; if you want to you can even power that device locally and not worry about volts coming back from the camera.

Old SVHS machines, the half line and archive ingest

As every superhero knows it's the second half of line 23, field one, that active content starts in a PAL signal yet come the start of field 2 line 336 starts as a full line with the corresponding half line at the end of field 2 on line 623. Consequently without a reference signal the only way to tell the difference between field one and two is by the half line at the top of field 1 (well, the broad pulses at line 3 vary but most equipment is field-locked by the time you get to that point in the scan).
We've been testing an ingest/archive solution at the workshop for an African state broadcaster who have a large analogue archive (SVHS and BetaSP). Capturing off the SVHS deck they'd provided for testing (a Panasonic AG-7550) we got some very strange effects. The route is this;

Analogue VT -> AJA FS/1 processor -> SDi into Content Agent uncompressed AVI

this is then compressed to 50 Mbit/s MPEG2 transport stream (I-frame only) and mux'ed into an MXF OP-1A and onto the shared storage. The file is then QC'ed on another machine with a Decklink SDi o/p running OpenCube MXF playback software. Both the input to the CA and the output of the QC are displayed on a JVC DT-V24 video monitor and Tek WVR5000 waveform.

All the clips captured off the SVHS were field-reversed by the time they got to the QC and so we assumed that there was some problem with the capture. After a lot of testing and head-scratching I discovered that the internal TBC on the Panasonic was removing the half-line at the start of field 1 and from then on the capture was marking F1 as F2 and vice-versa.



If you look at the output of the QC machine you can see the half line at the top of the frame, but the motion of the replaced video suggests the fields are reversed. In fact the fields are in the correct order but the capture card has marked them incorrectly and so by the time they are multiplexed and played out they are in the wrong order. The Tek shows the missing half-line at the start of field 1 (on input) and since there is a half line on replay it seems it must be there on field 2; hence the confusion by the capture card.

Turning off the AG-7550's TBC and relying only on the FS/1 showed the half line return to the start of field one and the problem disappeared.

Varying standards...!

Recently we installed five bays worth of equipment at a big facility - of course we did all the usual; Scope of Works, Method Statement, etc and when we'd finished all the usual test results - particularly electrical safety (since we all live in a 17th Edition world now). Now this machine room (one of several) was more than fifty cabinets and so you'd think they've sorted out all of their standards. However - when were handing over we were met with the following criticisms;

• "You've mounted all the storage chassis flush with the front of the bays; our standard is that they are proud of the rack-strip". I went around the all other cabinets and they were an almost 50/50 mix of mounted proud and mounted flush...?
• "You've attached the earthing straps to the front mounting-point of the bay, our standard is to the rear". I went around the all the other cabinets and discovered the only bays in the whole comms room with earthing straps were the ones we'd just installed.
• "Our standard for PDUs is for 10-amp IEC outlets - even when they're feeding C19 (16A input) equipment, you've installed 16A PowerConn outlet PDUs" - we have to certify what we do to appropriate standards (that pesky 17th Edition again!) - no death-leads when you use us...

So I left confused, they'd taken us on to do a job they clearly didn't want/weren't able to do but they had nothing but ridiculous criticism by the end of the job. There doesn't seem to be any camaraderie amongst engineers anymore.

iPhone vs Windows Mobile

I'm very pleased to have recently upgraded to an iPhone 4 handset. It really is much better at running the v.4 software than the iPhone 3 and reminded me of the reasons I was pleased to leave Windows Mobile behind and how some of those things are starting to bedevil Apple.

1. Windows Mobile was effectively unsupported from the moment you owned the handset. In seven years and six handsets I never had a manufacturer-pushed upgrade. On a couple of occasions I managed to find slightly hookey versions (from other networks etc) than upgraded versions slightly but it never worked well. This seems to be playing out again with WM 'phone 7 for mobiles (or whatever they're calling it!); a year out of the gate and there has been one update that bricked a lot of handsets! MS are still asleep at the wheel when it comes to cell 'phones.
   
2. With Windows mobile you have to hit the right hardware/OS version combo; A couple of handsets I owned were early in OS cycles (the first WM6 ones for example) and they weren't man enough for the new version of the OS. A couple (the HTC M1000 and the Vodafone V1615) were right before OS upgrades and worked brilliantly. The iPhone is starting to suffer this; nobody should run V4 on a v3 handset and expect the nice experience they remembered when they first got the iPhone.

It's an interesting time in smartphones; I wonder who will take third place after the iPhone and Android (or the other way around) - Blackberry or Nokia/Windows Mobile?

"Schneier's Law"

Anyone can invent a security system that he himself cannot break. I've said this so often that Cory Doctorow has named it "Schneier's Law": When someone hands you a security system and says, "I believe this is secure," the first thing you have to ask is, "Who the hell are you?" Show me what you've broken to demonstrate that your assertion of the system's security means something.

Bruce Schneier is such an insightful chap - his blog is required reading if you have any interest in security or crypto; and really that should extend to anyone who is involved in networks. The grain of truth I take from this law is that you have to have to certain level of understanding of a subject to recognise your own ignorance. My dad had an expression "...you're not even wrong" - being so far removed from the truth that you're not even on the same field as people who understand the problem (even if they've come to the right or wrong conclusion).

I'm so busy at work at the moment I'm not blogging to much.

Video post production a 'dying industry' - yikes!

This is a table nicked from the Wall Street Journal (via my pal Hugh - @hugh_waters on Twitter). On one hand it's very worrying but on the other hand it's what I've suspected.

1. The equipment required for TV post production is now a £1.5k laptop and not a room that cost a million quid to install (twenty years ago). It's why audio still makes money (you still need an expensive room even thought the equipment is cheap) and why OBs, studios etc will always be profitable.
2. Post production is largely run by owner-operators; folks who have an emotional attachment to it and will do work at a loss for the love of it and have a far too optimistic view of the future.

I don't know what the answer is - I'm going to try and concentrate on designing/building audio suites, TV studios etc and avoid edit rooms!

Side channel attacks with encrypted data

In cryptography, a side channel attack is any attack based on information gained from the physical implementation of a cryptosystem, rather than brute force or theoretical weaknesses in the algorithms. For example, timing information, power consumption, electromagnetic leaks or even sound can provide an extra source of information which can be exploited to break the system.

Several examples that I think are interesting are;

• Secure web applications; Bruce Schneier's excellent blog (which is required reading if you have any interest in security/crypto) describes the attack carried out on the IRS's (what they call the Inland Revenue in the US) online tax form site;

  ...it leaks a fairly accurate estimate of your Adjusted Gross Income (AGI). This happens because the exact set of questions you have to answer, and the exact data tables used in tax preparation, will vary based on your AGI. To give one example, there is a particular interaction relating to a possible student loan interest calculation, that only happens if your AGI is between $115,000 and $145,000 -- so that the presence or absence of the distinctively-sized message exchange relating to that calculation tells an eavesdropper whether your AGI is between $115,000 and $145,000. By assembling a set of clues like this, an eavesdropper can get a good fix on your AGI, plus information about your family status, and so on.

• Compromise of HDCP; The encryption used over HDMI displays is industrial strength and cannot be broken by brute force methods (not in this universe, anyway!) - instead by freezing the memory used by a software BluRay player you can be assured that the volume-key is somewhere in memory. By stepping through 128-bits at a time and having a try at decrypting the first few frames of video (which are very clear when they are decrypted) you quickly find the key for that BluRay or HD-DVD disk.
• The use of 'cribs' when decrypting Enigma traffic; Bletchley Park had typically less than a day to decrypt most traffic captured from German wireless telegraphy as they changed the rotor-positions in the Enigma machines every twenty-four hours. Apparently the intelligence gained by the French who were experts at recognizing the morse-key style of German operators (and hence were able to track which army group Fritz or Herman worked for) along with a knowledge the ten most profane German swear words and ten most common German girl's names meant they code-breakers had a head-start with seed-words which cut down the key-space to a manageable size that was process-able by 1942 mechanical computers!

Interesting though these examples are, the one that really peaked my fancy this week was the side-channel attach described by the Associated for Computing Machinery on the encryption used in VOIP systems. It turns out that most VOIP systems (Skype included) use variable-bitrate compression ahead of the encryption process (typ. AES at 128-bits). It turns out that by training a Markov Model with the encrypted data (yet knowing what the words spoken were) you can subsequently get around 50% accuracy with data streams from unknown talkers. Given that English has a lot of redundancy you could glean most of what was being said!

Read all about it here.

Blackmagic, have the courage of your convictions!

I've often bad-mouthed Blackmagic as they often build to price rather than spec. In the past when I've complained about their interpretation of the SDi spec they've always said that so long as they can light-up a monitor they're 'democratizing digital video' or some such(!) Anyhow - Joel showed me that they are now featuring screen-grabs from a Tek rasteriser on their website; they weren't so keen on it five years ago!

People's expense accounts depend on their unquantifiable skills!

In 1999 the Super Audio CD format was released - higher sampling rate and longer word-length than the venerable 44.1Khz/16-bit Red Book standard that traces it lineage back to the late seventies and the Sony F1 digital audio system.
I've spoken to audio engineers who have made a very good career out of there being a benefit in re-mastering recordings to this newer standard. Their contention is that the difference is "night and day" (please go back and read that post).
Anyway - in 2007 a couple of chaps from the AES did a double-blind test to see if audio professionals could tell the difference - it turns out they can do no better than random. Remember - that was audio engineers, dubbing mixers, and other people who know what to listen for in properly recorded audio. Mix Magazine did a very good write-up under the title of The Emperor's New Sampling Rate!

This all reminded me of a project I was involved in at Oasis TV in the late nineties where we were home-brewing an audio-FX server for the dubbing suites. At the time 9 gigabyte SCSI drives were £1,500 and so compression was implied! None of the dubbing mixers liked this idea and so I made up a CD of various recordings; spot-effects, different music styles, dry vocal recordings and finished mixed programme. The compression we were using was MP2 (so not as good as the now-ubiquitous MP3) at 128, 164, and 192 kBits per sec (as well as uncompressed).
Remember - these were the golden-ears listening on £10k matched amp/speaker combos. It turns out that somewhere between 164 and 192kBits per sec these guys dropped to about 50% accuracy in discerning the compressed audio from the original.
Actually I think it's a bit more complicated than what these two double-blind tests suggest; I store all my music at 192kbit MP3 encoded using LAME 3.9 - for 99% of my music I can't hear the difference. However;

• On some passages (typ. splash cymbals and some acoustic guitar parts) I am aware of compression artefact's.
• An old VT editor once told me (around fifteen years ago) that although he liked the look of (the then new) DV format he felt more tired after a day of editing DV footage compared to BetaSP - the differences aren't immediately clear but over time one is better (in some way?) than the other.

I do believe that you can only get to the truth of these things by statistical analysis - I place no faith in audio professionals who expect their view to be taken seriously without the numbers to back it up. Their salaries depend on them being able to 'hear' the differences - if they are there or not.