All the A4s have a distinctive chime whistle. Recorded on 27th November 2010 using a handheld Olympus LS-5. A windy and cold day.

More information about this engine from The Sir Nigel Gresley Locomotive Preservation Trust.

The only comparable unit I have is a Zoom H2 so I’ll use that as a baseline.

Compared to the H2, the LS-5 feels better to use, is less bulky and overall has the impression of being better quality all round.

In the box

In the LS-5 box, the unit, a set of batteries, foam windshields (which nicely click on the mic caps), a strap, a USB cable, a copy of the Olympus file manager ‘Sonority’ and printed manuals. The packaging is good and seems like Olympus are copying Apple to some extent. I did notice that the manuals are heavier than the unit.

In the H2 box (as far as I remember), the unit, a set of batteries, a stand, a mic holder adapter, a mains adapter, some printed manuals and a USB cable.

Start up time

When trying to capture a sound and the unit is off, startup time is of the utmost importance. The LS-5 takes about 3 seconds to start from cold, whereas the H2 takes around 15 seconds with an 8Gb SDHC card. The start up time on the H2 seems to depend on the SD card capacity – the higher the capacity, the longer the start up. One feature the LS-5 does not have that the H2 does, is a pre-record buffer.

Pressing the record button

Both units start promptly but there is a very slight delay on the H2. The LS-5 starts immediately as far as I can tell. The H2 has membrane buttons which are almost silent when used. The LS-5 has mechanical buttons which you can hear has a click at the start of each recording. The ending click on the LS-5.

Pressing the stop button

Stopping a recording is often overlooked. The quicker you can stop, the quicker you can restart. The LS-5 has almost no delay between stopping and being ready for restarting. The H2 does pause for a second or two after stopping and finishing writing to the disk.

Holding the unit

Again, often overlooked is how to hold the recorder. With the H2 mic arrangement I was never sure how to point the mics for best capture and it was certainly more convenient to hold it horizontally that vertically (which was probably the best orientation). Handling noise is about the same for both units.

Sound quality

Without the benefit of a scientific approach the LS-5 sounds ‘better’ and more ’rounded’. The mics seem less sensitive than the H2 (my H2 was used mostly on the mid gain setting at 100 recording units). Best overall setting on the LS-5 seems to be low mic sensitivity with recording gain at about 9.

Other features

The H2 has a 4 channel setting which the LS-5 does not. I’ve tried to use this in the past but with not much real success – post production being the main issue.
Both units have playback effects, not much use I would say.
The LS-5 has 2Gb of internal recording capacity and ‘zoom’ mic settings, although the zoom settings are not compatible with higher bitrate and resolutions.
Both units can record at a variety of resolutions and bitrates, and also MP3 (the LS-5 has the addition of WMA compression too).

Price

2 years ago, the H2 was bought at £150, the Ls-5 at £179, both from high street stores.

The LS-11 / 10 is 30% more expensive than the LS-5. As far as I can tell, you get the same unit but with less internal memory (that the LS-11), no DAW software, no case and no remote control. Apart from the remote control, the rest of these items is of negligible use.

LS-5 Sound recording samples

First sample is a train horn. Recorded using inbuilt mics, on low sensitivity with gain set at about 8/10.

Second sample is of 60163 Tornado steam locomotive. Recording settings as above. Train passed within about 30 feet of the recorder.

Both samples are untreated.

H2 Sound recording sample

For comparison, this recording is of another steam loco, A4 ‘Bittern’, using the Zoom H2 from a similar position to the recordings above. The H2 was set to Medium mic sensitivity with gain set to 100.

The announcement can be made for a number of reasons, and is used to alert transport staff to an incident without causing panic in what are usually, crowded and congested areas – the last place you need a stampede.

Wikipedia has an interesting article on the subject here.

Now one has changed.

The output of one particular station that broadcasts from near Povarovo, Russia (callsign UVB-76, AKA “The Buzzer” by its listeners ), has increased dramatically. It’s only deviated from it’s signal three times previously – in 1997, 2002 and 2006.

Wired reports “In early August, a garbled recording of a voice speaking Russian was heard by listeners. A few days later, on 23 August at 13:35UTC, a clearer voice read out the following message twice: “UVB-76, UVB-76 — 93 882 naimina 74 14 35 74 — 9 3 8 8 2 nikolai, anna, ivan, michail, ivan, nikolai, anna, 7, 4, 1, 4, 3, 5, 7, 4″, before returning to its normal broadcasting.”

Obviously this is a brilliant reason for speculation by conspiracy theorists. If you want to get some of that, more from Wired here.

Go to the iPad Audio Forum

I’ve been looking at this for a while to go alongside the other sound creations apps I currently use on the iPad.

Mixtikl iPad interface

It does take a bit of getting used to; I can’t say that the interface is like anything else I have used. I think that it’s simplicity is not reflected in the interface which is at some moments, counter intuitive. The interface works OK on iPad but it looks as if it has been ported from an iPhone app without doing anymore than stretching the UI and moving a couple of icons.

The default setting of 60bpm gives a clue to where the majority of the sound works; slow ambient, atmospheric, sparse piano – that sort of thing. With some experimentation, increasing the bpm provides mixed results – the ‘tribal’ pack provided with the app ‘scales’ well. Some of the provided sounds sound to fit their preferred bpm and for some of the sample based sounds, sample start and end clicking is apparant.

Audio effects

Good control over effects are provided but as you cannot skip sections or sound passages meaning you can’t get a good overall view of what is needed to account for all passages which makes a good effects mix difficult to achieve. A nice feature (once you have worked out it is there and how to use it) is to be able to chain a series of effects together with some rudimentary control elements for triggering or controlling effect parameters. Careful with the control blocks, these can have very unpredictable and loud results. Export to a DAW if you are fussy about that.

Online documentation comprises a couple of videos hosted on YouTube that appear to use a different version to that used on the iPad.

Once you get used to the interface, the results are good although as the sound is generative it can sometimes seem like quite some time (many seconds thinking “is it working?”) before anything happens. This also has an impact in that if you export a mix and edit in a DAW, there can be a delay between the DAW timeline start and the start of any sound in your sample.

What does it sound like?

The first sound sample is from the built in sampler that uses the iPad microphone. As the microphone doesnt have any controls for recording levels, this is probably much the same as any other app that uses it.

Audio sample one – from a selection of the Mixtikl ‘packs’

Audio sample two – a bit more ambient. Mixtikl has some interesting generative piano sounds.

Recording your efforts can be a bit hit and miss. When you record, you specify the amount of bars that the app will record, not based on a time period, then you must wait in realtime while the sound is rendered.

Once you work out how to access the export utilites (via ‘Share’ not as would be intuitive, the ‘Export’ dialogues), exporting the sound is via an in built web server (many apps provide this as a way of getting round the iPad’s locked in, no file system concept) which is basic and is prone to crashing (around 25% of the time for me). When the app crashes it does not recover your last workflow point, so frequent saving suggested.

The final results

I took a quick render from Mixtikl and uploaded into iSequence and then rendered that back out. This is the result.

Here are a couple of demos so far. Things are a little harsh with iSequence and there’s no eq to use so too much playing with the effect filtering can produce harsh results.

First one is the raw output from iSequence (LAME Mp3, VBR7)

Second has a touch of compression and some eq to try to deal with the harsh mids (or could be my new monitor headphones not yet broken in).

Not the best but still getting to grips with the workings. I used a couple of field recordings from a train journey recorded directly onto the iPad while creating the track.

If you too are getting distorted recordings I’ve found a workaround.

• Power off the iPad (hold down the power button and swipe off)
• Restart the iPad
• Start iSequence and record sound – result is distorted
• Shut down iSequence
• Start iSequence and record sound – sound is good to go! (within the limits of the hardware)

The iPad isn’t very configurable for recording yet and I don’t see any devices on the horizon to support plug in microphones however for rough and ready instamatic sound snapshots in the right environments, it has interesting possibilities.

iSequence for iPad Record

UPDATE – 5/8/10

A new update to this app is now available with a fix to the recording function, so I’m pleased to say the developers were listening!