How Does a Good Script Become a Bad Film?

Posted: 21st August 2020 by Cheap in Film, Filmmaking

I haven’t read enough film scripts. One I have read is the screenplay for Trumbo, written by John McNamara. I read it before having watched the film, and it moved me: to feel sadness, joy, and pride for the characters, and to shed actual tears. I’m no expect, but I would call that good writing.

Right after reading the script, I watched the film, directed by Jay Roach and starring Bryan Cranston. I was disappointed. The film was nearly lifeless. The moments I expected to be emotionally important were … not. The whole film was just kind of dull, and I would have been upset if I had paid to see it in a theater. (Evidently, not many people did.)

How does that happen? The film was cast with excellent actors who are firmly established as being capable of delivering great performances. So why didn’t they?

I can only assume it was a failure of direction. Jay Roach directed the “Austin Powers” films, which I liked, and the “Meet the Parents/Fockers” films, which I never saw. Those were comedies, and certainly directing a drama is not like directing a comedy. The only other film I’ve heard of is Bombshell, and that only because of its political relevance, and I haven’t seen it.

What else could it be? Editing? Probably not. Hostility on the set?

How does bad directing manifest itself? A director could give no direction and no feedback at all. Is this what would result? Could a director give harmful direction? (“Your performance just gave me chills. You better tone it down.”) Did the director develop a bad relationship with all of the actors?

Ultimately, I will never know, since I wasn’t on the set. The best I can hope for is to learn from bad films I have seen being made.

STIHL FS 131 for Trail Maintenance

Posted: 19th August 2020 by Cheap in Hiking, Trail Maintenance

The trail adopter’s manual for the Ozark Trail recommends the use of a hand weeder, which is a tool I’ve always called a sickle, though it really isn’t. It looks like this:

It works well enough, once you get into the swing of it (pun intended), though there is definitely an upper limit to what it can cut. Unfortunately, it gets to be exhausting, and it’s slow going, especially on the neglected trail I am working on. In nine or ten hours of working with it, I have barely covered two miles. Also, on the rocky trail I’m using it on, the blade is already getting fairly beat up.

The trail adopter’s manual also informs me that optionally, I can use a line trimmer. At home, I use a battery-powered trimmer made by Milwaukee, which (despite running on batteries) is by far the most powerful hand-held line trimmer I have used up to now. However, I hold no illusions that the battery capacity could power through miles of woody brush along the trail, and it is heavy enough that I would hate carrying it that long.

So after a bit of thought and a bit of research, I bought a STIHL FS 131.

After my experience with a STIHL chainsaw, there was no doubt I was going to select one of their line trimmers for trail maintenance. I’m glad I did.

I knew I would want a professional model, and I knew it would need to be more powerful than the bottom of the line. The FS 131 is 47% more powerful than the popular FS 91, but weighs the same and only costs a little more. The “bike handle” and the shoulder harness that goes with it is exactly what I need for extended use.

I plan to use it with the 250mm triple-bladed Brush Knife along the wooded parts of the trail, and the DuroCut trimmer head with heavy line for the grassy and weedy sections.

So far, it is amazing how simply and easily this thing starts. It sounds a little like a Harley when it idles. STIHL has done a good job reducing vibration, so it doesn’t feel as powerful as it really is.

Power: Wow. It spins that thick metal blade up like it’s nothing, which then coasts for an impressive amount of time, even while cutting through things.

The Brush Knife is more than adequate for the young woody growth that comes up on and along the trail. Most of it is approximately a quarter of an inch in diameter. The hand weeder needs a swift swing and be used just so to cut through them, and it barely manages. The trimmer with the Brush Knife slices through them like they aren’t even there. It goes through tough, woody, thumb-sized brush with a single cut every time. With some coaxing, it can saw through larger stuff.

I have not yet had a chance to test out the DuroCut trimmer head, but I expect to this weekend. There is a swath of grass at the trailhead of my adopted segment, and some very tall and tough grass where the trail wanders through a power cut. On other parts of the trail, which I intend to support with this, there are fields much like the power cut, and miles of very weedy bottom land. I plan to try both the serrated .120″ DuroCut line, and Oregon’s .155″ Magnum Gatorline Square line.

The shoulder harness works very well, and after running it for quite some time, I am neither sore nor fatigued.

One challenge has been fitting it in my car. It’s kind of long, and it doesn’t separate in the middle. If I was driving a truck to the trail, it wouldn’t be a problem, but I am driving a 5-door Ford Focus. If I fold down the back seats and remove the trimmer head, it just fits diagonally.

So far, I am very happy with this machine. It is my hope that I will be able to get caught up with the whole seven mile loop in one or two outings. After that, I will take a whack (pun intended) at some of the more egregious parts of the rest of the trail section, which is fairly well neglected. If I and the other people focusing on this trail section can whip it into shape, perhaps usage will increase.

Eccentric Characters, Growth

Posted: 5th August 2020 by Cheap in Fiction, Film, Filmmaking, Television, Writing

I started watching Top of the Lake, mainly because it’s a British police procedural, and because it has Elisabeth Moss in it. I’m only one episode in, but it’s definitely weird. I hope the dominant, entitled drug lord trope turns out to be less cliché than seems so far.

Anyway, there’s a character, GJ, who stands out, and it got me thinking about character development. She is evidently some kind of low key crazy cult leader. She enables (requires) her to act way outside the norm. She can say things that normal people wouldn’t. She hasn’t said much yet, but so far she has directly restated the core of the story’s drama for anyone in the audience who wasn’t following along. It will be interested to see, as the story progresses, how she will be used as a writing device to cut through the subtleties, and probably to create chaos.

The story has plenty to offer for sources of chaos. I suppose that’s the key to a good British police procedural, and otherwise they’d all be the same.

The local drug lord character is another one who can act outside the norms of human behavior (although somewhat within the pattern of a ruthless crime boss). Perhaps this is a key to making an interesting character: not just unimportant idiosyncrasies and eccentricities, but behaviors or capabilities that actually take the story in some direction. Sure, a character trait like Monk‘s OCD adds flavor, but it doesn’t make the story.

On a mostly unrelated note, I’m writing a screenplay, the working title of which is “The Atlantic Beach Quilting Club”. The surprising thing about this script is that I’m not stuck. Not yet, anyway.

Anyway, when I started, I figured it would be a short, maybe a longish short in the neighborhood of twenty minutes. However, it’s starting to get detailed, out of necessity to move the story. If this amount of detail continues, it may approach feature length.

One thing that hasn’t come yet is any kind of character growth for the protagonist. Only the shortest films can get away without character growth. The longer this one becomes, the more of a lacking this will be. Still, I’m not sweating it yet. The opportunity for growth may appear on its own as the story progresses. If not, I can probably go back and change or add to the story in some minor way to introduce the growth component.

No More WWII Films

Posted: 13th June 2020 by Cheap in Film, History

I think I am ready to add another genre of film to stop watching: World War II films.

I am seeing trailers for Tom Hanks’ Greyhound, and while it looks good, my reaction to it is completely ho-hum. That’s because I can’t imagine there being anything about it that I haven’t already seen. I have seen enough battleship movies to last me a lifetime.

I remember thinking the same thing about Fury in 2014, starring Brad Pitt. And I was right. To be clear, I didn’t hate the two hours I spent watching it, but I also could have gone my whole life without seeing it.

I get that WWII was the last “righteous” war (from the Allies’ perspective), and if you’re going to make a war movie that isn’t controversial, that’s the war to do it about. I also get that fascism is a relevant topic again, but a film that portrays battle and ignores politics is not relevant to the topic. The war ended 75 years ago. Everything there is to say on screen about the war has already been said. Almost no one alive today remembers it, so rather than being about our shared remembered history, it is about about legend.

Yes, CGI can portray battleship battles in a way that was not possible before, but if the story is worn out, then who cares?

When I first adopted a trail to maintain, I was given a manual that included a list of recommended tools to bring. Among the recommendations was a bow saw. I was expected to be able to cut branches up to six inches in diameter, and I would need a saw.

I had had plenty of experience with a bow saw when I was younger. I knew them to be, among other things, rather awkward. In recent years around the house, I had been using a pruning saw. So, I elected to buy a long, quality pruning saw with a belt scabbard. I used it with good results.

Recently I came upon a comparison of the benefits of the two types of saw. The bow saw should (theoretically) cut more efficiently, partly because of its longer stroke, but mainly because its thinner blade would cut a narrower kerf. This convinced me to try taking a bow saw for trail maintenance.

I hated it.

I didn’t do any empirical testing, but the bow saw did not seem to cut any faster or better than my pruning saw. (In some ways, it was worse, which I will get to.) In the end, the bow saw didn’t seem to cut more efficiently, and that pretty much closed out any advantages it might have over a pruning saw.

So let’s talk about the advantages that the pruning saw has:

1. The pruning saw was much more portable. It goes in its scabbard, which hangs on my belt. It’s right there when I need it, and it isn’t in the way when I’m not using it. In contrast, the only way to carry the bow saw that isn’t super-awkward is to carry it in your hand, which is fine, except that I need my hands for other things. There was no good way to hook the saw over my shoulder or over my pack that wasn’t in the way and liable to cut me.

2. The pruning saw cuts straighter. The narrow, flexible blade of the bow saw allowed the cuts to wander in directions that I can’t even explain. The comparatively stiffer blade of the pruning saw causes it to cut true and straight.

3. The pruning saw easily cuts from underneath. Even when a log is laying basically on the ground, it was easy to slip the blade underneath and cut upward.

I will go back to taking my pruning saw on trail maintenance outings. By the way, the saw I use is a Silky Sugoi 420.

The Naïve and Sentimental Lover

Posted: 13th May 2020 by Cheap in Fiction

Some of my favorite films are based on novels written by John le Carré. The Russia House, Tinker Tailor Soldier Spy, The Constant Gardener, and even The Tailor of Panama. So naturally I decided to start reading his novels.

I began with The Spy Who Came In from the Cold (I haven’t seen the film), and that was pretty good. After that, I decided to start reading his novels in order of publication. Somehow, I had assumed that all of his novels would be spy thrillers. The first two were really murder mysteries, but I didn’t really catch on, and they were alright. The next two were spy novels, but definitely not thrillers. Instead, they portrayed dysfunctional agencies. All the characters, including the protagonist, were unlikable, and I didn’t much care for them. Nevertheless, I sought to push onward.

Then I reached The Naïve and Sentimental Lover, and boy was that a surprise.

It’s not a spy novel of any kind. It’s not a murder mystery. I don’t know what to say it is, except perhaps literature.

The protagonist (Cassidy) is unlikable, not because he is wealthy, but because he is simultaneously full of himself and spineless and gullible. He doesn’t really seem to want anything, and he lets things just happen to him. He meets Shamus, who easily twists him around his finger. Shamus is unlikable as a character because he has no boundaries. They end up in some kind of bisexual relationship, with Cassidy throwing his money at Shamus. Helen is likable, simply because she is nice and agreeable, but she doesn’t have any will of her own.

Anyway, I struggled to get through the novel, and the only reason I kept going beyond the first couple of chapters is that I was waiting for it to turn into a spy story. It didn’t, obviously.

However, it did eventually demonstrate some redeeming qualities. It was very well written. le Carré’s style and prose were excellent. There were also moments of being very surreal, no longer telling a story but painting a picture of feelings and impressions.

I just wish he would write protagonists I didn’t despise.

Anyway, I’ve decided I no longer want to read everything he has written. Tinker Tailor Soldier Spy is next anyway, but after that, I’m going to skip anything that isn’t certain to be a good spy novel.

Machining, Yet Another Hobby

Posted: 16th April 2020 by Cheap in Metalworking

I have been wanting to do so for years, but I finally bought myself a Grizzly G4015Z combination lathe/mill. I don’t have huge plans to make things with it. Primarily, I intend to use this to learn about machining processes and get some experience doing them. Ultimately, I hope to help my brother open a gunsmithing shop, and part of that goal includes the design and manufacture of firearms.

My father worked in a machine shop during my childhood, and I was occasionally able to get a glimpse of things, but I didn’t know much then, and my attention was primarily focused on the very large parts they machined and the artful variety of metal shavings.

My interest in guns has often led me to wish for niche products that no one makes. That, combined with my brother’s need for a vocation, inspired us to send him to gunsmithing school. If we can figure out the financial and real-estate hurdles, we would like to get him set up with a shop.

In the mean time, there are plenty of things for both of us to learn and experiments to conduct. The G4015Z is a very small machine (weighing only 475 lbs!) and is limited in what it can do. However, the basic functionality is there, and it will give me the platform on which to learn what I want to learn.

Of course, buying the machine is only the beginning of expenditures. There are all kinds of cutting tools and measuring tools I will need to buy. And materials to machine are also surprisingly expensive.

I have acquired a Sound Devices MixPre-10 II, which is a big step up from the Zoom F4 I have been using primarily for nearly three years. This review will largely be written as a comparison between those two devices.

[Be warned that many of the features of the MixPre-10 II are not present in the MixPre-3 II or MixPre-6 II. Neither of those devices have TA3 outputs or 4-pin Hirose power connectors. The MixPre-3 II does not have combo inputs to accept 1/4″ TRS connections, and it just doesn’t have enough inputs to support ambisonic microphones. There may be software/feature limitations in those devices. This review is about the MixPre-10 II.]

32-bit Float

The big game changer is obviously the wide dynamic range and the ability to write to 32-bit float WAV files. This same functionality exists in the Zoom F6, about which I don’t have many specifics, except that it works similarly.

The important factor with 32-bit float files is not the amount of dynamic range, but that they can record values above 0 dB. This means your recordings won’t clip when the signal exceeds 0 dB; the samples will simply be written in the file as positive dB values. In your DAW, they will appear clipped, but as soon as you adjust the gain, a clean, undistorted waveform is recovered.

Sound Devices and Zoom both refer to this feature as making the recorder virtually unclippable. It is still theoretically possible to exceed the recorder’s dynamic range and clip the signal, but in practice, your microphone will reach its max SPL and distort before the recorder clips.

What does this mean? It means a location sound recordist no longer needs to be quite so careful to avoid clipping. It means it is no longer necessary to create safety tracks. It means limiters are irrelevant (and indeed, the MixPre-10 II disables limiter functionality when in 32-bit float mode).

Be warned, however, that this amazing functionality can be defeated by other devices in the signal chain. Specifically, the dynamic range of radios used for wireless mics can still be a limiting factor. I will have to do some testing to see where the ceiling is. The Sennheiser EW 100 radios I use have a lot more dynamic range than I’ve been using, so 32-bit float still increases the headroom. I’m just not sure how much.

32-bit float WAV files are larger and require more storage.

Anyway, if you were wondering: yes, 32-bit float absolutely works as advertised, and no it is not hype.


The one big disadvantage of this recorder versus the Zoom F4 is power consumption. Depending on configuration, this recorder draws two to three times as much current as the F4. This has a worse than expected impact on battery life.

While it comes with a AA battery sled, Sound Devices doesn’t really expect you to use it. The recorder hates alkaline batteries, because they can’t put out enough current. A fresh set lasts less than an hour. It loves expensive, non-rechargeable lithium batteries, but even with those, battery life is a little more than four hours. My F4 liked NiMH batteries the best, but this recorder only lasts about two and a half hours from a full charge.

One has little choice but to invest in a dedicated high-capacity battery system, which is then connected to the recorder’s 4-pin Hirose DC input.

I ended up purchasing an eSMART battery system, which cost me $660 to get started. That’s two Inspired Energy 98 Wh lithium-ion batteries, an Audioroot charger, and an adapter cable. Following the same testing methodology, a full battery lasted just over twelve hours.

I have been experimenting with the recorder, powered from the included AC adapter, through a power meter, and so far I have discovered the following tips for saving power:

  • Turn off unused inputs. Note that disarming the input channel is not enough. Go to the Input setting on the channel menu for each input, where you normally select Mic or Line, and set it to Off.
  • Do not leave a USB flash drive connected. Simply having it connected uses a noteworthy amount of power. Instead, connect it occasionally for backups, and at the end of the job to hand off to the client.
  • Turn down the LCD and LED brightness. The default is full brightness, which is intended to be useful in full sunlight, but there is a power cost. Indoors, having the LCD turned all the way down and the LEDs set to 2 seems to be sufficient, and it helps reduce the power demand.

User Interface

The user interface on the MixPre is much simpler than the one on the F4. This is partly due to the touch display, and also the usefulness of the Wingman app, but also because the MixPre simply offers fewer configuration choices than the F4. In hindsight, the F4 has numerous configuration choices that no sensible person would ever use.

There is only one configuration item I miss. The MixPre records polyphonic WAV files. That’s it. For film, I don’t care about MP3 or any of that crap, but I know that polyphonic files are frequently a bit of a challenge for inexperienced editors. There’s no option to record isolated tracks to their own files.

Anyway, the menu is laid out sensibly, and it’s easy to figure out how to set things the way you want. In fact, it’s so easy that Sound Devices didn’t bother to document much of it in the User Guide. (I own other products from Sound Devices, and they are all documented better than this; I am surprised by this cutting of corners.)

There is one big annoyance regarding the user-interface. Many functions are controlled using a jog wheel, the same jog wheel that controls the headphone volume, which is located on the side of the unit. When used in a bag (which must be how 95% of their customers use it), this is buried inside where it isn’t easily accessible. No one knows what Sound Devices was thinking, but it’s very inconvenient.

The power switch is similarly inaccessible on the other side. This, at least, seems like it could have some rationale behind it, since you wouldn’t want to accidentally turn the unit off.

Wingman App

The MixPre is intended to be used to be used with an iOS/Android app called Wingman, which connects via Bluetooth. You can, of course, use the recorder without the Wingman app, but then you’re missing out on a major source of simplicity, namely the ability to use the keyboard* to enter metadata. I have only used the Android app.

By metadata, I mean scene name, take number, labels for each track, and notes for the recording. You can edit this on the recorder’s screen, using the jog wheel that is impossible to reach, but no one is going to wait all day for you to scroll around the alphabet, picking one letter at a time.

The MixPre allows you to do one simple but profound thing that the F4 does not: edit metadata in a recording after it has been recorded. Not only can you easily fix things like the scene name, but you can also enter notes about the take after it has been recorded. On the F4, you can only enter notes before the take, which means you can’t make any notes about things that happened during the recording. (F8 users are telling me they can edit metadata on existing recordings, so that is apparently a limitation only to the F4.) This simple capability makes a huge difference because it means the sound report can actually be useful.

Speaking of the sound report, it comes as a CVS file, not an XML file, which means it is easily useful to people who aren’t XML coders.

The Wingman app does lack one simple feature that seems like it should have been obvious. You cannot create or rename the project from the app. This must still be done directly on the recorder, using the jog wheel keyboard, which is that much more painful since this is the only thing you need to do that way. I am hoping Sound Devices will fix this in a future update.

I find the Wingman app to be such a central piece of my workflow that I bought a dedicated tablet to run it on. Why a dedicated tablet? It’s so I don’t have to worry about radio interference, right next to my bag, from the much stronger cellular network radio transmitter in a smartphone. One can’t use airplane mode and still use Bluetooth, and there’s no other way to turn off that radio. A non-cellular tablet eliminates that radio interference.

(* The MixPre does theoretically support USB keyboards directly. I should try that out. However, I’m sure you can see how a USB keyboard would be useful on a cart and not so useful in a bag.)


The MixPre-10 doesn’t have fifty kinds of audio inputs and outputs. Instead, it basically has two of each, but the recorder makes good use of those.

There are eight XLR and 1/4″ TRS combo inputs. On the F4, the XLR is mic level and the TRS is line level. On the MixPre, mic/line level is menu selectable, and either type of connector can be used either way. Both XLR and TRS can be balanced inputs.

There is also a 3.5mm (1/8″) TRS stereo “Aux” input, for channels 9 and 10. This input can be set to mic, line, “camera”, and timecode. The mic mode supplies plug-in power, and it’s intended to be used with consumer-type microphones. The camera mode is intended to be used for camera return audio, and I guess it must operate at headphone level. The timecode mode can be used to input linear timecode audio (I don’t know why anyone would want to do that, given the presence of BNC connectors for that purpose).

That’s it for audio inputs. For outputs, there are two TA3 connectors and a 3.5mm X1/X2 connector.

The TA3 connectors output the left and right mix channels by default. I would have preferred XLR outputs, since that’s what the rest of the world uses. Even though I don’t intend to use them often, I felt obligated to buy a pair of TA3 to XLR adapter cables. They output at a line level, but you can adjust the output gain from -40 to +20 dB. (It’s worth noting that the F4’s XLR outputs, when configured for line level, output at a consumer line level, as used by consumer audio appliances. The MixPre outputs at the level used by professional audio equipment, like sound mixers.) The routing is fully configurable and can output any channel, either pre- or post-fader.

The X1/X2 connector is a 3.5mm stereo TRS connector. Like the TA3 outputs, this outputs at a line level, its output gain is adjustable from -40 to +20 dB, and the routing is fully configurable. I tend to use this output for my IFB transmitter. I have been meaning to test whether I can lower the gain and connect it to the mic input on a DSLR.

None of the rest of the connectors are used for audio. There is a Type-A USB connector, which can be used to connect a flash drive for backing up recordings, a keyboard for entering metadata, or a control surface. It can also be used with a hub to combine all three. There is a separate Type-C USB connector which connects the device to a computer and can also supply power (but not enough power to run all features).

There are two BNC connectors, one for timecode in and one for timecode out. There is a 4-pin Hirose connector for supplying power, using either the supplied AC adapter or an external battery pack or power distribution system. The back has a battery sled attachment, for which Sound Devices sells a few options. There is one SD card slot. There is a micro-HDMI input, which is yet another way to input timecode. The headphone jack is 3.5mm rather than 1/4″.

Overall Impressions

I love this recorder.

For one thing, it just sounds better. I realize this is a totally subjective statement, and I am unable to articulate why or even how it sounds better, but it definitely does. I’m not talking about noise. Every input seems hotter. Not louder, but more dynamic range? There is almost certainly a quantifiable difference in the way the inputs do A/D conversion, but I don’t have the tools or the knowledge to measure or demonstrate it. All I know is that the recordings sound noticeably better.

The 32-bit float recording is huge. It means (almost) never again having to tell the director that I clipped the take he loves, and it has to be re-shot. Heck, it means I barely have to look at the meter.

Going from 4 to 8 inputs won’t really make a big difference for me. However, never having to create a safety track is effectively a big increase in inputs, because on the F4 and F8, each safety track consumes a second input.

Buying the battery system was a drag, but now that I have it, power is one more thing I don’t really need to worry about.

Being able to produce meaningful sound reports is really nice, and it’s one more degree of professionalism I can offer.

The better user-interface is a somewhat intangible improvement, but I know (from screwing around with my old H4n) that if I ever go back to look at the F4, I’ll be disgusted with its clunky and complicated menu.

The bottom line is that I am very happy I made this step upward.

Using Mystery to Create Suspense

Posted: 27th February 2020 by Cheap in Uncategorized

As a reader, suspense is important for me. It is the most important factor that determines whether I will enjoy a novel or not, even more important than the characters. More than that, it determines how engaged and interested I will be, and even how quickly I can get through a book.

One way to create suspense is through mystery. In a murder mystery novel, mystery is at the center of the story. However, that’s not the only way that mystery can play a role.

Creating obstacles for the protagonist raises the stakes. However, a situation with no apparent solution creates a mystery for the reader. How is the protagonist going to prevail? Will the protagonist prevail? The more impossible and hopeless the situation seems to be, and the longer the impossible odds persist, the more suspense is created for the reader, who genuinely wonders how the conflict will be solved.

The opposite is also true. If the solution is obvious, the story is a disappointment. In The Dragonriders of Pern, Anne McCaffrey hinted early and often that dragons can probably travel through time. It was an obvious solution, but the characters agonized over what to do, until the last minute when they finally thought of it. This was very disappointing and frustrating.

Cop-outs are also disappointing. The first Dean Koontz novel I ever read was Sole Survivor. The protagonist kept looking for an explanation about how this person could have survived the plane crash, and I certainly couldn’t think of one. I was eager to reach the end and see what brilliant solution the author had devised. When, in the end, it turned out to just be magic, I was disappointed.

The solution doesn’t need to be terribly imaginative. In Tatiana, by Martin Cruz Smith, the answer to the mystery is actually quite simple. However, the author is careful to keep us thinking about other things. There were some facts presented right from the beginning that make more sense in the context of the answer, but we assume a different explanation (incompetence on the part of Smith’s corrupt Moscow police organization).

Zoom F4 Battery Life

Posted: 6th January 2020 by Cheap in Filmmaking, Technology
Tags: ,

In early 2017, I purchased a Zoom F4 digital recorder. Though I had been recording location sound for film for years, this is the first device I have owned that was designed for filmmaking. Prior to that, I had been using a Roland R-26, which was a step up from my Zoom H4n, which was designed for musicians.

Here is some important advice for anyone with new gear that runs on batteries that will be used on film sets or at events: understand the battery life. There is nothing worse than running out of battery in the middle of a production and not having sufficient replacements. How many spare batteries should you have? That depends on how long a good set of batteries lasts. You should find out from reliable sources, or you should perform your own tests.

(The manufacturer is not a reliable source. Case in point, Zoom suggests that the F4 will run for 3.5+ hours on NiMH batteries, but my own testing ran the device for over 10 hours.)

Eneloop Pro

Eneloop Pro NiMH batteries have a stated capacity of 2550 mAh.

For my battery tests, I set up the recorder to record on four inputs, supplying phantom power on each to my most power hungry microphones, and plugged in a set of headphones. I informed the recorder that it had NiMH batteries. (As far as I can tell, this only affects the battery display.) The XLR outputs were turned off (which, you will see below, is relevant). I set up a video camera to watch the recorder. With a freshly charged set of batteries, I turned on the recorder, started it recording, and then walked away.

After 2 hours 13 minutes, the battery status dropped down to 3 bars. Remaining time, 78%.

After 4 hours 28 minutes, the battery status dropped down to 2 bars. Remaining time, 56%.

After 7 hours 57 minutes, the battery status dropped down to 1 bar. Remaining time, 23%.

After 9 hours 50 minutes, the battery status dropped down to 0 bars, and a window began popping up every 30 seconds warning, “Low Battery!” Remaining time, 3%.

After 10 hours 9 minutes, the device shut down. The end of the file was properly saved.

Other Battery Types

I conducted tests using the same test scenario with non-rechargeable alkaline and lithium batteries. Unfortunately, I deleted the data once I found they weren’t useful. All I can really say for sure is that lithium outlasted alkaline, and NiMh outlasted both.

Configuration Matters

I was on a film set for which the Director requested IFB. I didn’t own a legitimate IFB radio set, but I hooked up a pair of Sennheiser EW 100 radios fed from one of the XLR outputs on the F4. As the day wore on, I was shocked at how quickly the batteries in the F4 ran down. I had a second set charged and ready, but I wasn’t even halfway through the day before I had to switch out to them.

The next time I used the recorder, the same thing happened. Then I realized that the XLR outputs were still turned on, and I disabled them. That seemed to resolve the battery life issue. My conclusion was that enabling the XLR outputs had a bigger impact on power consumption than anything else.

A few months ago, I learned a new trick. I have an AC power meter which measures power consumption in watts (and amps, but that isn’t as useful). If I have an AC adapter for a device, I can run it through the meter, and then I can experiment with different settings to see their effect on power consumption.

About the lowest I can run the recorder is at 1.7 watts. This is about 177 mA at 9.6 volts (the nominal output of 8 NiMH batteries). That is with one input enabled, and no phantom power.

Enabling more inputs matters. With all four inputs enabled, the power consumption goes up to 1.9 watts.

Phantom power matters. Enabling phantom power on the four ports brings the power consumption up to 2.3 watts.

Outputs matter. Turning on either the XLR or the 3.5mm output brings it up to 2.7 watts. Turning on both brings it up to 2.8 watts.

Microphones matter. Plugging in four (AKG P170) condenser microphones brought it up to 3.5 watts.

There may be other things that affect power consumption that I’m not aware of. The manufacturer specifies the power consumption at 12 watts, which must be a maximum, but I don’t know how they achieve it.

Things that did not make a measurable change in power consumption: recording, connecting headphones, the number of SD cards.

I could measure power consumption more precisely on the DC side using my Fluke meter, but I would have to rig up a way to tap into it. I would need to make a short 4-pin Hirose extension cable that routes one part of the power circuit through the meter. I may do that, because I have a Sound Devices MixPre-10 II on the way, and I will want to investigate these same questions for that device.


Use NiMH batteries. (Or use an external battery pack through the 4-pin Hirose connector.)

Disable the outputs (XLR “main” and 3.5mm “sub”) when you aren’t using them.

Disable phantom power where it is not needed (on radio receivers, for example).