Sound Devices PIX 240 ProRes/DNxHD Video Recorder with EPIC

[Matt Mayer]

And on an unrelated note:

For PIX ops looking for a simple way to MANUALLY trigger the PIX, I have discovered that chinese-knock-off Canon DLSR REMOTE SHUTTER RELEASE devices will control RECORD on-off on the PIX! (I am sure the Canon OE remotes will too, but at considerably more than the $5+ cost of the chinese ones). Zillions on ebay under "Canon remote shutter release".

It plugs into the LANC port (ie it is a WIRED remote) on the Pix and set the PIX START/STOP REC menu to Switch Closure.

Works like a charm and allows you to put the remote on, say, the tripod handle,etc and fire the PIX as needed. Nice if you just want to record JUST with the PIX and not have to be slaved to the Scarlet recorder/SSD. I would suggest you reprogram the RED fan to, say, 50% so the "idle" speed of the fan doesn't interfere with your audio. I just created a Preset that I can select that includes this fan setting when shooting with the PIX-only in this mode.

If I may suggest to Sound Devices, such a simple remote, only with a LITTLE wee button/controller-on-a-wire would be a great feature we could cable over to a camera-handle or some such trigger point on our rigs! Would happily pay the non-chinese price for this as so far the Chinese units are a bit bulky.


Stu Aull
Alaska

We had actually hoped that one of our vendors would start making them, but that hasn't materialized yet.

If one has even the slightest bit of talent at soldering, these are exceedingly easy to make. It's a 3/32" TRS jack. Tip is the switch closure contact, the Ring is an LED driver (very handy because you now have a Tally light that confirms record), and the sleeve is ground.

I bought about $5 in Radio Shack Parts, added some cable and heat shrink I had lying around and made my own in about 20 minutes. It isn't pretty, but it works well. IIRC, this is the switch I used:

http://www.radioshack.com/product/index.jsp?productId=3159591

You should be able to find something similar virtually anywhere in the world.

 

[Matt Mayer]

Just an FYI here:

Have just checked the SD PIX user forum and a suggestion made there by Tech Support is to use the REDs TC as a trigger, rather than the SDI-Flag I have been using (these are both trigger menu options in the PIX). This is detailed on PAGE 30 of the PIX USER GUIDE pdf v1.05 "External Control".

I am assuming that TC is transmitted also via the LEMO connector, and I don't have to run yet another cable for TC betw the 2 units...?

Haven't tried this yet but will post results...



Stu Aull
Alaska

Timecode is embedded in the SDI stream as well. The easiest method to use it would be to select EXT TC (SDI/HDMI) in the Timecode/Sync menu. Remember it requires that you are using Record Run timecode.

 

[stu aull]

Thanks - as always - for the info and clarifications Matt!

Stu Aull
SX 199
PIX 240

 

[Patrick Tresch]

Timecode is embedded in the SDI stream as well. The easiest method to use it would be to select EXT TC (SDI/HDMI) in the Timecode/Sync menu. Remember it requires that you are using Record Run timecode.

But will it be enough to have the PIX locked synch and to be accurate up to 0 frames unsynch? My understanding is that TC is not a synch method but just a sound info and this could lead to different TC between the recorders (Epic TC and Pix TC).

Pat

 

[Simon N]

But will it be enough to have the PIX locked synch and to be accurate up to 0 frames unsynch? My understanding is that TC is not a synch method but just a sound info and this could lead to different TC between the recorders (Epic TC and Pix TC).

Pat

This is our issue. We are not getting consistent results between Epic timecode & Pix timecode. In general using Red stop/start flags and sdi timecode into the pix, 80% of the clips match timecode & frame, however the other 20% match timecode but have a 1 frame difference in picture, even though timecode is the same.

I made a video to show my workflow and the resultant inconsistencies, if you are interested in seeing it you can have a look

http://www.youtube.com/watch?v=fygyxoy1DHY

I have altered a couple of minor settings since (sync out), which all appear to give the same result. However, the biggest change is to remove the sync cable which connected has caused a world of pain.

 

[Douglas Williams]

This is our issue. We are not getting consistent results between Epic timecode & Pix timecode. In general using Red stop/start flags and sdi timecode into the pix, 80% of the clips match timecode & frame, however the other 20% match timecode but have a 1 frame difference in picture, even though timecode is the same.

I made a video to show my workflow and the resultant inconsistencies, if you are interested in seeing it you can have a look

http://www.youtube.com/watch?v=fygyxoy1DHY

I have altered a couple of minor settings since (sync out), which all appear to give the same result. However, the biggest change is to remove the sync cable which connected has caused a world of pain.

Hi Simon

Many Thanks for taking the time create and upload this video - it has been extremely insightful.
I have read through all 39 pages and have seen the progress SD has made in its developments.

I would have thought it would have been safe to summarize a successful workflow between the Pix and Epic was to

1) Update to latest firmware on PIX
2) Boot Epic camera first before attaching PIX Device
3) Purchase 3BNC-TO-00 SYNC CABLE http://www.red.com/store/cables/prod...emo-sync-cable
4) Refrain from zooming in ( focus checking) on Epic whilst recording to the PIX as this can cause corrupted files.

Please correct me If I am wrong...
But from the general above steps, many users seem to have a "rock solid" system.

Just wondering how/why Simon is still having problems ?

And also, are users actually replacing the "offline" files from the PIX with the R3D files in their rock solid workflow.

Thanks.

 

[Simon N]

Our issue is the inconsistency, so based on 80% of the clips being correct, we are doing a power down of the camera, power up & re attachment of timecode every five or so slates which has pushed our success rate of clips matching up to about 92%.

We would obviously like to get to 100% without having to think about it, however we are getting closer.

 

[Douglas Williams]

Hi Simon

Please keep us informed of your progress.

I still don't get it, why does this work for some and not for others all things being equal ie, firmware, cables etc...

 

[Simon N]

We are changing our protocol every day or so trying to fine tune our workflow to get 100%. Our best results have been from disconnecting timecode, powering down the rebooting and reattaching timecode into the camera, every 4-5 slates. On B camera we have achieved 100% success for 2 days (whoopee!) and with A camera 2-3 clips per day are out.

One thing we hadn't done much of was to power down the pix on every camera battery change, so today I had the pix shut down every time we powered down the camera and restarted after the epic rebooted. Not exactly the same but it harkened back to the Red One days of a very long boot when you have to co-ordinate the powering off & switching on of 2 pieces of kit, especially when the first AD is shouting that we are running out of time. The result were less than impressive, out of 13 slates before we changed rolls, 10 were 1 frame out and 3 were in sync.

Matt Meyer posted on the thread that I started in the Epic forum that this issue was driving them nuts and I can see why - we now have 2 cameras that when using 2 identical peices of kit produce different results, one camera is now "rock solid" (2 days @100%) though my definition of that would be from now til we wrap (11 weeks) and another that is 90% of the time is 100% accurate ;-)

 

[Douglas Williams]

Hi Simon

Great admiration for soldiering on dude - seriously...
It seems your the only person on planet earth who has this problem, all those who have contributed dearly to the writing of these 39 pages - are flying sky high, so therefore need not write anymore including Sound Device themselves, wouldn't it be simpler for them to possibly send a new unit as this seems to be such a one off case ?

If you don't mind me asking, what is your role on the project ?

Anyway - hope you get your workflow sorted Trooper.

 

[Simon N]

We have 2 units both exhibiting the same issue. As per previous ,we have managed to stabilise one by a particular workflow, the other is still erroneous 10% of the time.

We have now got a recognised sync issue with both our cameras.

I am the DoP (and have been for 23 years )

 

[stu aull]

v3.2.8 in my scarlet - full frame recording and "EXT REC" features are making me and my v2.0 PIX240 _very_ happy!!

Stu Aull
SX 199
Alaska

 

[N_Villers]

The Pix would be perfect if it showed thumbnails of recorded media for playback.

 

[Nick Morrison]

v3.2.8 in my scarlet - full frame recording and "EXT REC" features are making me and my v2.0 PIX240 _very_ happy!!

Stu Aull
SX 199
Alaska

Hey Stu, so the clean record is working perfectly? Great to hear. We are seriously considering this device. How about slo-mo? Can it record 48fps, 60fps, 120fps, etc? How does that work?

Also, how about the audio? Are your sound engineers plugging directly into the Pix to give you clean sound....so your Proxies get the perfect audio? Would love to hear how this is working too, thx.

I think the Pix 240 can be a great alternative to the I/O module if all you are looking for is perfect sound.

 

[Shane Kelly]

Hey Stu, so the clean record is working perfectly? Great to hear. We are seriously considering this device. How about slo-mo? Can it record 48fps, 60fps, 120fps, etc? How does that work?

Also, how about the audio? Are your sound engineers plugging directly into the Pix to give you clean sound....so your Proxies get the perfect audio? Would love to hear how this is working too, thx.

I think the Pix 240 can be a great alternative to the I/O module if all you are looking for is perfect sound.

It will only record slomo if you play it back through the cameraand hit record on the pix. It will otherwise record the 24fps stream from the sdi. I tend to plug in to the pix with audio. Full size xlrs and line level. Great audio quality.

 

[HD Hildebrand]

It will only record slomo if you play it back through the cameraand hit record on the pix. It will otherwise record the 24fps stream from the sdi. I tend to plug in to the pix with audio. Full size xlrs and line level. Great audio quality.

So, in essence, you could record sync sound on the Pix 240 when shooting slow-mo (which you can't on the Epic or Red One when shooting slow-mo), then playback same clip from SSD to Pix240, hit record on 240 and you have one take with 24p sync sound and one slow-mo take. Correct?

 

[Shane Kelly]

So, in essence, you could record sync sound on the Pix 240 when shooting slow-mo (which you can't on the Epic or Red One when shooting slow-mo), then playback same clip from SSD to Pix240, hit record on 240 and you have one take with 24p sync sound and one slow-mo take. Correct?

That's correct, sound would be sync at 24 fps, or whatever your time base is. Could be useful for ramping. Shutter setting would need to be relative and you would have to use the sped up slowmo take in your final edit but for offline that would not be an issue. Unless you like the high shutter speed look, of course. Or you use an absolute shutter speed but personally, I like relative for slomo.

 

[stu aull]

Also, how about the audio? Are your sound engineers plugging directly into the Pix to give you clean sound....so your Proxies get the perfect audio? Would love to hear how this is working too, thx.

I think the Pix 240 can be a great alternative to the I/O module if all you are looking for is perfect sound.

Sound Devices are the Go-To-Guys for pro audio, which I suspect you already knew
so -yeah! excellent audio - really, the PIX is a fab digital audio recorder that also happens to record great video too.
I am sure the RED IO mod is fab, but the PIX price-point is pretty hard to beat....esp if you just need HD, but even as a proxy-recorder...
Stu Aull
SX199
Alaska

 

[Nate Clapp]

just an fyi before you feed audio directly into a PIX rather than the camera (any camera in this case, not just RED cameras) you will need to determine the audio offset. Usually the SDI video stream is slightly delayed compared to the audio going into pix directly. Also in many cameras, the audio coming out of the camera's SDI stream is offset (earlier) than the video stream. Sony F3 and Epic (but on previous build, each build needs to be tested to be sure) both do this, as well as many others. My testing with a now obsolete Build 3.0.0 was 80ms if audio is coming off SDI from camera (Epic) or 150ms if you feed audio directly to pix. also I set TC offset +1 so TC on pix and epic are frame accurate. good news is this can be saved as a setup file and loaded into pix when you use that camera, and everything will be exactly in sync. every new firmware suggests you should check whether everything is still valid.

 

[Nick Morrison]

just an fyi before you feed audio directly into a PIX rather than the camera (any camera in this case, not just RED cameras) you will need to determine the audio offset. Usually the SDI video stream is slightly delayed compared to the audio going into pix directly. Also in many cameras, the audio coming out of the camera's SDI stream is offset (earlier) than the video stream. Sony F3 and Epic (but on previous build, each build needs to be tested to be sure) both do this, as well as many others. My testing with a now obsolete Build 3.0.0 was 80ms if audio is coming off SDI from camera (Epic) or 150ms if you feed audio directly to pix. also I set TC offset +1 so TC on pix and epic are frame accurate. good news is this can be saved as a setup file and loaded into pix when you use that camera, and everything will be exactly in sync. every new firmware suggests you should check whether everything is still valid.

WOW. Super informative. THX