To P2, or not to P2? That is the question...

Redundant recording to SDHC has yes, greatly closed the gap in reliability.

I sincerely hope that's the case... but I am not yet convinced. I've heard no actual reports of how the cameras handle card failures... and it's hard to test! I tested, e.g., my 744t by yanking out the CF card in the middle of a recording, and indeed the HD recording was fine. The 130/160 stop recording when you open the card door, so you can't do a test like that... unless anyone wants to mod their cam to, say, just have a switch one can throw to indicate card door status... anyone?
 
Is there a reason Atomos hasn't come up yet? If we're talking about cameras in this price range, I think it's worth at least brushing on. Atomos produces attachments that let you use whatever 2.5" hard disk you please, and record the video from the uncompressed HD-SDI or HDMI stream (different models depending on your camera's output) directly into the highest quality native codec specifically for your editing platform (it pushes the heck out of ProRes, but that isn't the only option available).

Codecs don't have to be irrevocably tied to your capture medium, though; there's a whole world of cams out there. I do agree that I wouldn't touch AVCHD with a ten-foot pole if I could avoid it; of all the popular options, it's the lowest quality, and the most difficult to work with.
 
Is there a reason Atomos hasn't come up yet? If we're talking about cameras in this price range, I think it's worth at least brushing on...
... I do agree that I wouldn't touch AVCHD with a ten-foot pole if I could avoid it; of all the popular options, it's the lowest quality, and the most difficult to work with.

I've been considering an atomos ext recorder in order to manage 4:2:2 and help even the odds a bit more. But they look to be extra hassle.And extra expense, bringing the price closer together for a p2 camera vs avccam. Since you brought it up, how would an external recorder workflow compare to what we are talking about? Just copy the footage from the recorder to your NLE, right?
It seems working with p2 not having to even be copied to the NLE is a time and disk saver.
I'm curious, why is avchd harder to work with as u said? Because its much more compressed?
Isn't it as easy as copying it to your NLE, opening premiere, importing the footage and away you go?

Is the reason the hpx250 can't do 1080 60p a camera limitation then? It sounds like it's not a p2 limitation.
 
Here are the AVCHD "harder" issues... i don't find the AVCHD workflow harder myself, but others definitely do. It has to do with your personal workflow.

1. AVCHD is a highly compressed interframe codec... the computer must work much harder to decode, edit, and re-encode it. Each "frame" is computed based on differences and similarities to 17 or so other frames (number may be a bit off, but that's the idea of "long GOP")... so each time a frame is displayed, it has to reference these other frames and calculate the differences and similarities. When you generate a new frame, you have to regenerate all the other frames in the GOP that might reference it. So... much harder work for the processor. An intraframe codec invovles single, self-contained frames with no dependencies on the frames around them, so are much easier/faster to work with.

2. AVCHD can be edited natively, or transcoded to an intraframe codec. For some people, this is an extra step that gets in the way, and it also explodes the footage to a much larger data size. Some people seem to have a good experience working with untranscoded AVCHD. These seem primarily to be PC users, and some Premiere users on the Mac... I shoot pretty much only longform, and have not had a magical experience with native AVCHD editing in Premiere. For some reason, Premiere spends a lot of time processing or analyzing long imported AVCHD clips anyway. I can put them in the timeline and "start editing immediately", but I get no audio for a few minutes, and very jerky playback for a while. So, in my audio-oriented work, I can't start editing effectively for a while anyway. This is when I attempt to edit directly from the cards, btw. My time to transcode to ProRes in FCP7 is about the same as the time to transfer data OFF the SD card... so for me, it's the better workflow... the same time I would spend transferring data off the card onto the computer, or waiting for it to be analyzed by Premiere, I spend doing the transcode, which gives me a better subsequent editing experience. P2 on the Mac has the same issues... transfer off the cards or log'n'transferring from the cards takes the same time, so I don't find the process to be at all onerous.

3. P2 card reading is either more cumbersome or vastly more expensive than SD card reading. I have literally 10 or so SD card readers around, one in every camera bag, etc. I can count on other people having the ability to read SD cards. P2 cards either need pricey panasonic readers, the camera as a reader, or an old PC with Cardbus slots. There is the cheaper one-slot Panny reader... but at $300 for one card, it's still far more expensive; and I tend to needmore than one card mounted at a time for P2 codecs. So, in that way AVCHD is easier.

4. AVCHD (SD) media is so cheap that it's basically unlimited, and this has often been a great enhancer of my workflow. I've never had a situation where I needed to offload AVCHD data in the field. I don't have enough P2 to make that claim (and not enough instances of this to make buying more P2 worthwhile). You can get more SD cards in countless consumer storefronts; P2 in an emergency is very hard to find. So I call AVCHD easier in this regard, as well.

5. Editing from P2 cards is a nice idea, and again is great for many workflows, but to do so effectively for long-form requires an expensive multislot reader. Defineitely a plus if your workflow and budget allow it.

I'm outta steam... there was more I was goiung to say, but forgot...
 
Ullanta, thank you for the excellent explanation.
And thanks to everyone else too. I'm learning a lot reading what you guys have to say.
 
(mind you, don't disregard my earlier posts… I don't find the P2 workflow "easier" but I do prefer P2 for reasons discussed earlier in the thread… this last thread purely addresses only the workflow "easiness" in my world.)
 
So on long recording, say 1 hour of continuous recording, it sounds like the processing/analyzing of AVCHD could become an issue. Compound that in a multicam long event with 3 or 4 long files and then?
Also sounds like the workflow of p2 is only significantly easier if you either can afford multi card readers or can find and use some old computers to use as p2 readers.

So copying footage to the NLE to edit would likely edit faster with p2 footage due to less processing of the less compressed footage, but since the files are bigger to start with, it would take more time to copy to the NLE in the first place.


I'm re-reading everything in this thread mutiple times. Good info here. I think many people besides myself will benefit.
 
1080/60p is where the TV manufacturers would like to see things head. I have no idea if it will become a broadcast standard, I mean, it hasn't been all that long ago that we swapped over to digital TV at all. It might be five or ten or fifteen years before 1080/60p is broadcast commonly, and whatever camera you buy today will have little relevance to what you're shooting on 10 years from now.

Do you think there ever will be a standard again? Is it necessary and will it save the manufacturers anything in costs? Now that modern TVs can handle anything and adding a new format involves little more than flashing a new codec perhaps formats will be the future equivalent of choosing film stock? Sorry not part of the OP but curious to hear your view.
 
So on long recording, say 1 hour of continuous recording, it sounds like the processing/analyzing of AVCHD could become an issue. Compound that in a multicam long event with 3 or 4 long files and then?
Also sounds like the workflow of p2 is only significantly easier if you either can afford multi card readers or can find and use some old computers to use as p2 readers.

I'm sure some people have different experience... but I'd say, for longform, that's true... I'd go so far as to say its only significantly easier if you can afford the multi-card readers. The "old computer approach is workable, but not easier.

If your projects fit on a single P2 card, or on the 2 cards that fit in the camera, and you use that as a reader, you can have an easier P2 workflow. It's when you need more data for that in a typical project that P2 gets unwieldy without the multicard reader. 2 64GB cards may well cover your needs...


So copying footage to the NLE to edit would likely edit faster with p2 footage due to less processing of the less compressed footage, but since the files are bigger to start with, it would take more time to copy to the NLE in the first place.

That's a tricky question... P2 files are larger, but transfer rates are SUPPOSED to be higher (I have pretty slow transfers using my HVX as a reader). The point for me is not generally "which is a little faster", so much as "can I edit immediately or not." Generally I multitask around the transfer times pretty effectively... I'd love to be able to edit immediately, but at the moment don't have the setup for it; and using FCP7 or Premiere, it doesn't seem like there is really a setup for it. Well.. I HAVEN'T yet tried working with P2 directly in premiere... I will soon!

I'm re-reading everything in this thread mutiple times. Good info here. I think many people besides myself will benefit.
Very good! Remember that we all reflect our personal experiences, and that yours may vary. But information, taken non-dogmatically, always helps! It's good that you're thinking this through, and applying it all to YOUR workflow, to figure out what'll be best!

-Barry
 
A big thanks to everyone for helping so far in this thread. I've gone back and given "helpful" posts to everyone.

After digesting this further I see that P2 seems to be an ideal choice if using one camera where footage can be worked with straight from 1 or 2 cards from the camera. But if dealing with multiple cameras/cards it very quickly becomes significantly more expensive both in cards and muti-card readers, in addition to cameras costing more. Too much for my budget, but not for a larger post house. I can see where it'd be really nice to simply get back from a multicam shoot and then use a nice multicard reader to edit all the footage directly from the cards, then export the project, backup the clips you need and then onto the next project without even having to record all the footage to the HDD. Of course, even that wouldn't be possible on long post projects where you would need to quickly re-use the cards again and therefore have to copy all the footage to the NLE's HDD to free up the P2 cards for re-use before the previous project was finished. Unless you had so many cards you could let some sit around while using others. :)

The benefit of in-camera Intra-100 recording is a big plus, thus eliminating what I feel would be a hassle using external recorders with AVCHD cameras.

The "using old computers" as card readers seems like it may become less and less of a good workaround with time. It appears to me that requires finding and buying old computers can lead to other problems like failure, due to age, etc.. But I admittedly do not know exactly what is required nor how recently the appropriate technology was available.

I would feel pretty confident in shooting with a pair of P2 cards recording in sequence due to reliabilty strengths, but using redundant recording of SD cards also seems to give a peace of mind knowing that there are 2 copies, and each can record for a very long time.
 
Thanks for more clarification guys!

I'm specifically looking at the hpx250 and ac160.*

I think your biggest concern between these cameras (since you are OK with the functionality of both the 250 and 160) will be how long your record times are.

I may have missed it but, How long are your sessions of recording??

As others said:
P2, on the low end, is 20x more expensive than SDHC cards. So if you record anything lengthy like a concert, wedding, conference, multi-cam show, etc. SDHC is going to be much much more economical. If your recording a short or something where you can offload your P2 card frequently it may not be a issue.

Redundant SDHC recording is $60 or less per 16gb (2 x 16gb chips at $30 each) and yields 99 minutes of record time.

P2 is $600+ per 16gb and yields 32 min of record time. (edit: only $239ish now)
 
Last edited:
I may have missed it but, How long are your sessions of recording??

That's part of the problem. It's close to half short form and half longer form right now, and I'm not sure how the balance will sway in the future as we pursue more work. Some of my work is short form single camera chroma keying. Some is multi-cam 1+hours event (dance,etc..). Some is multi-cam weddings, which can be long. The longest events are 2 or 3 hours of record time.
I shoot everything ranging from talking heads to car chases and occassionally special effects. Also writing a series that I may attempt to produce, but I shouldn't weigh my decision on that. It's a wide range of demands so that's why it's tough for me to decide.

I like what I hear about the rubustness and reliability of P2. I like the on-board 4:2:2.
Obviously SD is clearly much cheaper, exponentially so when the recordings get long.
The 1080 overcranking of the AC160 would be nice sometimes.
The flash band compensation of the HPX 250 would be nice in events.

I'm thinking first of getting an AC160 so we can start shooting single camera HD. Testing green screening with it and overall performance to see if it all works well enough for everything we do. If it looks like we need better for the keying and heavier post work then as soon as possible getting an HPX 250 for primary camera with three 64GB cards and use the AC160 as "B" camera. Then as soon as we can afford it get an AC130 for a "C" cam. This is assuming that the HPX 250 will match closely enough for multi-cam work. If the AC160 seems to cover all ground good enough for us then "B" and "C" will be AC130s and I'll not be joining the P2 crowd.

The goal is eventually 3 solid cameras. We used to shoot 4 but cannot afford 4. Three can do a lot but will have to wait until I can raise the money. Luckily they all don't need to be HPX250 caliber, but they need to be able to work cut together. One HPX caliber camera can handle a lot of the corporate work by itself (though we are used to shooting most of it with 2 cameras), and the multicam stuff tends to not have to be as good nor require heavy grading or etc..
 
Last edited:
Also... I'm not sure if this has been mentioned... the data rate on the P2 cams is variable in a way that the data rate of the AVCHD cams is not... that is, certain framerates reduce the data rate of the P2 codecs to much lower levels. For DVCProHD, for example, 720/30pN gives you 2 minutes per GB (more than 2 hours with a single 64GB P2 card); 24pN gives you about 160 minutes on a single 64GB P2 card. (I THINK this is the same for AVC-Intra, but before believing me lets have someone else confirm it, since I don't have an AVC-Intra cam). So, if you can get away with using 720p in longform situations, P2 becomes more viable at a lower price. 2 64GB cards would give you almost 4 hours of recording time that you could all be on-line via the camera.... I think with AVC-Intra 50, it could be more than 7 hours.

note that this does not affect quality - the amount of data used PER FRAME is exactly the same. But the native recording modes save a lot of data when there are fewer frames per second!


I think a combination of HPX250 and a few AC130s is the best bet for you... if you have the HPX250 for the things that need it, I'm not sure that the 160 offers much benefit for your situation over the 130 (unless you'll need SDI during multicam shoots...?)
 
Last edited:
Also... I'm not sure if this has been mentioned... the data rate on the P2 cams is variable in a way that the data rate of the AVCHD cams is not... that is, certain framerates reduce the data rate of the P2 codecs to much lower levels. For DVCProHD, for example, 720/30pN gives you 2 minutes per GB (more than 2 hours with a single 64GB P2 card); 24pN gives you about 160 minutes on a single 64GB P2 card. (I THINK this is the same for AVC-Intra, but before believing me lets have someone else confirm it, since I don't have an AVC-Intra cam). So, if you can get away with using 720p in longform situations, P2 becomes more viable at a lower price. 2 64GB cards would give you almost 4 hours of recording time that you could all be on-line via the camera.... I think with AVC-Intra 50, it could be more than 7 hours.

note that this does not affect quality - the amount of data used PER FRAME is exactly the same. But the native recording modes save a lot of data when there are fewer frames per second!


I think a combination of HPX250 and a few AC130s is the best bet for you... if you have the HPX250 for the things that need it, I'm not sure that the 160 offers much benefit for your situation over the 130 (unless you'll need SDI during multicam shoots...?)

AVC-Intra50, at 720/24pN, takes up less space than AVCHD: 20 megabits.

Thank you both tremendously. I had forgetton about that option and wasn't even aware of the size surprise.
I don't forsee needing SDI for multicam shoots (hopefully). The closest I forsee us needing would be a 2 camera green screen, but even that can usually be worked around with one camera by repeating for differing framing. And if it's two camera it's likely to be at our greenscreen setup where the lighting will be better and more controlled and an AVCHD camera might still pull a good key as a "B" cam.

Getting help from such knowledgeable people is invaluable.
Now I'm hoping for a price drop after christmas/new years to get the first camera unless we land a job that necessitates a quicker purchase.
 
Probably a real dumb question but..Is there harm in always using a p2 camera as a reader? If I end up with one p2 cam, it's probable that much of the editing from it will be done online from a directly connected camera as a reader and I'll never buy a reader. Its almost hard to think of working that way because I'm still used to the old tape mechanism cameras where you wanted to save the heads and use a deck. But since it's more like solid state memory now, I'm guessing there's no harm in using a camera for that a lot.
 
I'm joining in late here, but my 2 cents all the same....if all you need is just a chip-based recording format, P2 is overkill. If, on the other hand, you need the unique features that P2 has to offer, go for it. Even though it is a little bit spendy, you will NOT be sorry! I am not wealthy, but I will definitely buy more P2 cameras in the future....
 
"Probably a real dumb question but..Is there harm in always using a p2 camera as a reader?"

Why would you not mount a $100 reader from Amtron in your PC? Just pop the card out of the camera into the reader and you can either edit from the card or copy the files into a folder. You can also view you card contents by using free "P2 viewer" software.
 
Last edited:
Back
Top