Is the a6000 worth buying an external recorder for?

T

ThatOneGuy

New member
I got the a6000 a little while ago, as it was the best thing I could afford. I know it does uncompressed hdmi output. I just had a birthday where I amassed a large amount of gift cards and the like. Enough to make an external recorder such as the ninja star look like a good buy. Is it worth it, or will the recorder not do much to the a6000's video?

I should note also that I am not a professional by any stretch of the imagination.
 
Sony has improved their video on these cameras so much that there will be only a very small gain to be had with an external recorder. But there is a gain. Especially in fine details like moving water, leaves on trees in the background, detail on buildings in a city-scape scene. One way I always spot it is in still-frames from a video clip. I like finding a great frame from the clip and exporting it as a still image, and this is a case where you can really notice the difference in the extra detail you get - even if it is less "noticeable" in moving video.
The problem with the Ninja Star is that it records Pro-Res, which is a little less compatible with Windows editing software. But the real killer is the CF Cards that thing uses...they are INSANELY expensive per gigabyte. You'll spend more on the cards than you will on the Ninja. That's why I decided against getting the Ninja Star.
 
How hard do you push your clips in post?

If you apply strong color grades, you may have had your images fall apart, they definitely did so on my NEX-5N: after pushing and pulling to get my desired result, I would start to see obvious macroblocking, particularly in dark parts of the image that had motion and had been pushed up. Moving to the a7S, with its much better XAVC-S codec, solved this for good. And ProRes would be even better, in the sense that it keeps small detail alive, that the internal codec blurs away.

Switching over to an a5100 would solve this too (it shoots XAVC-S), but you'd lose the lovely viewfinder. Ah, decisions, decisions...
 
I have watched countless videos of footage shot both internally and externally on those recorders. I never see a difference. Some claim there are differences in post, but I can't see it being significant because you don't get extra dynamic range if you go from 4:2:0 to 4:2:2, you just get less color repetition across pixel blocks. Stuff that most people won't see a difference with.

The external records are great if you want to shoot longer than you can on cards, in a format that is easier to put in you NLE. Also, if you want an EVF too. I DO NOT HAVE ONE. But I love giving opinions about that which I have no real experience ;) In short, external recorders are for professionals, where time is money.

As for the XAVC-S. The XAVC CODEC can provide greater color depth, but the version in the A5000 is the "S" version, only 50mbs (compared to 28 in the A6000). I believe the extra data only helps in motion compression. Again, this is just what what I see on YouTube.

So my answer to your question is NO. I'd get a lens, the wireless mic, or lights, or a stabilizer, really, there are a few things, maybe even 4 or 5 things, you could spend your money on. :happy:
 
I can see a clear difference from the NEX-5N codec (which is probably the same one from the a6000) and the a7S codec (XAVC-S, same one as the a5100). It's not a bitrate thing (28 Mbps vs 50 Mbps), it's a codec internals thing. Files look exactly the same straight out of the camera, but once you start pushing and pulling one falls apart much sooner than the other (in the sense that macroblocking artifacts start to appear; not in all shots, in any case: most often in those where WB has been corrected and then the shadows are pulled up and tinted). ProRes should hold even better.
 
Samuel H said:
I can see a clear difference from the NEX-5N codec (which is probably the same one from the a6000) and the a7S codec (XAVC-S, same one as the a5100)..
Click to expand...

This is nonsense post. The difference between 5N and A7s is due to sensor difference. A7s does full sensor readout, while 5N skips lines. Nothing to do with codec
 
There are lots and lots of differences between those cameras, and I can tell you: codec is one of them. Changing the sensor doesn't affect macroblocking even one bit, changing the codec does.
 
Samuel H said:
There are lots and lots of differences between those cameras, and I can tell you: codec is one of them. Changing the sensor doesn't affect macroblocking even one bit, changing the codec does.
Click to expand...

Yes, changing the sensor doesn't effect macroblocking, but its higher signal-to-noise ratio creates less color ambiguity. The cleaner the image, the less macroblocking to begin with, with any CODEC. Video CODECs have been around since the 1970s, but I'm happy to concede they didn't ramp up until the 1980s. So for 30+ years engineers have been working on maximizing the amount of image information one can COmpress into a file and DECompress on a viewing system. The better the compression the more channels a cable company can sell. The better compression the more digital projectors can be sold to theaters. The more cameras can be sold, etc, etc. IN SHORT, there is not much more quality that can be found by improved CODECs because a TON of money has been thrown at the technology. Even with a better chip-neutral CODEC, a CODEC running on a dedicated chip can do better. In the real world, there's more to video quality than the CODEC, a lot more!

If you can show me a controlled experiment that shows how 8-bit XAVC-S is better than H.264 then you can make a good argument for an external recorder. And I'd LOVE to see such a test because I TOO, like everyone else on the planet, what better image quality from my cameras ;)

I don't want to start a flame war, but as nice as Vitaley's stuff was for hacking those, now old, Panasonic G cameras, the differences were not that big, and relied a bit on using faster cards; that is the CODEC settings would not have worked using slow MEMORY HARDWARE. CODECs don't, and can't, live in a vacuum. The data can't move fast enough they don't work at higher qualities. If it's garbage in (image) it's garbage out.

I shoot both H.264 and RAW. I can show a huge difference between them. I have never been able to see much of a difference between H.264 consumer cameras. Again, I agree, there is some difference, but not enough to pick cameras based on CODEC alone.
 
Of course I'm not saying anyone should pick cameras based on codec alone. But the a5100 and a6000 are very, very similar to each other, except for the codec and the EVF. I don't have any of them so I can't run a controlled experiment.

I think XAVC-S makes a difference over Sony's H.264 implementation, but I don't find this all that interesting anyway, so, to each their own, off to greener pastures. Have a good day :)
 
Put simply - the hacks, the higher bit-rates, the better codecs that some folks say doesn't really make much of a difference - I assure you they do and I'll list a few cases where the difference can be very noticeable.

1. Leaves on tress in your background. -Your subject standing in front of the camera may look nearly identical, but the leaves will jump to a much clearer, less mushy definition. ANYTHING in your background really that is made up of lots of small things.

2. Running water close up. A creek or brook, or small river, waves lapping - these will all come out much more clear looking. Easier to actually see and focus on the movement.

3. Hair and fur. There is probably a good chance that your video may contain something that has hair or fur. Results = see item #1.

4. Stills. Taking stills from your video is an extremely common and demanded thing. A still from a video has drastically less megapixels than really ANY stills camera and this is where you are going to need absolutely every single little bit you can gain, because with still pictures people are naturally scrutinizing the image either somewhat closely or very closely. My stills quality and my leaves on trees quality jumped WAY up on my GH2 when I applied a high bit-rate hack.


Yes - RAW is a HUGE difference, but the gains realized from higher quality codecs and higher bit-rates are by no means trivial.

And as for watching on YouTube - the number of people watching YouTube at 1080P and on high-res monitors is growing extremely quickly.
 
We're all in agreement. I was only trying to explain why, for the original poster, an external recorder would not be worth it. I too, spent a lot of time comparing CODECs etc., but finally decided I wasn't interested in splitting hairs and have found "greener" pastures myself :cheesy:
 
Yeah, I think we may have been debating 2 different points.
That's the first thing I said to the O.P. - Sony's newest codecs are so much better than codecs from years gone by that an external recorder doesn't have much left to improve upon.

Sony's own awesome A7s sensor would still look like mushy crap with AVCHD at 22mbps I assure you.

For other cameras like the GH2 it was more than hairs that were getting split - with the 3 very common shooting subjects I mentioned it was readily visible.
Sony's new XAVC codec retains the same amount of detail at 32mbps that it took older cameras a bit-rate of at least DOUBLE that to equal.

Again, that person standing smack-dab in the middle of your frame isn't going to look any different - but macro-blocking on things like trees and bushes and busy backgrounds looks clearer in a way that even shows on YouTube compressed videos at their 480 setting.

Nikon's D800 is another great example. It's video quality was nothing to get it very much notice - until it was hacked to triple it's paltry 20-something mbps bit-rate.

Same goes for the D750 - It's got a fantastic native look from its great sensor - but it's foolishly low bit-rate leaves peripheral parts of your frame trashy macro-block mush. Hacked to get it up to a mere 60mbps will make that all go away. Without it the camera is NOT a semi-pro contender...with it, it absolutely will be.

So while sensors between cameras make a HUGE difference, a great sensor can and HAS BEEN crippled by its own codec or poor bit-rate. Sony didn't do that on its latest cameras - Nikon did.
 
(luckily nobody pointed out that the a7S actually shoots both XAVC-S and AVCHD, which would have forced me to shoot a controlled test; phew)
 
If you're not doing it professionally I do not believe it is a worthwhile upgrade for you. Take the money and buy yourself some more lenses or batteries. I do own the Ninja Star and an A6000. My ninja star never leaves my C100 but I too was curious recently and put it on the A6000 for giggles. You do get a slight increase in the small details but it's hardly things most people will notice. At least not one that justifies the cost of CFast cards. The internal codec is not perfect but it works and I've definitely worked with worse.
 
Samuel H said:
Of course I'm not saying anyone should pick cameras based on codec alone. But the a5100 and a6000 are very, very similar to each other, except for the codec and the EVF. I don't have any of them so I can't run a controlled experiment.

I think XAVC-S makes a difference over Sony's H.264 implementation, but I don't find this all that interesting anyway, so, to each their own, off to greener pastures. Have a good day :)
Click to expand...
Another big difference, if it's true, is full sensor readout.

The dpreview video review of the A6000 shows moire, mush and less detail in the A6000 images compared to the RX10 in the video stills image comparison tool. My guess is it's because the RX10 has full sensor readout even if it only has a 1" sensor.

A Sony press release and many reviews (perhaps based on that press release) say the A5100 has full sensor readout. But I have yet to see comparable images like the image comparison tool at dpreview that demonstrate this, if it's true.

I own the A6000, but I'd buy the A5100 as a video body if it had full sensor readout, since it already has the better codec.
 
Samuel H said:
Then I think you'll stick to the a6000. From Slashcam.de:




Sony a5100


Sony a6000
Click to expand...

Wow, I searched high and low to find a video stills comparison like this. Hat tip to you for finding it. I totally did not think (or know how) to search German sites! :thumbsup::thumbsup:

Well, I guess that sort of ends my brief hope that the A5100 might be substantially better and therefore worth getting. Would it be crazy to buy an RX10 just for video? It's wierd thinking that a 1" sensor camera actually has better video than an APS-C camera with a great sensor like the A6000, but the video stills don't lie, right? Am I missing something?
 
You must log in or register to reply here.
Back
Top