Personal View site logo
Make sure to join PV on Telegram or Facebook! Perfect to keep up with community on your smartphone.
Please, support PV!
It allows to keep PV going, with more focus towards AI, but keeping be one of the few truly independent places.
Red Scarlet versus GH2, any special requests?
  • Doing a week shooting with the scarlet with a bunch of gh2s. May have some time to do some side-by-side comparisons. If anyone has anything they are curious about, let me know and we'll see what we can do.

  • 20 Replies sorted by
  • Whatever comparison you have time to do will be welcome! Outside in sunny lighting would be good since so far the few worthwhile tests I've seen are either indoors or outside in gloomy overcast lighting (Philip Coltart's Vimeo test: vimeo.com/30751603).

    The difficult part with any test is keeping as many variables the same for both cameras. Certainly if any movement is involved it's good to run both simultaneously side-by-side on a double camera bracket (I use a Velbon but there are alternatives). For static tests things aren't as critical and two tripods would do.

    Looking forward to your results.

  • Oh...if at all possible please compare them for jello when using a longish tele lens with panning/movement.

  • Run tests that use lighted setups, not extreme lowlight or anyting testing the extremes.... If you can afford a RED Camera you can afford lighting, so lets test for real productions not for outside at night in complete darkness and see who wins the noise battle when adding 9 stops worth of light grading. Shoot a scene with both cameras would be cool. I don't care if you adjust lighting for each cameras strengths. Try to make each camera look its best. Thats the best way to see the differences when it comes to quality.

    Tests should be real world tests. I hate tests where people tests extremes. No real use when comparing a professional camera in setups that have any value or production value. WHo is using a RED without adequate lighting, and if they are, they are wasting their money.

    Tests should try to make cameras look their best. We should be picking which camera looks better, not finding ways to make one look worse.

  • @JPB1138 I disagree - one of the main reasons I used a RED on the shoot I did was to shoot really wide outdoor shots with a very wide dynamic range and dynamic motion (which I could not do if I took the ND Grad approach, for instance).

    @JDN You may want to read through this test first, just to see what has already been done. http://www.personal-view.com/talks/discussion/2074/a-knife-and-two-bazookas-to-a-gun-fight

    Here are some comparisons I was going to do but ran out of time.

    Matching framing with same lens for:

    • Scarlet 5K 24P or 30P vs. GH2 24H or 80%24H
    • Scarlet 4K 24P or 30P vs. GH2 24H or 80%24H
    • Scarlet 3K 24P or 30P vs. GH2 24H or 80%24H
    • Scarlet 2K 24P or 30P vs. GH2 24H or 80%24H

    Then for pixel level detail:

    • Scarlet 2K 24P or 30P vs. GH2 EX24H or EX80%24H

    And for higher frame rates:

    • Scarlet 1K WS 120P vs GH2 SH (NTSC)
    • Scarlet 1K HD 89P vs GH2 SH (NTSC)
    • Scarlet 1K HD 60P vs GH2 SH (NTSC)
    • Scarlet 2K 60P vs GH2 SH (NTSC)

    If you shoot something with a really wide dynamic range, make sure to test HDRx with 6 stops as well. Be aware that "Magic Motion" is not perfect, and if possible post a frame that is "just A" and just B" as well as the Magic Motion composite so that people on the site can demonstrate the potential results that could be achieved with other tonemapping tools. If you send me one, I'll be happy to that myself.

    I suggest using the RAW and Exposure tools when setting your exposure for the dynamic range of the scene. Seeing purple where you will get shadow noise and a visual indication of where you will clip can be very helpful in terms of setting aperture, etc. On the GH2, use Smooth or Standard (or Nostalgic if you do not mind balancing the color) to get a good dynamic range and remember that Standard and Vibrant appear to be the most color neutral.

    I recommend using Sedna A or C (AQ1 or Q20 are equally useful) as opposed to Canis Majoris Night, since Sedna is currently the most widely accepted "all-purpose" maximum quality setting. If you have issues with SH mode, Mysteron would be the second choice.

    If you can only do a smaller number of tests with each camera, take these matched framing ones from the above list.

    • Scarlet 4K 24P or 30P vs. GH2 24H or 80%24H
    • Scarlet 2K 24P or 30P vs. GH2 24H or 80%24H
    • Scarlet 2K 24P or 30P vs. GH2 EX24H or EX80%24H
    • Scarlet 2K 60P vs GH2 SH (NTSC)
  • @thepalalias I think it is obvious or should be... that the scarlet will beat the GH2 in those situations. Like extremely poor overhead outdoor lighting, or directly towards the sun... is it really a test?? Or just a confirmation of stuff we already know? I love wide outdoor shots, I tend to use a polarizer and a vari-nd in one. A Singh-Ray Vari-N-Duo which has both, that really helps cut down on bad highlights, but it still wont equal the DR of the Red RAW cameras.

    Again though I really only care about 24p vs 24p tests, as they are most useful for cinematic stuff on GH2, so Again RED will win the frame rate war, but we know that already.

    All I was saying is that we should test the best the camera's have to offer, not finding the weakest possible areas to examine. Outdoor test is fine, I think the Gh2 can look good outdoors, use a polarizer and VAri Nd to maximise its potential.

  • I would like to see if possible how the sensors react to panning and fast inside frame moving like someone jumping and running, like something fast in front of the lenses (shutter 50/24p). Also it would be nice to see how noise compares, in low light situation, like indoors no light skin tones. Also heavy detailed texture with light camera movement, like smoke, close ups of sea waves, those things that absolutely destroy compression. And how they resolve dynamic range in midday full sun and shadows wide shot.

    Tanks in advance.

  • @JPB1138 I guess it depends on the definition of a test. In the sense that you are testing the results so that people can get a more tangible comparison, I think it qualifies. I think most people know the gist of what the gap between the two cameras is like, but given the comments that people sometimes make on the forums about a GH2 clip having "RED-like dynamic range" or things like that, I figured it was worth showing.

    As far as the frame rate situation, with an Epic, there is very little question of just how far it surpasses the GH2 and certainly it allows you to do far more than you could with Twixtor, MVTools2 or other pixel motion software. But with the Scarlet, people debate how effective the low-resolution modes are or how the 2K 60P compares to the GH2 720P 60P so I think it may help a lot of people to see that.

    Anyway, I was happy to get the chance to see for myself instead of accepting others opinions, so that is why I mentioned it.

  • I would very much appreciate, a matched shot (as close as possible) showing highlight rolloff. I'd like to have a super short native clip of each. To me the ugliest part of the GH-2 and most DSLRs is the harsh video like roll off to highlights. Usually showing an artificial yellow edge. Red has better roll off, more dynamic range to capture the highlights and finally in the red software, once you select your white balance, the software can recreate the missing color data from the remaining color channels. I wonder if some similar technique can be implemented in a grading app to fix thoser highlights for GH-2 footage.

  • "I think most people know the gist of what the gap between the two cameras is like, but given the comments that people sometimes make on the forums about a GH2 clip having "RED-like dynamic range" or things like that, I figured it was worth showing."

    +1 I think it would be worth it as well. Just to see how much difference there really is in harsh conditions. Could be useful to see what exactly you have to do with the GH2 to expose it closest to something like a Scarlet or Epic.

    "RED-like dynamic range"

    The original Red One probably. Probably better than that actually. The original non-MX had like what 9 stops? GH2 is probably 10-10.5, shadow-biased.

  • @GH2UW no problem -- it's all outside in the hot florida sun (at least the real shooting is) so testing light will be no prob.

    @thepalalias exactly -- one of the big reasons we're auditioning the scarlet as our A cam for a 4 x 1 hour doc is its ability to handle huge DR and hold shadow detail which we can bring out in post. And thank you for links and advice!

    Also, we're definitely going to test 60fps and 48fps -- doing chopper stuff at 3k 48 fps to see how it looks (especially after that was the one thing so praised -- amongst all the scorn -- in the hobit screening)

    @leonbeas we'll see what we can do indoors and on high detail -- we're out on the water / nature, so definitely can find some situations where compression is stressed

    @disneytoy you got it.

    This is of course assuming we get some time to play around at the end of the shoot -- but have the camera for more days than we need it, so...

    Appreciate the feedback -- we're trying to make a call whether we can "get away" with the gh2 in many situations or whether to go up to the Red. I'm pretty sure I know what the answer will be (budget permitting) but it will be neat to see how they stack up, especially seeing as the scarlet is the only red remotely useable for run and gun.

  • @GH2UW no problem -- it's all outside in the hot florida sun (at least the real shooting is) so testing light will be no prob.

    @thepalalias exactly -- one of the big reasons we're auditioning the scarlet as our A cam for a 4 x 1 hour doc is its ability to handle huge DR and hold shadow detail which we can bring out in post. And thank you for links and advice!

    Also, we're definitely going to test 60fps and 48fps -- doing chopper stuff at 3k 48 fps to see how it looks (especially after that was the one thing so praised -- amongst all the scorn -- in the hobit screening)

    @leonbeas we'll see what we can do indoors and on high detail -- we're out on the water / nature, so definitely can find some situations where compression is stressed

    @disneytoy you got it.

    This is of course assuming we get some time to play around at the end of the shoot -- but have the camera for more days than we need it, so...

    Appreciate the feedback -- we're trying to make a call whether we can "get away" with the gh2 in many situations or whether to go up to the Red. I'm pretty sure I know what the answer will be (budget permitting) but it will be neat to see how they stack up, especially seeing as the scarlet is the only red remotely useable for run and gun.

  • PS: shooting next week so may not reply but will be following

  • @jdn Best of luck! Here are a few considerations I will add based on your comments.

    • For the helicopter shot, you will often need to do some stabilzation in post. So the higher the frame rate, the easier its for the software (and for pixel motion software if you want to do slow-mo). If you want the nice "floaty" feel, stick to 180 degree rule unless you plan to do slow mo in post.

    • For a documentary, keep in mind that the RED Scarlet is very loud when turned on if not shooting because of the fan. This is not an issue unless you are filming close subjects that can be alarmed. The noise is not an issue while recording.

    • Stock up on power, for sure! The bricks can last up to 2 hours but the RED Volts only last 20 minutes. Bricks are the way to get GH2 like record times (or 6 RED Volts).

    Hope that helps. :)

  • For heli shoots, make sure you have good mechanical stabilization (like gyros). The Scarlet is not free of RS, and it can look quite nasty with software stabilization.

    If you need to go less pro, use a RS fixer first and then the stabilization software.

  • Out of the subject, I don't know for what reason people would use the scarlet compared to the Sony F3 for documentaries. A good 1.5 stop better DR in all test I have seen, much better low light, much easier workflow, battery system etc. Nearly the same could be said about the Canon C300 form very small size and portability. In case of features perhaps the 4k would be a compeling reason, but documentaries does it make sense.

  • I have a Scarlet and without any accessories, being powered from V-mount brick (which you'd have to do for documentary), a super-lightweight DP4 monitor and a not that heavy 17-55 zoom I feel it's already way too heavy for any long running doc stuff. Then there's the backup and storage thing, so for doc stuff there are a few better choices.

  • @Nomad Or use DeShaker, which does both. It is PC only and takes a bit to learn but it is one of the most powerful available and in the right hands gives performance superior to many of the commercial engines. http://www.guthspot.se/video/deshaker.htm

    Plus, it is the only engine so far that can (under many circumstances) do it without a loss in reduction.

    Unfortunately, I am not very good at using it yet. :)

    @danyyel You cannot just look at the dynamic range, you have to look at where it is weighted as well.

    http://www.zacuto.com/the-great-camera-shootout-2011/episode-one

    Look at 16:45-17:00 - you will see that if you expose for skintones, the earlier RED One MX completely outperforms the F3 in the highlights without question. Now, the situation is reversed for the underexposure test at 11:05-11:17. Both cameras have a great dynamic range, no question about it and I would be hard pressed to use the basic dynamic range to suggest one over the other. Now if you look at how you're recording that dynamic range, the resolution you can get and the HDRx option, the RED gets the advantage again.

    But your point about high ISO performance is likely valid. I have never compared the two but I know the F3 is one of the stronger performers.

    As for the C300, depending on how the test went, the Zacuto Revenge test later this month may or may not help to answer that question. One of the nice things about the C300 is how quickly you can set things up, but the RED once again has a recording format that really lets you make the most out of the recorded dynamic range at a whole different level.

    @stip Definitely legitimate concerns. The weight wasn't much of an issue for me while using. I shot handheld with a 13 lb Cooke lens, a v-mount with a brick and sure, I would need a short rest at the end of each 6 minute shot. But with a lighter lens, it felt like it would have been pretty reasonable (and that is without any form of rig or shoulder mount).

    That said, it certainly is not as light as a GH2.

  • Drop tests from various heights . . . something like 1m, 3m, 5m, 10m, 20m, 50m, etc. perhaps onto different surfaces . . . waterbed, glass panel, concrete, linoleum, salt and fresh water, etc.

  • @danyyyel @stip here's the thing: we gotta have 422 50mbit or higher slow motion as a good chunk is being shot that way and need to clear broadcast threshold. And there's only one line of cameras that does that.

    We did a bit shoot with the f3 and a nano -- gh2 kicked its ass. Seriously. Not an F3 fan at all. Cannot handle highlights. Gets fringy and videos in bokeh. Erganomics suck. Really wanted to like it -- and really want to like the F700 -- seems to have fixed some of those problems. Will try it.

    @thepalalias yep, ks-6. Any thoughts on mounting it? It perfectly counterbalances the scarlet on a shoulder rig, but I feel like when you only have one gyro, you want it right under the camera.

    BTW all, any tests we do are only in the field but we are renting a scarlet for a future shoot in a weeks time and have some play around time so we will do more studio type tests then.

    After one day, what I can report is: it takes longer to shoot with the scarlet. Duh. This makes GH2s even better as b cams given their ease of use -- save the scarlet for wow stuff. And finally, were able to shoot in mid day sun in florida (had no choice) without any clipping. I know people talk a lot about erganomic / weigh drabacks of a scarlet for doc, but that is a big plus.

    Gh2s are still rocking it though. Don't get me wrong.

  • @jdn Your experiences echo my own, both in terms of the relative speed of the GH2 vs the Scarlet and in terms of the ease in shooting in natural light. If you want to talk about gyros, @rambo would know a lot more and has been nice enough to provide some advice in the past.

    Here is a short discussion he and I had in the comments of my blog a while back. http://perlichtman.com/pasdenapulse/wordpress/2011/07/07/transpac-2011/