Personal View site logo
Driftwood ClusterX series 2:│moon│ЅріzZ│nebula│drewnet│Slipstream │Redshift
  • 1053 Replies sorted by
  • Shot using Moon Trial 5 and various lenses.

  • I tested Moon Trial 5 by shooting my parents new dog, and i really like the result in terms of quality.

    Shot with the 14-140 mm, smooth -2,-2,-1,-2 and at ISO 800. Only issue with the sandisk extreme 45 MB/s card, I was not able to play videos on my camera longer than approx. 30 seconds or so.

    I also shot some videos on my crane and while I panned "fast" from left to right, some frames skipped several times. The SD card could be the problem, or does anyone else have the same issue?

  • @onionbrain I haven't started using it in comparisons yet but I plan to shortly. How big a step forward would you say it is from Cluster 7 and where do you notice the improvements the most?

    I remember at high ISOs with Cluster 7 AN Sharp 2, the 60P modes yielded I-frames similar in size to CM Night but much lower than Crossfire 3 CM Night. If you get a chance to test 60P before I do, could you take a look at whether there's been much change in that area? I'll probably compare 24H modes first. :)

  • @kkorr I have no clue what I just watched -- but for 13 minutes I was hypnotized by whatever the heck was going on. Looked great!

    Okay, my thoughts on Trial 8 of Cluster X DREWnet.

    Sample footage...

    Regarding 1080/24PH...

    Visually -- it is absolutely the equivalent of a 160 Mbps intraframe stream. At least. It handles all the motion you can throw at it, all the detail, and all the motion and detail. It handles all of it as well as AVC-intra100. And, you can mess with it in post until the end of the world -- and it holds up. I would have no problem using it for professional distribution applications -- in fact, I already have.

    Frankly, in terms of handling motion -- I dare anyone to compare it to Moon and seriously respond that Moon beats it. My instinct tells me it would do better than Moon (especially for people using Premiere Pro CS6) -- but at the very worst the difference would be minimal -- like 400% zoom in type comparison minimal.

    So, in conclusion -- I'm still using it, and I like using it -- and that's really the best testimonial I can offer.

    Concerning other modes (720, interlaced anything, etc.) -- I haven't messed with them.

  • @joofmagoof Whats your transcode method?

  • @mastroiani @cjdancer

    Sony says "up to 43 MB/s" write speed on the specified part.

    @cjdancer Panasonic says 60 on theirs?

    In either case, the SandDisk 64GB 95MB/s is rated at "up to 90 MB/s".

    Note that I sometimes get better spanning performance on a relatively slow Transcend 64GB SDXC Class 10 than I do with a much quicker Delkin 633x 16GB card, assuming both are fast enough to record with the setting in question. Whatever brand you test, make sure to try with a 64GB card if you are trying to confirm the spanning viability of a new product line.

  • @cjdincer sony SDHC cards are cheaper than SanDisk as well. Although haven't tried them myself

    http://www.amazon.com/Sony-UHS-1-Memory-SF64UX-TQ/dp/B008PO60F2

    94mb/s read, but doesn't specify write speed

  • @mastroiani once in a while or after heavy use, a low level format on a mac or pc refreshes the card. I have one of those cards and it sometimes does weird things, kind of flakey. new toshiba series are cheap. 90read/60write plan to try one soon

  • @cjdincer yes, it's 32GB 45mb/s. I just got home and trying to copy the private folder using my home computer. So far it looks that it will complete the job! Perhaps it was the hard drive on the other Mac? I'm keeping my fingers crossed.

    EDIT: The files copied just fine! What a relief. I thought it was the SDHC card

  • @mastroiani is it a sandisk 32gb 45mbs? try copying mts files individually see which one is causing error

  • I shot one continuos 23 minute take using moon 5 and when I tried copying the Private folder from my SDHC card (32GB 45mb/s) onto my desktop it gave me an error and couldn't finish copying the file. I tried several times do copy but get the same error. I will post the screenshot when I get the error again.

    Files spanned fine and the camera didn't stop during recording. There are several 4GB files - is this how it's supposed to be?

    EDIT: This is the error message I get.

    Screen Shot 2013-04-02 at 4.16.29 PM.png
    394 x 111 - 22K
  • man, now reading all this i have to try moon settings, curiosity will kill me :\

  • @spacewig The finest noise patterns depend slightly on the subject matter and shooting style. For a high detail subject at a moderate ISO to high ISO, Moon 5 has been having the finest noise pattern in my testing (with the difference most noticeable in the lower right hand corner, for reasons Driftwood explains earlier in the thread). In lower detail scenes IntraVenus V1 has the most aggressive bitrate of the modern settings I have tested (though others like Sedna and IntraVenus V2 are also pleasing and not too far behind depending on the material). Canis Majoris Night has a slightly different approach - it is a little less gradual in some of the color but better at separating noise from detail in lowlight. It is slightly more prone to banding (at least it anecdotally seemed that way, I need to confirm).

    All of these settings are GOP1, which means they have the same amount of detail available to use on each frame regardless of how much motion there was in the frame before. This makes them just as good at high motion shots as stationary ones and especially good at noise patterns since they do not try to interpolate noise between frames, which can lead to inconsistent update patterns. Higher GOP settings allow for greater potential detail in stationary shots where not too much changes from frame to frame.

    So for noise, I would suggest trying Moon 5, IntraVenus (v1 for more detail or V2 for smoother), Sedna and CM Night. Right now I use Moon 5 the most.

    Edit: By the way, the bitrates for Moon vary dramtically by content. I average 72 mbps on moderate detail (IntraVenus V1 was a lot higher) but over 140 mbps on high detail (where IntraVenus V1 was the same or lower). I also have not tried earlier versions of Moon, only Trial 5.

  • Moon has one of the finest noise structures you can expect from a GH2.

  • @bheath Yeah, I think I'm going to single out problem areas in post with masks and deal with them individually. How far under 0 EV would you say the noise starts becoming a problem? (BTW, reference to Noir just for the sake of discussion as it involves extreme lighting) Thanks for your input.

    Is there a patch that has a finer noise pattern than others in underexposed areas? I ask here as there are just way too many patches to read every single thread and try to map people's evolving opinions are patches are/were released.

  • @spacewig - @thepalalias has a good approach for exposure. I do that too with this camera.

    Certainly video cameras like these have a narrow latitude, but it can be solved with lighting and choice of shot. Also, in $ modest situations, there is often no money for large fill sources [of light] so these kind of surfaces become opaque, therefore becoming a [sounding board] for noise. Even 35mm film stocks will choose these areas to show grain.

    If a stylized, noir effect is desired, I still fill the hell out of the shadows and tone them down in post. It doesn't make for a very moody feel on set, but nothing usualy does.

    You're right, a spot meter on those dark little devils will always tell you where the noise will appear.

    One thing to do save the shot in post is to mask the dark areas and bring them down even further or de-noise them, as they say.

  • I had something new happen over the weekend with my GH2 using drewnet beta 8. Everything records fine, but the video file didn't show up in the playback menu. the GH2 just ignored the file, even though it was in there when I imported the files to my PC later. I thought they were gone/deleted until I did the import.

  • @cjdincer I hear you with regards to a well lit scene. However, sometimes you'll want a scene to be more dynamic in terms of lighting choice, i.e. like noir. Does this mean all parts in the shadows will have noise? I think some noise is inevitable but at ISO200 I really wasn't expecting this much.

    Perhaps I need to start a new poll about people's favorite low-noise patch...

  • Here was my first trial using the DrewNet trial 8 but like my last drewnet patch, this keeps happening. Anyone know why this is happening or how to fix it? (artifacts, pausing, etc.) Here's a link to the video.

    Why does this happen? GH2 from Jay on Vimeo.

    Patch: Trial 8 DrewNet, Panasonic 20mm panny

  • @spacewig I think gh2 is better with reflective surfaces (I maybe wrong, too) Dark fabrics may need extra attention with GH2. similar stuff happened to me in broad day light shooting an ant nest, EV was less than -2 or around -2 and when I got home, there is an annoying noise in shadows which did not take my eye when shooting and shocked me at the time. but this happened to me with a lower bitrate setting. your having this problem with moon is very upsetting. I can't pinpoint what really is causing this. with the latest settings I don't see any noise at iso160-200 anymore in a well lit and exposed scene. Thanks to this forum (VK) and the information which allow us to learn tips and tricks of GH2 and other tools. I do feel your disappointment. need to go more down the rabbit hole to learn how to avoid it.

  • @thepalalias Thanks for taking the time to answer thoroughly, it is much appreciated.

    @cjdincer I am trying to understand what is meant by this. Does it love light as an absolute quality or does a sensor fare better with objects that reflect more light? I'm just having a hard time understanding because those monitors were not in the dark or shadows. I regret now not having taken a spot reading of them as now it's too late to tell how many stops lower they were than the incident reading, though I can confirm - having calibrated my light meter to the GH2's sensor on a number of different profiles - that smooth has a DR of 6.3 stops at ISO200 (.1 of a stop less than ISO160 which makes it a no-brainer default setting). Shian was dead right about this figure and has been repeating it since before he even made his Expose in the Zone (or whatever it was called) video.

  • @thepalalias Thank you for explaining this to me in a manner which I can understand. I am using 64gb 45mb/s Sandisk Extreme cards. I know this is not the correct card to use for Driftwood patches, but I already have them and they work great on the Nebula 6B and Slipstream#3 in all modes and also work great in 24h modes on the other Driftwood patches.

    I guess my thought process was wrong in regard to the bit rate, but you have cleared that up for me. Hopefully in the next month or so I will be able to buy the 95mb/s cards and be able to use the higher end patches for my GH2's.

  • @dado023 Yes, that is exactly what I suggest. I've done it several times with great results, even though I use the standard 24P output from the mode far more often. As long as your software is setup properly and doesn't try to drop/blend/interpolate frames, there is no loss in quality. Just remember that the mode does not record sound.

    As an example of how to use it: in Sony Vegas all you have to do is click on "Properties" for the clip in the timeline, check the box for "disable resample" and change the playback rate to 1.250. The reason is that 30/24 frames second = 1.25.

    Anyway, as far as HBR - yes, it just is not as robust as 24H/24L and settings authors will normally have far more optmized 24H/24L settings available.

    But if you need 30P with synced sound, HBR is your only option on the GH2.

  • @dado23 I'd say give moon t3 a try. has a higher HBR bitrate than moont5. I am using it right now. HBR is really good but sometimes does not playback. I turn the cam off and on. it plays fine. try and see for your self cause I think I am getting rather good results. @spacewig all sensors love light so I usually keep the EV around +2 or sometimes more ignoring whatever the histogram says, then I pull it down in the post as necessary. This way, we are (probably) getting better readings in darker areas of the frame.

  • @Azo You didn't mention which card you are using. Are you using the SanDisk 64GB 95MB/s or something else?

    The settings are specified independently - authors can specify crazy 24H settings and moderate HBR or 720P mode settings as they see fit and there are many, many more variables than bitrate that are specified in each setting.

    In general, I have had much better luck with 720P modes recently than late 2011, for instance. At that point in time, I routinely ran Driftwood settings in 24H that would not work 720P modes - even when the bitrate was much higher (as you mention) in 24H.

    Anyway, the best advice I can give in understanding it is to start thinking of bitrate as only one factor in the difficulty presented to the camera/memory card during recording, as opposed to the single determinant of success. The number of frames recorded per second, the frame limit, the quantizer, etc., are all variables that can make a difference as well.

    By the way, have you tried using 24L in those settings? Often it has a better chance of spanning.

Start New Topic

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Sign In with Google Sign In with OpenID

Sign In Register as New User

Tags in Topic

Top Posters