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.
AVCHD: Extreme settings
  • 278 Replies sorted by
  • @stonebat

    Good test, but I have only one body :)
  • Let's all agree to disagree.

    Determining quality of motion is not easy. Unless one has two GH2 with different GOP settings, it's almost impossible to test... unless someone can tell how to test quality of motion easily with one GH1 body.

    I have an idea for motion test using two GH2 bodies. One w/ GOP3 and the other w/ GOP12. Record a fast moving subject at the same time. Slow down both clips to 10% speed by using Twixtor or AE Timewarp. It will make it easier to spot motion artifacts since the software will exaggerate motion artifacts if there is any.

    But who got 2 bodies? No one? Then we gotta agree to disagree.
  • @awaterman
    @danyyyel
    @mpgxsvcd

    Please do not start this Low GOP discussion stuff.
    As it can go for ages without any useful result :-)
    We already had all this.
  • Well put danyyel!

    Now keep testing!!!!!!!!
  • It is extraordinary how people want to have the end word. They can't accept that we have choice and people might look at things differently. For example, when someone says that he wants a stable 44mbits file because he feels it is enough for him as he does not want the file to be too big. Should I tell that he is an idiot in doing so because I want the best quality possible even if because of diminishing returns even if I can get a small gain for 2 to 3 times larger file. We have all types of threat and avenue of research here. From the extreme to the stable, from low gop to hdmi etc... what is your problem with that. We are only at the start of it, no one knows where all this will end, so why should we draw conclusion now, because you say so!!!!!!!!!!!!!!!

    Even Vitaliy was not a big supporter of the low gop at the start, I don't know what he thinks now but he did not stop it. The innovation we are seeing now is all for people who had broaden their minds and that are happy to experiment.
  • GIVE. IT. A. REST.
  • @Kae

    I really would like to test the requests you have made.

    "Has anyone posted a clip with HEAVY motion (in front of camera + moving camera) with new stable 3.62 settings?"

    However, I need to understand what you define as "Heavy motion(in front of camera + moving camera)". Are you talking about something like a group of people moving quickly in the distance or a large object moving back and forth in the near field. Also how realistic is shooting @ 24 FPS with heavy motion and a moving camera?

    Wouldn't you expect the motion of the video to fail just because the frame rate is too slow? I wouldn't expect RAW 24 FPS footage to look good in the situation you described.
  • @awaterman

    I have provided plenty of tests(The 720p B frame setting, the original 32 mb/sec settings, the highest stable MJPEG settings, the current highest stable 720p AVC-HD settings). The difference is I have demonstrated how I came to all of those conclusions.

    There hasn’t been any evidence to back up what Kae is claiming. Of course high bit rate almost I frame only AVC-HD looks good. However, that doesn’t mean that it affects motion of the video.

    It really doesn’t matter what the hypothesis is. If it has no evidence to back it up I am going to attempt to validate it one way or the other. The problem with validating what Kae is saying is that it is completely undefined and ambiguous. “It just looks better”.

    I would gladly do the testing if I knew what to test and what I was looking for.
  • @mpgxsvcd - I find it amazing that you have the time on your hands to question other peoples findings, without offering anything of any substance yourself. I personally use settings submitted by @kae and find the quality exceptional, so therefore have no need to search for this "holy grail" you seem to seek. We are all here willing to test settings to offer comments on these settings to assist others in finding what they want and we do this in our own time for the love of what we do.

    There may be no benefit at all to low GOP, although I personally agree with @kae on this one, but the quality of the shots I am taking means I am happy with the settings.
  • @mpgxsvcd - please don't derail this thread like that. Kae doesn't have to 'defend' his findings. He likes what he's seeing, and that's that. I was using his 3gop for a while, now I'm liking 12gop with 3.62. But why lay into him him like this. This is just starting a war of words and doesn't help.

    Please keep testing, people.
  • @Kae

    I was just trying to understand what you are seeing. Your explanation clears that up. To me it sounds like you have seen a benefit to motion with a low GOP. You have seen clearer and more consistent motion when you have used the shorter GOPs. However, you can’t fully explain why it happens or what the difference is.

    You accept that the differences you see are caused solely because of the shorter GOP and increase of I frames without any empirical evidence to back that up. That is fine there are a lot of people out there that accept things based solely on faith.

    It looks like you want to share this enlightening experience with everyone else as well. Again that is fine. There are a lot of people out there looking for a quick fix that can be the ultimate savior for their camera. They don’t have the desire to find out how everything works. They want someone else to just tell them the answer without ever considering what the question really is.

    You have stated that you don’t want to debate the validity of your findings. Everyone can choose for themselves whether they want to use the lower GOP or not. You are right we are all free to choose what we believe in.

    You have a loyal following that appears to like the results they are getting. That is great they have seen the light and now they unquestionably believe they have found the correct path. What I say won’t change that. They are willing to accept your teachings on faith alone. That is something that I don’t choose to do. However, we all have the choice. Some of us just have different criteria that we base those choices on.

  • It's hard to imagine the Pentax resolving more than the 20mm, as that lens is just SO sharp and contrasty at f8.

    I'm thinking 66ish may be the happy medium here for now. 44 is not doing it for us yet, so we now have our low set to 48... still have the high at 88. I'll try to do some 66 vs 88 tests today .
  • Ian_T >"to the motha!!!"

    LOL!
    That's me bro!

    Ian, are you shooting at 88Mb/s or 66Mb/s (or none of this shit)?
    I've had no card speed errors (Sandisk Extreme 32GB 30MB/s) with the 20/f1.7 or the 14-140.
    I'm using 66Mb/s FB1,FB2,FL at 3X settings with AQ4.

    Hmm... It sounds like your Pentax 50mm may have a bit more resolving power that's driving the motha to the limit! Although, I thought the Panny lenses are on the high rez side. It may be interesting for you to frame up a rez chart and check it out.

  • @proaudio ...to the motha!!!

    Talking about camera crashes...I have also been using Cosimo's settings. The strange thing is when i use it with the 20/f1.7 it's all good (so far) but as soon as i put on my Pentax 50mm Super Tak manual lens it continuously crashes (card speed error). I don't get it. I'm talking within the first few seconds. I have to move the camera around for it to record at least a little something. This is all indoors BTW.
  • I was using cosmio's 88mbs settings and had card crashes if I started with movement but if I held the camera still for 3 seconds before moving it would record without problems. What would cause this?
  • @Jspatz

    SanDisk Extreme Pro / HD Video / all operate the same in the GH2... marketed 30Mb/s write speed -> effective 25Mb/sec. The Transcend Class 10 has a marketed write spreed of 20Mb/s -> effectively 19Mb/s. So yes, you should see a difference of 25% in speed.
  • "What SDHC card is capable of 45Mb/s? "
    SanDisk ExtremePro 32GB. Despite what others have reported, it handles with ease what fails on Transcend Class 10.
  • Well put, Kae. Each to their own.
  • @proaudio4

    None. Only the SanDisk pro but the GH2 can't handle that writing speed (30mb cards achieve the same performance). So only advantage is writing to your pc via a fast card reader.
  • @Jspatz > What SDHC card is capable of 45Mb/s?
  • 132mbs 3 gop works great on 45mbs card other than no spanning. The same with 12 gop crashes the card.
  • I find it futile and childish to ask someone proof of what they like and don't like. If everybody has choice why should one force the other into his. In french there is a saying like "les gout et les couleurs" Taste and colours cannot be explained, so what, less than 200 hundred years ago all the painters where painting fat woman, go and tell them that their paintings are pieces of shit because nowadays it is more the skinny models than are in vogue.
  • Looks like a long and short GOP rivalry. btw i need 3.61 so a can put Kae too!
  • Well put Kae!