ac6000cw

18 minutes ago, John Matthews said:

I'm fairly certain all these cameras will have moiré patterns at some point whether it be 1080p, 4k, or 6k. 

I agree, but some are better at minimising the impact of it e.g. like the GX85/G95/GX850 you mentioned above. Moire in 1080p is more noticeable on the Oly E-M1 ii & iii than on those, despite having a higher-res sensor (and noticeably soft 1080p video).

1 hour ago, kye said:

The 6600 does not have an OLPF, and the moire is present the moment the light is collected on the sensor. 

This is the same comparison with 4k video - https://www.dpreview.com/reviews/image-comparison/fullscreen?attr18=daylight&attr13_0=panasonic_dcgh5&attr13_1=panasonic_dmcgx85&attr13_2=sony_a6600&attr13_3=panasonic_dcg95&attr15_0=jpeg&attr15_1=jpeg&attr15_2=jpeg&attr15_3=jpeg&attr16_0=200&attr16_1=200&attr16_2=200&attr16_3=200&attr171_1=1&normalization=full&widget=478&x=-0.5653754912361596&y=-0.0730609051110552

...and this with still images (same test image and same cameras, so same sensors and filters) - https://www.dpreview.com/reviews/image-comparison/fullscreen?attr18=daylight&attr13_0=panasonic_dcgh5&attr13_1=panasonic_dmcgx85&attr13_2=sony_a6600&attr13_3=panasonic_dcg95&attr15_0=jpeg&attr15_1=jpeg&attr15_2=jpeg&attr15_3=jpeg&attr16_0=200&attr16_1=200&attr16_2=200&attr16_3=200&attr171_1=1&normalization=full&widget=478&x=-0.5653754912361596&y=-0.0730609051110552

Note the A6600 moire is much reduced in 4k video and is almost non-existent in stills mode.

So the bad 1080p moire on the A6600 isn't coming from "the moire is present the moment the light is collected on the sensor" (which would show up in stills mode), it's being caused by the way data from the photosites is being collected, sub-sampled and processed. I suspect the image data used for 1080p video on the A6600 is being line/pixel skipped on sensor or as it's read out from the sensor and then de-bayered. The GX85 and G95 1080p is probably pixel-binned on-sensor, so it's better but has some moire, whereas the GH5 generates 1080p by reading the data and de-bayering it at high resolution then resamples it (using super-sampling) down to 1080p.

1 hour ago, kye said:

Once that happens there's nothing you can really do to get rid of it.

I agree - although I think some cameras/manufacturers do a better/cleverer job of minimising it at the de-bayering stage

On 10/26/2023 at 5:59 AM, John Matthews said:

So what I was saying earlier about a correlation between pixel density and moiré is true. The less space there is between pixels, the less chance of moiré. In this case, 12k capture is better than 4k when it comes to moiré.

On the G100, it really sucked for 1080p moiré at precisely the distance one would vlog at. Go back 50 cm and moiré would be gone due to the frequency of the patterns. I imagine the GX80/85 has a similar problem in 1080p.

On 10/26/2023 at 9:27 AM, kye said:

I don't think it works that way though.  A 12K sensor would be sensitive to moire if there were repeating patterns that happened to align with the gaps between the pixels, just like a 4K sensor.  

It might be that common causes of moire are around a certain size and therefore impact one combination of sensor resolution / sensor size / and focal length more than other combinations.  Also, lower resolution sensors might be more prone to moire as they're typically older and there were larger gaps between the pixels than there are now.

Lower resolutions are likely to have issues on cheaper cameras too, due to the camera line skipping and therefore effectively creating very large gaps between the active pixels.  

Sadly, there's lots of different ways to create moire, and many of them tend to come from strategies to make the product more affordable!

The older Sony APS-C cameras like the A6500/A6600 are/were terrible for moire in 1080p - this is a DPreview 'video still' test chart comparison of GH5, GX85, G95 and A6600 at 1080p - https://www.dpreview.com/reviews/image-comparison/fullscreen?attr29_0=panasonic_dcgh5&attr29_1=panasonic_dmcgx85&attr29_2=sony_a6600&attr29_3=panasonic_dcg95&attr72_0=1080&attr72_1=1080&attr72_2=1080&attr72_3=1080&normalization=full&widget=494&x=-0.4377130790929456&y=0.008264403455490291

(The GH5 shows how sensor to video processing should be done, the A6600 shows how not to do it!)

In modern 'stills' cameras based in Bayer pattern sensors, the moire is basically caused by the large gaps between the different colour sites, so you get serious colour aliasing (when there's no effective OLPF in front of the sensor) if the light pattern happens to match the photosites of a single (R, G or B) colour.

From https://www.redsharknews.com/production/item/5209-what-really-causes-moire-in-cameras :

2 hours ago, John Matthews said:

I already have most of the other lenses you mention with the exception of the 35-100; instead, I have the 40-150 Pro f/2.8. I'm not sure it's a better lens.

At twice the weight and 1.6x the length of the 35-100 F2.8, I didn't even contemplate the 40-150 Pro F2.8. I did consider the Oly 40-150 Pro F4 as it's about the same size & weight as the 35-100 F2.8, but decided I'd prefer the faster aperture.

Compromises...

Basically my target weight limit for body + lens is around 1 kg.

4 hours ago, John Matthews said:

Just thinking through a big camera and small camera setup in terms of lenses, I think it makes sense to alternate between "pro" lenses and "consumer" lenses throughout the popular focal lengths. For example in MFT,

fisheye, 10, 12, 17, 25, 42.5, 75, 150, 300 (or more)

This would allow to mix and match better and have the right tool for the job. In the past, I thought I'd just copy the focal range for the pro and consumer camera, having the same focal range in both, understand better the focal length and essentially working consistently. For example,

Pro: 12-35, 35-100, 200 (don't have any of these lenses)

Consumer: 12-32, 35-100

Conversely, one could just choose the 10-25 and 25-50 (don't have those either) to cover most of the focal range, then choose small primes on the consumer side.

Maybe some could double as a lens in both situations (example, use the 9mm for both pro and consumer if it passes mustard).

Fact: the G9 ii is big for MFT; maybe the features will mitigate this, but the need for a smaller camera (for me) is necessary. The question is: which focal lengths should be pro and which should be consumer and not let the size of the g9 ii to get "out of hand".

I just curious to know what you would do if you need two systems (big and small), what would you choose/have you chosen?

 

If I had to choose just one lens to keep out of my collection, it would be the Pana 14-140mm F3.5-F5.6 - for the range it provides it's relatively small (75mm long) and light (265g), it supports dual-IS2, focus breathing isn't noticeable and on my copy at least the zoom ring is reasonably smooth. It's been my most-used lens ever since I got it as a bundle with a G6 years ago.

If I could keep a few more, then...

The Oly 12-40mm F2.8 - just a great all-round, reasonably bright, moderate wide to moderate telephoto, no noticeable focus breathing, 85mm long and 382g. (The Pana 12-35mm F2.8 is a bit smaller and lighter and gets you dual-IS - I was considering both a while ago and then a really good used deal on the Oly came up so I went for that).

The Oly 75-300mm F4.8-F6.7 - what I use most often for wildlife, pretty small (117mm long) and light (423g) for the focal range, and decently sharp for something that small/light/cheap.

Low light? usually just chuck the Pana 25mm F1.7 into the camera bag or pocket in case I need it - cheap/small/only 125g.

I also own the old Pana 20mm F1.7, but that's noisy when focusing, so I might swap it for a 17mm or 15mm fast AF prime at some point (Oly or Pana Leica probably).

I've recently acquired a used Pana 35-100mm F2.8 (to complement the Oly 12-40mm F2.8). Decent lens and not too large/heavy for a 'pro' lens, but it's got noticeable focus breathing so not ideal for video C-AF use.

(and yes, the Pana 12-32mm pancake is great when you want to keep things as small as possible, especially if you're taking two cameras on a trip, so that would be a 'keeper' too).

17 hours ago, John Matthews said:

When I look at the 4k 120fps files, their only 100 Mbps in HVEC. Can a v30 card handle 5.8 open gate?

According to the G9ii specs, 4k/C4k 120p is 300Mbps HEVC.

These are the open-gate specs:

...so it's 200 Mbps up to 30p 4:3 and 300 Mbps up to 60p 17:9.

I've captured Olympus C4k at 237 Mbps and G9 150 Mbps 4k50p to Sandisk Extreme (non-Pro) 128MB & 256MB V30 UHS-1 cards with no problem, so 200 Mbs should be fine to a high-spec V30 card (V30 rating => 30 MBs => 240 Mbps sustained write speeds). In reality I've found the high-spec, larger capacity (128GB or larger) V30 cards are very fast, limited as much by the UHS-1 interfaces as anything else, so you may find that 300 Mbps works fine.

Datasheet performance figures for a UHS-1 Sandisk Extreme (non-Pro):

...and the Extreme Pro version:

Quote

59.94Hz [C4K] 4096x2160

119.88p, 300Mbps (4:2:0 10-bit LongGOP) (H.265/HEVC, LPCM)*

Quote

59.94Hz [4K] 3840x2160

119.88p, 300Mbps (4:2:0 10-bit LongGOP) (H.265/HEVC, LPCM)*

(Quotes from Panasonic UK G9ii specs. It supports 100p as well)

1 hour ago, John Matthews said:

Am I correct in saying the G9 ii does 120fps 4k, but the GH6 only does C4k 120fps?

According to the specs here https://www.panasonic.com/uk/consumer/cameras-camcorders/lumix-mirrorless-cameras/lumix-g-cameras/dc-g9m2.html#specs yes, the G9ii does 120p in both UHD and C4k

Re. S5 ii moire and aliasing tests, the test-chart videos here are pretty comprehensive - https://www.optyczne.pl/70.4-Inne_testy-Panasonic_Lumix_S5_II_-_test_trybu_filmowego_Jakość_obrazu.html

For comparison, these are the GH6 tests - https://www.optyczne.pl/62.4-Inne_testy-Panasonic_Lumix_GH6_-_test_trybu_filmowego_Jakość_obrazu.html - and this is for the OM-1 - https://www.optyczne.pl/59.4-Inne_testy-OM_System_OM-1_-_test_trybu_filmowego_Jakość_obrazu.html

Unfortunately they haven't tested the G9 ii yet...

11 minutes ago, John Matthews said:

I learned a couple of things after watching a Lumix live stream. DFD is still used for S-AF because it's faster than PDAF. I've also heard that when working in low-light and small apertures (f11), only contrast is used due to PDAF not being efficient- this is true amongst all brands.

Yes - as far as I know, all mirrorless cameras with PDAF use some form of contrast detection to 'fine tune' focus after PDAF has got it close (and more so in low-light and with small apertures - I guess diffraction affects PDAF in that situation). Also I remember Olympus saying that C-AF in video (on their cameras with PDAF) uses a mixture of PDAF and CDAF.

But in reality, I think with modern cameras the main C-AF issue is how they decide what to focus on (and then track it) - the 'Ai' part - rather than how they do the focusing.

23 minutes ago, John Matthews said:

I do believe, perhaps falsely, that pixel density has a major role- the more, the better; I think M43 on the GH6 or G9 ii definitely has that going for it.

Provided the camera can process all the pixels on the sensor to produce the final video stream i.e. no line-skipping or pixel-binning,, then yes, I think in general a higher density is better. It's quite noticeable that the Panasonic S cameras are prone to it (but the high-res S1R not so much), but their micro4/3 cameras are not - but how much of that is due to processing differences and how much to pixel density plus lens resolution I don't know.

But if slow-mo is important to you, I guess the 4k @ 120p and FHD @ 240p on the G9ii also has to swing things in that direction?

4 hours ago, John Matthews said:

The one thing that bugs me about the S5 ii is possible moiré. I've never seen it on the GH6 and I bet the G9 ii will be the same.

That's one major reason I decided against an S5 ii - the others were the size & weight of long telephoto lenses and the APS-C crop in 4k50p/4k60p (I normally shoot everything in 50p, so it would effectively be an APS-C video camera for me).

So sixth months ago I looked seriously at more upmarket APS-C cameras for video instead, but having played with an R7 and an XH2s I decided they felt too large and/or awkward in my hands (the Sony A6700 wasn't announced then, and the FX30 has no viewfinder so ruled out). Then the XS20 was announced and I pre-ordered it... but before deliveries started a used OM-1 turned up at a really good price so I bought it and cancelled the XS20 order.

So in the end the fact that I already had micro4/3 lenses and the OM-1 just feels 'right' in my hands (and is weather sealed, with top-notch IBIS and a lovely 5.76M dot EVF) kept me in the micro4/3 world a while longer...

9 hours ago, MrSMW said:

But then again, there is something to be said for seeing an entire shoot in B&W through the viewfinder and only later revealing the results in colour…

I assume that's really the main purpose (beyond being a marketing gimmick) of having a dedicated switch position - enabling a quick flip to B&W for composition purposes, especially if intending to generate some B&W pictures from the RAWs afterwards.

1 hour ago, Dustin said:

Interesting test but I don't think I'd want to go to M43 at this time.

I didn't think you would, but I was surprised at how bad the Fuji XH2 was (even in IBIS + DIS mode) and how good the Canon R7 was, compared to the others.

There's an interesting IBIS comparison including the XH2, R7 and S5II and several micro4/3 cameras here: 

On 9/21/2023 at 11:34 AM, BTM_Pix said:

I'm not concerned about flat out image quality as the power zoom trumps that for this particular application but the distortion makes me curious so a non-corrected RAW to jpeg conversion would be great if you can possibly manage it.

I've PM'd you 🙂

I own the lens (bought used a few years ago, don't know how old it actually is).

I haven't used it recently, but yes, from memory the power zooming is reasonably smooth but my copy is a little bit noisy (but quieter than the Pana and Oly pancake PZ lenses) - from comments I've seen, there seems to be some sample variation in noise levels. I can't remember what the focus breathing is like.

Do the BM cameras apply the lens distortion and chromatic aberration corrections?

It'll be a few days before I can do it, but I could put it on my OM-1 or G9 and upload a short test video if that would help you decide?

IIRC, around the time of the GH6 release (18 months ago), there was live stream/press discussion with some senior Panasonic people, where one of them basically said that including PDAF hardware on the sensor would have delayed the sensor development too much (for the GH6 timescale). 

So my take on it is the sensor in the G9ii is effectively the 'fully developed' version of it, and the GH6 has an intermediate version without the PDAF hardware capability. I think it would have been commercially crazy to not launch the GH6 with PDAF if they could have - they'd have sold more in the intervening months at higher street prices.

We'll have to wait and see, but I think there just has to be a GH6ii or GH7 soon.

1 hour ago, sanveer said:

I wonder if the G9 ii overheats,  considering they removed the fan on the S1ii size body. 

I've just watched/listened to the PetaPixel (Chris and Jordan) and Geeky Nerdy Techy YouTube reviews, and both of them reported that at room temperature anything below 4k120/4k100p mode recorded until the battery died (which I think for 4k60p was around 1h 20m - 1h 30m). 4k120p overheated after about 25-30 minutes.

So pretty good I think.

Interestingly Jordan said the G9ii might become his favourite everyday video camera - it would be a bit ironic if the G9ii turns out to be a serious GH6 killer, despite being supposedly aimed at stills shooters...

Both reviews said the automatic (versus manual on the GH6) 'dynamic range boost' function works well, and also the IBIS is extremely good (as in 'best we've ever used').

One thing I am a bit surprised at is not upgrading the viewfinder - it's the same 3.68M dots resolution as it's predecessor, whereas the OM-1 has a (really nice to use) 5.76M dot panel.

I also have to agree with dpReview's comment:

Quote

So it's not obvious that Panasonic has answered the prayers we've heard from Micro Four Thirds shooters with the otherwise solid-seeming G9 II. Even though it's chockablock with video goodness, it's not the 'GH6 with better autofocus' fans were clamoring for, and even though it has some slight stills advantages over the OM-1, it's hard to imagine anyone passing up that more compact camera for this bigger option with worse battery life.

Initial review and specs on dpReview - https://www.dpreview.com/reviews/panasonic-lumix-dc-g9-ii-initial-review-the-g9-series-matures

30 minutes ago, BTM_Pix said:

Its always been there in the background of course

Yes - most of the camera manuals I have say something like 'recording times may be reduced in high ambient temperatures'.

31 minutes ago, BTM_Pix said:

and the R5 was a wild ride while we were investigating it but it’s back with an increased vengeance recently.

I wonder if some of this is because the R5 situation bought the general issue to the attention of a lot more people (as it was discussed ad-nauseum at the time)?

Consequentially influencers & reviewers now have it on their 'must test it and/or say something about it' list - and of course they tend to choose the most computationally demanding video modes to test overheating, rather than the modes that probably 95% of users might choose day-to-day.

47 minutes ago, BTM_Pix said:

I’m curious because manufacturers seemingly show no interest in tackling it definitively and I would have to guess that it is because they are comfortable with the amount of, erm, heat they receive from dealers and customers about it.

I assume it's just that making the camera smaller and lighter (combined with a long enough features list) probably sells more product than putting a fan in it (to make it more appealing to a minority of video-orientated customers). I know a reasonable number of people who have and use reasonably 'serious' cameras, but mention to them that I use mine a lot for video and the reaction is often almost 'what's video' or that they have no real interest in shooting it...

I'm sure the design engineers could fix the thermal issues, but if that would make it noticeably larger and more expensive they probably get overruled by the product management & marketing side of the company (unless it's a hybrid camera that's very much aimed at video users, e.g. like the GHx, S5iix and S1H). Even the (original) G9 has a 10 minute 4k60p limit whereas it's very close relative the GH5 has much longer record times...

10 minutes ago, BTM_Pix said:

Namely, the GX80 and the LX100.

Putting a mic jack on them and giving them 10 bit 4:2:2 internal and VLOG would be a piece of piss for Panasonic and they could sell them all day long at their equivalent launch prices.

They might have to push the boat out and put IBIS in the LX but I don’t even think that is a deal breaker.

...Or even put IBIS and PDAF in a version of the G100... (but the GX80/GX9/LX100 are much better looking cameras from a style point-of-view).

If Sony can sell small, expensive and thermally-compromised cameras I'm sure Panasonic could if they had the right product...

4 hours ago, IronFilm said:

Yes, the way I understand is that:

1) G9mk2 is announced and immediately (or fairly soon afterwards) available to purchase

2) while "the development of" the GH6X/GH7 is announced, with their available being either late this year or perhaps next year 

As the G9ii looks like it's using a version of the GH6 25Mp sensor with PDAF added, a GH6ii/GH7 is probably not far away...

As an OM-1, G9 and GX80 owner, I'm not really interested in the G9ii, but a GX80/GX9/EM5/OM5 size camera with mic jack and the 25Mp PDAF sensor inside would be very interesting!

Now released:

A7C ii - https://www.dpreview.com/reviews/sony-a7c-ii-initial-review-not-so-small-improvements

A7CR - https://www.dpreview.com/reviews/sony-a7cr-initial-review-high-resolution-in-a-small-package

Sony certainly seems keen on filling every niche they can with a camera...

If Premiere supports VST audio plugins, could you use a delay plugin to time-shift one of the audio clips to fix the phasing issue, instead of shifting it directly on the timeline?

Another thought is that there are auto-align/auto-phasing VST plugins around, designed specifically to deal with multi-mic phasing issues in a multi-track mix e.g. like this one from Melda - https://www.meldaproduction.com/MAutoAlign

(Note I've never used that particular plugin, but I use other Melda plugins so I happen to know it exists).

Alternatively (but more work initially) could you fix the phasing issues as you do already, then export the fixed & mixed audio for the clip to a WAV file, then put it onto the timeline to replace the original audio for the clip (then group it back with the video)?