kye

The results of my morning walk yesterday.... Not an official entry (I may as well just put the camera in a box now in preparation for shipping it..) just a kind of proof of concept test. I went for the "it's rubbish, so mess it up and make it look like early film" but I know very little about early film, so there you go. Equipment used: Panasonic GF3, Super Takumar 55mm f1.8, generic 0.7x m42 to m43 SB. Graded in Resolve. Things I learned: It's difficult to focus on a fixed LCD screen with no MF assist functions It's kind-of pocketable and the tiny fold-out Manfrotto Pocket worked well for low angles on the ground, benches, rubbish bins, etc People really get freaked out... one guy walked into the middle of my shot, stopped, pulled out his phone and using selfie mode had a good look at me and took a photo. All the time I'm just standing there looking at the screen on the camera and thinking "get out of my shot! why couldn't you have just walked though it like a normal person!" I should have added way more film-grain in post to survive the YT compression The camera has no manual controls in video mode, but if you set the camera to manual stills mode, 1/50 and base ISO then you can adjust ND/aperture to expose correctly, then when it auto-everything's in video mode it should get a 1/50 with base ISO.... it should do, anyway. This wasn't shot with an ND, just in dark conditions, and some of these shots weren't set up right so shutter angles will be all over the place. I took a photo (RAW still) and video from the same location, then overplayed them in post to compare. Whomever manages to get 4k will absolutely kill everyone else.

The entire point of an eGPU is that it's an expansion for someone who needs a laptop (for whatever reason) but wants to edit quickly when they're 'docked'. I'll use my eGPU to create proxies, and then to do final editing and grading on the original footage when I've finished the edit. It will also be useful for doing quick edits. Surely someone here has gone down this path?

Resolve is moving towards GPUs over CPUs, and v16 is absolutely killing my MBP 13" with its 'you-might-as-well-not-have-a-GPU' integrated Intel Iris 550, so it's time to get an eGPU! I've done a bit of research, but no-where near enough, so I have questions. I've seen the Razor Core X chassis recommended lots of places, does anyone here use it? I've heard about bottlenecks due to the Thunderbolt connections, and potentially that it's Mac drivers and might be fixed, so is it worth getting a super-duper card? The 1080ti was the card to beat for ages, is it still right up there? or is there another card that is better value? Who else has taken the plunge?

On a recent trip my wife took pictures with her iPhone and I took video with the GH5. We often took images from the same vantage points one after the other so there was ample room for comparison. One thing I noticed when I started grading the images was how lovely the iPhone photos were in terms of colour and processing. I've got it on my list to try and break down the iPhone colour science to learn some stuff. It's an interesting thing, the debate about convenience vs output. A friend of mine who is into stills photography still uses his Canon 40D (or maybe it's even a 30D) because although he's done several major trips (eg, weeks going through Europe) and really tried to use his iPhone, he says that "the only photos I give the slightest f@#$ about came from the DSLR". For him the matter is closed - no smartphone picture has ever done it for him so taking a picture with a smartphone is basically like throwing the moment away. It's not like it's about anything other than the image either: he's a minimalist, hates the size and weight of the DSLR, he prints the odd picture but mostly they go out in a regular email update to friends and family, and he part owns and runs a tech company doing programming and virtualisation of services in the cloud. He's the perfect person for a smartphone, a parent who values convenience, is completely connected online and has disposable income to buy whatever smartphone he'd like, but nope! Do you think it's the shutter delay that gets in the way? or something else specific that you can think of? I've done street photography with both and apart from stealth, the real camera wins in every other department IMHO.

There are, they're just not what you'd call 'competitive' Australia has historically been subject to the 'Australia tax' which makes new goods about 40-80% more expensive due to lack of competition through wholesalers having a monopoly across a geographic area, and while this is currently diminishing, I still get a lot of "this item does not ship to your location" on Amazon, and retailers here haven't really adjusted to internet pricing. In terms of local second hand goods, shipping costs somewhat kills price competition: People here who are spending even a few thousand on luxury items like electronics, fashion, or even art, often fly to Asia to buy and then bring things home in their luggage as the total cost is less than buying locally! It's a good thing though, it somewhat curbs my bower-bird-like curiosity ???

Nice images!! You win the $150 challenge! oh, hang on...

I took my GF3, M42 to M43 Chinese SB, Super Takumar 55/1.8 and Manfrotto Pocket desktop tripod for a walk this morning and grabbed some shots as a sort of concept test. The GF3 has terrible IQ, no controls in video mode except WB, and a fixed screen with no MF assist function, so focusing it when it was on the ground looking up was a little.... challenging. Especially wearing full corporate uniform! But, I shot something when I otherwise wouldn't have, so that's a good thing I think "below average" might be being kind! But I bought it as a stills camera and for that it didn't disappoint. Combined with the pocketable 14mm f2.5 pancake lens mine has racked up enough flier miles to go around the globe more than once. I bought mine new with the kit lens, 14mm f2.5 lens, and all the extra batteries, filters, screen protectors, rocket blowers, and practically everything else available from china that might be useful for just under US$400. When I dropped it and it landed on the corner where the touchscreen was closest to the edge the metal chassis took the shock and apart from a slight wrinkle in the body there was no other damage. 2 weeks! 2 weeks!?!? You're hilarious!! Shipping to Australia often costs more than the item and takes MONTHS. When I was buying lots of old valves/tubes from the eastern block the ebay feedback time limit would close before I received the goods in the post! I had items take 10 weeks sometimes!! I received a package on Monday that I'd ordered in mid-May from JAPAN. A modern country with effective logistics infrastructure, strong trade relations, and is even in the same timezone as me! 2 weeks.....

I'm one step ahead of you. I already have MFT lenses, so... Find list of Panasonic m43 cameras <check> Ebay search GH2 under $150 <nope> Ebay search GH1 under $150 <maybe....> Ebay search G7 under $150 <nope> Ebay search G6 under $150 <nope> Ebay search G5 under $150 <nope> Ebay search GHX1 under $150 <maybe....> Ebay search GF6 under $150 <nope> Ebay search GF5 under $150 <maybe....> Find list of Olympus m43 cameras....... But hang on, how good are these cameras? Crap, I don't know anything about any of these, except that my GF3 video quality is gonna get laughed off the internet.....

@Andrew Reid - good job on the rules I also kind of like the idea that if a cameras current value is less than $150 then someone can still use it, considering that someone else who didn't own it can buy it and use it, and also that any camera currently devalued to that probably isn't in use any more! but happy with your call. That's because resolution really matters!! Look.... ??? Also, and completely unrelated question for the group. I have a particular new project coming up and am looking for recommendations for a camera that is as cinematic as possible and that I can buy for less than $150.....

Wow.. 8 weeks and that big. He's gonna be huge!! Better start writing your masterpiece so you can afford to feed him!

Oooh, I like the manual lens only idea!!

Never apologise for being a nerd. Never.

As cliche as they are, Gorillapods are very popular with vloggers for many good reasons. Lots of vloggers walk about talking to their DSLRs on Gorillapods, with hardly any bothering with a gimbal, and people aren't going crazy talking about stabilisation for those shots.

Lots of cool ideas and ways that this could be done. Another idea is that it can only be $150 (or whatever the amount is) for everything in production. That would make people trade off how big their SD card is with how fast their lens is with what mic they could use, etc... it might need to be more than $150 though. Also another suggestion that people can enter an unlimited number of times. This would encourage people to make more films and be less precious about them, which is also a good thing. Cool idea. I'm kind of reminded of the Top Gear challenges where they buy old sports cars that are now falling apart If we do standardise on a singe lens, the Helios is a good choice - everyone can use a crazy-bokeh 50mm FF lens, after all!

True. Anamorphic doesn't seem to tickle me in quite the same way that it seems to tickle others though. I quite like the aspect ratio though.

Awwww! How cute! How old is Bruno? (my wife wants to know)

Every technical choice has an aesthetic association. It might not be the same association across different groups of people, but everyone will have some kind of association. It would be the perfect choice if you wanted that 80s video look, the 90s home video look, or you were creating something where there was altered perception involved, like someone with a migraine goes outside into the sunshine. Everything depends on the outcome you're trying to achieve. Including trying to impress people on the internet with your camera tests....

It will likely be here sooner than we expect, but I don't expect it for a decent amount of time, so even if it's a lot sooner it will still be a while Those demos were sort-of interesting, but some were considered standard computer science graphics algorithms and were being given as assignments back in the early 90's. I'd certainly hope that the processing power would be available to do such things, being that Moores law has had almost three decades to work on it! We do have face-detection built in, which is a large part of it, but I recall that the harder part was getting computers to do object recognition in the less-than-perfect conditions. I remember watching a 'latest tech' type show on TV many years ago and the guy was talking about computer vision, and how the computer could recognise a spoon pretty reliably, as well as many other kitchen cutlery they'd programmed into it, but when they pointed it at the drawer with all the random serving spoons and tin-openers that every house has, it was pretty much useless in partial matching things. I'd suggest that there's a similar problem with recognising things in a 3D environment when they're not in focus, especially if you only have one lens to try and extrapolate depth from. It's worth noting that the depth extraction algorithm was from a stereo pair of images, which probably had a very deep depth of field. Humans, on the other hand are pretty good at depth perception using only one eye, using things like size, relative motion, parallax, contrast, and other factors to build a 3D model from a series of 2D images. Maybe the AI of the future will have a few cameras on the front of the camera for analysis, or even like phones do where one is a low-res depth sensor.

I haven't been following this thread as this isn't a camera for my style of shooting, but wow.... those specs! Perhaps the most interesting mode for me would be "6144x2592 @75fps max (DCI 6K 2.4:1)".. 6K, widescreen, 75fps is a very interesting resolution! and who knows, there might be a later firmware update to get higher frame rates from the lower resolutions perhaps?

I shoot handheld with the GH5 and manual non-IS lenses, so I'm familiar with the territory. There are lots of effective options: Keep the GH5S and pick any combination of: put it on a rig (eg, shoulder-rig, easy-rig, gorillapod, etc) that gives more stability / ad an IS lens / shoot a wider lens and add stabilisation in post Get a GH5 and pick any combination of the above Go Olympus which is meant to have the best stabilisation and use any combination of the above Choose a FF / S35 camera and use any combination of the above I'd suggest you also take into account the whole workflow... Keeping the GH5S will ensure maximum compatibility of files and image / colour in post, getting a GH5 will keep similar image / colour and also share the same lenses, Olympus will share the same lenses and have better stabilisation but different colour, S35 / FF will need different lenses but have different image / colour, etc.. It's all a trade-off.

I've bought Wasabi batteries and not had an awful time, but I can't claim that I've really tested them hard either. There's no reason that original batteries will always be better. The problem is that after-market batteries are probably price-driven and are therefore likely to be of lower quality, but there's no reason that an after-market supplier can't make a superior product if they decided that there was a market for it. Some after-market batteries claim to have larger capacity or are more durable, and a company could do well and get a reputation if they actually delivered on this. On the other hand, manufacturers often compete on pricing on the camera body, and make more profit on the peripheral items such as lenses or batteries. Yes, they have a brand name to protect with reliability etc, but in any market with an established name the reason that newcomers often grow and steal market-share is that incumbents are often price-gauging...

I think that blown out highlights might be the worst aspect of poor quality digital, for me at least. Worse even than poor resolution, over-sharpening, or low-bitrates with macro-blocking. That's one of the primary benefits of shooting with 10+ bit depth, you can expose for the greatest DR and then choose your middle exposure in post while applying a nice highlight roll-off, and do it without breaking the image. To this end, I haven't really noticed the DR of my GH5 being too low as I can normally get the exposure I want and also not clip highlights.

All of the above, plus many more things you didn't list. Saying "Shouldn't a mAh be the same" is kind of like saying "Shouldn't all 4K be the same"..... Battery science is complicated.

I completely disagree. I shoot very fast run-n-gun using auto-WB and available (normally mixed) lighting, so I know all about what you have to do to correct WB in post when it's not right. When I was shooting my 700D I struggled to get barely usable images, and I thought it was me until I did an A/B under controlled lighting to discover that if the WB is right then the image is fine with zero adjustments but if it's off, even by a little bit, then you're basically screwed. I was also dissatisfied with the sub-720p resolution from it, did some research and bought the XC10 for its might higher video quality and bitrates, but that suffered these same problems as the 700D. The C-Log on the XC10 was better than the more normal profiles, but still hard to work with, and I still found the image to break apart easily. I now shoot with my GH5 and the difference is like chalk and cheese. The GH5 in 10-bit is glorious to work with and you have to be brutal in grading to even degrade the image slightly, and WB adjustments in post are straightforward. The GH5 even in 8-bit modes is much more forgiving and I have now realised that my problem wasn't me, it was the difficulties associated with Canon CS when not white-balanced appropriately. Ironically, I found Canon with ML to be a completely different beast and suffers none of the problems grading shots with WB problems, so it's not the Canon hardware that is causing these difficulties, but the image processing - the magical Canon CS.

I agree. I think their theory that AI and image processing will eventually be the winning tech is probably correct, but we're likely a good amount of time away from that, enough certainly to implement phase detect and develop the DFD technology in parallel. The two technologies kind of compliment each other in a way, with AI learning what should be the subject of the video and which direction the focus should be in, and PD confirming which direction to go for each area of the frame. Still, who knows what this stuff costs to implement, so maybe the numbers actually do stack up, they'll be laughing as they pull ahed of the other manufs when the DFD works well, and they'll have saved some money along the way, but who knows