VanDaven

I do not even dare to ask why the designers of the FS5 have included a gain / ISO switch in their plans... I remember the times I was shooting ENG material on this extremely robust Canon XF305 that was REALLY not intended for low light shooting. I bought it back in 2010, and it had a gain switch. If you'd switch the camera to +12 dB of gain you would see noise. Mainly noise. It didn't have any macro-blocking-squaredance party to put some relief on the codec in order to break the image whatsoever. It also didn't have much banding. It was using a codec very similar to the C300, 50 mbps for 25p of 1920x1080. The BBC bought lots of them if I remember correctly. When there was noise in the image, you would either leave it because it fit the situation it was shot in or you would try your best with NEAT (which often worked). 

Honestly, from deep in my heart, the FS5 is the most disappointing product I bought in a very long time, especially after reading some of its defender's statements recently. It's good enough for what I use it for, so I will keep it (as I have already sold the tools I used for those specific jobs before). Don't get me wrong, there are a lot of good things about it apart from the image problems (the hot topic of yesterday): Ergonomics, 

But from the rattling handle (sorry: smart grip) to the incompatibilities with certain 3rd party batteries and the "built-in fixed non-removable" noise reduction it makes the "worst camera of 2015" quite a star.

It's ISO 3200 in SLog, All other Gammas have different base ISO values (between 650 and 1000).

Very nice, was the FS5 one shot in SLog? Really wondering why there is so much more detail than in the 1DC shot...

Rolling shutter is also nowhere nearly as bad as on the A7SII. 

It's a different sensor. It's a different product. 

It won't happen at ISO3200 in SLog because that's the base ISO of the profile. Like ISO1000 is the base ISO of Cine4. It will only happen at higher ISOs (= adding gain to base ISO). In SLog, which is Sony's approach to mimic CineEI / film stock properties, you can only apply a sensitivity boost by applying Gain. 

In all values above base ISO, no matter what Color mode or Picture Profile. I am very certain it has to do with the sensor readout + noise @ UHD. Did you check my Vimeo link earlier in the thread? My 4k upgraded F5 does not show this problem at similar or higher gain settings.

Great thing that you will report this to Sony directly, so thank you for that. If you report to them, then please include the information that it ALSO does the edge ripping in SLog just like I stated before (you might either over-read this information or just simply ignored it). Just switch it to Gain instead of ISO and put the gain over a value of +9dB and edge tearing will start to appear. Again I'd like to point out that the edge tearing will only happen once the camera is set to a sensitivity beyond base ISO. 

No, it's a different sensor. The sensor inside the FS5 is practically the same they also use in the F55 / F5 / FS7.

The aliasing in those Canon 5D3 shots really gets me every time. Prime reason why I sold mine and only shot very few commercial projects on it. 

It occurs in SLog, but, of course, not at base ISO, I already stated that earlier. If you want to get a higher sensitivity in SLog, just choose GAIN instead of ISO, boost it and see for yourself. Hopefully it can be fixed by a firmware update. See, there's a lot of different processing between the different cameras, that's why you get different ISO ratings on the F55, F5 + FS7 and FS5. A 200% crop of a UHD file on a HD timeline is basicaly 1:1 pixels, by the way. I also stated that it's noise disguised as macroblocking, not true macroblocking from the codec. It doesn't appear in any of the base ISO settings of whatever gamma curve you choose. 

And don't get me wrong, just because I pinpoint things I discovered for myself doesn't mean I don't like the camera. I purchased it for my business and I am very happy that I finally found a portable little beast that shoots in broadcast-compatible quality (HD). As mentioned earlier, the button layout is really superb, features like Clear Image Zoom, VariND and Auto Lens Correction (on Sony E-mounts only at the moment) are unique in this class plus the audio quality is really good too. 

In case you wanna see them, here's a quicker waver test: https://vimeo.com/150138698

Note problem is the noise disguised as macroblocking

Congratulations on your purchase! I also bought the FS5 shortly before Christmas for a really good price and I absolutely amazed by the possibilities. It bridges the gap between the really superb but artistically "crippling" S16 Canon XC10 and my larger Sony F5 (with 4K upgrade): Still lightweight enough for doc and news in a faster-paced environment and compatible with my smaller slider and crane from the 5D3-days. 

Color-wise the F5 / F55 and the FS5 are very similar (gamut is the biggest difference, the more you pay the more you get), and what's good enough for Netflix' Marco Polo and Big Bang Theory should also be good enough for us, shouldn't it ? I think the SLog color temperature profiles originate from the CineEI mode of the larger models, kinda like choosing film stock. SLog is rarely used here, my dearest profile on the F5 will always be HyperGamma7 (CineGamma4 on the FS5 goes a bit into that direction). 

I personally think the best advantage in comparison to C100/300 etc. is the button layout, it is quite hard to hit the wrong button on the FS5 even without looking. 

Beware of third party batteries: I bought mine together with two SWIT S-8U63, and if you insert them fully charged, there is a high probability that the internal clock (including timecode on free run) will be reset. As long as the charge is not shown on the display, formatting SDXC cards is also not possible. Other users on the facebook group also reported other third party batteries to do the same thing. 

8bit UHD SLog will result in a lot of sky banding, so it's better to stick to CineGamma. XAVC-L has the same problems as on the FS7, especially at higher-than-base ISO settings. From my opinion, it's still usable.

Apart from that, enjoy the FS5, waiting for your review,

Happy New Year, 

Martin

Theoretically it should even be possible on SDXC, at least with SanDisk Extreme Pro cards...

Like with the XC10, I wouldn't believe all those horror stories you hear about it on the web. If you have any specific questions, feel free to send me a PM. By the way, the FS7 also has some deficiencies in comparison to the F5 / F55, they are just not as obvious. 

As an owner of both the XC10 and the Sony FS5, the only thing that I can tell you is that the XC10 is way less disappointing than the Sony. Not that the FS5 is a bad camera, but the expectations were much higher...

Ummm... no macro-blocking, nope. That is the way shallow DOF looks on a S16 sensor. In comparison, a S35 sensor has a much larger area, while S16 of XC10 is closer to the "traditional" 2/3" PDW sensor. Very nice for TV news or mini-docs. 

Makes sense. Well, I almost purchased a C300 last year for exactly the same purpose that I got the XC10 now (cheaper and more portable than the F55). I would be quite annoyed if I would have. Btw, I delivered (sundown / rather low-light) 4K b-roll footage from a regional event to a well-known Austrian broadcaster a while ago and everyone thought it was from the F55... 'nuff said.

Thank you very much for your input on this matter. While I understand that all sensor designs have their aliasling limitations, real-world testing of a C300 (large format single sensor) vs. XF305 (3-chip design) a while back showed that the C300, even though utilizing oversampling via a 4K sensor for HD recording, was way more sensible towards moiré and aliasing than the XF305. 

In regards the XC10, I can only tell you and the rest of the community that it has been the most versatile addition to my image gathering tools  in a long time. The footage, no matter if 4K or FullHD, is easier to work with than whatever comes out of my GH4 and color-wise (especially in regards to reds) it's also more forgiving than my F55 (which, on the other hand, is 10 bit footage which is undoubtedly an advantage). 

One could argue all night whether it's "true" 4K or not, but in the end, it's a magnificent little camera that makes my work easier. 

Erm, ok, I think there are some substantial things that are important at this point. First off, a 3chip system does not debayer (so also no moiré / aliasing etc.). 2nd off I was talking about 1080p performance (which is important to me because I almost never deliver in 4K). It performs better then the C300s I have worked with (still show some aliasing / artifacts) En plus, a detailed EBU assessment of the camera and it's resolution exists, which, in detail, can tell you about the XC10s performance, no need to argue about that. 

Thanks, it's very useful and put my previous ENG gear to rest thanks to the XC10s ultra-portabililty. 

The LED light is a Sony HVL-LBPC, extremely versatile and bright. 

And sturdy it is... actually, after having owned a couple of "dslr-like" video cameras (7D, 5D3, GH4), the XC10 is the first one where I don't think that a cage or rig is necessary thanks to the amazing stabilization and the ergonomic design.

It's not soft, it's just that the Canon Log picture style does not add any in-camera sharpening, so you need to sharpen in post. If you use any of the other picture styles, it is as sharp as my 3-chip XF305. 

The (original) Beachtek version is actually stronger than you might think it is. Using this one every day and if you tighten them screws you can beat someone with that setup  

No, unfortunately, like most Canons, it is not. 

I doubt that this example actually presents the XC10 in the best light. To be honest with you, this video makes the camera look more like a GH4 (with a profile right out of the box). It doesn't have the "hard" look  at all in any of the profiles.

Would it be possible to show some demo footage of the before / after also in comparison to the 1DC? 

Thanks. 

Thanks for the comparison, the DR is clearly top notch in comparison to the two other cameras. I don't know what you think about the low light test, but with a bit of post work (sharpening and NR), I daresay that it even produces usable results beyond ISO4000. 

Up to now I haven't noticed any moiré or aliasing (unlike the C300, which would show aliasing in many fine detail shots). 

From the time the XC10 was announced to actually trying it out in the field for a couple of weeks now, my opinion about this camera has turned 180°. 

Tonight I was feeling bored so I tested out the XC10 low light capabilities with a bit of post processing (denoising / sharpening). Take a look: