[Sam Lee]

>>… a dual LTO-7 drive library. <>Formatting an LTO-7 tape as M8 gives ~8.5TB and is a clever idea, but the fact that it exists suggests that true LTO-8 media will be expensive and/or delayed ?<< Although it's an economical plan, I'm just not sold on long-term compatibility. Rather stick with the real LTO-8 media than LTO-7 M8 for now until M8 has been proven in the real world for compatibility. I'll take your advice and won't skip LTO-7 due to large existing LTO-6 tape media. After further reading, LTO-8 and LTO-6 media are not interchangeable (MP & BaFe), I'm speculating that wide adoption of LTO-8 will be very slow from all of the customers primarily because of backward compatiblity . This in turn will keep the high drive and media prices much longer than normal. 12 Tb native on LTO-8 is very appealing but it also will need further investment of LTO-7 and LTO-8 drives for existing and future tape libraries. I have thousands of LTO-6 tapes which I do not want to discard them to LTO-8. LTO-6 tapes are still very useful for many situations despite the larger # of tapes needed. I'm now spending thousands per month for 12 Tb Barracuda Pro hdds. They're fast but wasteful for less accessed data. Investing in LTO-7 & LTO-8 drives is still much cheaper in the long run when data approaches petabyte scale - which I'll achieve within a year at about 200 Tb/month average. For my situation, it will have to be LTO-7 for compatibility on LTO-6 media and LTO-8 and beyond for future larger capacity. When LTO-8 is ~$70, I'll migrate from LTO-6 while still using LTO-6 media on LTO-7 drives. This should give me maximum return of investment when LTO-6 was about $100/tape back in late 2013.

[Sam Lee]

Thanks for the tip. Will experiment with this in great details before can deploy it in bulk.

[Sam Lee]

Depending on the final output destination, frame rate conversion is never good. Either dropped frames from 29.97 or blended frames (via hardware based standards conversion) will arise from any frame rate conversion. Other problem from blended frames process is ghosting artifact. The question is how tolerable the platform is. Many providers allow small section of the entire program only. Others absolutely have to be all in native frame rate.

[Sam Lee]

I have similar problems using consumer level drones with 8-bit, long-gop drone originated footage. The most common is slight dithering and color banding that originated in the highly compressed 60 Mbps MPEG-4. It’s not just me. I see numerous of other productions that originated in consumer level drones shot in UHD and downsampled to 1080p. There are simply too much video artifacts. Did you try to shoot in RAW? At this level, any moire pattern and other common artifacts should not exist. But the costs will be quite high.

What helped was shooting in 1080p. But video artifacts are not 100% though due to the moderate amount of video compression already exist before the file is processed in FCP X. You can blame the compression originated in the raw .mp4s for these artifacts.

Unfortunately these artifacts will fail qc if submitted to major OTT platforms (Netflix, Amazon Video) are good examples.

[Sam Lee]

Hmm. Not sure if you want do to this in the long-term. Downsizing anything always seems to have trade offs. There are small aliasing artifacts and chroma noise as a result when doing the software/desktop method when footage are not shot in RAW.

[Sam Lee]

Thanks for the tips. I’ll look into the terminal commands.

I’m curious to know if BRU is capable of performing sequential tape volume archives with two LTO-# drives? Let’s say I have two HP LTO-6 drives and I don’t use the doubler mode. I don’t always need to do 2 set for certain type of data. I simply alternative between tapes when doing multi tape archive. This saves a bit of time when I have to put in the 2nd and beyond tape. Is this possible at all OR it must be a true tape library hardware & firmware?

[Sam Lee]

The technology for deinterlacing has never really perfected. There’ll always be small combing, blending, weaving, half-sizing, or other variants. It can never be the same as true progressive scan camera originated footage. You can clearly see this on Amazon Video, Hulu and other OTT platforms that have interlaced originated programming.

1440x1080i appears to be DVCPRO HD PAL.

[Sam Lee]

Easiest way is to by going in auto save and open that prior to the import of the sound library. Then rename that auto save project to the current one and then delete the current one after everything’s OK. Worked for me every time. Not the most elegant way but does work if you don’t want to manually remove the thousand of imported files.

Alternatively you can add Smart Folder to filter out audio only and select the imported ones. That’ll take more work than just opening the library prior to importing.

[Sam Lee]

I dealt with PsF in the past with the video inspector by having to manually set it to progressive instead of interlaced. It’ll be similar to your stuff XA 30p footage.

This will not be exact to your situation but similar enough to try it out: I’d customize your sequence and set it to 1080-30p. Then I click on the video inspector. Rate conform this to the default Floor. That should do it for your situation. No need to do reverse telecine in Compressor. This method is very similar to using the discontinued Cinema Tools. Then export the file to see if it works out.

[Sam Lee]

You can’t reverse telecine a 30 fps native clip back to 23.976 because there’s no 3:2 pull down applied on the 30fps (likely 29.97 fps) clip. Compressor will not let you proceed.

If I’m not mistaken, you have a 29.97 fps originated footage and simply want to conform it to a 23.976 timeline? If you do that, there will always be dropped frame from the 29.97 fps originated to a 23.976 fps timeline. The other method to reduce dropped frames that I’m aware of is using a hardware standards converter. That’s just a lot of work. I see the end result is blended frames on a 23.976 file. Whatever the case, you can never get a true 23.976 fps from a 29.97 fps footage. It’s either blended or dropped but never in its native form. True native frame rate generally has to be shot in the correct frame rate from the camera and not post processed.