Jimmycrackcrack

I recently bought an external PCIe enclosure so I could make use of a specific PCIe device in an editing setup. One of the nice things about this particular enclosure is that it also happens to come with an m.2 slot for NVME drives as well.

Usually when I edit with my home set up, I'm provided with the storage by the client, and even if not, at the very least, video media, plus backups takes up a lot of room and NVME drives are expensive so I'd usually opt for something cheaper as the actual location for the footage and assets. I figured then that it might be take advantage of an NVME drive of a smaller, more affordable capacity and use it just as a location for video render cache that I just clear after every project wraps. The high speeds of these drives seems like it would be a good fit for this purpose.

However I've heard that SSDs, including NVME are famously short lived and have particularly short life spans in terms of number of write operations. Is that still the case and would the constant writing and clearing of relatively small video files actually be kind of the worst use of one of these drives?

My understanding between TB4 and TB3 is that they're essentially the same, it's just that the standard of TB4 essentially mandates that the device must do all that TB3 maybe could do. Minimum bandwidth is increased and I think I read something about power delivery minimums as well. This eGPU chassis I bought came with it's own TB4 cable, which is actually the first Thunderbolt cable I've seen that specifically says "4" on it.

I assume the reason they supplied this is because, given what it does, an eGPU chassis is going to need to support some pretty bandwidth for a GPU. In my case though, I'm actually using this chassis not for a balls to the wall kick ass Graphics card, but actually to allow me to attach an old and very humble i/o card from Blackmagic. It's currently working just fine for that purpose.

Thing is, the supplied TB4 cable is pretty short and the chassis along with the ATX power supply mounted on it makes for a pretty hefty desk-space consuming setup. I'd like to move the whole setup somewhere fairly far off from the laptop to save me some precious desk space. I looked up 2m thunderbolt 4 cables which I understand is the longest distance you can get for TB4 and still maintain bandwidth and while it's not too bad, the prices are high for a cable. It occurs to me though that since I'm barely using a fraction of the available bandwidth anyway, could I use other, cheaper, long cables. USB4 comes up a lot in my search for 2m TB4 cables for example. (although they are mostly from AliExpress so don't know how good an idea it is to buy from them). If the chassis has TB4 controllers in it, as does the laptop to which it's attached, can one just put a USB4 cable between them? Are they physically different?

For that matter, since my bandwidth needs are so tiny, could I just find cheaper, longer TB3 cables?

I don't know my terminology very well. I just bought this eGPU enclosure. It also comes with an m.2 slot I suspect that's probably what this 4 pin power slot is for.

I have a spare ATX PSU to power this thing with and it's not modular, the cables come out of the PSU box in a big messy bundle and there's no where to detach or attach cables. There's lots of different connectors that come out of this bundle but alas no square arrangement of 2 rows of 2 pins as needed by this chassis.

There are however 2 such connectors that are kind of joined together through a little plastic catch, but in a manner where you can slide them apart. It's clearly intended that you can be able to separate these if you want to, but them being attached to each other in the first place has me a little worried.

The cable from which they each branch has TKG written on it and each of the connectors has L and R printed on it respectively. If I separate them, I can definitely fit one in to the slot, but is there any reason one shouldn't do this?

UPDATE: It works!! Initially the chassis wouldn't power on but I discovered that if I simply don't plug in the 4 pin slot at all then it does. I'm pretty sure that slot is for powering an m.2 drive if you have one and that was one of the things that made me decide to buy this particular chassis so it doesn't look great but I'm hoping that if I actually had an m.2 drive to test it with, that plugging in that PSU connector to the 4 pin slot would work, but at the moment, when there is no such drive connected, the entire chassis doesn't power on. Even better still, the blackmagic card works!! This is great because the manufacturer actually responded to my email asking if it would work too late and I had already ordered it and they said it wouldn't work so the fact that it does is a big relief. Word of advice for anyone testing this with standard computer monitors instead of proper reference monitors like me, it might say "out of range" or similar on your monitor for a lot of standard video frame rates, but for testing purposes, I was able to get it to work at 60p. No good for a real project, but hopefully with a real reference monitor that wouldn't be an issue.

Asking on another's behalf. I don't want to give too many details including the car make and model.

This person's new car has the ability for you to 'start' it and also 'turn off' when you finish the journey. It's confusing what the turn off really means. It keeps lights on for some time after you disembark the vehicle whether you want it to or not, and if you open a door it turns on the instrument cluster screen to display a diagram of the car with a door being opened. You can also turn on the infotainment screen and access some but not all options. The manual has some warning about not plugging things in when the car is off as it could drain the battery.

Is there some physical state the high voltage battery is in when the car is 'started' that's different to the state it's in when it's 'off'? Does it have some effect on wear when the battery cycles between those states too often?

This issue came up when they were thinking of buying a dash cam. The dashcam was designed mainly for ICE vehicles and has a feature called 'park mode' where the camera can be in a kind of standby off state while a vehicle is parked and the car engine is off, but can switch itself on if it detects some kind of movement or impact like if someone drives into your parked car. The dashcam website has some warning saying that for EVs, you should buy a separate battery pack for it because this 'parked mode' doesn't work if the dashcam is installed hard wired in to an EV. This is confusing because the 12v battery should always be accessible regardless of the car's "on/off" state and I would have thought worked just like it does in an ICE car, whereby the camera continues to draw some small amount of power to power the standby mode and allows the maximum power draw the camera could need if the camera is triggered by an impact. In ICE cars, this usually only works when something is hardwired because somehow the cigarette lighter outlet doesn't work without the engine running, (I guess by design so you don't drain the battery with accessories and can't start?) but it sounds like from the manual in this EV it continues to work whether the battery is considered "on" or "off" but conversely somehow if you hardwire an accessory to it doesn't?? It's unclear as well whether that means the dashcam's park mode would work if you plug it in to the cigarette lighter outlet of the EV rather than hardwiring, or if it just doesn't work in EVs no matter what you do and requires its own battery, which seems unlikely but is not spelled out anywhere.

I have sequential downloads enabled on my torrent client, I have a download speed that is fast enough that the ETA for the full download of the media is shorter than the duration of the media itself, and I can watch it in IINA or VLC, but, unfortunately Jellyfin doesn't recognise any new media in my designated library folders until a decent amount of time AFTER the entire file is downloaded and has it's correct extension.

Is there some way to watch as one downloads using Jellyfin?

I occasionally do some paid editing work in my home suite. I use a MBP and I just use whatever storage I have left on external drives or buy new ones as the project budget permits. Most of the time, my work is done on-site using a production company's facilities so it's not a big time operation here at home.

I also like to download and watch video over my wifi to to TV or my phone in other rooms of the house (don't typically move the laptop much). I tend to use the laptop's internal drive for that.

I'm beginning to outgrow my storage for both purposes, but only just. I could continue as I am for quite some time, deleting media at home after I watch it, and buying physically fairly small drives to put away in cupboards for work. However, I'm thinking I could fix both storage needs for a very long time by spending a bit bigger (but not MUCH), and getting a proper RAID. My mind immediately went to NAS, but it occurs to me that, that mightn't necessarily be the most cost effective or efficient way to go given the limited scope of my needs.

My home network is very slow consumer equipment, and I have no ethernet infrastructure at all. I thought I could maybe just hook the NAS up to the laptop via ethernet but then at that point, isn't that just DAS with the extra complications of networking? Would I need a switch between the 2? My home streaming is just done over wifi, since everything is compressed media anyway.

If I buy a decent thunderbolt DAS RAID and expose it to the wifi network via the laptop, would the costs stack up in terms of power consumption and wear and tear of the expensive lappy (given it'd be powered on nearly constantly)? Are there NAS devices that I can directly attach to the lappy for editing, but leave on and connected to wifi for home streaming? Would it need any additional networking equipment in that use case? Can I run jellyfin on it? I feel like a NAS doesn't make sense but would like help puzzling this out.

Excuse the basic questions but I'm not having much luck web searching for answers. I have the server running on my laptop which is also where the content itself if and I have an android phone with the mobile client installed via f-droid.

I can't seem to cast to chromecast with Jellyfin from either the laptop itself, or the android client app. The client app lists streaming to chromecast specifically as one of it's features in the description on f-droid.

Just trying out Jellyfin for the first time. I'm also just trying out media server software for the first time, having downloaded Emby 2 weeks ago so forgive if I'm misunderstanding some fundamental concepts.

I have a series on my hard drive that has been incorrectly identified as something else, the Title is wrong, the posters are wrong, the casting information is wrong and I'd hazard a guess the subs are probably wrong too. That's fine, Emby actually got this particular series wrong as well. The difference here though, is I can't figure out what to do about it. I've seen lots of forum posts saying you can enter an imdb ID number but this is a problem because that only seems to be possible for individual episodes, not the whole series, and in any case, it doesn't appear to DO anything when I apply to any one given episode. More frustrating still, each episode in the series has somehow taken the name of the series as its episode name so they all have the same name and you can't tell which episode is which.

How do I remove the incorrect identificaiton and replace it with a manually selected correct ID? Also, importantly, will supplying a correct IMDB number or whatever else it is I need to do to correct this misidentification, cause the correct subs to be downloaded?

I've often thought it would helpful if the thing I was cooking on was close to as wide as the oven itself. In Australia ovens are usually 60x90cm. I often see and use American recipes because they're so common on the English speaking web and they quite often refer to sheet pans or baking sheets, which seem not to be a very common thing here. They look bigger than the types of things I can commonly buy here, which tend to be cookie pans that are really small. I used to think those American baking sheets were literally as big as the oven and slid in as racks but on further research it seems they're not actually that big and also need to sit on racks themselves and aren't as wide as the typical American home ovens.

I guess my theoretical baking rack would need it's rims to be less wide than the distance between rack grooves otherwise the food would touch the oven walls and baked goods that rise would might rise up to those grooves which would be no good either, but still that should only be a few cm. I actually sort of already have what I want as it came with the oven. It's a rack, that's not a wire and is a solid continuous sheet of metal that slides in to rack positions. The problem is, it always produced weird results when baking and seems to burn the bottoms of cookies and also has a large shallow ramp at the front that messes up what you can put on it. I read my oven instructions and discovered you're not actually supposed to cook on this thing and it's for catching drips. That's super weird to me since on occasion it's been used for this purpose accidentally and it's singularly unsuited to the task as any drips immediately bake right on to it and are impossible to remove and produce lots of smoke on the next use of the oven. I guess it's sort of better than nothing since I can theoretically clean that off when I take the rack out to clean it as opposed to the oven floor, but it's only marginally better since the effect of the baked on drippings is so thorough that it's near impossible to scrub off. Anyway, point is, while it's for whatever reason unsuited to the task presumably because of whatever it's made of and it's slightly odd shape, it's proof in my mind that the concept makes sense and can be done, and yet I can't find anything designed for this.

You can buy additional wire style racks, but seemingly not continuous metals sheets of appropriate size to fit in to the rack grooves.

I have been trying in vain to do this in both automator and shortcuts.

The trouble seems to be happening right at the very start. I can't seem to figure out how to get selected files from finder to be passed as input to a shell script running exiftool.

I actually thought this might be a good thing for me to test using chatGPT for as it's meant to be good at this type of thing and while I assume the shell scripts it was generating were probably good, it couldn't seem to get me passed this basic first step.

I've tried making the shortcut a quick action, which by default adds the 'receive' action to the shortcut, but somehow it seems to be impossible to get the output from that to be the input for the shell script, nothing works. This was tested with a few debugging steps to log the output and it definitely looks like that first step is where things are going wrong. I really don't get it. This was way harder than I expected.

I wouldn't want to find out the hard way. I have a BMD decklink 4k mini monitor PCIe card. I used to use it in a PC, but I upgraded to a laptop. To replace with an external input device is too expensive unless I downgrade capability significantly.

PCIe chassis are more expensive than expected but I've noticed ones that specifically call themselves 'eGPU enclosures'. For some reason when they're marketed to that specific purpose, they cost a lot less, probably because they often don't come with power supplies (which I actually have spare).

I'm looking at 2 such eGPU enclosures and they are a decent price and I think they should work, but I'm a little scared by them specifically saying "eGPU". Would I likely have any problems buying one of those for my PCIe device rather than for a graphics card? Or is PCIe, PCIe regardless?

I'm trying to avoid having to throw away my decklink mini monitor 4k that I used in my old PC build. It's a PCIe gen 2 4-lane card.

To replace this card with something new of similar function that uses external ports, I have to either buy something quite a bit worse in terms of function for a little bit more than my current PCIe version currently costs, or something equivalent in function for way, way more than the 4k decklink costs.

I figured best bet would be to just get a PCIe enclosure to keep using my old card with the new laptop but the costs of those enclosures are STAGGERING, just unbelievably high. The cheapest I could find is 2nd hand and so old it uses TB2 ports. I thought that might be okay with thunderbolt adaptors for modern connectors but it occurs to me that Apple Silicon driver support could be an issue. Any idea if it would even work?