WEBVTT 00:00:00.160 --> 00:00:08.080 after months of performance hiccups several blue screens and us literally 00:00:05.759 --> 00:00:12.160 being able to bring down the entire core of our operation 00:00:09.599 --> 00:00:17.440 just by ingesting video footage to our storage server i'm finally ready for a 00:00:15.280 --> 00:00:21.359 change admittedly maybe 00:00:18.480 --> 00:00:27.680 running new new wanik our main video editing server on Windows 00:00:24.000 --> 00:00:30.000 was a mistake i had my reasons some of 00:00:27.680 --> 00:00:35.520 which were valid but it's clear at this point that you guys were right the 00:00:32.960 --> 00:00:40.800 advantages of running Windows are far outweighed by the problems that we've 00:00:37.920 --> 00:00:44.960 encountered so it is finally the day so many of you have been waiting for when 00:00:42.640 --> 00:00:50.640 we remove the last Windows machine from our server room 00:00:48.399 --> 00:00:54.399 we're going Linux baby we're doing it right thanks cablemod for sponsoring 00:00:52.559 --> 00:00:58.559 this video cablemod allows you to personalize the look of your pc with 00:00:56.160 --> 00:01:01.920 custom colored sleeved cables try out their configurator and build your cables 00:01:00.320 --> 00:01:07.400 exactly how you want them with their realistic cable preview we're gonna have 00:01:03.760 --> 00:01:07.400 it linked down below 00:01:14.000 --> 00:01:20.159 this should be easy i've got my Linux install usb right here all we got to do 00:01:17.520 --> 00:01:25.119 is slap it into wanik's server and boom we're off to the races right 00:01:22.560 --> 00:01:28.799 you see the thing is unlike most of our videos where the hardware is just going 00:01:26.720 --> 00:01:34.000 to get torn down immediately afterward wanik server is our one and only 00:01:31.759 --> 00:01:38.479 production network editing server we can't just shut it down format it and 00:01:36.159 --> 00:01:44.320 reinstall anytime we want because a we wouldn't have anything to store our 00:01:40.880 --> 00:01:46.320 videos on or or edit off of and b we 00:01:44.320 --> 00:01:49.920 would lose all of the work we've done on our active projects that means we either 00:01:48.320 --> 00:01:56.159 have to do this whole process on a weekend or we have to migrate at least 00:01:52.880 --> 00:01:57.920 temporarily to a different server as you 00:01:56.159 --> 00:02:02.640 can see it is not the weekend so the plan is to use my home nas machine as a 00:02:00.240 --> 00:02:07.759 test bed and then eventually migrate what we've done over to new new wanick 00:02:05.759 --> 00:02:11.920 one three new pneumonic my intention is for this to eventually end up at my 00:02:09.599 --> 00:02:16.160 house but that may not end up happening for a number of reasons not the least of 00:02:13.760 --> 00:02:20.480 which is that Jake has become very enamored with the kind of performance 00:02:18.080 --> 00:02:24.239 that we get out of these gen four keoxia NVMe drives he and i are gonna have to 00:02:22.879 --> 00:02:28.480 fight that out a little bit later in the meantime 00:02:26.560 --> 00:02:32.800 let's uh let's build up a server shall we actually that's a bit of a lie this 00:02:30.800 --> 00:02:36.879 server is already built so what we need to do is we need to 00:02:34.879 --> 00:02:42.080 upgrade it we'll start by taking out this epic 7402 00:02:39.519 --> 00:02:45.920 i got it 24 cores is plenty for what i'll be doing at home but as we learned 00:02:43.680 --> 00:02:50.400 the hard way you basically can't have enough CPU performance if you 00:02:48.319 --> 00:02:54.160 want to get the most out of these gen 4 NVMe drives oh and if another 8 cores 00:02:53.040 --> 00:02:59.040 doesn't end up alleviating the bottlenecks we've found we've got a 7742 00:02:56.959 --> 00:03:01.360 64 core that we can play around with later 00:03:00.000 --> 00:03:06.560 another thing you can't have enough of in a storage server like this is RAM and 00:03:04.239 --> 00:03:10.400 i mean that both in terms of capacity and in terms of speed when we were 00:03:08.480 --> 00:03:15.280 deploying this at my house we put the smallest crappiest kit that we possibly 00:03:13.519 --> 00:03:22.560 could i mean obviously without compromising on reliability so this is 00:03:17.760 --> 00:03:24.000 64 gigs of 2666 ddr4 truthfully the 00:03:22.560 --> 00:03:28.159 sticks i'm putting in now aren't really ideal either they're running at 26.66 00:03:26.400 --> 00:03:34.080 mega transfers per second compared to epic's maximum of 3200 00:03:30.959 --> 00:03:36.080 but we're gonna have 256 gigs rather 00:03:34.080 --> 00:03:41.440 than 64 gigs which is going to make a big difference especially for using zfs 00:03:39.120 --> 00:03:46.959 which we are as much as there might be some kind of GPU accelerated 00:03:44.319 --> 00:03:50.879 storage something uh i've never heard of it so we're gonna pull this 1050 ti out 00:03:49.280 --> 00:03:54.879 here that was supposed to be for plex transcoding and we're gonna replace it 00:03:53.280 --> 00:04:02.000 with this gorgeous melanox connect x6 100 gigabit 00:03:59.280 --> 00:04:06.959 per second network card uh do we really need these one terabyte drives for boot 00:04:04.080 --> 00:04:10.720 we have like a dozen of those uh those octane m.2s that are like 32 gigs and 00:04:09.439 --> 00:04:13.360 they're perfect there's actually not that many left server boot drives oh 00:04:12.560 --> 00:04:18.720 really oh all right well i guess i keep using them for stuff like this where it just 00:04:17.040 --> 00:04:23.919 really doesn't matter how much capacity you have and you just want something reliable speaking of our installation 00:04:21.919 --> 00:04:29.680 media there was a fair bit of deliberation when it came to choosing an 00:04:26.080 --> 00:04:31.840 operating system ubuntu server seemed 00:04:29.680 --> 00:04:37.120 good except that the current long-term supported version or lts for short 00:04:34.400 --> 00:04:41.280 doesn't have open zfs 2.0 which features a bunch of performance improvements for 00:04:38.800 --> 00:04:47.360 zfs which is the file system we intend to use and then db and 11 sounded good 00:04:44.080 --> 00:04:49.680 it has open cfs 2.0 and a version of 00:04:47.360 --> 00:04:54.840 samba the software used to host Windows compatible smb network shares from this 00:04:52.000 --> 00:04:58.880 year which is great except it's from march um which is not great so the 00:04:57.360 --> 00:05:03.919 beginning of this video ended up being a bit of a lie because while we could have 00:05:01.280 --> 00:05:09.039 still gone with Linux just find a distro that's got a new enough kernel uh 00:05:05.759 --> 00:05:11.919 potentially build open zfs 2.0 and samba 00:05:09.039 --> 00:05:16.479 from scratch there is an easier option trueness it's got opencfs 2.0 a very 00:05:14.720 --> 00:05:22.240 recent samba and a really nice web interface for managing your storage it's 00:05:18.720 --> 00:05:25.280 not Linux but it is based on a unix-like 00:05:22.240 --> 00:05:27.840 operating system called freebsd so we're 00:05:25.280 --> 00:05:32.479 we're back on the used to be free nas now truenas train i don't know what it 00:05:30.000 --> 00:05:36.400 is about their product but pretty much every time i go to use it something 00:05:34.560 --> 00:05:40.880 blows up in a way that is totally inexplicable i've ended up escalated to 00:05:38.400 --> 00:05:44.720 like directors and vps of engineering multiple times with nobody ever able to 00:05:43.039 --> 00:05:49.440 explain why the product just won't work properly for me but hey fifth time's a 00:05:47.840 --> 00:05:53.919 charm works properly for Jake yeah as long as 00:05:51.840 --> 00:05:58.240 it works properly for Jake you know our backup servers are running truenast for 00:05:56.160 --> 00:06:01.840 like a year now i i know i know it's a good product i've recommended it to 00:06:00.160 --> 00:06:06.319 countless people i've just never been able to get it working the way that i 00:06:03.919 --> 00:06:10.400 need it to work did it just turn on do you have it set to just restore power on 00:06:08.160 --> 00:06:14.080 ac probably okay let's just give it a sec here that's what servers are 00:06:11.680 --> 00:06:17.120 supposed to do okay where's that usb we don't need it oh we're gonna do it 00:06:15.600 --> 00:06:22.000 properly i thought you told me we need a usb uh just for just for visuals are we 00:06:19.520 --> 00:06:25.280 gonna go ipmi af and use an image and everything yeah look at us doing things 00:06:23.759 --> 00:06:29.039 properly today yeah i didn't download truenast but i 00:06:27.120 --> 00:06:31.919 will now wait what happened to my laptop i put it away you don't need it oh you 00:06:30.639 --> 00:06:36.800 told me to bring it i didn't realize there was a keyboard here ah the problem 00:06:34.240 --> 00:06:40.560 with mac ipmi is i can't like press f10 properly like doesn't work press f10 to 00:06:38.720 --> 00:06:45.280 pay respect doesn't work i can screen record on this thing for 00:06:42.639 --> 00:06:49.919 like an hour and only use like 30 battery and its touch pad is the 00:06:47.039 --> 00:06:53.120 equivalent of being hung like a horse it is look at it 00:06:51.680 --> 00:06:57.599 look at this let me show you my touch pad okay look oh okay it's a touch pad 00:06:56.160 --> 00:07:03.120 actually what i was really going to show you was my underwear from lttstore.com 00:07:01.520 --> 00:07:07.440 store lttstore.com it comes with cat hair comes no it doesn't that's from 00:07:05.280 --> 00:07:12.400 your cat i just walk out in the hallway and she's like what's up bro 00:07:09.520 --> 00:07:15.280 virtual cd-rom enter here we go boys this is the smart way to do it you 00:07:13.759 --> 00:07:21.199 shouldn't actually ever have to plug a monitor into a server you ready install 00:07:18.560 --> 00:07:24.800 wow wow that was inspirational look at it go okay hold on hold on pick our door 00:07:23.599 --> 00:07:29.520 hold on yeah we don't want to install the wrong drive now this is cool even 00:07:26.639 --> 00:07:34.400 though AMD epic servers do not have any innate support for raid that means you 00:07:31.840 --> 00:07:39.360 can't even install two boot drives like we did and then just configured in the 00:07:36.160 --> 00:07:41.919 BIOS to run raid freaking simple one 00:07:39.360 --> 00:07:46.080 trueness has a quick and easy way for you to install your operating system to 00:07:44.000 --> 00:07:49.280 two drives so in the event of a hardware failure you're just 00:07:47.919 --> 00:07:53.919 ready to go again immediately yeah it literally just shows up as a separate boot entry you can just pick which one 00:07:52.080 --> 00:07:58.639 to boot from okay time to go to lunch see you later i'm asking if i can turn it on well 00:07:56.639 --> 00:08:03.280 that's too bad because i already did ah roasted 00:08:01.039 --> 00:08:07.440 yeah what's up now what are you gonna hard power it off yeah 00:08:05.039 --> 00:08:10.080 what's up no ha got him now who turned it on 00:08:08.879 --> 00:08:16.319 this guy wanna turn it off again 00:08:13.199 --> 00:08:16.319 i'm turning it off first 00:08:17.280 --> 00:08:23.039 this is bad hardware practices hey wait mother 00:08:20.960 --> 00:08:26.400 man it's not plugged in all the way oh my god snapped out must have been when 00:08:25.039 --> 00:08:31.280 you unplug the power i'm turning it off first 00:08:29.520 --> 00:08:34.880 why is there is there blinking on the port ip 00:08:33.279 --> 00:08:39.919 no we have to take it to the server room later anyways let's just take it there now this is the way to do it i mean did 00:08:38.320 --> 00:08:44.080 we say we were gonna do everything properly today 00:08:41.440 --> 00:08:47.120 uh no okay cool because that's definitely not what's happening hey i 00:08:45.760 --> 00:08:51.120 got 100 gig you won't believe how much gigabytes you 00:08:49.120 --> 00:08:54.399 can put in this bad boy there's a bunch of hard drives under there stop slapping 00:08:52.959 --> 00:08:59.920 it let's go back around i don't want to wait okay let's create a pool 00:08:57.200 --> 00:09:03.040 name name it's called lambo so do i click them in order to add the 00:09:01.680 --> 00:09:06.800 video no you don't um just click all and then unselect that 00:09:05.120 --> 00:09:11.040 oh yeah we don't want this dao i think that's the virtual 00:09:08.880 --> 00:09:15.120 thing from ipmi 12 selected just click over okay so i did a little bit of testing 00:09:13.279 --> 00:09:20.080 beforehand it doesn't seem like it makes a difference between having one v dev or 00:09:17.519 --> 00:09:24.640 two for our workload okay raid z is fine yeah raid z is going to give us one 00:09:22.640 --> 00:09:28.399 failed drive before we actually start to lose data i'm not expecting to lose a 00:09:26.720 --> 00:09:32.880 ton of these drives and this data is going to be replicated using our snapshot to another server that's over 00:09:31.279 --> 00:09:37.200 in unit 101 yeah and then it'll be replicated again off-site to kamloops 00:09:35.040 --> 00:09:41.519 once Jake gets around to fixing the new kamloops server yeah there we go oh 00:09:39.760 --> 00:09:45.920 that's 72 terabytes that's not bad that's not bad usable okay so we got to 00:09:43.360 --> 00:09:51.279 make a oh here first thing go to that yeah we have to turn auto trim on pool 00:09:48.839 --> 00:09:56.800 options oh why is that not by default well most people don't use ssds with 00:09:53.120 --> 00:09:59.040 trueness that's fair we got to make a 00:09:56.800 --> 00:10:03.200 new data set add data set and then let's call this a z drive turn 00:10:01.279 --> 00:10:06.240 compression off we don't need that pretty much all the data we're going to 00:10:04.480 --> 00:10:09.519 write to this thing is incompressible video footage so there's just no benefit 00:10:08.240 --> 00:10:12.720 we could turn it on later because there are other things like word files but for 00:10:11.680 --> 00:10:17.839 now because we're going to do some performance testing too let's leave it off turn a time off we don't need that 00:10:15.760 --> 00:10:21.200 either that's like the access time is that a festival store the last time a 00:10:19.519 --> 00:10:26.320 file is accessed you might use that to see if your kids have been looking at your 00:10:24.560 --> 00:10:31.360 share type smb that's gonna set it to be case insensitive okay um we're gonna 00:10:28.959 --> 00:10:34.720 leave the record size at 128 kb now you could theoretically change it to one 00:10:33.200 --> 00:10:37.760 megabyte that might be better if you're like have a plex server because those 00:10:36.320 --> 00:10:41.279 are bigger chunks which means less overhead i mean our chunks are pretty 00:10:39.360 --> 00:10:43.760 big but for whatever reason when i was playing with 00:10:42.560 --> 00:10:47.760 premiere it seemed like it was a little worse so at least in scrubbing 00:10:46.640 --> 00:10:51.440 yeah because it will increase your latency stuff oh we gotta do some zfs 00:10:49.920 --> 00:10:55.360 tuning yeah because why would you want things to just like work immediately 00:10:53.440 --> 00:11:01.360 look it's not made for NVMe we're kind of doing it dirty right now primary 00:10:57.760 --> 00:11:02.959 cache all one word equals 00:11:01.360 --> 00:11:06.000 metadata and then we'll do 00:11:04.720 --> 00:11:09.279 space lambo slash z drive 00:11:09.600 --> 00:11:15.120 okay enter aha okay 00:11:12.839 --> 00:11:18.480 okay that just means our RAM now stores only metadata yes 00:11:16.720 --> 00:11:22.480 which makes sense because our storage is so fast 00:11:20.079 --> 00:11:27.120 that there's pretty much no point using our system RAM as a more conventional 00:11:24.959 --> 00:11:29.920 cache for it for most use cases though i haven't actually played with it in 00:11:28.320 --> 00:11:33.760 premiere it might actually like the lower latency and when we deploy to 00:11:32.160 --> 00:11:37.600 actual wanik we'll have a terabyte of RAM so maybe 00:11:36.079 --> 00:11:43.279 it might actually make sense but for now we're just gonna leave it off okay wow it's really poorly formatted but 00:11:40.959 --> 00:11:49.040 lambo z drive users metadata on the next line cool it worked all right now we got 00:11:46.000 --> 00:11:50.880 a tomb samba so on truenast smb 00:11:49.040 --> 00:11:54.880 samba is not really configured for the type of performance that we're going for 00:11:52.720 --> 00:12:00.000 we're talking like 20 30 40 gigabit on our 100 gig card not you know gigabit or 00:11:57.440 --> 00:12:03.279 10 gig so we have to make a few changes now by default for some reason at least 00:12:02.079 --> 00:12:06.880 according to this dude on the internet that has this article about trueness and 00:12:05.200 --> 00:12:13.120 the testing that i've done it seems like they set the the default threads for smb 00:12:10.880 --> 00:12:17.040 multi-channel to like one read one right whereas like the default in samba is 00:12:14.880 --> 00:12:20.399 like a hundred right there's probably a reason for this and maybe they've 00:12:18.320 --> 00:12:23.760 changed it back but from testing before changing it to testing after it was a 00:12:22.320 --> 00:12:27.839 huge difference i wonder if it's one of those things where for the average home 00:12:25.680 --> 00:12:31.519 user repurposing an old machine and just like throwing a couple hard drives in 00:12:29.200 --> 00:12:36.320 which closet yeah yeah maybe it like overwhelms you know that the athlon xp 00:12:34.639 --> 00:12:40.079 then this is also like the community edition right so if you were to buy a 00:12:38.800 --> 00:12:44.079 machine from truenas one of their pre-built i imagine they would already configure this stuff for you anyways 00:12:42.959 --> 00:12:47.839 we're just going to copy paste this stuff into our config 00:12:46.160 --> 00:12:54.000 just need one more zero in there a lot of bytes yep now the thing is these 00:12:51.360 --> 00:12:57.440 drives are capable of like tens of gigabytes a second of raw reads and 00:12:56.480 --> 00:13:03.040 writes but as soon as you start trying to run a 00:12:59.839 --> 00:13:05.519 file system on them like zfs performance 00:13:03.040 --> 00:13:08.320 takes a bit of a kick in the balls especially when you have to calculate 00:13:06.800 --> 00:13:11.680 parity which is a whole other thing that's why we upgraded the CPU hopefully 00:13:10.160 --> 00:13:15.839 we can get a little more parody in let's see where we end up okay 00:13:14.639 --> 00:13:21.839 we're going to do a right test first 50. here we go 12 jobs because we have 12 drives laying 00:13:19.760 --> 00:13:26.480 out some files that's pretty good wow 00:13:24.079 --> 00:13:30.480 that's more than 100 gig it kind of fluctuates a bit 00:13:28.160 --> 00:13:34.800 but that's pretty damn good that is not bad as long as we're above 100 gig yeah 00:13:33.120 --> 00:13:38.079 then we can't possibly be bottlenecked by the drives and 00:13:36.480 --> 00:13:41.600 i think we're at bottleneck by CPU right now too i bet if we go look at the CPU 00:13:39.920 --> 00:13:46.720 oh yeah it's probably getting absolutely crashed 100 00:13:44.079 --> 00:13:50.480 41 you know it's probably a little bit slower in terms of writing then i bet 00:13:48.720 --> 00:13:55.760 you if we do a read test it'll be 100 oh okay but i think it's just because it's calculating parity and whatnot true nas 00:13:53.519 --> 00:14:00.079 is not built for these speeds we're using them for these speeds 00:13:57.839 --> 00:14:03.279 but properly yeah all right read eight gigabytes a second you know i was 00:14:01.680 --> 00:14:08.240 getting a little higher than this before i'm a little surprised i wonder if the 00:14:05.279 --> 00:14:12.160 CPU is slower so far beyond good enough now something to note is that when 00:14:09.760 --> 00:14:17.519 you're benchmarking your desktop system changing the queue depth to get like 00:14:14.639 --> 00:14:21.519 massive numbers is not representative of the real world because an individual 00:14:19.519 --> 00:14:27.440 user sitting in front of a computer could never use it hard enough to reach 00:14:24.399 --> 00:14:30.399 q depths of you know eight probably let 00:14:27.440 --> 00:14:35.199 alone 16 or 32 which you might do in benchmarking however for a machine like 00:14:33.040 --> 00:14:40.079 this where literally dozens of people are accessing it that is conceivable and 00:14:38.240 --> 00:14:43.120 that is a reasonable way to test it but we were only testing with two yeah so 00:14:41.519 --> 00:14:46.800 that's why i said it's turning it up is not it's not a hack to just show you 00:14:44.800 --> 00:14:50.399 guys a bigger number it actually is applicable to this user i wonder if we 00:14:48.240 --> 00:14:54.399 try more threads we have more threads so let's try 24 2 00:14:52.399 --> 00:14:56.720 give me all the threads 20. 00:14:56.880 --> 00:15:00.480 give me more threads i want a bigger number 00:15:00.880 --> 00:15:08.320 whoa hey there it is 00:15:05.519 --> 00:15:11.040 18 gb bytes per second interestingly CPU usage doesn't really 00:15:10.160 --> 00:15:15.839 go up tldr it's very fast very fast it's way 00:15:14.079 --> 00:15:19.760 faster than we ever need and it's way faster than one so let's 00:15:18.000 --> 00:15:24.399 try fishy you know it's good that we actually left the one terabyte in there 00:15:21.920 --> 00:15:28.639 because of those ingest issues so that was another problem that 00:15:26.399 --> 00:15:32.959 actually that was what really prompted me to go okay yeah forget it we're done 00:15:30.720 --> 00:15:37.759 with this because when we would ingest footage from the stations over there 00:15:35.360 --> 00:15:42.079 if it was greater than one terabyte it would fill up less than that is it it 00:15:40.399 --> 00:15:46.639 basically like when you write to storage spaces it just goes to RAM and then it 00:15:44.560 --> 00:15:50.240 flushes to disk slowly and when we had the ursas those drives could do like one 00:15:48.560 --> 00:15:53.680 and a half gigabytes a second you know times a few if you ingested fast enough 00:15:52.240 --> 00:15:56.320 you'd fill up the RAM and storage spaces only lets itself use half of your system 00:15:55.279 --> 00:16:01.759 memory right and then at that point it would just crash the network share like for 00:15:59.600 --> 00:16:06.000 everyone and there's no evident way to turn it off yeah because 00:16:04.480 --> 00:16:12.000 our disks are fast enough we could just go straight to the disks but we couldn't 00:16:08.560 --> 00:16:13.680 turn off the stupid round caching 00:16:12.000 --> 00:16:16.519 and especially when you've got an os that blue screens every once in a while 00:16:15.600 --> 00:16:24.880 having 500 gigabytes of data potentially in a 00:16:19.839 --> 00:16:24.880 RAM cache yeah one does not simply 00:16:26.000 --> 00:16:30.560 so this is kind of dog poo what we need to do is go into one of the editors 00:16:29.360 --> 00:16:36.800 stations that is not Windows server running storage spaces and like copy these files 00:16:34.320 --> 00:16:41.279 or something okay so i actually oh my god i didn't touch it 00:16:39.600 --> 00:16:46.160 i'm sorry it's been a stressful week 00:16:44.240 --> 00:16:50.639 what's up can i borrow this of course okay let me drive for a sec here 00:16:48.320 --> 00:16:56.720 okay so i want to copy this to my local drive 00:16:53.279 --> 00:16:56.720 huh that is not very good 00:16:57.680 --> 00:17:00.079 Jake 00:17:01.360 --> 00:17:06.880 it's possible that the issue is just Windows file transfer being Windows file 00:17:05.120 --> 00:17:12.319 transfer so we're pulling out the big guns here bringing out show easy copy 00:17:09.760 --> 00:17:15.120 look at that 20 gigabit oh well that's better 00:17:15.439 --> 00:17:22.880 uh where are you going to and from right now i'm going from wanik to 00:17:20.799 --> 00:17:27.760 to the other the test one yeah and i'm sure we open this 00:17:24.720 --> 00:17:27.760 lots of blinky lights 00:17:27.839 --> 00:17:31.280 so there are okay 00:17:31.360 --> 00:17:37.520 yeah there's your CPU right there okay 00:17:34.400 --> 00:17:38.559 i'm copying over a red mag 00:17:37.520 --> 00:17:43.840 boom pinned to 495 megabytes a second which 00:17:41.679 --> 00:17:49.280 is the speed of a red meg okay what else we got red map brandon is 00:17:46.400 --> 00:17:49.280 what's yours going at 00:17:49.760 --> 00:17:56.000 499 okay so we've got a total of a Gigabyte a second heading over there i 00:17:54.080 --> 00:17:59.600 need to find some more media to copy here what you'll see is sometimes it'll 00:17:57.600 --> 00:18:03.280 slow down when it's going between clips red footage is a bunch of clips so yeah 00:18:01.520 --> 00:18:06.320 between each clip Windows file transfer will like dip for a bit yeah we're 00:18:04.880 --> 00:18:11.919 definitely running into some kind of bottleneck on this system where it's maxing out at 500 megabytes a second no 00:18:09.600 --> 00:18:16.320 matter what we plug into but still overall the ingestations are 00:18:13.760 --> 00:18:21.840 performing better than with old wanik which is new 1-3 so this will be new new 00:18:19.280 --> 00:18:25.120 new wanik anyway the point is i paused all the transfers and we're 00:18:23.440 --> 00:18:30.000 going to have three or four editors now try to edit and then i'm going to start these again and make sure everyone's 00:18:28.960 --> 00:18:33.520 smooth our guinea pigs are going to be ed mark 00:18:32.480 --> 00:18:37.679 and emily over here and you guys are 00:18:35.679 --> 00:18:39.840 pretending to edit right i mean that's basically all everyone in here does 00:18:39.039 --> 00:18:44.799 right if everyone a prime pretending to edit what are you 00:18:43.440 --> 00:18:48.640 pretending to edit right now i'm not pretending you know oh okay well we've 00:18:46.799 --> 00:18:53.679 got at least okay we got at least one editor actually editing fantastic 00:18:52.160 --> 00:18:57.360 hoffman that's a beautiful picture of david all right so guys keep pretending 00:18:55.600 --> 00:19:01.200 to edit andy you stay here i'm gonna go turn on all those file copies just like 00:18:59.280 --> 00:19:04.640 before we got about 10 gig on the one on the right five gig on the one on the 00:19:02.559 --> 00:19:08.880 left and how are your projects going it seemed fairly normal or very normal or 00:19:06.960 --> 00:19:12.640 better than normal it's not better but it's normal not better but normal yeah 00:19:10.799 --> 00:19:17.919 it's pretty much exactly exactly the same all right thanks teddy yeah it's 00:19:15.520 --> 00:19:21.360 good it's acceptable is it it's not worse 00:19:19.520 --> 00:19:25.200 that's good but it's not better it's 00:19:22.559 --> 00:19:29.520 it's hard to say to be honest okay yeah okay it's it's 00:19:27.520 --> 00:19:32.640 i could i could have it with this yeah okay 00:19:30.960 --> 00:19:37.280 really wasn't the goal today the goal today was hey this is going to be perfect and amazing you guys are going 00:19:35.200 --> 00:19:40.960 to love it so uh is it no no 00:19:38.799 --> 00:19:44.960 the goal was the same okay ed how's yours doing 00:19:43.280 --> 00:19:50.160 um i think this is doing well historically i think it's about on par 00:19:47.360 --> 00:19:55.200 with uh what performance was like before but as of late this is a lot better um i 00:19:54.320 --> 00:20:00.640 used to have runaway footage pretty pretty often 00:19:58.640 --> 00:20:04.559 in the last uh little bit for those of you not familiar that means when you 00:20:02.880 --> 00:20:08.799 stop the playhead it like keeps going for a bit or yeah yeah okay like the 00:20:07.039 --> 00:20:12.240 last couple weeks right yeah the last couple weeks when i would play back 00:20:10.400 --> 00:20:16.000 footage especially uh the footage from the sony's it would 00:20:13.760 --> 00:20:19.120 just like run away for me for a few seconds it's hard to make an accurate 00:20:17.440 --> 00:20:23.280 car that way yeah okay now it seems fine 00:20:21.200 --> 00:20:26.720 okay all right confirmed we're gonna move ahead with it and confirmed we're 00:20:25.200 --> 00:20:31.600 gonna tell you about our sponsor thanks to current for sponsoring this video current is a mobile app and debit card 00:20:29.840 --> 00:20:35.360 that helps you spend and save better so you have the freedom to do you their app 00:20:33.600 --> 00:20:39.919 makes it easy to see where your money is going and set different savings goals to 00:20:38.000 --> 00:20:45.200 keep you on track you can withdraw your money without fees across all 40 000 all 00:20:42.960 --> 00:20:49.200 point atms in the u.s and there are no overdraft fees up to a hundred dollars 00:20:47.520 --> 00:20:53.360 you can send money easily to friends and sign up takes less than two minutes so 00:20:51.039 --> 00:20:57.039 head to current.com Linus tech tips and sign up for current today if you guys 00:20:55.280 --> 00:21:01.520 enjoyed this video maybe check out of some of our previous server vlogs it's 00:21:00.240 --> 00:21:06.240 the last time we tried to deploy this server how about that yeah 00:21:04.720 --> 00:21:08.320 someday we'll get it right maybe this is it