Gadgetory

everyone welcome to another ask GN episode this is the third one of three I've shot today it's getting real lights about 3 a.m. and after this I have to edit the rest of the David Cancer interview how to make some changes to the day 4 vlog go to sleep wake up in a couple hours and move this thing into the new office so I just went over all that in the patreon episode on patreon.com slash gamers and axes if you want some behind the scenes info it will be in there we had audio issues and that one hopefully they're not too bad but anyway there's a lot going on right now and it's hard to keep up with so this has Jana's always post your questions in the comment section below if you have them we'll try and get them next time this is likely the last sgn I will be shooting in this location we'll pick up the series in the new location shortly after this and may have already done so by the time this one goes alive so let's get into it before that this video is brought to you by the Corsair strafe RGB mark to mechanical gaming keyboard the strafe mark 2 uses Cherry MX which is available in MX red and MX silent and uses the elevated key cap design that has become part of course there's keyboard identity elevated key caps make the keyboard much easier to clean with a blast of compressed air and limit dirt build up the keyboard use a metal body design of have received praise from us in the past in old reviews for high build quality learn more about the Corsair strafe mark 2 at the link in the description below first of all quick housekeeping stuff thank you very much to all the programmers last episode I was answering a question about optimization I gave some of my thoughts on like GP and CPU optimization and then I asked any programs in the audience I know there are a lot of you out there to please count with your own experience below and you did and man there's some really insightful stuff there I learned a lot from it and I hope others did too so thank you very much for sharing we should basically start at elegy on series 2 that some of you have joked about so that I can learn from you all who are experts in various fields anyway thank you for doing that and also quick note stored at Karen's Nexus done that still has a lot of stock finally of the GN tear down crystals where we have it's a really sweet like 3d logo with really high resolution print in it of the GN logo except torn down with various PCB components on it like MOSFETs inductors fans stuff like that so if look at it kind of closely you'll see all that detail so that's on store deck here in Texas net let's get into it first one the crimson pinch says are there any plans to restock oh well there's a store question any plans to restock the GM pint glass they were gone before I could get some yes we have a lot on the way demand was really high super happy to see it sold out in like 16 hours or something so we have more on the way they will be here within a week of this going live if not already in so thank you for asking though we we like them a lot and we'll have more first real question then Steve Strazza says what does watts per meter Kelvin mean to a consumer is it directly correlated to performance for a given Tim ug a higher number it means better thermals is it something that casual overclockers should concern themselves with or is it mostly for the benefit of extreme overclockers the term has come up several times and talking about various Tim's and it would be useful to know how we as consumers should be using that information this is a pretty good one to start the episode off with I think it's a great question so like anything first off measuring the thermal conductivity Watts from your Calvin it's not universally accepted how that's done between the different OEMs and manufacturers and thermal pacemakers so just like anything else measuring fan DB stuff like that you might have some variants and how they do it so numbers aren't necessarily directly comparable between each other although there are absolute standardized methods to take throwing conductivity measurements doesn't mean everyone does it that way so there's no like overarching governmental standards saying you have to do it this way for thermal paste so that's the first note second note strictly speaking know it is not always directly correlated to perform so here's the thing if you have a perfectly controlled environment which never happens with a computer if you have like absolute perfect conditions there's a straight theory in straight theory yes something with a higher specific thermal conductivity then its competitor will outperform said competitor all else being equal so when I say all else being equal this can this contains things like viscosity of the paste contains things like the curing time of the pace how long does it take this to cure and harden and that can affect its performance either positively or negatively depending on how it's designed this includes the literal compounds within the compound is it silver diamond ceramic whatever and and stuff like that so we're we're assuming a lot of things are identical and if you assume everything's basically identical the only difference is the thermal conductivity specifically don't account conductivity then yes the one with the higher number will be better but that's not how it works in the world of computers so in the world of computers and we've tested this so some of this we actually we haven't published it maybe one day but so we've done testing on thermal paste I've done some some testing for individuals who had specific questions about how they perform and I found in some actually a lot of scenarios something like a 5 watt per meter Kelvin paste will perform the same as a 12 watt per meter Kelvin paste it's not the same either one's better or worse but we're ignoring everything like curing time and how well does the age and how easy is it to spread and stuff like that I'm is speaking a lot because I'm really tired and 3:00 a.m. and we're moving an entire studio sorry so all of that notwithstanding like I said we've tested four or five white community Kelvin pastes the same in performance as the higher ones and the reason for that is because computers are really complex there are a lot of variables there's a really good chance that unless you're pushing a lot of watts through the CPU that is and maybe you're deleted or soldered and you have a good interface and a good cooler and good cooling and air flow and everything unless you have all that stuff including the high wattage going through the CPU you might not actually realize a difference in performance I mean it's it's one of the last things you can change in the line of cooling to try to improve performance because you kind of do the obvious thing first you get a big cooler and then you get a CPU that makes sense thermally get a case with good airflow and maybe you deal in maybe put liquid metal all of these things you would do first before just only buy and hire and paste and I don't mean like literally because buying higher-end pays is so easy that it might be a first thing you'd try for a lot of people so it's just it's it's there are a lot of other things that can sort of bottleneck the performance and you'll see the same performance and so then it's all kind of irrelevant other than things like curing time and aging so certain pace age a hell of a lot better than others you might notice that and I think someone asked about this in one of the episodes might notice that video cards if you've had it for a few years you take it apart you put new thrown paste on it it's an easy way to immediately improve the thermal performance by a couple degrees in the least under the same thermal test conditions and that's because the paste ages and hardens and cracks and gets useless so curing an aging is a huge consideration but we're not talking about that are we're talking about the thermal conductivity so other notes here that I took on this other than all the various all the variables I had a note from vsg who these days does a lot of work for techpowerup but also has his own site thermal bench and is a thermal dynamics expert we'll call him so VSG noted you know we should actually define what does watts per meter Kelvin really mean and it is thermal conductivity but what does that mean so his note to me was was this I was asking him what's the best way for me to illustrate this and he was saying the M as in watts per meter Kelvin for define all the letters the M is a single length dimension that comes from a / L service area and thickness thermal conductance is basically the amount of heat in watts passing through a given time on a material of 1 meter squared area through one meter of thickness that has two sides l offset and temperature from each other by 1k so breaking that down their own conductivity important part here thermal conductivity is conductance times L over a where L over a are going to be thickness and service area in that order and also note that K in this case is delta T and that the watts part here so heat is measured in just joules for energy but to express it for this equation expressing it as a rate of change of heat in watts by keeping track of the time so once you go from joules to watts by taking a time element and tracking that so over a period of time we're from joules to watts so watt spoon your Kelvin is is more or less defined as that if you want to visualize it hopefully that helps you out so going back to the core to the heart of your question though and thank you VSU for the answer going back to the heart of your question of is it directly related to thermal reforms for given Tim yes except in the world of a million variables in a computer at which point you're gonna run into so many other things first that you might not even realize again or you might have a huge gain it just depends so for example last example given before moving on I noted that there are times we've seen same performance 12 + 5 watts per meter Kelvin pace the opposite also happens you can do pace directly under the diet of of a CP or sorry directly under the IHS on top of the die of a CPU and see bigger differences there then you might see with pasted on top of the IHS with a lidded cpu with limited cooling capabilities or something like that cuz just gonna run into so many other problems and things you can improve first also margin of error is huge for that kind of testing because you have things like human error and application of the pay's - so anyway a lot of things to consider with testing pastes just watts per meter Kelvin doesn't make it better just like just horsepower on a car doesn't make it better or just watts on a power supply doesn't make it better there are a lot of other factors to consider and the biggest one is how well does it age and does it harden over time so some pastes are better built for that than others next one Brom says when asking go okay that's just a joke on a more serious note Brom said regarding the office can you share some plans previews of how you want to decorate the sets are you gonna go all stylish like LTT all humblebrag like a.j or something else you've mentioned the tables and the tool wall and in previous video you mentioned Adam stab his workshop and he plans to try building something like that to build something like Adam Savage's workshop I think you need to be in the business for like decades and have millions of dollars in a lot of space I do really like his workshop it's cluttered as hell but it's organized chaos and he seems to know where everything is and it suits his style perfectly I'm not trying to at me like that I admire it and find it very interesting and fun to look at everything he has and think it's a great set but so first thing we're gonna do is move this stuff all over so we will at least be starting with the same set except this shelf is going to be a functional shelf elsewhere in the office so I don't think this shelf will be in the shot we might have something it's place I don't know I don't know if we're gonna be shooting out this like kind of 45 degree angle anymore I really like it it's different it's interesting but we have a space to shoot straight against a wall now which means we can have some depth to the shot so I can have cool stuff on a table behind me and a table in front of me maybe with cool stuff on it to a wall livestreaming station with an amazing 8 foot table plus a sick husky workbench behind it so we have a lot of options decorating I am sticking to the sort of industrial look that we have and I really like it I find I find that function makes a lot of sense I like that the sets are places I can work and do things that I would do when the cameras are off and and I think that looks great because it's so natural that is just everything makes sense like if I put tools on the wall that I actually use and I know where they are it feels like it's not like a it's not a being a poser it's something we actually work with daily and I find that more relatable when you're viewing the video that like oh these guys actually use that stuff and I don't know it's not just a prop so that's my approach we are getting some wall art and stuff like that for around the office I don't know how much will be in the sets I ordered some silicon wafers that are really cool so we have silicon wafers coming in from old AMD parts I don't even I couldn't tell you what they are to be honest they weren't listed on there but old AMD parts old Texas Instruments pard stuff like that so that'll be cool wall art and probably a couple posters I was kind of playing with the idea like vintage posters that might be kind of funny today like advertisements for hard drives that are 1 megabyte or something like that I thought that'd be kind of interesting but I don't know if it'll be in the sets or not it might just be kind of a cool thing to walk into when you come into the office so yeah I don't know it's more or less gonna be the same to start same wall color and everything and then we'll expand it if you have ideas of like I don't run wall art to have physical like Hardware pieces that might be interesting in the background let me know but we are keeping a functional look to start with I consider projecting our logo onto the wall and then tracing it with paint like I mean just like the not this logo but the tear down logo that's on the mod match considered projecting out to the wall and painting over it you know then turn the projector off obviously and we have a cool logo painting on the wall but I don't know that I have the patience for it I think it'd be difficult and I'm not particularly trained in that so I don't think we'll do that but anyway some stuff I thought about next one Swift so it's silver said if I passively cool my whole PC CPU GPU power supply will it heat up my room anywhere at all only insofar as reducing your cooling efficiency might have some more power leakage and it's so minimal that it's nothing to consider so this is actually a really interesting question and I think it's kind of hard to initially wrap your head around this concept and I'm sure most of you already know the answers to this but so let's start with power leakage you if you cool a CPU less efficiently for every 10 degrees difference there's roughly 4 percent power leakage so yes technically your room will get hotter but you're almost definitely not going to notice you're talking like potentially single-digit watts depending on the power of the seat of the component that you're cooling so let's just let's just say no let's just say the answer to the question is functionally no any increase you will not notice but this is a bigger topic of cooling I mean I remember years and years ago before we started doing any kind of testing before I was really doing this I remember learning that cooling your CPU better didn't make room cooler or didn't reduce the amount of heat coming out of the PC beyond the power leakage and I just thought that was the coolest thing because really any heating or cooling system all it's meant to do is move the heat around to get it the hell away from the silicon components they can operate efficiently and out of the case and into whatever the environment is room or out the window or whatever so yeah if it's not going like the Seabee running cooler doesn't mean that the room will be cool or it's producing the same amount of heat again ignoring leakage producing the same amount of heat using the same amount of power and all that's happening is you're getting that heat away from the CPU faster and more efficiently and so the CPU can run cooler and the heat still getting dumped into the room and that's all there is to it you're moving heat around that's all any kind of cooling system is basically ever you're you can't destroy the thermal energy you can't destroy the heat so it's going somewhere and that's going to be into your room but yeah the the difference is again it's gonna be about four percent on every 10 degrees on the CPU alone so you're potentially single-digit Watts maybe double digits with a really high on CPU it's pushing a lot of power so now real difference we did test by the way it was one of our least feud videos from last year sadly because it was a really cool one it was about like impact of I don't know multiple GPUs on room temperature or something like that and it was interesting so I set up like a mining rig for a couple days and we do a test turn it off let the entire room cool down for a few days whatever and test again and what I did was keep adding GPUs and it wasn't the most scientific test in the world but it was a test of how much heat gets put out into the room via power consumption and at what point does that affect the like the human comfort level in the room when do you start actually getting room temperature increases and we did some tests where we got like nearly runaway scenarios with a thousand 1500 watts I can't remember I think I want to say I want to say it was like a thousand watts or something was one of the worst ones we did with 5gp as I think and after just a couple hours of that they see on and with the room all closed it was something like 10 degrees warmer which is just really interesting I think personally it's it's not 100% really it is your question but it's an offshoot that is interesting it might be worth watching that one when it was only like 2 GP is so the heart of the test for that one was how does a you know we people always talk about 10 70s in Vega 56 back then anyway and talk about how well the 1070 draws less power so my room will be cooler we saw a lot of that on reddit and elsewhere and I was like okay but I want to test that is that true is there any validity to 50 watts difference making that much of a difference in your room temperature the answer is basically no so that was a fun test but anyway that's completely on sort of underneath it's your question might be something you guys find interesting though next one Savage said should you replace the thermal paste on graphics cards that are a few years old even if they're high-end models also would the thrown pads be reusable or with new ones of similar thickness need to be purchased there are all pads first you should always replace they're all pads on your GPUs with same thickness than all that if you must replace them use the same thickness pad if you use it a let's say it's like let's say it's a one millimeter pad originally and you put 1.5 on there because that's all you have around you can start creating Boeing and the PCB when you remount the cooler because it'll it'll apply pressure it was not designed to apply in the area where that pad is and you can have things like if it's got HBM it's pretty easy to mess up the HBM I don't know if it's like if it's a cracking within it within the inner poseur or if it's like traces are getting messed up or cracked or whatever's happening I'm not positive what precisely happens but I've done it personally where if you put a pad and that's too big and you clamp it down and it's too much pressure in one spot that's got HBM especially you can start getting video artifacting that's unsolvable because there's now a physical hardware defect from Boeing of the PCB so it's it's something we've done and I would advise against it you want to replace the pads with the same size pads if you do my if you do have to replace them and replace them showing really do if they're so torn up after tearing it down the card that is that the pads are so torn up that they're not really reusable or if they have a bunch of dust on them that happens a lot if they feel hardened if they don't if they don't produce grease anymore they're supposed to when when they get hot then those are all valid reasons to replace them or if they just suck let's get one two they don't paste should you replace the oil paste that are a feat on cards a few years old I don't know if he should it depends on how good the cool the card and the cooler were originally just do your do some kind of test and decide if you didn't have log data before that kind of sucks because then you can't make a perfect comparison but do a test ad hoc test decide if you're happy with it or not and if you're having other thermals and who cares leave it alone there's no reason to take it apart unless you really want to and if you're not having other thermals give it a shot because there's a good chance yes it will improve so should you the answer is no you well the answer is you don't have to should you depends on if you like doing that kind of thing I would because it's fun and it's not difficult and if you've never taken a part of video card it's it's so easy it's a couple screws the hardest thing you could do is use the wrong tools which I did on this channel years ago when I started taking apart cards so I get it but you know just use the right tools and it's pretty easy and it's fun high-end models yes high-end models are plenty capable of using cheap their own pace we've seen it next question is from medalists who says your opinion about corrosion from liquid metal on a copper heatsink I'm getting the new throttle put pro nicely done I want to squeeze out every bit of performance I can get but I have many things I've read many things about corrosion on copper heat sinks I don't want to damage the expensive thing first of all for a laptop you don't have to use a liquid metal like if it's any kind of direct I cooling really liquid metal is not necessary at all you can just use like a high-end thermal paste that has a long lifetime like it doesn't age very much it's not conductive that's a big thing something like cryo not I'll note they are an active advertiser with us but it is something I would genuinely use hence asking them to be an advertiser or not someone else so cryo not or an equivalent page from a competitor would be good for that just avoid stuff that is electrically conductive things that have silver in them avoid things that have any kind of like metal in them and it'll be fine to use like it's not gonna hurt anything because you're not gonna short anything liquid metal really comes into play when you're bridging the die to an IHS to a cooler because then you have a lot of thermal interfaces and it's going through more layers of stuff which means it's less efficient at transferring so it can really use the help your direct dye you're already so close to the to the CPU with the cooling element that you should just use thermal paste in most cases so liquid metal might help let's say you you know what you're like I but I want the best and I'm willing to do it and it's not a big deal so okay if you're willing to do it corrosion that is the question kuroh damaging metal with liquid metal is certainly going to have and if you apply it to aluminum let's skip that all the way first if you put liquid metal like gallium indium conductor not mixture or something like that on aluminum it will eat away at the aluminum there will be damaged and you won't like it and it won't be good so don't do that if you put on copper I'll remind you an IHS is copper its nickel plated so the nickel plating on the IHS helps keep the the worst you get is some visual just visual artifacts like has a visual deformity where with an IHS that's nickel plated you wipe the liquid metal off it comes off 80% of the way you have some just gray left behind it's not a big deal doesn't impact on all performance really but it's off behind and I could see how that would concern people but there's no pitting or anything like that if you start putting it on copper so we did this with the rocky cool I don't see it because we're moving but there are cool IHS the just basically pure copper one the exposed copper one I should say with no nickel plating you put on copper it will stain it it shouldn't typically I don't I haven't tested every liquid metal but for conduct and I you shouldn't get physical damage with like pitting and corrosion and stuff like that you will get staining for sure but you can wash most of it off with acetone you'll still have a silver stain there basically forever once you apply it it's not gonna hurt the whole performance in any meaningful way we've we've looked at that in our own testing never published it because didn't really see a need to then do it for publication we just did it to see if that was something we needed to be worried about when we were doing controlled thermal tests and things like that so no I I would use something like cryo not or whatever's equivalent from competitor non conductive and good and doesn't doesn't crack too soon but if you usually could metal you know it'll it'll stay in it but it's not going to corrode it just keep it off of the aluminum and keep it away from anything you can't short or arc to more tin to Stith says I have a few questions about power efficiency and CPUs would it be more efficient to run a CPU at a higher percentage load at a lower clock or at a lower percentage load at a higher clock for example there's a two gigahertz CPU running at one hundred percent load use less power than the same CPU at four gigahertz under 50 percent load assuming this hands high workload we're assuming a whole lot of things here more than just workload but it's fun questions so let's do that so for power efficiency with CPUs we have a great content piece on rise in 2000 series where we did a volt frequency curve and the efficiency of that curve so we basically Patrick Latham did a lot of this work he's on my team and plotted the voltage requirement to sustain a given frequency and then what is the minimum voltage required to maintain those frequencies as we continue to overclock the the chip and what we found was that it's more or less an exponential curve but it kind of if you plotted it out it kind of goes like this and eventually just gets so steep and voltage requirement voltage on this axis frequency on this axis eventually the voltage requirement gets so steep that you can no longer increase the frequency without something like - or exotic cooling and their Bower did his own version of this test with Alan Zoo and found also an exponential curve but with a much higher frequency than we were doing but it applied pretty much universally so getting back to your question then power efficiency always goes out the window when you start overclocking so especially if you're overclocking by like a gigahertz plus so if you if you're overclocking it requires exponentially more power to keep or Veeck or I should say to a voltage to keep pushing the frequency and you lose a lot of your efficiency there it's it's never going to be well I shouldn't say that in most cases let me give you a real example just just to cover my cover my answer here because I'm not gonna say it's never gonna be more efficient or less whatever because there could be a CPU out there where it is more efficient to do that but let's just take like an ad 86 KR u 7r okay if you take that from 4.7 gigahertz stock with max boost and push it to 5.1 I can't really think of a scenario where the the performance gain from overclocking is going to outweigh the efficiency loss from overclocking it that doesn't mean it's not worth overclocking it doesn't mean it's not worth getting the task done faster it might well be but it's not going to be more power efficient if you take that specific processor in that scenario and do it because the power require the voltage requirements it's so extreme once you start pushing those higher clocks that efficiency goes out the window so yeah you're talking about on a let's say a thirty to forty minute render test which we've done for overclock into max potential clock often it's on an order of magnitude of like 10 percent improvements at best in a lot of cases some CPUs are more extreme but that's kind of what you're looking at whereas power consumption could easily be 2x so hopefully that answers your question pretty well it doesn't mean you shouldn't do it just means that you don't do it because you think you'll be saving energy and 7 trekkie last one I just saw that you show your patreon revenue publicly how many creators do this and I you for it do you think this level of transparency will go away in the future as you grow this is something I've thought about a lot the answer is if people abuse it then yes so we've we've had some things we used to publish a lot more information about what we do and how we operate and stuff like that and people in community abused it and tried to use it to attack us without actually validating whether what they were saying made any sense an example would be when we were doing 30-second test runs for a lot of the games rebenchmarking and the whole AMD community piled on us basically saying that all of our testing was invalid regarding the Rison launch because the 1,800 X was completely unimpressive for my price to performance standpoint all even though I said the 1,700 was an amazing CPU but that's that's the community so the thing is I didn't publish it until a year later for competitive reasons but a year later we started the what was it called bench theory series and in that video and article series I showed that with our testing with the games we test the 30-second tests produced the same results as testing for a minute or five minutes or longer and that's because you know believe it or not we actually do look into these things we don't just pick a number and go so it actually was validated before we ever started doing the tests but the community in that instance was looking for something to bludgeon us with and we published a lot of our testing methodology because we believed in that transparency and people use it to attack us without anyone checking to see whether they knew what the hell they were talking about turns out they didn't because I was able to prove that the methodology was completely valid and I already knew that but we just we didn't publish that data till a year later for competitive reasons because we didn't want to publish it until we changed everything in our testing methods which are now different than they were then but I haven't published that either again for competitive reasons and once we revise that method we'll publish the previous year's or current year's methods in this case so that's an instance where providing more information although probably a net good in the the area of test methodology and something that I do want to do it can bite you and it and not even in ways that are justified so that yes transparency goes away as people abuse the transparency and there are competitive things to consider there's stuff we don't publish their stuff that we don't share for how we test because we don't want competition to do the same thing and catch on and you know it gives us an edge lets us do what we do and be unique and do our own thing and create our own market and all this stuff so there's always going to be stuff that we don't share like software that we've built for console testing for example we do console benchmarks now we're not going to publish that and give it away because we spent years figuring it out like in terms of just like building the experience to build the software I'm not saying this the software didn't take years but the experience did so we don't just share all that stuff but I am slowly starting to add more and more to the methodology sections back in even after getting burned for it before because I do think it's the right thing to do I do like the idea that if something's wrong people can in the community can validate and help us figure it out so that we can correct the information if there needs to be a correction even though it's it's really extremely rare but the fear of course is that when you're asleep or on the other side of the world or whatever you can adjust the problem someone's going to take that information and run to the outrage culture machine on the internet and create this big story out of nothing and bludgeon you with it so it's a balance between do I want to share because the right thing to do well there's three things to balance if you could have a scale with three things on it a triangle scale I guess you have competitive concerns you have community concerns and then you have doing the right things and concerns and it can't just all be doing the right thing because then you lose all your competitive advantage and you potentially expose yourself to get attacked in ways that actually aren't based in reality or anything else like there's difference between getting criticized for doing something incorrectly or maybe being able to improve something and getting shamed and attacked publicly for something they're not doing wrong so you got to balance all those things in it I'm constantly learning how to do it better and we're trying to get back to sharing more of the methodology because I do like sharing that and I think it's very interesting for people who don't test Hardware all day to know what goes into it so yeah the all of that from your question of patreon revenue publicly I like mostly I'm not gonna like post our revenue information or how much people get paid or anything like that because it's not anyone's business but I will I am happy to show the patreon revenue publicly because it is a community-funded platform and I think if the community is giving that money then the community has a right to know how much money they're giving us so that when they see us doing all the stuff we're doing there's a really clear link where you're like oh I'm helping pay for that I'm helping pay for them to do fan testing I'm helping them pay for them to hire people and hope you pay for them it's get an office where they have actually have space not bump into each other physically so I that I think makes the most sense to keep open and transparent as long as people don't abuse that transparency that's kind of all I'm gonna really look really looking for so anyway also come with money and knowing how much someone gets is that what you don't know is how much we spend so there's something - talked about before where sharing information like patreon revenue publicly if you're someone like what's Jim sterling or someone who makes $12,000 a month on it we don't know his expenses his profit could be pretty low or it can be really high so people will see that number of 12,000 a month for someone like Jim sterling and thank holy crap that guy's loaded but I don't know what his case is but in our case there are a hell of a lot of expenses so there is a bit of a danger where people will see a number where at something like I don't even remember to be honest I think right like 5,000 public revenue on patreon right now per month and it looks really good and it's very helpful but you have to remember that you know I I'm spending multiples of that on things like moving payroll inventory travel all that stuff monthly so anyway yeah that's it for this one I want to start moving this over at at the David Cantor interview thank you for watching patreon.com slash Cameron sexist hopes how directly you can go to the store now Cameron's Nexus Nets pick one of our mod mats shirts cubes cool video card Anatomy posters we still have a couple in there that are randomly signed you might get lucky you get when I signed ten and yes subscribe more I'll see you all next time