Gadgetory

everyone welcome back to another episode of ask GN as always if you have questions for next episode leave them in the comment section below or on the patreon discord if you support us through patreon you can leave questions in the ask GN channel it's been a couple weeks since we did one of these we were at CES so there were a lot of questions logged I've got a good amount of them here but I probably won't get to all of them so if I miss you posted again below before that this video is brought to you by the gamers Nexus patreon and our patreon backers if you want to help us out directly you can go to patreon.com/crashcourse 'us where we have access to a patreon only discord where you can chat with the GN team or you can support us at $5 or higher and get access to behind-the-scenes videos as we release them once or twice a month learn more at the link in the description below I want to start out though with micro Center and I guess it's just graphics cards in general but man micro Center what's going on over there this is like we're the champions of PC builders we give you the best deals and hugely discounted CPUs or cheap or free motherboards with the purchase of all these other things and then I want to come to graphics cards they're like you know what this is a this is a trend we can get behind $1500 ASIS 1080i Strix that sounds fair so before I get into this one the we just posted that GPU video what do miners or what do manufacturers think of crypto miners and that talks about a lot of elements outside of crypto mining like memory prices and things like that what I mentioned in there also is that the difference between primary and secondary sellers on Newegg and Amazon means that you'll see 1500 or $2,000 GPUs but they're not sold by Amazon and Newegg they're sold by third-party sellers on Amazon and Newegg marketplaces so it kind of looks the same but it's not there's a gigantic difference when the cards are in stock they tend to be less insane than those prices they tend to be closer to MSRP maybe slightly above you don't really see the outlandish prices in the u.s. anywhere with those two websites except for the third-party sellers Micro Center we had a GN team member visit one in Michigan recently and they had a 1080i strikes for $1,500 limit one per household why would you even limit them at that you're like raking in so much money let them buy as many as they want what do you care you paid way less than that you paid less than $800 for that card so I don't it's just it's it's just stupid it's like trying to appear like you're doing something to limit the consumption of the card so you multiply the price by some factor and then you limit it to one per household it does not make any sense you can buy Titan X Peas for cheaper than 1080 eyes and you should it's looked like the whole last year or like don't buy this heightened XP for game by 1080i well I ideally you don't buy either cuz they're not good value for right now unless they're like $800 but if you really need one today screw the 1080i at $1500 either go camp at a store not Micro Center and buy one for a reasonable price or a buy different card I guess see a micro center they had GTX 1060 FTW gaming six gigabyte cards let me just let me just repeat that gtx 1060 what was the price of that $250 or thereabouts that card is at this micro Center for six hundred and thirty dollars limit one per household because while we're taking three times the original price of it we still want to make sure we appear like we're serving gamers so yeah they have apparently a BYOD I think it's build your own discount that one of the G on patreon members was talking about and I confirm this online with a couple others as well so Mike or senator claims that you can appear in the store you have to find a manager and they will apparently give you a better price on the video cards than what's shown on the shelves if you buy other components there which is actually reasonable because if you're a PC builder you're buying multiple parts for a PC then yeah okay you're buying Pew anyway you know what tells micro Center you're buying one and hopefully they give you closer to MSRP for the GPU that sounds great sounds like they're being the champions of PC gamers again we tried that at this Michigan location and what they told us was let me find the quote here so regarding the $1500 issues 1080i Strix when we said what about a discount if you buy multiple components they said that's what it's selling for on Newegg and Amazon so I think I hear mileage may vary depending on the store and who got the memo and who didn't either way here's the difference if a retailer is selling them at those prices as opposed to a third-party reseller there's a it's it's kind of an egregious breach of trust for the community and the customers you have that's it's completely outlandish this is like absurd if it's just it was just Newegg who has a third party seller listing a 1080i for $1,500 whatever you might as well go to ebay and complain about prices there that are outrageous too but when it's micro Center directly that's wholly unacceptable and I think they should be condemned for it but hopefully they'll wake up and maybe all the stores can get the memo on the build-your-own discount or whatever they're calling it but yeah that um this it's pretty see at this point knowing what it cost to make a GPU and knowing everything we do from the GPU what do manufacturers think of miners video go watch that if you haven't before you comment but knowing all that stuff and the cost change the retail prices we're seeing here whether or not it's micro centers fault someone in there is gouging the hell out of everybody and I would have to say it's probably closer to the retail side of the chain than not so definitely very disappointed in micro center but let's get into some of the questions so this one is from discord Billy O'Neill says ask GN regarding the recent GPU prices video fitting did any of the manufacturers with whom you spoke consider using their position as a board maker to run miners themselves thanks we actually we joked about this a bit with a couple people so I I probably would not be given fully honest answers if anyone did have a mining operation in their warehouse so keep that in mind but depends on who I'm talking to something we have better relationships with some of them than others but I we kind of joked about you know these cards make so much money once you guys just mined with them it doesn't really work that way there's investors involved and things like that and so the investors want to see the product moving off the shelves that said there was a lot of joking at CES about actually with with I think at least two vendors about a you know we should do is we should give away a heating unit like just a space heater we should give them away with an Ethernet cable tell people to plug it in and let it mine and then it minds the currency to the manufacturer and the heating element was free which was just joke but actually kind of brilliant an interesting way so that's that's about the extent of the question you're asking I think not really most of the most the individuals we spoke to at the companies you have to remember these are individuals I would say have a tenuous belief in mining in general so obviously that's not the company that's the individual but that should pretty much answer your question as to what why they would not do that next question is Yusuf who said I've been I've been mining with my gigabyte 1080 water force for almost a year now on my main rig it still runs perfectly it's an i/o card I want to bring this one up to address a to address an important point that that seems to be misunderstood failures don't typically happen like degrading performance you could it Those matter if it's mine you can be running fire strike for a year and your score might be the same in a year as it is today variants notwithstanding so when you see failure depending on the type of failure it's more likely that it it to quote a reply to that comment on YouTube it runs perfectly until it doesn't which actually makes a lot of sense because it runs perfectly and then it stops there's very little indication of failure in between there the the typical indications of failure might be fans dying so you'll see thermal throttling you'll see clock dropping stuff like that in an instance of fan sign yes you will see a decrease in performance you'll be able to look at the card identify the failure the fan and replace it if that's not the point of failure you have other options like memory failure from overheating pretty uncommon or more likely a power component of some kind just reach an end-of-life electrolytic capacitors are a great example of this where if you have one that's rated for 105 degrees Celsius 5000 hours and it's running close to that I this is an instance where with capacitors the closer and closer you run to its spec it will have an impact on lifespan whether or not you ever know depends on what it's rated for because yeah of course you're wearing down the life of the component just by using it but if it's within spec if it's running pretty cool you're probably gonna replace the system or the component before the usable life of that capacitor expires so not commonly a problem however if you're putting it in a high load high-stress environment with poor cooling which not the case for a liquid-cooled card but there could be capacitors in there that aren't liquid cooled if you have something in there that starts running hot say it's getting close to the 100 degree mark which is not unreasonable for a graphics card as we've shown in many tests if it's getting near that mark you run it for 5000 hours so you know maybe get a couple of years out of it or something if it's if it's on and off and not 100% load 24/7 but as you approach the end of that rating and it's different for every specific part but as you approach the end of it is possible that one of those capacitors pops they start ballooning so you look at the capacity if there's some that are kind of like bulged out and looks like that are gonna explode then it might be that the card is nearing a point of failure and when it does pop it's probably just gonna stop working this is something that you could see with some power stages that were used on the EVGA ACX cards I guess two years ago now before they issued the fix where there was a bad supply of components from a component supplier to EVGA and those could blow up under adverse conditions basically because there were videos of it with ACX cards so yeah I think the point here I'm making is that if you're like well I've been rendering with or I've been mining with my GPU for two years and it's fine why do people talk about parts tine sample size of one and it you're not gonna see worse performance it's just gonna stop working and then then you'll stop commenting about how great it's been working so that's not to say you should be concerned or and you can just keep thermals in check see the previous ask GN and we have a long discussion on thermals in there everything comes back to heat he is the catalyst for all failure in computers for the most part heat and voltage so yeah I think that pretty much sums up my points on that one but basically keep it cool and if you're feeling feeling cocky because your card hasn't failed it doesn't mean that you're out of the woods next one Pike said Pike said this shit has been bugging me to no end if a 1080i puts about 300 watts overclocked and a nice out an 8700 K puts out less than that even with our hard overclock then why can a 1080 TI be cooled with a 120 millimeter cooler but the 8700 K needs a 240 millimeter plus cooler this is a great question a lot of this comes down to two main things die size or dissipation area and direct contact the GPU in that instance and before anyone not familiar with these tests Pike here is correct you can put a 120 millimeter CLC on a GPU like a 1080 or something 1080i and you'll be operating at 40 to 50 C with an ambient of I don't know maybe 20 or so it's actually it works very well if you put the same cooler on in 8700 k it's gonna struggle hard it's you're gonna be nearing the pan your voltage your you're definitely way higher temperature than the GPU would be comparatively so the reason for that primarily is because of the direct dye cooling the GPU diet is exposed directly to the cold plate the CPU has an integrated heat spreader on top of the die between which there is thermal compound and then on top of which there's thermal compound and you've got nickel plated copper IHS and between them so it's got more mediums more media to transfer through and when you have more transfer media you're dealing with just worse and worse transfer of that energy as you keep pushing through all those interfaces so that's part of it we've kind of demonstrated this by deleting CPUs and using liquid metal look at any of our liquid metal tests you can drop here from 8 to 20 degrees Celsius depending and by adding liquid metal so that's one component is the the thermal compound between the IHS and i is not very good and then the other component is there is an IHS that adds to it and then you have the die size a gtx 1080i has a larger die than a lot of the CPUs on the market if not most them and that's more area for the heat to be distributed through so it's not as concentrated on just whatever few cores there are on the die on the intel cpu in this instance so that's most with those thermal transfer direct eye contact and anticipate or die a size and dissipation area thermal density basically is what it comes down to no rise in 2000 a new hope says what's the most important tool GN uses that you would recommend for most enthusiasts to have I would say the most important tool is a software tool and it's hardware info 64 and everyone should have it it's very good I encourage you to go into the options of it and not just use it out a box as the options are pretty powerful you can do a lot with it it's probably my favorite software solution for looking at any stats at all really on the computer not just thermals but it's got all kinds of sensors for the high end motherboards that you may not even know we're in there like vrm sensors stuff like that it's pretty sweet to really dig into it other than that a current clamp would be high up on my list of must-haves digital multimeter would be high on my list and I like keeping around different types of thermal compound liquid metal things like that just to just to have in case I need it and I normally keep like a this is the good stuff this is the okay stuff and then if I know a cool is just gonna be on something for a couple hours or whatever while I do some unrelated non cooling focused test I'll put the cheaper stuff on there so that's kind of what I like to do in terms of accumulating tools stuff like thermocouples not really needed for most enthusiasts yeah I'd say hardware in 464 current clamps digital multimeters and a nice toolkit like the iFixit pro tech toolkit we like that one a lot or their 128 bit kit those will allow you to open pretty much anything on the market next one contractor three one six what are the three biggest or most-wanted areas of test methodology pieces of equipment that time and money and practicality currently prevent you from pursuing I would say power supply testings up on that list power supply testings interesting it's it's requires very expensive equipment to do it properly actually monitor testings in here to I'll get to that in a moment so the power supply testing short of calling it an overview opening the box and talking about the cables there's not a lot you can do with power supply testing without buying some good equipment and the former isn't a review so we don't do it it's an unboxing so you start looking at stuff like chroma rigs or what are they sun moon I think they're called stuff like that it gets expensive fast and the problem with power supply reviews is they don't get a lot of traffic it's not a highly searched item a pretty small group of enthusiasts really care about their power supplies enough to look up professional reviews of them as opposed to just check in aggregate Newegg score so the money that we can make from a power supply review is pretty low and if we can even we probably probably make money on it up until the cost of the equipment required is factored in at which point it may be more or less impossible to profit on it any time in the foreseeable future I mean like they're talking more than a year our high it's pretty bad so especially considering we have a limited amount of time that is a fixed resource I can't create more of it so if my time can be spent testing other things that are one I'm more of an expert in - they are more accessible three they're more interesting to everybody on the internet and then four it's just it's the tools are more accessible then obviously the time is better spent there like it's pretty much it's I think everyone can if you think through the process of why you would choose to learn how to test for example fans in depth versus testing power wise you can see where the cost difference is to get into it it's not even just ROI it's can you even afford the tool to start the testing and a lot of that stuff the answer's no I gets really expensive really fast tens of thousands of dollars so that that's one thing we're looking at the power supply testing it may happen but it's not easy so that's definitely one aspect another thing I would like to test is monitors so we talked to the Vesak group and displayport people at CES and they showed us some monitor testing to really do it right you need expensive software and you need you need a bunch of tools and calibration tools and maybe high-speed cameras ideally to double-check the manufacturer specs and if you're not testing all of those things it's not really an objective review you're leaning a lot on how does it look and that's I mean it's already pretty hard to convey through a camera because the camera might not shoot at the same quality the monitors putting out stuff like that so and then the viewer doesn't have a monitor good enough to see the quality of the monitor they're trying to watch your views for so I really need to object to the numbers and we're just not there yet I think there's one a thermal chamber I would love to have but that takes a lot of space and is also expensive even if you buy an old one so yeah those are the three key things I think they're all very expensive wants in the case of monitors is just a lot of time and then let's see Soren asks I was recently told by a teacher that you should never hold hardware buy copper contact because you can get oil on them or ESD into them and destroy the entire piece of hardware this is true why do nearly all tech tubers hold hardware such as video cards and RAM with their fingers on the copper contacts so a couple things there's two sides to this one is there it can go kind of either way the teacher is correct electrostatic discharge is a thing that's real it exists and anyone who says otherwise is an idiot because it's science like it's very easily proven so electrostatic discharge is a thing that happens now you can shock components I I could have a charge on me right now if I didn't just touch this but if I had a charge and I touch the back plate chances are if I discharge electricity into that video cards back plate it'll probably be fine the same is true for the shroud or really the heatsink most things there's a lot of ground plane in there if you touch the GPU core there might be a problem or some sensitive SMD surface mount device on the PCB there could be a problem and we've I mean 10 years doing this I've killed a lot of components from an electrostatic discharge one of our oldest how do motherboards work videos from many years ago we probably spent about an hour filming it and it was it was me and Patrick Stone spent an hour filming it handling the same board the whole time never grounded ourselves on carpet and it was dead at the end of the shoot and it worked fine at the start of the shoot the only reason that happens is ESD so it's definitely something that happens that said there are specific conditions really kind of required to get an electrostatic discharge that's fierce enough to hurt something and also conditions where you have to touch the component in the right place because it's possible that it finds its way out to ground rather safely and doesn't hurt anything it's also possible you touch the exact Ron thin and destroy it so first let's think of the voltage requirements in order to feel an electrostatic discharge or see it like when you touch a car door it's got to be thousands of volts to really notice it as a human so that's pretty high the components that are most sensitive like a GP or CPU are used to dealing with things like one volt by the time it's all filtered down and gets to them so you can see why there would be a problem but again you really have to touch the exact wrong thing for it to be a problem as for oils and handling contacts and things like that it is bad practice to pick up the card and hold it by the PCIe slot for example right there it's another another form of potential damage that's pretty the oils on your skin probably aren't going to cause problems it's possible if you're somewhere like Taiwan and we've seen this in person where the humidity is just insane all the time and you have like USB ports there's a valid use case for gold-plated ports and things like that and a high humidity environment because of corrosion touching the PCIe slot isn't gonna cause corrosion unless you're an alien so the the only real concern is ESD into the contacts or something like that oxidation doesn't isn't really that much of a concern you know just kind of make sure you're grounded ideally you don't touch them pick things up by like the back plate the shroud the expansion slot things like that where you're less likely to cause damage when I retouch in a bare PCB as possible you could hurt it that said and we get hands-on with this stuff constantly even without grounding throughout the whole process you just kind of tap it once and most the time it's fine I have killed stuff though so it does happen but yeah I think the reason I'm at taking it on to answer that is because I don't want people to think that it's blown out of proportion in terms of a there there's like two camps to it there's ESD isn't real which is stupid and there's being a hyper paranoid of destroying anything by touching it which we also see when I move PCBs around and tear downs people are like why are you so hard on the components they're fine so that just find a middle ground between those two calves and it'll be good but yeah let's good question though see last one is timber meadow who said still interested to hear about IT certifications and this is also previous job experience I guess if I remember correctly from the question ask GN 66 so yeah I forgot to answer that one then I don't have any IT certifications I have work experience so other than doing this I long time ago did web development I did test engineering so worked at Dell for a while during test engineering and and validation so job there was basically whatever components we were testing we would develop a test plan for how to break them through reasonable user means or extreme means attempt to break the component not like with a hammer but like three years and then report those problems to the developers or the engineers to get them fixed there was a we did a lot of stuff a lot of projects going on but yeah I was mostly developing tests following through test methods creating test cases where you have a spreadsheet that's like step one two three four pass or fail for all of them we did there's some other cool like shock and vibe testing equipment that was in there which is stuff that you'll see in the industry where basically a shaking table will rattle the hell out of laptops and stuff like that to see at what point does a component come loose or does stuff break or lose connectivity I also did validation testing for CAE software computer-aided engineering parametric analyzer software I tested worked with a signal integrity software and testing of signal integrity software to make sure things like dielectric materials had the right values insulators have the right properties attached to them worked on field solver testing and stuff like that so basically it's all testing all the previous experiences it's all testing all of it so that's it for this one though subscribe for more as always if you want to leave a question post in the comment section below or in the patreon discord if you want to join the patreon discord I've got a patreon.com slash gamers Nexus thank you for watching I'll see you all next time