Gadgetory

okay so we built seven gamers one CPU a $30,000 computer running on raid as an operating system and using it and some really cool open-source virtualization technology to run seven discrete gaming virtual machines at once all fully capable of like high performance gaming so I was cool thirty thousand dollar computer two thumbs up very awesome ten @ @ n IG and etc then what we did was we showed it off at CES okay so we got that out of the way the world was able to see it in person then after that we brought it back to the office converted our entire conference room here into a LAN Center and had an interoffice gaming competition complete with taran being terrible at video games yes the very worst one here except he did kill me once and you can check that video out here so you might be asking yourself why is line is still milking this damn thing what could possibly be left to do and the answer is of course we're going to overclock it intel brings ddr4 to the mainstream with their core i7 6700 K and Core i5 6600 K processors check out the link in the video description to learn more so the first step anytime you do any performance tuning is to get a baseline measurement because if you don't then you have no idea if the dials you're turning are doing anything at all so I'm going to introduce you guys to the benchmarks that we're going to be running Cinebench r15 is going to tell us what we need to know about cpu performance we are going to be using crystal disk mark to tell us what we need to do about our SSD array performance I 264 cache and memory benchmark is going to be interesting especially once we run it on all machines at once more on that later and we're going to be using fire strike extreme to see how our GPUs are doing some of the stuff that's open is just temperature monitoring and stuff like that and finally we're going to be having a look using our handy dandy wall power meter doodad here at what kind of power consumption we see when we're hitting the whole system at once but let's start with a single VM baseline with all the other ones sitting idle and I'll show you guys on screen hopefully right now how I plan to go through this entire test suite and show you guys the scores so here we go we're getting a Cinebench score of around 600 we're getting crystal disk mark scores in the area of sort of you know two gigabytes per second reads and around 400 megabytes per second writes to our array our memory scores we're seeing about 5 gigabytes per second so on average reads and writes to our quad-channel ddr4 memory and our 3d mark firestrike extreme score is 61 48 so the next test and this one I will need some helping hands for is going to be to hit all seven VMs at once for our baseline score to see what happens when they have to actually share resources which especially for things like memory it's not very pleasant sometimes okay so I've assembled a crack squad of helpers whose intelligence and physical coordination I'll lead me to believe that it is probable nay likely to both kind of the same thing that they will be able to perform the skills that I set out for them today tasks actually not skills don't worry too much about that basically what you guys have to do is all press a button at the same time not gonna work okay Colton you're fired again thank you okay so two of you will have to press two buttons at once at the same time Johnny you got this okay Johnny you're on machines one and two you've on you're on machine 3 Colton you're here Nick you're here and I'll go over there and I will tell you which button to press and you will all press it at the same time you're going to click the button that says run next to CPU in the Cinebench r15 programs everyone got your cursor position ok so I'm gonna say that don't do it yet I guessed Aaron's not here I don't have to give instructions like that ok 3 2 1 go ok so you can see here with our seven instances of Cinebench running we are actually pinning the entire CPU here at 100% usage that was exactly our intent and will tell us without any extra turbo boost for only activating a couple of cores at a time how all of our VMs can run in a worst-case scenario where they are sharing as much resources as possible okay so we're scoring around 500 points which is about a hundred lower than we did when we were just running it on a single VM fascinating ok everyone ready 3 2 1 go ok so this is interesting on the read side of things really there wasn't much of an effect from the numbers I've looked at so far but on the right side of things some of our VMs were so starved for disk time that they actually timed out on their sequential write benchmark very very interesting this isn't something we're gonna retest after we overclocked because we can't overclock storage but I just wanted to have this as just a for my information sort of number I 264 cache and memory benchmark it's a gray window this one I'm really interested in because desktops really don't care much about their memory bandwidth the today's I mean dual-channel quad channel ddr3 ddr4 it's all fast enough but if you're trying to hit something with seven VMs at once I'm expecting the performance to be affected somewhat it's this is really interesting so you can actually see the systems are running the benchmarks at different speeds mix VM over here so this is player number five is actually still running the memory benchmarks or at least it was before I and mine is almost done running all of the level 1 and level 2 cache benchmarks so it does seem like their ability to even run these tests is affected by the fact that we're hitting them all at once here we go last one position your cursor over run three two one fire strike extreme thank you clickable random times like okay I guess that's it three to load final exam okay you guys are dismissed for now thank you I'll call you back once I overclocked the system so check this out this is what our memory and cache benchmarking looked like this is what our benchmarking looks like in firestrike extreme so with seven instances of it running you're actually getting up near 100% utilization closer to like 90 here okay very interesting so 3dmark scores not affected much by our cpu and ram sharing because 3d bark is mostly dependent on the graphics card in each VM has its own r9 nano so all this left now to do is to record our scores so we can create some average score graphs for you guys and establish our baseline so that we can find out if our overclocking helps now most server motherboard makers do not have any overclocking features but this is a soos and this is technically AWS or workstation board so in the AI Tweaker menu we've got there AI overclocked tuner which has two settings neither of which I have played with at all manual and OC tune so I'm just going to go let's OC tune it level three and find out if the system works there's a very good chance it's not going to work I have not tested this at all so this is actually a really bizarre quirk of this system it starts up here oh if it turns on at all it starts up here oh wow yep that's not posting let's clear seamless and then we will go from there let's see if that does anything oh that was it I need this bar to get past like me here well that wasn't it like I I know what it is because we did ASIMO's clear that bio setting that I need the 4G decoding setting is not enabled so we are not able to boot with all of these PCI devices so maybe by unplugging all the video cards I will be able to have them not get picked up and I will be able to get into the bios this is how you remove the graphics cards from seven gamers one CPU there we go okay they're out so now can try again okay so that was it a problem with having a system that doesn't post in its default configuration okay so let's try manual this time um my understanding is that on these types of boards you can expect somewhere in the neighborhood of like a hundred and five megahertz so just for lulz let's try 105 and see what happens I'm guessing not a whole lot and that's what I got okay then okay so after four attempts we're back in the BIOS are setting did stick but obviously didn't work so let's try a more conservative 103 which would be a three percent overclock okay so because we have no video card we have to use VNC to get into our VM and find out if everything is working as intended well that is not reporting correctly at all so that's a pain so we're gonna have to run a benchmark and find out okay I tell you vm's really screw up hardware reporting both cpu-z and i 264 are reporting numbers that would indicate that our CPU is not overclocked but Cinebench is reporting that it is overclocked and even though these scores are not directly comparable because we don't even have any of the other VM booted we're getting 900 in Cinebench which would indicate that everything's working properly at the very least so now we can proceed to the next step and put our graphics cards back in okay all our video cards are showing up which basically means we can start the array and fire up our VMs did it just do that or did you do that Hey all right so now it is time to overclock our video cards now the r9 nano isn't really known to be much of an overclocker oh but we're gonna go ahead and increase the power limit as high as it goes and we're gonna increase our clock tonight I don't know 1050 and see what happens hey graphics 1050 megahertz just like that I guess water cooling helps a little bit with the whole overclocking Murai nano situation very nice he'll afterburner is detecting all seven GPUs that shouldn't be possible because we've passed them through two separate PM's it's also detecting Wow afterburner doesn't know what's going on so it's got them all being used while we're running the benchmark so it knows they're seven but it doesn't know that it's only using one of them and it's just cloning this core clock graph which actually has the card not reaching uh not reaching a thousand megahertz anyway so I guess all that's left now then is to run some benchmarks and find out if we got any benefit for our trouble everyone ready overclock Cinebench benchmark what damn it Nick I've wanted someone to just start the benchmark early out of that stared out please work no no what that's good huh player whatever it is so it's one two three four five six seven and then you put OC scores Cinebench what my keyword says don't worry about that okay actually that does seem to be faster yeah forty points better on my everyone ready three two one run it please don't crash why it's okay give it time power draw 1600 watts so did everything but next lunch okay so after much struggling we managed to achieve whopping over clocks of anywhere from 0 to 70 megahertz on our graphics cards yielding fire strike extreme scores in the very very close to what we were already getting range leading us to believe that this whole thing was a terrible idea but I'm really glad that you guys got to come along for the adventure with us because that is what we do on this channel that is what we do this is a sponsor spot I have been looking forward to for quite some time we recovered a take on what they do is pretty obvious they recover data and why we're affiliated with them will be pretty obvious if you checked out the data recovery vlog which if you haven't already is a pretty a pretty exciting piece of content as far as they go on this channel anyway they gave me some bullet points to talk about and I will do that but then I'll get into why they're so special to me so first of all they're one of the only data recovery companies to invest in R&D so if the technology to recover your data doesn't exist they create it they're fast they're tech savvy they have friendly customer service available 24/7 and that was indeed my experience their data recovery guarantee policy means that if they don't recover any data there is no recovery fee they have worldwide emergency data response options including on-site remote and in lab that's 24/7 365 they've got state-of-the-art equipment utilizing today's most advanced technology with proprietary in-house developed tools they've got ISO for class 10 and ISO 5 class 100 clean rooms to enable them to provide a optimum solution for a recovery environment and basically what it boils down to for me is they got my data back it was an extremely complicated raid array not recommended rate array so I had three raid fives that were hardware raid fives on separate LSI controllers and then I had those striped together and raid zero for an effective raid 50 so three raid five stripes in Windows I thought it should be pretty safe because I can lose up to one SSD per controller before I should run into any issues but what I didn't account for is losing a controller outright so one of the controllers failed causing the array to go into an unrecoverable state and we recover datacom had me plug in all those 24 drives to a host bus adapter and they were able to remotely using their own Linux tool reconstruct the whole thing allow me to copy it off and basically save my turkey which thanks to those guys and if you guys have a data recovery emergency then check them out the link is at the video description I heartily endorse we recover data comm so thanks for watching guys if you just liked this video hit the dislike button but if you liked it and you thought it was crazy and awesome and extreme hit the like button get subscribed and maybe even consider supporting us by buying a cool shirt like this one changing your Amazon bookmarks one of our affiliate code instructions for our up there or just contributing directly through our all new Linus tech tips forum freshly updated for 2016 which you can check out at the link in the video description and now that you're done watching this and you're wondering hey what should i watch next maybe check out the video where I show you how to code rupal your network speed for a hundred bucks that is assuming that you have a couple of really good computers already and you have like networking from eating etcetera etcetera cetera terms and conditions may apply Thanks up