Jump to content
Search In
  • More options...
Find results that contain...
Find results in...


  • Content Count

  • Joined

  • Last visited

Posts posted by S2000Gan

  1. My build page is so out of date.
    Ive since gotten an EVGA GTX 770 "Founders Edition" and then an EVGA GTX 1080 FTW and upgraded my case to a Define R6, custom water cooling, and then finally a platform upgrade to a 9700K and Maximus XI Code with 4x8GB DDR$ 3200 Trident Z RGB
    I'm not great at taking pictures of stuff. here's a crappy pic of it's current state.
    Here's my facebook album with more pictures of the various states my computer(s) has been in over the years



  2. I recently got a 2nd GTX 770 for my old i5-2500K PC and I tested that both GPUs work individually but when I put them together in the same system and enable SLI the Nvidia driver crashes and restarts every time I try to launch anything GPU-demanding like a game or benchmark. It works fine on the desktop using DisplayPort at 1440p 120Hz though.
    Oddly enough SLI works when I use an HDMI cable, but using HDMI I only get 1440p 60hz whereas with DisplayPort I should be able to get 120Hz (it seems like I cant get 144/165Hz using the 770)
    It even crashes when I have the monitor connected over DisplayPort set to 60Hz and disable Gsync and DSR.
    It's not a HUGE deal since this system will most likely only be connected to 1080p displays over DVI or HDMI and my 1440p 165Hz Dell S2417DG will mostly be connected to my GTX 1080 system but I'd still like to figure out this issue if possible.

    Anyone know whats going on?

  3. Hello again everyone. 
    I recently got some parts to build my first watercooling loop!
    Unfortunately, my Lian Li Lancool K63 is not really designed to mount much other than 140mm fans...
    So I picked up a new Fractal Design Define R6.
    My plan is to keep my Black/Red build but now It'll be in a white case! 

    Why the switch to a white case? I'm going for a "Type R" styled theme. :D
    Once I get the system built and running I plan on picking up a Red Honda badge and a Type R badge for fun and putting those in there somewhere. After seeing some pictures of systems in white cases I've decided I'm going to avoid putting a bunch of red LED stuff in the computer because I'm not a big fan of how all the white surfaces reflect the red light into kind of a pink-ish-ness. Instead I'll be sticking with the white LEDs in the reservoir and CPU waterblock and potentially adding a white LED strip if I think it needs more lighting.
    I'll probably keep the green GEFORCE GTX on for fun, although we will see how the LEDs on the Ballistix Tactical Tracer RAM look and whether or not I end up having those LEDs on since they can only be Green or Red on the ones I have.

    I'm still running my 2500K although I'm interested in seeing if the water cooling (especially on the VRM) helps my temps enough to lower the voltages I use for my current overclock. I'll probably end up taking this whole computer apart and replacing the current platform if the 9900K and 9700K really end up being 8 core CPUs (which kinda of makes the Gene-Z VRM waterblock seem like a waste, but hey I got it on clearance for really cheap. Also while I love the GTX 770 I have (Thanks EVGA for making a Titan cooler version at the community's request!), I am probably going to upgrade it when the 1180 finally releases (possibly a Hydro Copper version? Or if not at least a Hybrid?)

    For now it's just soft tubing although I would like to do rigid tubing at some point. I've bent PETG before (Thanks to BS Mods for coming to LTX 2017 and 2018) so its really just a matter of building up enough funds to spend on all the stuff I need for that. Probably wont happen for a while. Priority is saving up money to upgrade the platform (CPU/Board/RAM) and graphics card.

    Anyway, here are some pics of my system in my old case, and what it looks like at the moment!

  4. On 6/27/2018 at 11:55 AM, AlexTheGreatish said:

    I figured we would use TG Kyronaught, since it is longer lasting and doesn't have the potential for disaster like liquid metal.  We will have some Conductonaught and conformal coating on hand though if you are feeling adventurous.

    Try asking in this thread?

  5. 2 hours ago, AlexTheGreatish said:

    Kryonaught normally performs about 4°ish worse than LM, so you're still looking at around a 15C improvement.  That said, I've had an 8600K at 5.1GHz with a Noctua air cooler and stock TIM so with a custom loop you can probably skip delidding and just go for it stock.

    I have seen that the 8 series can manage 5ghz on air pretty easily and from what Ive seen, custom WC loops already do a much better job keeping CPUs closer to ambient. (I think due to the thermal mass of the water?)
    Anyway, I guess my next question would be does deliding (with either LM or kyronaught) make much difference if you are using a custom WC loop? Or does it only really matter on cooling solutions with less heat sinking abilities? I can imagine that better TIM means better thermal conductivity regardless, but I can see how the TIM's rate of thermal conductivity would be less important in coolers that already stay cooler such as custom WC loops.

  6. On 6/27/2018 at 11:55 AM, AlexTheGreatish said:

    I figured we would use TG Kyronaught, since it is longer lasting and doesn't have the potential for disaster like liquid metal.  We will have some Conductonaught and conformal coating on hand though if you are feeling adventurous.

    Im going to be doing a custom water cooling loop and might end up replacing my CPU/Chipset/RAM platform by getting an 8600k before LTX (that I hope to run at 5GHz) and if so was thinking of bringing it for delidding. I dont recall if LTT has ever done a video literally comparing a stock CPU to a delided one with Kyronaught or Conductonaught. If i recall the 6700k deliding video yielded little improvement but in the the $1000 CPU video the results were near 20 degrees improvement on the 7700k from deliding and using Kryonaught. And the Ultimate compact Titan XP video wasnt really testing the performance of the liquid metal. The Liquid Metal Laptop cooling video also showed a 20* improvement. Do they both perform the same??

  7. I'll need to replicate your test to see.  Can you provide step-by-step instructions on what you did to test this so I can do the same?


    Im using a i5 6600k

    Asus Z170 Deluxe

    GTX 770

    GTX 970

    2x8GB DDR4

    3x3TB WD Red Hard drives

    I installed UnRAID

    formatted 2 drives and put them into an array

    set the 3rd drive as a cache drive (but didnt format it or tell any of my shares to use the cache)

    made all the folders linus mentioned in his video put them all on the array

    Made 2 VMs

    installed windows 8.1 on one VM and windows 7 on the other

    noticed the hard drive performance seemed slow at times on each of the VMs

    Downloaded CrystalDiskMark

    when I run CDM on one VM with the other turned off I get 1.8-2.4GB/s Read speed and the test starts within a normal amount of time as any of my normal computers maybe even a little faster

    when I run CDM on one VM with the other turned on I get 50-100MB/s Read speed and the test takes significantly longer to start than my normal computers

    when I run CDM on both VMs at the same time I get about 25MB/s Read speed and the test takes very long to start on either VM


    at some point during the tests I realized my cahce drive wasnt formatted and wasnt set to be used by any of my Shares, I formatted it and told the vdisks share which I had the OSes installed on to use it but it doesnt seem to make a difference in the CDM numbers, do you have to reboot UnRAID after changing the Use cache disk setting?

    For the record both my desktop and laptop run off of SSD which get about 450MB/s Read speed.

    I only mention read speed because I havent payed as much attention to their write speed since its similarly fast (or slow) in relation to the read performance numbers.

    Need me to get any other info from anywhere?

    UnRAID info? Screenshots?

  8. Yes sorta...

    MS stopped messing with the providers and started selling through their own store front instead. Verizon can go shove a stick ups its *** with the way it treated windows phone owners. T mobile isn't to far behind these days. I am glad MS has decided go the way it has with selling directly to the consumer. Tmobile and Verizon are stuck on primarily selling the iPhones as if they are worth more than gold.

    Except they havent stopped messing around... I just said in my post that the 950 is exclusive to AT&T and its only the 950XL that is available unlocked through the microsoft store and even then only for AT&T or T-Mobile :/ I might settle for the 950XL even though I think its too big and wanted the 950. Either that or ill just switch to android :/ I really do prefer the windows phone platform though :/

  9. I was on at&t, I have a Lumia 920, only verison got the Icon and 735, my family switched to tmobile, and only at&t get the 950 from what it sounds like, the 950 xl is going to be available on at&t and tmobile through the microsoft store, but I dont want the XL, too big...

    Microsoft's biggest problem is not everyone who bloody wants their phones can even get them...

  10. Im testing this out on a system consisting of:

    i5 6600k
    Asus Z170 Deluxe
    16GB DDR4

    GTX 770

    GTX 970

    three 3TB hard drives one as the cache and 2 as the array

    I tried doing it on a i5 2400 and a Asus Z68 Maximus IV Gene-z but I found that even though the chip supports VT-D it seems like only Q series, X series and Z170 support VT-D

    Im trying to create a aVM but im getting this error after enabling the ACS override because both GPUs are on the same group, it seems like only X series boards have ACS, not sure about Q series.

    Not sure where to go from here.

    internal error: early end of file from monitor: possible problem:
    2015-10-25T18:46:30.182775Z qemu-system-x86_64: -device vfio-pci,host=00:1f.3,bus=root.1,addr=01.0: vfio: error, group 9 is not viable, please ensure all devices within the iommu_group are bound to their vfio bus driver.
    2015-10-25T18:46:30.182808Z qemu-system-x86_64: -device vfio-pci,host=00:1f.3,bus=root.1,addr=01.0: vfio: failed to get group 9
    2015-10-25T18:46:30.182823Z qemu-system-x86_64: -device vfio-pci,host=00:1f.3,bus=root.1,addr=01.0: Device initialization failed
    2015-10-25T18:46:30.182837Z qemu-system-x86_64: -device vfio-pci,host=00:1f.3,bus=root.1,addr=01.0: Device 'vfio-pci' could not be initialized


    EDIT: I figured out that it was getting this error from me trying to give the VM the onboard audio, is there any way for me to give one of the VMs the onboard aubdio?


    Im also wondering if it is possible to have the VMs be able to have new USB devices plugged in without allocating them on the VM creation page. Do the mouse and keyboards I have plugged in have to stay in the exact same USB ports they are currently in?


    EDIT2: Why is CrystalDiskMark reporting 1.8-2.4GB/s read speed from my VMs when i only have one of them running? I only have a 2 drive array and 1 drive cache. All are HDD not SSD. This doesnt seem right at all. I dont think I even have my cache setup properly. Because I dont have SSDs I put my vdisks share on the array and whether I have use cache disk set to yes or no I get these crazy speeds in CrystalDiskMark. But when I have both VMs on then CrystalDiskMark reports 50-100MB/s read speed. Both the read speed im getting when using only one VM and the speed im getting when using both VMs seems to be unaffected by switching the use cache disk option. Far as I can tell the only difference is when only one VM is on the CPU utilization in the UnRaid Dashboard is lower and when both VMs are on and I try running a CrystalDiskMark run its at 99% Also the test takes alot longer just to start when both VMs are on. When I run CrystalDiskMarkon both VMs at the same time they report only about 25MB/s each.