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

-iSynthesis

Member
  • Content Count

    2,591
  • Joined

  • Last visited

Awards

This user doesn't have any awards

7 Followers

About -iSynthesis

  • Title
    Veteran
  • Birthday Nov 19, 1999

Profile Information

  • Location
    Munich, Germany
  • Gender
    Male
  • Interests
    Anything about Computers

System

  • CPU
    Intel Xeon E3-1230V3
  • Motherboard
    ASRock Z87 Extreme 4
  • RAM
    16GB G.Skill RipjawsX 1600MHz CL9
  • GPU
    ASUS ROG Vega56
  • Case
    InWin 103
  • Storage
    RAID0 Kingston A400 480GB x2 / RAID1 Seagate Barracuda 1TB x2
  • PSU
    Fractal Design Integra 650W
  • Display(s)
    BenQ XL2411Z / BenQ RL2450H
  • Cooling
    Enermax ETS-T40
  • Keyboard
    Roccat Vulcan AIMO
  • Mouse
    Steelseries Rival 310
  • Sound
    Sony MDR-1A
  • Operating System
    Windows 10

Recent Profile Visitors

1,918 profile views
  1. To add a bit more, storage is almost always measured to the base of 2. One byte is 2^3 bit, 2 byte, 16 bit, is 2^4 and so forth. Kibibyte, Mebibyte etc. are also to the base of 2. One Kibibyte is 2^10, Mebibyte is 2^20. With 64 bit systems, the word length of a storage element is 64 bit or 2^6 bit. This means, that the logic unit of a processor can process a 64 bit long piece of information in one step. To answer your question, we use multiples of 8 because 8 bits, or a byte, is the amount of information needed to encode a single ASCII letter. Our system is based on tha
  2. You are taking a Grand Tour video and applying it to usps. It just doesn't work. A usps route is pretty much the perfect scenario for an electric. Literally. Lots of stop and go means you can get a good amount of energy back. Short routes that should be well within current electric capabilities. This combined means that a single charge a day would probably get you through most routes. The charging infrastructure at the depot wouldn't have to be insanely fast charging, it would just have to charge them over night. This is neither expensive in comparison to other spending
  3. It depends on the power supply, but most ones reach peak efficiency somewhere around half of it's max power draw. However, from that point up until full power draw the decrease in efficiency is pretty small, usually only a couple %. Power supplies operate at least efficiency at low loads. This added together means that your peak input power will be at full load and with a slightly decreased efficiency from what it's max rating is, for an 80+ gold PSU you'll be above 80% close to the 90% mark. Refer to the chart in this wikipedia article to estimate efficiency at different levels of
  4. So the temps have always been trash, right? Also with the old cooler? It's an odd issue with everything you've tested, the only thing that would come to my mind is that they didn't apply the Thermal Interface Material properly when they made the CPU (not sure if those Xeons were soldered). Other than that, maybe try a full reset of the bios and check stock temps...
  5. This strongly depends on the battery inside. Some specially designed Lithium-Ion batteries may operate a bit below -50°C even. But for normal ones you shouldn't go below 0° when charging (storing it colder is another topic). This may / will decrease it's capacity irreversibly. They will also charge substantially slower as the internal resistance rises with lower temps. I myself have used power banks at below that, they were fine afterwards but I'd recommend putting the power bank into a warm bag or something, tightly wrap it maybe (It should produce a bit of heat itself so if it's 0°C out
  6. Probably a whole lot of input lag if I had to guess^^ In all honesty though, most servers you mentioned are not built for graphics rendering. They'd probably be pretty crap at it and trying to get thousands of gpus to work together would be a nightmare as 4 way SLI/Crossfire never really worked well in the first hand. And more would be a nightmare or impossible to work to create single frames together...
  7. That's the debug tool, not sure what it shows but it's not artifacting. Frametimes maybe, no clue... Which modpack are you playing? What is your CPU / GPU usage like? (Detailed on the CPU part, per-core usage etc.) I'd recommend tweaking around the optifine settings some, some settings can have huge impact on fps. Minecraft with that many mods I'd recommend at least allocating 8GB of ram and make sure to enable multi-core rendering. There's a setting under Performance I believe about rendering chunks more frequently, I'd suggest turning that off. I'll check tomorrow but
  8. Motherboard GPUs are a relict of the past, pretty much the only application they're still used for is server motherboards for basic Displays. They have many disadvantages; They can't be powerful as that would require active cooling. They would increase the cost of the motherboard as it would be another processor that would have to be soldered onto the motherboard. Putting it onto the CPU is an easy way out; The GPU is just another addition to the existing silicon. It's connected to the memory anyways and it's actively cooled. It just needs a few extra phases power delivery on the boa
  9. Yes, that's the exact instructions I followed. I did not let it get hot before, I found it not really being necessary but if you're willing to go the extra step it surely can't be bad... Just don't burn yourself^^ If you need any help understanding steps taken I can try and translate the important stuff from the article (or you're german yourself?) Good luck!
  10. MX4 is just fine, that's what I used too. The STRIX Vega has one row of VRMs, I ended up getting a thermal pad replacement for it (3mm Thermal Grizzly Minus 8). It was not needed but got the VRM temps down quite a bit. It kind of depends what your temps are there, ASUS really f-ed up their design so they ran hot as hell but you might not have to...
  11. To find out if heat is the issue try turning up your fans to 100%, this should get your hotspot temp lower. If it still crashes it's something else. My vega is running on 1x8pin (PSU side) to 2x8pin (GPU side) without a problem since I repasted. Even overclocked shooting over 330W. Repaste, use a lot of Thermal Paste on Vegas. Start off by tightening the two screws further away from the PCIe slot FIRST, only when they are bottomed out screw in the other two till the end. This should drop your hotspot temp, which may be the reason for the crashes you're experiencing (This happ
  12. I believe there was an old undervolt being applied. One of the above pics showed a max core voltage of 1.044V, stock is 1.25 I believe. You can undervolt a good amount but mine wasn't terribly stable below 1050... That'd also explain the low temps
  13. I've had something similar happening with my Vega 56 I believe. It shut down both monitors and the fans ramped up to 100%, temps were fine seemingly. For me it was the Hot Spot temp (it's shown in hwinfo). While the GPU temp was around 70 or 75 the Hot Spot went up to 118 and shut down afterwards. I used a method of tensioning the cooler that Igor's lab used. Basically you want to put quite a bit of paste on the GPU (The die on vegas is really large and sometimes uneven). Then you mainly tension the upper two screws of the core (The one further away from the PCIe slot) an
  14. Yeah, they're pretty rare. And tbh usually completely overkill for anything else than hardcore LN2 overclocking^^ But good
  15. The short look I could take at ASUS's German website (before it went down for some reason) it only supported Sempron, Athlon II and Phenom II CPUs. Also it's not worth it. And even if it did work you would probably fry the hell out of that poor VRM with anything more than an fx 43xx Edit: AM3+ chipsets take AM3 and AM3+ CPUs, AM3 chipsets take Am2/2+ and AM3 CPUs. AM3+ has an extra pin for the FX CPUs, there's no use trying to, you'd just break off the pin.
×