Jump to content

DarkAlpha_Sete

Member
  • Posts

    17
  • Joined

  • Last visited

Reputation Activity

  1. Informative
    DarkAlpha_Sete reacted to Lord Vile in CPU temps keeps jumping 10ºC while idling on desktop   
    My 3600x does it too, fairly normal and it doesn't do it under load. Jumping mid 30's to 40's doesn't really bother me especially when my cooling curve is set to ramp up at 60
  2. Informative
    DarkAlpha_Sete reacted to Haro in CPU temps keeps jumping 10ºC while idling on desktop   
    That's normal for zen 
    It happens all the time
  3. Informative
    DarkAlpha_Sete reacted to Stahlmann in Will plugging 3 pin fans with different specs into a hub work?   
    You can connect up to 4 fans to 1 header using PST.
    If you use different fan sizes, each will run at their specific RPM matching to the PWM signal.
  4. Informative
    DarkAlpha_Sete reacted to Stahlmann in Will plugging 3 pin fans with different specs into a hub work?   
    Where are you from? In germany you can buy them for about 7 € each online.
    They are actually one of the cheapest and still awesome performing fans in most cases.
  5. Informative
    DarkAlpha_Sete reacted to HenrySalayne in Will plugging 3 pin fans with different specs into a hub work?   
    The DeepCool FH-10 is designed for PWM controlled (4 pin) fans. It is unable to control 3 pin fans and they will turn with full speed all the time. The 3 pin models are controlled by voltage, not a PWM signal. So adding two (or more) different fans to a single fan header (via a y-split cable), will work, but each fan will react differently to voltage changes, which is not ideal.
     
  6. Informative
    DarkAlpha_Sete reacted to Stahlmann in Will plugging 3 pin fans with different specs into a hub work?   
    The fans are daisy chained and have a splitter build into the cable. Like this:
     
    With a Molex connector you can plug the fans directly into the power suplly bypassing the mobo.
    But this way you have no control over the speed and they will always run full speed.
  7. Informative
    DarkAlpha_Sete reacted to Stahlmann in Will plugging 3 pin fans with different specs into a hub work?   
    With molex u cant adjust the fan speed and they will always be running at full speed.
  8. Informative
    DarkAlpha_Sete reacted to Stahlmann in Will plugging 3 pin fans with different specs into a hub work?   
    I'd suggest to buy 2 "Arctic P14 PWM PST" fans in your situation.
    The "PST" stands for "PWM sharing technology".
    You can plug 2 or more into 1 header.
    They are also much cheaper and perform very close.
  9. Informative
    DarkAlpha_Sete reacted to mariushm in Did I damage anything by not giving the PC enough watts for 15 seconds?   
    No, that's not how it goes.
     
    The power supply takes the input AC voltage and produces output voltages the components need. It also monitors the quality of the input and output voltages. ... if they go outside some ranges the psu shuts down to protect the pc.
     
    When the pc consumed so much power, the ups wasn't able to keep up, so the voltage it produced probably went down below the range where the psu can function, so the psu shut itself down.
     
    No, nothing bad happened to your hardware, to your processor or video card or motherboard.  The worst that could happen is you losing some data due to sudden shut down
  10. Informative
    DarkAlpha_Sete reacted to Boomwebsearch in Did I damage anything by not giving the PC enough watts for 15 seconds?   
    This is true for standby UPS units (some will put on backup when the voltage becomes too low/high), although line-interactive UPS units contain transformers to adjust the voltage if the wall power is below or above a certain threshold. Any decent power supply should shutdown if the voltage is too high/low to protect your components and if there is a short detected.
  11. Informative
    DarkAlpha_Sete reacted to mariushm in Did I damage anything by not giving the PC enough watts for 15 seconds?   
    No, the video card expects that power is available.
    If the video card demands too much and the power supply can't provide that much (ex running a rx 580 card that consumes up to 225w on a 300w power supply, where the cpu and motherboard also consume power) then at some point the built in protections of the power supply will trigger and psu will shut down.
    If you have a cheap psu with minimal protections, it may happen the psu lowers output voltage, for example you'd see it output 11.6v instead of 12v, in an attempt to keep up. At some point, the voltage will go too low and psu protection will shut down the power supply.
     
  12. Informative
    DarkAlpha_Sete reacted to mariushm in Did I damage anything by not giving the PC enough watts for 15 seconds?   
    Also most UPSes  don't power the computer directly unless there's some power event (AC voltage too low or too high, frequency fluctuations , ac voltage loss etc)
    Normally, the UPS just monitors the line and the pc is connected directly to the mains.
    IF there's some flaw, the UPS quickly disconnects the mains cables and puts itself in between and starts its own internal generator, converting 12v from the internal battery to 110v AC or 230v AC the power supply expects.
    The whole process lasts usually less than 10ms ... and power supplies are supposed to function for at least 1 AC cycle (~16ms) at their maximum rated power without incoming energy, so things work out like that most of the time.
     
    The expensive on-line UPSes are exception... these always power everything connected to their mains outlets from battery, all the time.
  13. Informative
    DarkAlpha_Sete reacted to Boomwebsearch in Did I damage anything by not giving the PC enough watts for 15 seconds?   
    I picked up an old APC UPS from a thrift store that was advertised for a higher wattage than it could handle, the UPS shutdown and beeped loudly + continuously until I powered it off. Nothing was damaged, and I returned the unit, got a larger unit later on.
     
    The line-interactive UPS units have transformers to lower/boost voltage and continuously monitor it to adjust it whenever needed, although if it falls way too low then it will start running off of battery, line-interactive UPS units have a shorter switching time than standby units (usually). 
  14. Informative
    DarkAlpha_Sete reacted to mariushm in Did I damage anything by not giving the PC enough watts for 15 seconds?   
    It's more or less a side effect.
    Trying to keep it as simple as possible.
    You have a VRM , a dc-dc converter circuit which converts 12v into the lower voltage the cpu needs, which can vary between let's say 0.5v and 1.4v, depending on the frequency of the cores (to save power, when the cpu can lower all the cores' frequency, it ca also lower the voltage required to operate, because there's no need to keep the voltage up)
    So the circuit plays with two parameters : voltage and current ... voltage x current is power, how much the processor consumes.
     
    This converter circuit works by turning on and off those phases thousands of times a second, varying the amount of time each phase stays on and off, to get the exact voltage the cpu wants and supply the total power... think of it like each phase squirting a bit of water into a bucket trying to keep the maximum level fixed while someone either sucks a bit of water through a straw, or suddenly taking a big gulp of water.... the vrm controller reacts quickly as the water level changes in a bucket and either squirts or dumps a bit of water to keep everything at constant level.
     
    So let's have an example:  let's say the processor wants 1.1 volts and 50 watts.
    The controller chip sees 12.05v at the input and sees that the processor wants 1.1 volts and it quickly does the math and knows it has to turn on and off each phase 35400 times a second to sustain that 1.1 volts at 50 watts of power consumption.
    If the processor consumes a bit than 50 watts, let's say 60w, the controller chip will detect that the output voltage went below 1.1 volts (the water level has fallen because someone sucked too much water from bucket) and does the math again : it reads the input voltage again and now maybe it's 11.98v instead of 12.05v (if power supply sucks and has high voltage ripple) and the math says it has to increase the number of pulses from 35400 times a second to 38000 times a second to keep the voltage at 1.1v
    These recalculations happen tens to hundreds of times a second.
     
    From time to time, especially when overclocking, you can have situations where the processor had all cores at 3.0 Ghz and needed only 1.1v to function and cpu consumes 30w, and after a few  ms the cores are at 4 ghz and now the voltage is 1.4v and cpu has an instant power consumption of 100w
    The controller chip does the math and reacts quickly and changes the number of pulses to quickly satisfy the cpus demands and does so super fast, but if the power supply is of lousy quality and the output voltage fluctuates a lot (voltage ripple), the controller chip has to constantly recalculate and adjust itself to work with the crappy input power, so it takes a bit longer for the output voltage to the cpu to stabilize to the desired parameters.
    So if the cpu needs 1.4v to reach 4 ghz , the vrm may only be able to get 1.39v because the input voltage fluctuates too much, or it may take 10ms instead of 2ms to stabilize to 4 ghz, at which point the cpu maybe says "oh shit, the power quality is too poor to try for 4 ghz"  ... because there's not enough input voltage on the cpu, the cpu can't reach that 4 ghz and will have to settle for let's say 3.8 ghz
     
    So a cleaner voltage means just the potential for higher overclocking, and for smoother voltages all around, which could mean a sustained overclock, or getting higher turbo/boost frequencies.
     
    argh not sure if I've explained it well, but it's 1:30 am here and i have to get up at 6:30 to go to work, so I have to stop here for today.
×