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


  • Content Count

  • Joined

  • Last visited


About porina

  • Title
    Slightly Salty, Fairly Fishy

Contact Methods

  • Twitter

Profile Information

  • Location
  • Occupation
    former electronic/acoustic engineer


  • CPU
  • Motherboard
    Asus Maximus VIII Hero
  • RAM
    G.Skill Ripjaws V 3200 2x8GB
  • GPU
    Gigabyte 1650
  • Case
    In Win 303 NVIDIA
  • Storage
    Samsung SM951 512GB, WD Blue 1TB
  • PSU
    Corsair HX750i
  • Display(s)
    HP LP2475w
  • Cooling
    Noctua D12
  • Keyboard
    Logitech G213
  • Mouse
    Logitech G403
  • Sound
    Cheap Amazon speakers that aren't bad at all
  • Operating System
    Windows 10 home
  • Laptop
    Asus FX503VD (7300HQ, 1050, 16GB ram)
  • Phone
    Pixel 3a

Recent Profile Visitors

34,173 profile views
  1. It is enabled by default. It should be understood that turbo speeds are opportunistic. It will be reached only if a combination of load, power, thermals allow it. The base speed is the minimum you should expect, and any above that is a plus.
  2. I was unable to verify that when looking at images. Couldn't find anyone taking a shot side on while installed on the mount, and it might be perspective effects, but it looks like it was nudging the nearest ram slot, like the Arctic 34 does. Fuma 2 shows more of a possible gap. Since you've gone through the bother to provide a list, I'll go through the list although the decision has already been made and the Fuma 2 ordered. Hard to read chart, does suggest U12A slightly ahead of Fuma 2, both far behind D15 U12A far better than D15 far better than Fu
  3. Fuma 2 ordered, expected to arrive Monday although Amazon are sometimes early. Did quickly scanned some written reviews. With a little uncertainty it does seem a bit better than the Noctua U12A but not quite as good as the D15, although there isn't that big a gap between any of these. One review put it equal to a D15S which I didn't even know existed until now, just a single fan version of the D15. Fuma 2 is rated at 137mm width so should be narrower than the borderline Lucifer I got in. It is unclear if they are counting the fan clips in that. Worst case I can bend the
  4. Looks to be around same class as the Arctic 34 and I'd like to go a bit higher. "budget"? Well, compared to Noctua, maybe? The Noctua U12A is £90 and the Fuma 2 is £70. To me, that's still a fair chunk of change for a cooler, but it is a one time cost. Sounds near perfect then.
  5. Ooh, I like the looks of that. This is exactly the sort of thing I didn't know existed. I'll need to do some more research into its cooling potential relative to others. Of note is the Arctic 34 and Noctua D15. Somewhere in between would be ideal. In looking around the Noctua offerings, the U12A seems to be the best performing medium cooler but I'm still a little concerned about ram clearance based on some images I found of it in use. U12S is probably not that different from the Arctic 34.
  6. Yesterday was a frustrating day. I decided to put more ram in my system with Asus B560M-A mobo. This is micro-ATX so limited in where I can move stuff. With 4 sticks of ram fitted, the stick nearest the Arctic Freezer 34 eSports duo cooler was pushed on by the fan. I didn't like that. The cooler was semi-temporary with the intent I'd move in a better cooler like a Noctua D15, but that was no go. In front-back orientation I couldn't use the only CPU connected PCIe slot. In up-down orientation the heatpipes hit the ram. After some thinking I swapped in a Deepcool Lucifer V2, but this isn't great
  7. Standard DDR4 and DDR3L both run at 1.2v, so it would seem the IMC was designed around that operating point. 1.35v DDR4 is only commonly found on XMP modules, so something that impacts enthusiasts but not systems for the masses which will use standard ram. I've not looked into it, but my 6700k system built in 2015 recently started crashing. I found by trial and error the only thing that improved it was setting ram speed slower. It was running XMP 3200 fine for many years. That ram is now moved into another system that can use the speed, where it is working fine at 3200. So, I do wo
  8. I forgot that part. The difference of DDR3L was the standard working voltage dropped from 1.35v to 1.20v, which then matches standard DDR4. I don't think running higher voltage DDR3 would be a problem if you really wanted to, in the same way most higher performance DDR4 XMP kits ran at 1.35V also.
  9. Will be interesting to see which boards will actually be made in volume and thus can be purchased. Support of two generations of ram also applied during the previous transition from DDR3 to DDR4, with Skylake supporting both. Then, the outgoing DDR3 supporting boards were rarely seen. Will we see a repeat of this? I do feel that most of the focus will be on DDR5 boards. Why would you go for a cutting edge system and use old ram? About the only reason I can think of for using DDR4 will be for cost reasons. DDR5 will likely be at a premium for quite some time. Perf comparisons betwee
  10. It is interesting, but usage of ARM based chips is quite different from usage of x86 chips, so the equally big question is what AMD intends to do with ARM? Also ARM is less about general purpose compute, and its usage is more into targeted use cases. Could AMD design a high(er) performance ARM core and re-license it out? This is likely also a safety net for the future. x86 might not remain widely used indefinitely. Intel have a far more diverse product portfolio than AMD so would be less impacted by that. The way they're integrated into products as a whole makes the
  11. 9900k would be the easy option and the one I pick, since I hate reinstalling. No reinstalls or configuration. Just drop it in and go with how the software is already set up. The main negative I see is I wouldn't want to buy one new today given their age. See if there are any used options. 5800X would likely provide further performance, but the extra mobo cost and work to switch platforms may be a consideration. If a new platform is open then Alder Lake might also be a consideration, due by end of year. If you're going to move platforms, make a bigger move. Define swe
  12. Good point. Following that thought, if a photon were to hit something that doesn't do anything particularly special with it, and does not get reflected, does it just get converted to heat? With thinking like this, maybe it was for the better I didn't try for a physics degree while younger. Went into engineering instead where you get away with "close enough" a lot more
  13. More resource beyond what you need doesn't really help. 8GB is going to be a natural optimisation point for game devs as so many current and recent past GPUs have that much. It might not offer ultra or max settings in all cases, but you're not going to have a bad experience with it. The question then is, if game devs go past 8GB, how far do they go? Depending on the specific card, nvidia have 10GB, 11GB, 12GB cards. AMD have 12GB and 16GB cards. Where do you draw the line? 10GB would be safe to cover those options, but it isn't that much more than 8GB. 12GB could be another optimisation point,
  14. Random thinking: I recall the energy of a photon is planck's constant multiplied by the frequency of light. If we apply that to E=mc^2 and rearrange for mass... Take a 550nm wavelength photon (green-ish), that works out around 4*10^-36 kg. Not zero but pretty close to it for practical purposes.
  15. The memory bus has a certain width which makes it have optimal capacities. Deviating from that will drop performance significantly as you cripple the memory bus. The 3070/3070Ti/3060Ti GPUs have natural steps at 8GB and 16GB. The 3060/2060 have natural steps at 6/12GB. If nvidia decide that 8GB is the minimum, then 12GB is the next step. Also it is not like 8GB GPUs will become useless any time soon, I'd say unlikely for the next generation or two it'll still be plenty. If game devs choose to have options using more than 8GB going forwards, you can be sure there will also be lower