Saturnus

  • Content Count

    4497
  • Joined

  • Last visited

Community Reputation

3864 Excellent

About Saturnus

  • Rank
    Duplicant Freedom Fighter
...

Badges

Recent Profile Visitors

10021 profile views
  1. I see your Clint and raise you a Croc No, not this croc This Croc It's pretty obvious it's Crocodile Dundee that is the inspiration due to the fangs on the band.
  2. It was a common cause of frustration for new players that didn't know how their dupes had gotten stuck.
  3. Another way to improve performance that only very marginally reduces precision would be to divide the map into 4 or more frames with the physics calculations for each frame running simultaneously on different threads. It would create minor inaccuracy in the cells where frames meet but it would increae the performance almost by a factor equal to the number of frames the map is divided into as long as that number is lower than than the number of logical CPU threads available. However, I don't think it's possible using the current API.
  4. First row beneath bunker tiles or bunker doors have always taken damage from meteors hitting them.
  5. Yeah, AMD already proved with Infinity Cache on their graphics cards that having an extremely large cache basically eliminates the need for ultra high bandwidth. Memory timings will still be somewhat relevant but it means that one doesn't have to invest in fast memory to get at least most of the benefits from the CPU upgrade. Furthermore, other tests have shown that since the 5800X3D runs at significantly lower voltage and uses significantly lower power than the standard 5800X, you also don't have to pair it with a higher end motherboard. Most low midrange and up B550 boards is fine, or even a good quality B450. It's the perfect upgrade CPU to almost anyone that already have an AMD system, and plays sim games.
  6. As far as I can tell from reviews is that 3800 is the maximum MT/s setting you should attempt with the 5800X3D. To increase performance beyond that you have to tweak memory timing. The easiest and safest way to do that is to use the DRAM Calcuator but note that voltage settings are somewhat optimistic unless you have a really high-end motherboard. I usually allow 0.05V more than the DRAM Calc suggests to get it fully stable in all use cases.
  7. That entirely depends on your perspective and what you want to achieve but even for many 5800X owners it's actually worth considering. If you do it now and don't wait too long. That's because right now you can get about $250 for the 5800X on the second hand market, so the net cost of the upgrade is $200. If you mainly play the type of games that benefits from the 3D V-cache like ONI, Factorio, Age of Empires, Civ6 and so on then paying $200 for 28% performance uplift on average is actually incredibly cheap. People routinely spend far more than that for less performance upgrade. The longer you wait the more the used price on your 5800X will drop making it a less attractive option. If on the other hand you're satisfied with what you got and don't have $200 to spend on upgrading your gaming system then it's obviously not a good idea.
  8. I see you still have literally no idea what you're talking about. DDR5 doesn't come in CL14, CL16, or anything near it. As I wrote the used DDR5 memory is the most premium DDR5 memory money can buy at the moment. It's $550 for a 32GB kit and that's after a recent $120 price cut.
  9. Except tests shows that hypothesis is incorrect. Note: All tests with Ryzen CPUs are done with the exact same DDR4-3200 as used for the DDR4 test with the12900K. Also note how going from DDR4-3200 to the most premium DDR5-6400 currently available does almost nothing compared to the incredibly massive jump the extra 64MB L3 cache on the 5800X3D gives.
  10. No. There's a very long thread on the Factorio forum on the topic. After the latest performance upgrade which significantly improved performance this was the conclusion: Like ONI, the main CPU usage on Factorio is running the same calculations over and over again, and the faster you can move the data into the CPU the better. The best is to have lots of L2 cache, second best is lots of L3 cache, after that effective memory transfer speed (which is not the same as just the memory clock speed or bandwidth), and bottom of the list of things that matter is CPU clock speed, or really executions per second.
  11. Slicksters convert CO2 into crude oil or petroleum. Also most maps have a vast abundance of infinite water sources you can convert into hydrogen for power. Each electrolyzer produces 698W surplus power out of 1kg/s of water when accounting for 120W electrolyzer, 54W gas pump, and 24W liquid pump.
  12. Not yet sadly but Cities: Skylines, Factorio, and Age of Empires 4 have all been tested. All 3 games runs about 80-85% on a single core with minor utilisation of up to 3 more cores, nothing beyond that. All 3 games heavily favours large cache sizes and fast memory but in particular the former. So they all match the same criteria as ONI.
  13. If it's for gaming then save another $150 and get the 5600 (non-X) model ($199 current US price). The difference is minimal, and don't be fooled by Ryzen being overclockable, it means next to nothing in actual gaming performance. Just leave it up to the automatic built-in PBO 2 and you get the best performance with the lowest power consumption 98% of time. At best I see 2-3% gain from OC'ing. It's in memory tunings you see the big differences in gaming performance for Ryzen CPUs. I see 10%-15% better performance by optimising memory timings and subtimings. And it's not about raw clockspeed either. Often lower clocked memory performs better if the timings and subtimings can be tweaked instead. However, testing on reputable tech sites shows that the 5800X3D is around 12% faster than a heavily overclocked 5800X in gaming on average. In single core heavy sim games like ONI the 5800X3D is 28% faster than a heavily overclocked 5800X on average. So if you are after the ultimate gaming CPU then at the moment the 5800X3D is it.
  14. Is funded 100% by Intel. CPUbenchmark and Passmark even used to share business address with the Intel Santa Clara Head Quarter. They've now moved to different countries to obscure that all their employees are directly on the Intel salary. Anyone that has been interested in PC hardware for more than a decade would know this.